diff --git "a/data/en001/text/00000027.txt" "b/data/en001/text/00000027.txt"
new file mode 100644--- /dev/null
+++ "b/data/en001/text/00000027.txt"
@@ -0,0 +1,9597 @@
+oVy6tV49qNI-00000-00000032-00000115 - Hi, everyone.
+oVy6tV49qNI-00001-00000115-00000505 I wanted to show you how to embed an image on a Canvas page.
+oVy6tV49qNI-00002-00000505-00000779 So I am here at Unsplash, which is a website
+oVy6tV49qNI-00003-00000779-00001048 where all of the images are in the public domain,
+oVy6tV49qNI-00004-00001048-00001393 which means they're openly licensed and free for us to use.
+oVy6tV49qNI-00005-00001393-00001684 So I can click, I've found this image,
+oVy6tV49qNI-00006-00001684-00001887 and I can click the Download menu
+oVy6tV49qNI-00007-00001887-00002230 and I'm going to download the small version of this image.
+oVy6tV49qNI-00008-00002230-00002326 When I download that,
+oVy6tV49qNI-00009-00002326-00002547 it does give me a suggested attribution,
+oVy6tV49qNI-00010-00002547-00002752 which I may come back to.
+oVy6tV49qNI-00011-00002752-00003159 And if I go over to my Canvas page and click Edit,
+oVy6tV49qNI-00012-00003159-00003644 I can find the Image icon on the Rich Content Editor.
+oVy6tV49qNI-00013-00003644-00003990 And let me find where I want to place this image.
+oVy6tV49qNI-00014-00003990-00004196 So if I put it, say, at the top of my page,
+oVy6tV49qNI-00015-00004196-00004320 there's my cursor,
+oVy6tV49qNI-00016-00004320-00004775 and then I can click Upload Image using the Image icon.
+oVy6tV49qNI-00017-00004775-00004958 And here I can just drag and drop
+oVy6tV49qNI-00018-00004958-00005168 the image I just downloaded.
+oVy6tV49qNI-00019-00005168-00005382 Notice I could also have searched right here,
+oVy6tV49qNI-00020-00005382-00005810 directly searching Unsplash from the Image Editor,
+oVy6tV49qNI-00021-00005901-00006015 but I already have my image,
+oVy6tV49qNI-00022-00006015-00006303 so I'm going to drop it in, there it is.
+oVy6tV49qNI-00023-00006303-00006584 And then I can scroll down and enter my Alt Text,
+oVy6tV49qNI-00024-00006584-00006913 which is really important for the accessibility.
+oVy6tV49qNI-00025-00006913-00007088 Just describing your image.
+oVy6tV49qNI-00026-00007202-00007468 And then I'm going to click Submit.
+oVy6tV49qNI-00027-00007468-00007742 It's going to embed the image on my page.
+oVy6tV49qNI-00028-00007742-00008242 It's pretty big, so I might want to make that smaller.
+oVy6tV49qNI-00029-00008277-00008678 If I highlight it, I can drag and drop the corners
+oVy6tV49qNI-00030-00008678-00008826 and make it smaller.
+oVy6tV49qNI-00031-00008826-00009011 I can also click Image Options
+oVy6tV49qNI-00032-00009011-00009387 and I can use the sizing menu here to change the size.
+oVy6tV49qNI-00033-00009387-00009633 So whatever you prefer.
+oVy6tV49qNI-00034-00009633-00009832 So just make it a little bit smaller.
+oVy6tV49qNI-00035-00009832-00010126 And you might also want to kind of float the image
+oVy6tV49qNI-00036-00010126-00010316 to the right of the text.
+oVy6tV49qNI-00037-00010316-00010655 So if you want to do that, again, with the image highlighted,
+oVy6tV49qNI-00038-00010655-00010849 come up here to the Rich Content Editor
+oVy6tV49qNI-00039-00010849-00011270 and use the Align menu, select Align Right.
+oVy6tV49qNI-00040-00011270-00011449 It's going to push it over for you.
+oVy6tV49qNI-00041-00011449-00011886 And sometimes you have to adjust it a little bit
+oVy6tV49qNI-00042-00011886-00012061 and maybe add some padding
+oVy6tV49qNI-00043-00012061-00012353 if you don't want the words so close to the image.
+oVy6tV49qNI-00044-00012353-00012550 So if you want to add padding,
+oVy6tV49qNI-00045-00012550-00012961 click the HTML Editor, find your image.
+oVy6tV49qNI-00046-00012961-00013255 Here it is, "image style, float: right."
+oVy6tV49qNI-00047-00013255-00013494 And right after float: right, semicolon,
+oVy6tV49qNI-00048-00013494-00013894 I'm going to say, space, padding, colon,
+oVy6tV49qNI-00049-00013894-00014394 I'm gonna add 20 pixels, 20px, semicolon.
+oVy6tV49qNI-00050-00014457-00014700 And then I'm going to click back over
+oVy6tV49qNI-00051-00014700-00014956 and see that it did add some space.
+oVy6tV49qNI-00052-00014956-00015066 And I am going to go ahead
+oVy6tV49qNI-00053-00015066-00015261 and add an attribution for this image.
+oVy6tV49qNI-00054-00015261-00015369 It's in the public domain,
+oVy6tV49qNI-00055-00015369-00015688 so I don't technically have to, but I like to.
+oVy6tV49qNI-00056-00015688-00016049 So I'm going to copy their suggestion, come back over
+oVy6tV49qNI-00057-00016049-00016542 and paste it at the bottom of my page, and click Save.
+oVy6tV49qNI-00058-00016703-00016786 There you go.
+oVy6tV49qNI-00059-00016786-00017059 And if you want some other examples of attributions,
+oVy6tV49qNI-00060-00017059-00017333 look around our course here in the DE Academy.
+oVy6tV49qNI-00061-00017333-00017504 You'll notice we give attributions
+oVy6tV49qNI-00062-00017504-00017679 at the bottom of our pages,
+oVy6tV49qNI-00063-00017679-00017828 and those are some good examples
+oVy6tV49qNI-00064-00017828-00018224 of what different attributions for images look like.
+oVy6tV49qNI-00065-00018224-00018369 All right. I hope that helps!
+obyHl-gndw4-00000-00000570-00000798 Robert Wagner: Greetings, and welcome to the Introduction to
+obyHl-gndw4-00001-00000798-00001110 Physical Science. In this lecture, we are going to talk
+obyHl-gndw4-00002-00001110-00001524 about a couple of concepts in physics: work and kinetic energy
+obyHl-gndw4-00003-00001524-00001902 and how they relate together. And we'll see that work is a
+obyHl-gndw4-00004-00001902-00002202 little differently defined in physics from a physics point of
+obyHl-gndw4-00005-00002202-00002718 view than it is in everyday life. So let's go ahead and get
+obyHl-gndw4-00006-00002718-00003201 started here. And let's look at what we mean by work. Well work
+obyHl-gndw4-00007-00003201-00003618 is the product of force, and the distance through which the force
+obyHl-gndw4-00008-00003618-00004092 acts in the same direction. So if an object moves, in this
+obyHl-gndw4-00009-00004092-00004509 case, say it moves to the right, and you have a force pushing to
+obyHl-gndw4-00010-00004509-00005076 the right, then you have a work that is being done. So that's
+obyHl-gndw4-00011-00005076-00005418 the force and this is the velocity, the object is moving
+obyHl-gndw4-00012-00005418-00005718 to the right, and the force is in that direction, that force is
+obyHl-gndw4-00013-00005718-00006291 doing some work on it. Now, in an actuality, the work is equal
+obyHl-gndw4-00014-00006291-00006698 to the force times the distance times the cosine of the angle
+obyHl-gndw4-00015-00006698-00007056 through which the force acts. And that allows us to look at
+obyHl-gndw4-00016-00007056-00007530 forces that might be acting in other directions at some angle
+obyHl-gndw4-00017-00007530-00007947 relative to the motion. For this class, we are not going to do
+obyHl-gndw4-00018-00007947-00008373 that. So in our case is cosine of theta equals one. And that
+obyHl-gndw4-00019-00008373-00008973 means that work is equal to the force times the distance. Now in
+obyHl-gndw4-00020-00008973-00009366 order for the force to do work, it has to be in the same
+obyHl-gndw4-00021-00009366-00009735 direction. So we have this object here, gravity would be
+obyHl-gndw4-00022-00009735-00010149 pulling down on it. But the gravitational force doesn't know
+obyHl-gndw4-00023-00010149-00010542 work, because it is not in the same direction as the motion of
+obyHl-gndw4-00024-00010542-00010992 the object, the gravitational force can only do do work if the
+obyHl-gndw4-00025-00010992-00011397 object is moving up or down. In this case, the object is not
+obyHl-gndw4-00026-00011397-00011835 doing that. So if the object does not move, the object is not
+obyHl-gndw4-00027-00011835-00012234 moving in the same direction, then no work is done. And if the
+obyHl-gndw4-00028-00012237-00012693 object does not move, then no work is done. So for example,
+obyHl-gndw4-00029-00012693-00012990 you can sit there and push against a wall all day as hard
+obyHl-gndw4-00030-00012990-00013469 as you can, you were doing absolutely no work. Because the
+obyHl-gndw4-00031-00013469-00013922 distance moved is zero. If it does not move, then no work is
+obyHl-gndw4-00032-00013922-00014484 being done in a physical sense. Now, the SI units for this are
+obyHl-gndw4-00033-00014487-00014910 the Newton-meter or the Joule. So one Joule is one
+obyHl-gndw4-00034-00014913-00015258 Newton-meter. And it's a relatively small amount of
+obyHl-gndw4-00035-00015258-00015716 energy that would lift a small Apple, a distance of about one
+obyHl-gndw4-00036-00015719-00016122 meter up against gravity. So it's not a very large amount.
+obyHl-gndw4-00037-00016122-00016524 But we'll see that we'll use this as our unit for energy and
+obyHl-gndw4-00038-00016527-00017172 work. So let's go ahead and continue. And let's look at an
+obyHl-gndw4-00039-00017172-00017490 example of this. So let's look at the example here, of the
+obyHl-gndw4-00040-00017490-00017832 person pushing the lawnmower. Now again, any when you push the
+obyHl-gndw4-00041-00017832-00018075 lawnmower, you'd push it an angle, you don't push it
+obyHl-gndw4-00042-00018075-00018429 straight to the right. For our purposes, we're just going to
+obyHl-gndw4-00043-00018429-00018885 look at that portion that is going in the direction. So we
+obyHl-gndw4-00044-00018885-00019305 are not worrying about this cosine or the fact that there is
+obyHl-gndw4-00045-00019305-00019656 an angle angle here. For our purposes, we're just going to
+obyHl-gndw4-00046-00019656-00020262 assume that that 60 Newtons is all going directly to the right
+obyHl-gndw4-00047-00020262-00020637 hand side. So let's keep our picture up there and write down
+obyHl-gndw4-00048-00020637-00021087 what we know. We know the force 60 Newtons, we know the distance
+obyHl-gndw4-00049-00021090-00021717 25 meters. And once we put those together, we then have our
+obyHl-gndw4-00050-00021717-00022131 equation work equals force times distance. And we put in our
+obyHl-gndw4-00051-00022131-00022638 numbers that we know 60 Newtons and 25 meters and multiply them
+obyHl-gndw4-00052-00022656-00023589 to get 1500 Newton-meters, or 1500 Joules. So we can do that
+obyHl-gndw4-00053-00023589-00023991 do that. And note that one question here with our answer is
+obyHl-gndw4-00054-00023991-00024375 that these had three significant figures, our answer has three
+obyHl-gndw4-00055-00024375-00024798 significant figures. So we'd have to remember that if we were
+obyHl-gndw4-00056-00024798-00025299 working with this for another part of a problem. For example,
+obyHl-gndw4-00057-00025299-00025641 how many significant figures does it have, because it looks
+obyHl-gndw4-00058-00025641-00026055 like it only has two. But in reality, if these both had three
+obyHl-gndw4-00059-00026055-00026466 significant figures, the one the five and the zero are all
+obyHl-gndw4-00060-00026466-00027033 significant. So let's go ahead and move on and look at our next
+obyHl-gndw4-00061-00027033-00027342 section here and talk about what happens to the work that is
+obyHl-gndw4-00062-00027342-00027714 done. So going back to our example of a lawnmower if we
+obyHl-gndw4-00063-00027714-00028107 push that at constant speed, what is happening, if we
+obyHl-gndw4-00064-00028107-00028500 something moves at constant speed, there is no net force
+obyHl-gndw4-00065-00028502-00028833 acting on it. However, we know we are pushing on this object.
+obyHl-gndw4-00066-00028866-00029208 So we are causing it to move where is this energy going?
+obyHl-gndw4-00067-00029247-00029550 Well, the energy is being converted to friction and then
+obyHl-gndw4-00068-00029550-00029868 to heat. So what is being dissipated
+obyHl-gndw4-00069-00029962-00030339 with in the atmosphere within the mechanism of the lawnmower
+obyHl-gndw4-00070-00030339-00030742 itself, and within the wheels within friction with the ground.
+obyHl-gndw4-00071-00030766-00031099 So that is all changing temperatures slightly in the
+obyHl-gndw4-00072-00031105-00031456 making kicking up the energy that way, when we look at
+obyHl-gndw4-00073-00031456-00031861 carrying an object upstairs, the work is being converted into
+obyHl-gndw4-00074-00031861-00032275 potential energy, which is a stored energy and can later be
+obyHl-gndw4-00075-00032275-00032689 released. So it's a little bit different, for example, then
+obyHl-gndw4-00076-00032689-00033010 this energy, which is just a dissipated out into the
+obyHl-gndw4-00077-00033010-00033484 environment, carrying an object up allows you a way to store the
+obyHl-gndw4-00078-00033484-00033853 energy. Now the other thing we wanted to look at here was the
+obyHl-gndw4-00079-00033853-00034393 network and the work energy theorem. So net work is the net
+obyHl-gndw4-00080-00034393-00034897 force. So we could have multiple forces on an object, we could
+obyHl-gndw4-00081-00034897-00035506 have a force this way. And we could have a smaller force this
+obyHl-gndw4-00082-00035506-00035923 way to the left. So a large force to the right, a small
+obyHl-gndw4-00083-00035923-00036580 force to the left, the net force is the difference between those
+obyHl-gndw4-00084-00036580-00037510 two, it would be F1 minus F2. So that would be the net force,
+obyHl-gndw4-00085-00037513-00037984 that is what gives us the net amount of work done. So we look
+obyHl-gndw4-00086-00037984-00038269 at that, and then the distance would just be the total distance
+obyHl-gndw4-00087-00038269-00038668 that it moves off to the right, we will also be looking at the
+obyHl-gndw4-00088-00038668-00039046 work energy theorem. And that uses kinetic energy, which I'll
+obyHl-gndw4-00089-00039046-00039475 discuss in more detail in a minute. But the network is equal
+obyHl-gndw4-00090-00039475-00039904 to the difference in the kinetic energies of the object at the
+obyHl-gndw4-00091-00039904-00040309 end, after it's been after the work has been applied, as
+obyHl-gndw4-00092-00040309-00040948 compared to before, so the kinetic energy is given by one
+obyHl-gndw4-00093-00040948-00041872 half mv squared. So that's the kinetic energy again, I'll talk
+obyHl-gndw4-00094-00041872-00042298 about that more in just a moment here. But the the network is
+obyHl-gndw4-00095-00042298-00042676 just the difference between those two energies, this is the
+obyHl-gndw4-00096-00042676-00043618 energy before. And the other is the energy after. And the
+obyHl-gndw4-00097-00043618-00044074 difference between those tells us the amount of work that was
+obyHl-gndw4-00098-00044074-00045034 done. So kinetic energy is given by one half mv squared, where we
+obyHl-gndw4-00099-00045034-00045523 have the mass of the object and the velocity of the object. This
+obyHl-gndw4-00100-00045523-00045826 is what we call the translational kinetic energy.
+obyHl-gndw4-00101-00045865-00046399 And any object with a mass moving at a velocity will have
+obyHl-gndw4-00102-00046399-00046804 some kinetic energy. So if the object has no mass, no kinetic
+obyHl-gndw4-00103-00046804-00047404 energy, if it is not moving, it has no kinetic energy. The
+obyHl-gndw4-00104-00047404-00047767 kinetic energy of your note depends on the square of the
+obyHl-gndw4-00105-00047767-00048292 velocity. And that means that an object moving twice as fast has
+obyHl-gndw4-00106-00048292-00048832 four times the amount of kinetic energy. So an object moving at
+obyHl-gndw4-00107-00048847-00049255 two meters per second has four times the kinetic energy as an
+obyHl-gndw4-00108-00049255-00049807 object moving at one meter per second. So let's look at an
+obyHl-gndw4-00109-00049807-00050059 example here. And we're going to do this problem in multiple
+obyHl-gndw4-00110-00050059-00050422 parts. So let's start off with the first one. And what we have
+obyHl-gndw4-00111-00050425-00050836 is an object on a conveyor belt. And it's a package
+obyHl-gndw4-00112-00050902-00051313 moving at point five meters per second. Now our first step is
+obyHl-gndw4-00113-00051313-00051748 just finding the kinetic energy. So we have our sketch here, we
+obyHl-gndw4-00114-00051748-00052036 want to look at what we've known, and what are we trying to
+obyHl-gndw4-00115-00052036-00052414 find. So we know these values, we know the mass, we know the
+obyHl-gndw4-00116-00052414-00052801 velocity, and we're trying to find the kinetic energy. So we
+obyHl-gndw4-00117-00052801-00053154 can put in the equation for kinetic energy, one half mv
+obyHl-gndw4-00118-00053154-00053655 squared, put in the values that we know from above, and then
+obyHl-gndw4-00119-00053671-00054208 calculate out that the kinetic energy of this object is 3.75
+obyHl-gndw4-00120-00054211-00054700 Joules. So very basic, there, we're just using the definition
+obyHl-gndw4-00121-00054700-00055159 of kinetic energy to determine the amount the kinetic energy of
+obyHl-gndw4-00122-00055159-00055633 this object right now. Now, we can continue using the same
+obyHl-gndw4-00123-00055633-00056008 object to do another problem. And now we're going to look at
+obyHl-gndw4-00124-00056008-00056296 afterwards, this is how it was moving, but now we're applying a
+obyHl-gndw4-00125-00056296-00056791 force to it. And it's a force of 120 Newtons over a distance of
+obyHl-gndw4-00126-00056791-00057385 point eight meters. So as we have as we have that, and we
+obyHl-gndw4-00127-00057385-00057820 also have the frictional force opposing this. So as with
+obyHl-gndw4-00128-00057820-00058108 before, we will have our picture. We want to put our
+obyHl-gndw4-00129-00058108-00058486 known values in there. And what are we trying to find what we're
+obyHl-gndw4-00130-00058486-00058879 trying to find the net amount of work done? Well, in order to
+obyHl-gndw4-00131-00058879-00059173 find the net work, we're going to have to first find the net
+obyHl-gndw4-00132-00059173-00059536 force. Now if we look at this, we have forces going in two
+obyHl-gndw4-00133-00059536-00060013 directions. We have 120 Newtons to the right. Five Newtons to
+obyHl-gndw4-00134-00060013-00060373 the left, so the net force, we are going to subtract those to
+obyHl-gndw4-00135-00060373-00061092 get 115 Newton's, and that would be to the right. Now, we know
+obyHl-gndw4-00136-00061092-00061441 that the work the network is the net force, which we just
+obyHl-gndw4-00137-00061441-00061905 determined here, times the distance through which it moves,
+obyHl-gndw4-00138-00061905-00062236 which we know from appear. So now we can go ahead and put our
+obyHl-gndw4-00139-00062236-00062620 values in for this one and solve this and find out that the
+obyHl-gndw4-00140-00062620-00063361 network is 115 Newton's times point eight meters, or 92.0
+obyHl-gndw4-00141-00063364-00063997 Newton meters, or joules. So 92 Joules would be the net amount
+obyHl-gndw4-00142-00063997-00064528 of work that is being done by the force. Now, as a reminder,
+obyHl-gndw4-00143-00064530-00064900 no work is being done by the gravitational force here,
+obyHl-gndw4-00144-00064980-00065593 pushing down, or the normal force pushing up, so that those
+obyHl-gndw4-00145-00065593-00065938 forces are not acting in the direction in which the object is
+obyHl-gndw4-00146-00065938-00066676 moving, and therefore no work is being done. Alright, so let's
+obyHl-gndw4-00147-00066703-00067129 continue with the last part of this example. So now we want to
+obyHl-gndw4-00148-00067129-00067606 find the final velocity. So same, same setup, every nothing
+obyHl-gndw4-00149-00067606-00067909 has changed. We all we know all of the same things that we knew
+obyHl-gndw4-00150-00067909-00068248 from before. So what I'm putting up here is two reminders of
+obyHl-gndw4-00151-00068248-00068599 things that we've learned so far about this system. And we've
+obyHl-gndw4-00152-00068599-00068968 learned about the initial kinetic energy, and the net
+obyHl-gndw4-00153-00068968-00069271 amount of work that has been done. That's what we calculated
+obyHl-gndw4-00154-00069271-00069583 in the previous two portions. And we're going to use those to
+obyHl-gndw4-00155-00069583-00070216 help us find the final velocity. So let's go ahead, and we know
+obyHl-gndw4-00156-00070216-00070747 that the initial kinetic energy is 3.75 joules, that's one
+obyHl-gndw4-00157-00070747-00071389 point, or sorry, one half mv0 squared. We know that the
+obyHl-gndw4-00158-00071389-00072001 network is equal to one half mv squared, that v is what we're
+obyHl-gndw4-00159-00072001-00072463 looking for right here, minus one half mv0 squared. Well, we
+obyHl-gndw4-00160-00072463-00072958 know what this is already. Here. We know what the network is over
+obyHl-gndw4-00161-00072958-00073636 here. So the only thing we know the mass from previous problem,
+obyHl-gndw4-00162-00073636-00074005 that's 30 kilograms. So the only thing we don't know in this is
+obyHl-gndw4-00163-00074005-00074359 the velocity. However, first, we have to rearrange this a little
+obyHl-gndw4-00164-00074359-00074887 bit to be able to solve so we move things around and get this
+obyHl-gndw4-00165-00074887-00075280 portion over on the left hand side. And that is equal to the
+obyHl-gndw4-00166-00075280-00076003 network plus one half mv0 squared, at 92 Joules plus 3.75
+obyHl-gndw4-00167-00076003-00076870 Joules gives us the value of that one half mv squared equals
+obyHl-gndw4-00168-00076870-00077542 95.75 Joules. And now we have to solve for the velocity. And we
+obyHl-gndw4-00169-00077542-00077839 can do that by saying the velocity is equal to the square
+obyHl-gndw4-00170-00077839-00077959 root of two
+obyHl-gndw4-00171-00077993-00078623 times 95.75 Joules divided by the 30 kilograms. And that is
+obyHl-gndw4-00172-00078623-00079169 equal to 2.53 meters per second. Again, everything in this
+obyHl-gndw4-00173-00079169-00079451 problem we've used three significant figures throughout.
+obyHl-gndw4-00174-00079469-00079988 So our final answer will have three significant figures. So
+obyHl-gndw4-00175-00079988-00080348 we've been able to use this one setup to solve a number of
+obyHl-gndw4-00176-00080348-00080807 different things and to solve a multiple step problem. So let's
+obyHl-gndw4-00177-00080807-00081254 go ahead and finish up as we do with our summary. And what we
+obyHl-gndw4-00178-00081254-00081623 looked at today was that work is defined to be the product of the
+obyHl-gndw4-00179-00081623-00081998 force on an object, multiplied by the distance through which
+obyHl-gndw4-00180-00081998-00082517 the force acts. So work equals force, times distance, and
+obyHl-gndw4-00181-00082517-00082922 remember that they have to be in the same direction. So even
+obyHl-gndw4-00182-00082922-00083294 though gravity is pulling on an object sitting on the floor, it
+obyHl-gndw4-00183-00083294-00083789 is doing no work because the object does not move. We looked
+obyHl-gndw4-00184-00083789-00084125 at kinetic energy of objects, which is the energy due to
+obyHl-gndw4-00185-00084125-00085040 motion and kinetic energy was equal to one half mv squared.
+obyHl-gndw4-00186-00085154-00085538 And the work energy theorem relates the net work to the
+obyHl-gndw4-00187-00085538-00085937 change in kinetic energy. And we use that in the last problem to
+obyHl-gndw4-00188-00085937-00086333 be able to find the velocity of that package moving along the
+obyHl-gndw4-00189-00086333-00086774 belt. So that concludes this lecture on work and kinetic
+obyHl-gndw4-00190-00086774-00087107 energy. We'll be back again next time for another topic in
+obyHl-gndw4-00191-00087107-00087599 physical science. So until then, have a great day everyone and I
+obyHl-gndw4-00192-00087599-00087797 will see you in class.
+odN92kI-rX0-00000-00000033-00000497 - This micro conference is about thermal zones hierarchy.
+odN92kI-rX0-00001-00000733-00001062 So this topic was already addressed last time,
+odN92kI-rX0-00002-00001062-00001508 we discussed but we don't reach a consensus.
+odN92kI-rX0-00003-00001508-00001802 We don't agree on something
+odN92kI-rX0-00004-00001802-00002122 and I wanted to discuss again about that
+odN92kI-rX0-00005-00002122-00002329 so we can have some progress,
+odN92kI-rX0-00006-00002329-00002514 on how we can represent the thermal zone.
+odN92kI-rX0-00007-00002514-00002898 So it's not about dealing with all the thermal zone
+odN92kI-rX0-00008-00002898-00003271 and taking decision, but just hope we can organize that,
+odN92kI-rX0-00009-00003271-00003771 to have in the resulting code to be clean.
+odN92kI-rX0-00010-00004046-00004546 So why do we need to to make this more order?
+odN92kI-rX0-00011-00004578-00004901 Is because yesterday if we look at the code,
+odN92kI-rX0-00012-00004901-00005241 we have upstream, I'm not talking about out stream
+odN92kI-rX0-00013-00005241-00005741 out of tree Linux connect but just the upstream connect,
+odN92kI-rX0-00014-00005804-00006282 we have one or two thermal zones defined,
+odN92kI-rX0-00015-00006282-00006570 and now we have boards coming with more than,
+odN92kI-rX0-00016-00006570-00007070 the biggest number I saw is 21 thermal zones.
+odN92kI-rX0-00017-00007173-00007456 So it's a lot of sensors and we have
+odN92kI-rX0-00018-00007456-00007618 to deal with all the sensor list
+odN92kI-rX0-00019-00007618-00007950 and then same time and take some decisions.
+odN92kI-rX0-00020-00007950-00008375 So what we want to do with all these sensors
+odN92kI-rX0-00021-00008375-00008772 is first we want to have some kind of a description
+odN92kI-rX0-00022-00008772-00009272 giving a topology which is, even if it's not perfect,
+odN92kI-rX0-00023-00009302-00009492 at least it gives us an indication
+odN92kI-rX0-00024-00009651-00009953 which is close to the reality.
+odN92kI-rX0-00025-00009953-00010269 Also by giving this relationship
+odN92kI-rX0-00026-00010269-00010626 between all we build a new hierarchy,
+odN92kI-rX0-00027-00010626-00011067 we can also organize the thermal zone
+odN92kI-rX0-00028-00011067-00011313 and detect where are the hot spots
+odN92kI-rX0-00029-00011313-00011665 and if there is a relationship between the different nodes
+odN92kI-rX0-00030-00011665-00012017 and if they can interact with the temperature.
+odN92kI-rX0-00031-00012170-00012498 Also by grouping the thermal zone with this hierarchy,
+odN92kI-rX0-00032-00012498-00012997 we can, they have one governor using one node,
+odN92kI-rX0-00033-00013140-00013640 while beginning to, under all the descendant of this node
+odN92kI-rX0-00034-00013733-00014194 it's not representing a thermal zone and deal with them.
+odN92kI-rX0-00035-00014194-00014510 And also it's a way to represent what we have,
+odN92kI-rX0-00036-00014510-00014872 legacy system where we have a spanning sensors.
+odN92kI-rX0-00037-00014872-00015290 So if we want to extend the current framework,
+odN92kI-rX0-00038-00015466-00015796 we need to create a tree representation
+odN92kI-rX0-00039-00016005-00016398 so that the good thing is,
+odN92kI-rX0-00040-00016398-00016656 it's compatible with the current representation
+odN92kI-rX0-00041-00016656-00016831 which is a flat representation.
+odN92kI-rX0-00042-00016975-00017231 This representation give us a good indication,
+odN92kI-rX0-00043-00017231-00017699 about the topology and also it's a way
+odN92kI-rX0-00044-00017699-00018016 to give a centralized view to a specific governor.
+odN92kI-rX0-00045-00018127-00018627 We said, we will be able to act on all the nodes,
+odN92kI-rX0-00046-00018674-00018859 we dug into this tree.
+odN92kI-rX0-00047-00018999-00019402 That means for a current framework,
+odN92kI-rX0-00048-00019402-00019689 we have the thermal zones
+odN92kI-rX0-00049-00019689-00019930 and this thermal zone can be without sensors.
+odN92kI-rX0-00050-00020038-00020423 So we break this paradigm saying,
+odN92kI-rX0-00051-00020423-00020583 we have a thermal zone, a sensor
+odN92kI-rX0-00052-00020868-00021194 and also a cooling device assigned to it or a governor.
+odN92kI-rX0-00053-00021194-00021539 So that means that this entity is alone
+odN92kI-rX0-00054-00021882-00022041 and we can add a sensor
+odN92kI-rX0-00055-00022215-00022413 or cooling device or governor, but it's optional.
+odN92kI-rX0-00056-00022413-00022840 And also we cannot give a weight to the nodes
+odN92kI-rX0-00057-00023000-00023378 of the thermal zone saying how much they contribute.
+odN92kI-rX0-00058-00023378-00023480 Yes?
+odN92kI-rX0-00059-00023480-00023729 - [Attendant] So, you know we discussed this many times.
+odN92kI-rX0-00060-00023729-00023875 - Yes.
+odN92kI-rX0-00061-00023875-00024065 - Why you want to call thermal zone,
+odN92kI-rX0-00062-00024065-00024255 you should call thermal sensor.
+odN92kI-rX0-00063-00024255-00024434 And let Zone B you know.
+odN92kI-rX0-00064-00024434-00024553 - [John] What is the sensor?
+odN92kI-rX0-00065-00024553-00024813 - A sensor, a thermal zone without sensor,
+odN92kI-rX0-00066-00024813-00024935 it doesn't make sense right?
+odN92kI-rX0-00067-00024935-00025212 Thermal zone is basically its own with many sensors.
+odN92kI-rX0-00068-00025371-00025480 Like you know.
+odN92kI-rX0-00069-00025480-00025563 - [John] No.
+odN92kI-rX0-00070-00025563-00025714 - Because if zone is that you know like an ACPI
+odN92kI-rX0-00071-00025714-00025931 if you define like, a zone is a part of like
+odN92kI-rX0-00072-00025931-00026127 it's a surface is a zone, right.
+odN92kI-rX0-00073-00026127-00026234 - [John] Yeah.
+odN92kI-rX0-00074-00026234-00026383 - That surface can have five sensors.
+odN92kI-rX0-00075-00026383-00026776 - Yeah but we keep the relationship one to one,
+odN92kI-rX0-00076-00026776-00027054 that means one thermal zone is ones and so--
+odN92kI-rX0-00077-00027054-00027232 - [Attendee] But that's the--
+odN92kI-rX0-00078-00027232-00027732 - I agree we can, we can say okay to sensor
+odN92kI-rX0-00079-00027774-00027961 but in this, in this representation
+odN92kI-rX0-00080-00027961-00028300 is the thermal zone as one sensor and we build a tree.
+odN92kI-rX0-00081-00028433-00028755 - So the question, maybe one of the
+odN92kI-rX0-00082-00028755-00029087 primary questions that I have with this approach is,
+odN92kI-rX0-00083-00029087-00029354 why do you need to represent the topology?
+odN92kI-rX0-00084-00029532-00029755 I mean let's step back and go on
+odN92kI-rX0-00085-00029755-00030042 problem statement first, right.
+odN92kI-rX0-00086-00030042-00030267 Why do you need to represent that
+odN92kI-rX0-00087-00030267-00030477 the topology of sensors here?
+odN92kI-rX0-00088-00030477-00030668 What you are gonna do with that information?
+odN92kI-rX0-00089-00030819-00031225 - So if we can have give a weight to thermal zones
+odN92kI-rX0-00090-00031225-00031426 and you present, you represent
+odN92kI-rX0-00091-00031526-00031970 the different thermal zones with this hierarchy
+odN92kI-rX0-00092-00031970-00032319 then you can have a representation
+odN92kI-rX0-00093-00032319-00032743 of how the different sensor
+odN92kI-rX0-00094-00032846-00033171 are grouped for a single entity
+odN92kI-rX0-00095-00033171-00033505 like cluster zero which in turn
+odN92kI-rX0-00096-00033505-00033737 can be grouped the odd cluster for CPU.
+odN92kI-rX0-00097-00033926-00034054 Let me give an example.
+odN92kI-rX0-00098-00034542-00035042 So let's imagine we have one sensor on each course
+odN92kI-rX0-00099-00035226-00035496 but we don't have a sensor for the cluster
+odN92kI-rX0-00100-00035496-00035709 and we want a way to group them,
+odN92kI-rX0-00101-00035709-00036026 we can do a virtual sensor also, but see--
+odN92kI-rX0-00102-00036026-00036109 - That's the point.
+odN92kI-rX0-00103-00036109-00036240 That's why we were saying if you
+odN92kI-rX0-00104-00036240-00036444 do present those bottom circles
+odN92kI-rX0-00105-00036444-00036729 as a sensor and thermal zone has weight
+odN92kI-rX0-00106-00036729-00036951 of dependency on sensors,
+odN92kI-rX0-00107-00036951-00037162 you can define individual weights of each sensor
+odN92kI-rX0-00108-00037162-00037486 and you can have a virtual sensor a pair clustered.
+odN92kI-rX0-00109-00037486-00037706 - Yes, but at the end we end up
+odN92kI-rX0-00110-00037706-00037886 with a new hierarchy, right.
+odN92kI-rX0-00111-00037886-00038243 - [Attendee] Yeah, yeah, you will have a new hierarchy.
+odN92kI-rX0-00112-00038243-00038339 - So you mean it's a hierarchy of sensor
+odN92kI-rX0-00113-00038339-00038528 or it's a hierarchy of thermal zones?
+odN92kI-rX0-00114-00038528-00038698 - A hierarchy of sensor and zones.
+odN92kI-rX0-00115-00038783-00039134 The sensor is like thermal zone, can't have optional,
+odN92kI-rX0-00116-00039134-00039417 like as I said before, don't have to have temperature.
+odN92kI-rX0-00117-00039417-00039557 - [John] Don't, I'm sorry.
+odN92kI-rX0-00118-00039557-00039892 - Don't have to have temperature, right.
+odN92kI-rX0-00119-00039892-00040242 So thermal zone is a combination of sensors.
+odN92kI-rX0-00120-00040242-00040412 So each sensor will have a definitely
+odN92kI-rX0-00121-00040412-00040663 have must have a set temperature
+odN92kI-rX0-00122-00040663-00040848 and the cluster can be virtual, John.
+odN92kI-rX0-00123-00040962-00041045 - Yes. - Right.
+odN92kI-rX0-00124-00041157-00041340 - Yeah, I mean the way that he is going
+odN92kI-rX0-00125-00041340-00041547 is a little bit simpler than what you're proposing.
+odN92kI-rX0-00126-00041547-00041884 I still don't see why, you may depend on
+odN92kI-rX0-00127-00041884-00042065 what's the post that you want to build over there
+odN92kI-rX0-00128-00042065-00042317 but if you have a hot spot on the plus zero
+odN92kI-rX0-00129-00042317-00042483 and you want to represent that, you can just
+odN92kI-rX0-00130-00042483-00042813 have a little sensor that is say set, I mean,
+odN92kI-rX0-00131-00042813-00043163 it's based off of a set of the sensors that he has here.
+odN92kI-rX0-00132-00043163-00043332 I mean it doesn't need to be limited
+odN92kI-rX0-00133-00043332-00043593 to the sensors that are on the cluster though.
+odN92kI-rX0-00134-00043593-00043717 Even if you have like something
+odN92kI-rX0-00135-00043717-00043946 which represents the hot spot which is even outside,
+odN92kI-rX0-00136-00043946-00044158 some I have seen that weird stuff already.
+odN92kI-rX0-00137-00044281-00044709 People put like a thermistor close to the chip
+odN92kI-rX0-00138-00044803-00045303 that represents better the hot spot than the internal ones.
+odN92kI-rX0-00139-00045355-00045704 So I mean it's not necessarily that you have to
+odN92kI-rX0-00140-00045704-00045991 build like a hierarchy to represent the hot spot.
+odN92kI-rX0-00141-00046149-00046458 I still don't see why you want to build in a hierarchal way.
+odN92kI-rX0-00142-00046616-00046959 I see that, yeah, it does represent how the system is built.
+odN92kI-rX0-00143-00046959-00047157 But what you're gonna do with it?
+odN92kI-rX0-00144-00047157-00047470 - So now if you want to have a governor dealing with
+odN92kI-rX0-00145-00047470-00047795 all this sensors, thermal zone, whatever,
+odN92kI-rX0-00146-00047971-00048292 because of all these design the current framework.
+odN92kI-rX0-00147-00048292-00048465 We have at thermal zone, we have a sensor,
+odN92kI-rX0-00148-00048465-00048624 we have a governor.
+odN92kI-rX0-00149-00048624-00049018 By using the thermal zone hierarchy,
+odN92kI-rX0-00150-00049018-00049197 which is a representation,
+odN92kI-rX0-00151-00049197-00049532 it's the encapsulation of the sensor.
+odN92kI-rX0-00152-00049745-00050009 - Yeah, today we have a one one bad representation.
+odN92kI-rX0-00153-00050009-00050092 - Yes.
+odN92kI-rX0-00154-00050092-00050406 So the idea is we we keep the thermal zone
+odN92kI-rX0-00155-00050406-00050851 which is virtual and it aggregates thermal zone
+odN92kI-rX0-00156-00050851-00051226 having one sensor, and that's the example.
+odN92kI-rX0-00157-00051226-00051591 For example, here.
+odN92kI-rX0-00158-00051591-00051788 - So one clarifying question for
+odN92kI-rX0-00159-00051788-00052111 the notes that I'm trying to take.
+odN92kI-rX0-00160-00052111-00052611 Do you mean, do you want to have one sensor per zone?
+odN92kI-rX0-00161-00052727-00052897 Is this what you are proposing?
+odN92kI-rX0-00162-00052897-00052980 - [John] Yes.
+odN92kI-rX0-00163-00052980-00053342 - One sensor per zone and then have a hierarchy of zones?
+odN92kI-rX0-00164-00053342-00053451 - [John] Yes.
+odN92kI-rX0-00165-00053451-00053554 - That's what you are saying?
+odN92kI-rX0-00166-00053554-00053690 - Okay. - Yes.
+odN92kI-rX0-00167-00053690-00054141 And for example here, you can have all the course
+odN92kI-rX0-00168-00054141-00054641 having a sensor and then cluster zero does not have a sensor
+odN92kI-rX0-00169-00054653-00054915 but when you read the temperature of cluster zero,
+odN92kI-rX0-00170-00054915-00055190 it aggregates so that the temperature of all the other.
+odN92kI-rX0-00171-00055319-00055727 We can, of course, we can create a virtual sensor
+odN92kI-rX0-00172-00055727-00056153 inside this thermal zone, aggregating the sensor
+odN92kI-rX0-00173-00056153-00056365 of the child zone, if you just--
+odN92kI-rX0-00174-00056450-00056648 - There are even like drivers that do that already.
+odN92kI-rX0-00175-00056648-00056731 - Sorry.
+odN92kI-rX0-00176-00056731-00056966 - There is drivers that do that.
+odN92kI-rX0-00177-00056966-00057160 - There are drivers doing that
+odN92kI-rX0-00178-00057160-00057407 but if you look at the code of its design,
+odN92kI-rX0-00179-00057407-00057658 it's about design because it's the same,
+odN92kI-rX0-00180-00057658-00057859 at the end is the same how we represent that.
+odN92kI-rX0-00181-00057859-00057942 - Right.
+odN92kI-rX0-00182-00057942-00058265 - But if we look at all the thermal zone,
+odN92kI-rX0-00183-00058265-00058633 the thermal zone can be built in hierarchy
+odN92kI-rX0-00184-00058633-00058941 and when you call get thermal zone temperature,
+odN92kI-rX0-00185-00059055-00059385 it's much more the impact on the code
+odN92kI-rX0-00186-00059385-00059885 in the core framework is less than having a virtual sensor
+odN92kI-rX0-00187-00059946-00060178 and trying to do or get temperature
+odN92kI-rX0-00188-00060178-00060362 from one sensor to other sensors,
+odN92kI-rX0-00189-00060362-00060552 to a virtual sensor to other sensors.
+odN92kI-rX0-00190-00060552-00060866 It's much simpler to write a thermal zone
+odN92kI-rX0-00191-00060965-00061404 saying when I asked to get me the temperature,
+odN92kI-rX0-00192-00061404-00061788 it will ask the older child thermal zone temperature
+odN92kI-rX0-00193-00061788-00062064 and do the aggregation without having a virtual sensor.
+odN92kI-rX0-00194-00062206-00062532 - Like a comment, if you say that something is easier
+odN92kI-rX0-00195-00062532-00062679 than something else it would be good
+odN92kI-rX0-00196-00062679-00063151 to give an example of when it takes place.
+odN92kI-rX0-00197-00063151-00063428 Like do you have a specific example in mind,
+odN92kI-rX0-00198-00063428-00063795 like when an implementation of a zone may be simpler
+odN92kI-rX0-00199-00063795-00064129 than an implementation of a visual sensor?
+odN92kI-rX0-00200-00064129-00064530 - [John] I think it's about code, so I have a patch.
+odN92kI-rX0-00201-00064530-00064824 I can send the patch but.
+odN92kI-rX0-00202-00064824-00064979 - Okay, so.
+odN92kI-rX0-00203-00064979-00065161 - So let me get example, like for example
+odN92kI-rX0-00204-00065161-00065401 in your cluster zero with one sensor.
+odN92kI-rX0-00205-00065401-00065598 Also, you know like in your case cluster,
+odN92kI-rX0-00206-00065598-00065811 like in our case example, I give you,
+odN92kI-rX0-00207-00065811-00066019 we have it like same temperature sensor
+odN92kI-rX0-00208-00066019-00066313 on a CPU but that temperature sensor
+odN92kI-rX0-00209-00066313-00066637 also affects like a surface skin.
+odN92kI-rX0-00210-00066722-00066938 So same sensor will be in cluster one
+odN92kI-rX0-00211-00066938-00067163 and cluster zero in your example.
+odN92kI-rX0-00212-00067258-00067481 How do you, how do you replicate here or?
+odN92kI-rX0-00213-00067598-00067733 So suppose one sensor here--
+odN92kI-rX0-00214-00067733-00067930 - [John] Something like that?
+odN92kI-rX0-00215-00067930-00068084 So you have a sensor on cluster zero
+odN92kI-rX0-00216-00068084-00068183 and sensor on cluster one.
+odN92kI-rX0-00217-00068183-00068606 - Same sensor is used in multiple, in the middle.
+odN92kI-rX0-00218-00068606-00068714 - [John] It's the same.
+odN92kI-rX0-00219-00068714-00068890 - So you would have an arrow also, for example,
+odN92kI-rX0-00220-00068890-00069132 from cluster one to two-- - Also on the cluster one.
+odN92kI-rX0-00221-00069132-00069215 It's here.
+odN92kI-rX0-00222-00069215-00069415 - No, from cluster one, you will have an arrow
+odN92kI-rX0-00223-00069415-00069695 from cluster one to one of the children of cluster zero.
+odN92kI-rX0-00224-00069987-00070111 So across--
+odN92kI-rX0-00225-00070111-00070214 - [Attendee] No.
+odN92kI-rX0-00226-00070214-00070298 - Cross reference, sorry.
+odN92kI-rX0-00227-00070298-00070427 - [Attendee] Yeah, from here.
+odN92kI-rX0-00228-00070427-00070511 - Either way?
+odN92kI-rX0-00229-00070511-00070652 - [Attendee] Yes.
+odN92kI-rX0-00230-00070754-00071057 So if you represent them as a sensor
+odN92kI-rX0-00231-00071057-00071285 and then the zone as a formula of sensors,
+odN92kI-rX0-00232-00071285-00071455 then you can, it's logical.
+odN92kI-rX0-00233-00071455-00071613 - [Attendee] It's more logical, yeah.
+odN92kI-rX0-00234-00071613-00071778 - Right.
+odN92kI-rX0-00235-00071778-00071885 - [Attendee] I mean I think the--
+odN92kI-rX0-00236-00071885-00072073 - You'll achieve the same purpose.
+odN92kI-rX0-00237-00072073-00072471 - The 101 representation, it kind of limits us, right,
+odN92kI-rX0-00238-00072561-00072868 and drives us to do these kind of weird hacks of,
+odN92kI-rX0-00239-00072868-00073202 okay, let's have multiple zones on top of each other.
+odN92kI-rX0-00240-00073202-00073524 But if we split the zone from sensor concept,
+odN92kI-rX0-00241-00073675-00073886 I think evenly makes easier, for example,
+odN92kI-rX0-00242-00073886-00074386 to aggregate or to connect with half or more, for example,
+odN92kI-rX0-00243-00074446-00074656 if we have a better representation
+odN92kI-rX0-00244-00074656-00074957 of a set thermal sensor that's using in thermal zone.
+odN92kI-rX0-00245-00074957-00075196 - [Attendee] And you will achieve the same purpose, right.
+odN92kI-rX0-00246-00075196-00075401 - I think we've reached the same goal.
+odN92kI-rX0-00247-00075401-00075576 - [Attendee] Yeah, same goal, yeah.
+odN92kI-rX0-00248-00075799-00075949 - So instead of calling that
+odN92kI-rX0-00249-00075949-00076042 thermal zone. - Thermal zone.
+odN92kI-rX0-00250-00076042-00076155 - You want to call that
+odN92kI-rX0-00251-00076155-00076320 sensors? - Thermal sensors, yeah.
+odN92kI-rX0-00252-00076320-00076558 Which has only temperature with no thresholds,
+odN92kI-rX0-00253-00076558-00076868 no, what's the called, trip points, no trip points.
+odN92kI-rX0-00254-00076868-00077062 - Trip points finish and so on.
+odN92kI-rX0-00255-00077062-00077204 - Yeah.
+odN92kI-rX0-00256-00077204-00077573 You can have a temperature notification limit.
+odN92kI-rX0-00257-00077573-00077712 That's fine, but not the,
+odN92kI-rX0-00258-00077802-00078109 the trip point set as at the zone there.
+odN92kI-rX0-00259-00078109-00078451 - So basically your suggestion will be to define sensors
+odN92kI-rX0-00260-00078451-00078623 as something that measures temperature
+odN92kI-rX0-00261-00078623-00079035 and zones, that something that applies certain limits
+odN92kI-rX0-00262-00079035-00079191 or trip points on those thing.
+odN92kI-rX0-00263-00079318-00079402 - Sensors, combination of sensors.
+odN92kI-rX0-00264-00079509-00079701 - On the combination of sensors, okay.
+odN92kI-rX0-00265-00079701-00079789 - Right.
+odN92kI-rX0-00266-00079789-00080128 - If we take the example of, we have--
+odN92kI-rX0-00267-00080128-00080470 - And then again, I mean, you can plug in
+odN92kI-rX0-00268-00080470-00080966 a sensor on multiple zones as well.
+odN92kI-rX0-00269-00080966-00081208 - So doing spanning sensors.
+odN92kI-rX0-00270-00081208-00081433 - Yeah, that's a lanoff.
+odN92kI-rX0-00271-00081433-00081675 - Sensors are expensive, right, people need to pay money,
+odN92kI-rX0-00272-00081675-00081933 in the virtual is free so they can,
+odN92kI-rX0-00273-00081933-00082299 so that's what they've tried to do, same sensor, use, reuse.
+odN92kI-rX0-00274-00082426-00082926 - So from the program programming okay,
+odN92kI-rX0-00275-00083011-00083172 recording inside the kernel.
+odN92kI-rX0-00276-00083172-00083255 - Yes.
+odN92kI-rX0-00277-00083255-00083582 - So let's imagine you have cluster zero.
+odN92kI-rX0-00278-00083582-00083752 Cluster zero is aggregating.
+odN92kI-rX0-00279-00083752-00084036 You have a virtual sensor there, okay.
+odN92kI-rX0-00280-00084036-00084484 And then you ask the temperature of this thermal zone.
+odN92kI-rX0-00281-00084484-00084799 It will call the temperature of the sensor,
+odN92kI-rX0-00282-00084799-00084984 which is a virtual sensor.
+odN92kI-rX0-00283-00084984-00085431 And this virtual sensor, what will it call?
+odN92kI-rX0-00284-00085431-00085667 The directly the sensor will get temperature call back,
+odN92kI-rX0-00285-00085667-00086109 or will we call to get temperature from the thermal zone?
+odN92kI-rX0-00286-00086109-00086531 - So essentially my kind of vision at this point
+odN92kI-rX0-00287-00086531-00086785 would be that, you have a zone
+odN92kI-rX0-00288-00086785-00087218 that has a list of sensors and then there is a function.
+odN92kI-rX0-00289-00087218-00087301 - Right.
+odN92kI-rX0-00290-00087301-00087470 - Which you have to apply to those sensors
+odN92kI-rX0-00291-00087470-00087585 to get the temperature.
+odN92kI-rX0-00292-00087834-00088031 - And that function can be you know polynomial
+odN92kI-rX0-00293-00088031-00088213 or set off the sensors.
+odN92kI-rX0-00294-00088312-00088662 - [John] So you, the paradigm of having one sensor,
+odN92kI-rX0-00295-00088662-00089025 one thermal zone, you break this this paradigm.
+odN92kI-rX0-00296-00089025-00089161 - You can have it.
+odN92kI-rX0-00297-00089161-00089262 This can still represent.
+odN92kI-rX0-00298-00089262-00089475 - I mean you would have like one thermal zone
+odN92kI-rX0-00299-00089475-00089849 has one aggregated value at the moment, right,
+odN92kI-rX0-00300-00089849-00090085 which is the output of that function.
+odN92kI-rX0-00301-00090085-00090399 - Yeah if you want you can have one doing two.
+odN92kI-rX0-00302-00090399-00090737 - But it's not the one thermal zone to one sensor anymore.
+odN92kI-rX0-00303-00091219-00091559 - And do you represent the physically
+odN92kI-rX0-00304-00091559-00091934 the meaning of the thermal zones, the sensors?
+odN92kI-rX0-00305-00091934-00092086 - Then that's back to my first question,
+odN92kI-rX0-00306-00092086-00092256 why do you want to represent that?
+odN92kI-rX0-00307-00092385-00092789 - Conceptually I find it's a better representation.
+odN92kI-rX0-00308-00092789-00093096 - I mean unless, I mean, yeah I kind of agree with that
+odN92kI-rX0-00309-00093096-00093426 but if you are not solving any problem, why brother?
+odN92kI-rX0-00310-00093578-00093855 What's the problem that you're solving, right?
+odN92kI-rX0-00311-00093855-00094086 You have like, what, at the end of the day
+odN92kI-rX0-00312-00094086-00094356 what you want to do here, is like represent hot spots.
+odN92kI-rX0-00313-00094441-00094547 - Okay.
+odN92kI-rX0-00314-00094547-00094662 - And then take control of that.
+odN92kI-rX0-00315-00094662-00095009 - So if we have a governor and we want this governor
+odN92kI-rX0-00316-00095009-00095470 to deal with different thermal zone or different sensors.
+odN92kI-rX0-00317-00095470-00095687 - Right, then I think then that question
+odN92kI-rX0-00318-00095687-00096013 becomes a similar question of,
+odN92kI-rX0-00319-00096013-00096513 how do we aggregate to a governor, multiple zones, right.
+odN92kI-rX0-00320-00096655-00096936 Then I think to get to the point where you want to solve
+odN92kI-rX0-00321-00096936-00097126 the problems that you are proposing here,
+odN92kI-rX0-00322-00097126-00097338 then we would have like a thermal zone,
+odN92kI-rX0-00323-00097338-00097587 can have aggregation of multiple sensors,
+odN92kI-rX0-00324-00097587-00097911 and a governor should have access to multiple thermal zones.
+odN92kI-rX0-00325-00098182-00098265 Then you don't need
+odN92kI-rX0-00326-00098265-00098634 the physical representation of the hierarchy.
+odN92kI-rX0-00327-00099475-00099929 - So now you have to apply some cooling effect.
+odN92kI-rX0-00328-00099929-00100185 An example, how do you split that, I mean,
+odN92kI-rX0-00329-00100185-00100445 you say that you want lead the cluster zero
+odN92kI-rX0-00330-00100445-00100911 to all the CPU to only consume a number of what?
+odN92kI-rX0-00331-00100911-00101044 And then you will have to split that.
+odN92kI-rX0-00332-00101044-00101253 If you don't know how is the hierarchy,
+odN92kI-rX0-00333-00101253-00101541 how you can split them in this barber jet
+odN92kI-rX0-00334-00101541-00101917 in smaller barber jet for each zone, a sensor,
+odN92kI-rX0-00335-00101917-00102070 if you don't have the hierarchy?
+odN92kI-rX0-00336-00102192-00102542 (faintly speaking)
+odN92kI-rX0-00337-00102542-00102678 No, it's not across the zone.
+odN92kI-rX0-00338-00102678-00102825 - Mic, mic.
+odN92kI-rX0-00339-00102825-00103003 - So the power locator governor,
+odN92kI-rX0-00340-00103003-00103258 you already have a different one budget
+odN92kI-rX0-00341-00103258-00103709 with many zones, right.
+odN92kI-rX0-00342-00103709-00103948 Like yeah like CPU, CPU you're already doing.
+odN92kI-rX0-00343-00103948-00104263 - No, I agree for that but then between the CPU
+odN92kI-rX0-00344-00104263-00104598 are you split the CPU budget between cluster zero
+odN92kI-rX0-00345-00104598-00105033 and cluster one, between CPU zero and one, up to eight.
+odN92kI-rX0-00346-00105033-00105297 At the point, if you don't have the dependency
+odN92kI-rX0-00347-00105297-00105605 between the CPU sensor to the cluster,
+odN92kI-rX0-00348-00105691-00105806 how you can split that?
+odN92kI-rX0-00349-00105956-00106059 I mean what you have right now,
+odN92kI-rX0-00350-00106059-00106266 you have three independent thermal zone
+odN92kI-rX0-00351-00106266-00106501 which are CPU graphic and peripherals,
+odN92kI-rX0-00352-00106501-00106690 and you split that quite easily.
+odN92kI-rX0-00353-00106690-00107080 But then if you go, if you want to go one more level below,
+odN92kI-rX0-00354-00107080-00107333 I mean for all the CPU, how do you,
+odN92kI-rX0-00355-00107333-00107672 how can you split the budget between all the CPU?
+odN92kI-rX0-00356-00107672-00107948 Because maybe one is operating more than another one,
+odN92kI-rX0-00357-00107948-00108163 so maybe the power budget can be different.
+odN92kI-rX0-00358-00108263-00108563 If you don't have the topology, you don't have any,
+odN92kI-rX0-00359-00108563-00108753 you can't make any decision like that.
+odN92kI-rX0-00360-00109029-00109438 - I mean, the way that I see this, you would have,
+odN92kI-rX0-00361-00109581-00109918 once again back to your example right, so you would have
+odN92kI-rX0-00362-00109918-00110418 like multiple thermal zones, one per cluster, can you,
+odN92kI-rX0-00363-00110779-00111014 one per cluster and then you would have like a,
+odN92kI-rX0-00364-00111254-00111409 as a Shunua was saying, right,
+odN92kI-rX0-00365-00111409-00111811 on the power locator we have the power consumption entity
+odN92kI-rX0-00366-00111811-00112226 that represents the consumption of each zone.
+odN92kI-rX0-00367-00112484-00112885 It doesn't, I mean, I don't think you would need to have
+odN92kI-rX0-00368-00112885-00113180 the aggregation or the contribution of each sensor,
+odN92kI-rX0-00369-00113180-00113526 but you would need to have the power consumption
+odN92kI-rX0-00370-00113526-00113805 for each zone and then that's it
+odN92kI-rX0-00371-00113805-00113955 at the level of the zone.
+odN92kI-rX0-00372-00114129-00114483 (faintly speaking)
+odN92kI-rX0-00373-00114483-00114607 Right.
+odN92kI-rX0-00374-00114607-00114958 - Yeah, we can't hear you.
+odN92kI-rX0-00375-00114958-00115417 - At the level one your whatever your power locator governor
+odN92kI-rX0-00376-00115417-00115713 will divide in CPU graphics and peripherals,
+odN92kI-rX0-00377-00115713-00115958 and the second level power budget of CPU
+odN92kI-rX0-00378-00115958-00116128 will be divided among them, right.
+odN92kI-rX0-00379-00116128-00116315 So there you have, governor is doing
+odN92kI-rX0-00380-00116315-00116716 multiple level basically with the hierarchy, right.
+odN92kI-rX0-00381-00116927-00117124 - So basically I think the problem is that
+odN92kI-rX0-00382-00117124-00117533 the hierarchy of the sensors doesn't have to reflect
+odN92kI-rX0-00383-00117533-00117850 the hierarchy of devices that you want to control.
+odN92kI-rX0-00384-00117850-00118057 - [Attendee] That's my point.
+odN92kI-rX0-00385-00118057-00118390 - And then so you basically end up
+odN92kI-rX0-00386-00118390-00118563 with two hierarchies to represent,
+odN92kI-rX0-00387-00118734-00119213 and that, I'm not sure it's the model for applying
+odN92kI-rX0-00388-00119213-00119497 constraints in the inner case that you
+odN92kI-rX0-00389-00119497-00119754 and a string of us were proposing
+odN92kI-rX0-00390-00119754-00119992 like when with the roof thermal zones
+odN92kI-rX0-00391-00119992-00120151 having multiple sensors in them,
+odN92kI-rX0-00392-00120151-00120455 so each zone will also have a list of devices
+odN92kI-rX0-00393-00120455-00120655 that it controls or something like that.
+odN92kI-rX0-00394-00120759-00121089 - Each zone will have sensors to which is multiple device,
+odN92kI-rX0-00395-00121089-00121394 its cooling devices are part of different,
+odN92kI-rX0-00396-00121394-00121580 right, cooling devices are different.
+odN92kI-rX0-00397-00121580-00121663 - Yes but they are important too right,
+odN92kI-rX0-00398-00121663-00122058 so you have to represent so in each zone
+odN92kI-rX0-00399-00122058-00122354 there will be a list of devices to control
+odN92kI-rX0-00400-00122354-00122597 and a list of sensors to read from,
+odN92kI-rX0-00401-00122597-00122727 something like that, okay.
+odN92kI-rX0-00402-00122947-00123270 - Which is true now, with the device.
+odN92kI-rX0-00403-00123270-00123635 - It's not the best representation, I agree,
+odN92kI-rX0-00404-00123635-00124135 it doesn't reflect the hierarchy of the the hardware
+odN92kI-rX0-00405-00124258-00124705 but maybe I'm just failing to see what would be the benefit.
+odN92kI-rX0-00406-00124705-00124944 I mean, if you guys are really thinking of having like
+odN92kI-rX0-00407-00124944-00125340 multiple hierarchies here and you guys are really using that
+odN92kI-rX0-00408-00125340-00125550 like the contribution of which hierarchy
+odN92kI-rX0-00409-00125550-00125739 up to the top hierarchies.
+odN92kI-rX0-00410-00125739-00126081 But I mean, at the end of the day as I said,
+odN92kI-rX0-00411-00126081-00126276 what you want to do is like you have a hot spot
+odN92kI-rX0-00412-00126276-00126472 you want to control that, right.
+odN92kI-rX0-00413-00126583-00126968 - It's not that you really want to control the hot spot,
+odN92kI-rX0-00414-00126968-00127186 you just, you want to split the power budget
+odN92kI-rX0-00415-00127186-00127468 and according, I mean, because--
+odN92kI-rX0-00416-00127468-00127757 - No, I mean, that's from the perspective
+odN92kI-rX0-00417-00127757-00128044 of how you are solving the problem, right.
+odN92kI-rX0-00418-00128044-00128356 But the problem statement is you have a hot spot,
+odN92kI-rX0-00419-00128356-00128646 you want to remove that hot spot from your silicon, right.
+odN92kI-rX0-00420-00128733-00129014 I mean, that from from a thermal management
+odN92kI-rX0-00421-00129014-00129127 perspective, that's what we have got.
+odN92kI-rX0-00422-00129127-00129350 - We can't afford to keep the hotspot.
+odN92kI-rX0-00423-00129350-00129661 - Right, that's the whole point, I mean you have to,
+odN92kI-rX0-00424-00129661-00130059 you know you do your thermal simulation,
+odN92kI-rX0-00425-00130059-00130178 your thermal characterization,
+odN92kI-rX0-00426-00130178-00130365 you find where your hot spots are,
+odN92kI-rX0-00427-00130365-00130694 and you do representations of, as Shunua was saying,
+odN92kI-rX0-00428-00130694-00130968 you have like sensors who are expensive.
+odN92kI-rX0-00429-00130968-00131085 Then you place sensors around
+odN92kI-rX0-00430-00131085-00131310 that hot spot to represent them.
+odN92kI-rX0-00431-00131310-00131461 And at that point you already know
+odN92kI-rX0-00432-00131461-00131663 what's the contribution of those sensors
+odN92kI-rX0-00433-00131663-00131798 to represent that hot spot,
+odN92kI-rX0-00434-00132111-00132396 right, depending on on the workload that you are running.
+odN92kI-rX0-00435-00132614-00132910 - I'm not sure to follow what you want to mean.
+odN92kI-rX0-00436-00132910-00133120 My point is just that having this hierarchy,
+odN92kI-rX0-00437-00133120-00133308 just that when you have several cooling
+odN92kI-rX0-00438-00133308-00133609 device of different way to cool.
+odN92kI-rX0-00439-00133609-00133863 For example, if you take a cluster of CPU
+odN92kI-rX0-00440-00133951-00134239 you can either scale down the frequency of all the CPU
+odN92kI-rX0-00441-00134239-00134596 or you can inject idle on only one CPU
+odN92kI-rX0-00442-00134596-00134800 to get the same power budget.
+odN92kI-rX0-00443-00134800-00134930 How you make the decision?
+odN92kI-rX0-00444-00135170-00135388 I mean, at some point you must know that you have
+odN92kI-rX0-00445-00135388-00135628 a power budget for this cluster and then decide.
+odN92kI-rX0-00446-00135789-00136066 - Yeah but with one hierarchy is not enough for that.
+odN92kI-rX0-00447-00136249-00136391 Is it?
+odN92kI-rX0-00448-00136391-00136474 - Sorry?
+odN92kI-rX0-00449-00136474-00136781 - One hierarchy is not enough for that because you need
+odN92kI-rX0-00450-00136781-00137261 to know what sensors to contribute to your measurement,
+odN92kI-rX0-00451-00137261-00137648 and what devices you can control anyway.
+odN92kI-rX0-00452-00138223-00138450 - Well we'd say that I am not shocked to see
+odN92kI-rX0-00453-00138450-00138766 why you need, it's not enough, one hierarchy,
+odN92kI-rX0-00454-00138766-00138913 because in each thermal zone you will
+odN92kI-rX0-00455-00138913-00139145 have some cooling device attached.
+odN92kI-rX0-00456-00139145-00139333 That's the cooling device represent
+odN92kI-rX0-00457-00139333-00139582 how you can cool this thermal zone.
+odN92kI-rX0-00458-00139582-00139789 - Well it depends on how you define
+odN92kI-rX0-00459-00139789-00140057 the thermal zone but we are running out of time
+odN92kI-rX0-00460-00140057-00140321 for this topic, so let's just pencil it
+odN92kI-rX0-00461-00140321-00140821 for the buff slot and let's just move on to the next one.
+odN92kI-rX0-00462-00141014-00141362 - You mean you've run already out of time, or?
+odN92kI-rX0-00463-00141362-00141611 (applause)
+odN92kI-rX0-00464-00141611-00141903 (faintly speaking)
+odN92kI-rX0-00465-00141998-00142167 - Yeah, okay.
+odN92kI-rX0-00466-00142167-00142368 - I mean, I agree that we need to move on.
+odN92kI-rX0-00467-00142478-00142604 (faintly speaking)
+odN92kI-rX0-00468-00142604-00142900 - Yeah we need to move on.
+odN92kI-rX0-00469-00143003-00143161 (faintly speaking)
+odN92kI-rX0-00470-00143161-00143306 - Let's get something, yeah.
+odN92kI-rX0-00471-00143418-00143710 (speaking faintly)
+odN92kI-rX0-00472-00143840-00144340 - Let's go back to this one.
+odN92kI-rX0-00473-00144540-00145040 - Oh sorry.
+odN92kI-rX0-00474-00145778-00146175 - Hello, microphone working, no.
+odN92kI-rX0-00475-00146175-00146296 - Now yes it is.
+odN92kI-rX0-00476-00146296-00146421 - Working now, okay good.
+odN92kI-rX0-00477-00146682-00146894 I'm Morten, I work for ARM,
+odN92kI-rX0-00478-00146894-00147323 and I have a couple of slightly more open-ended problems
+odN92kI-rX0-00479-00147323-00147563 that I haven't actually done any work on or anything.
+odN92kI-rX0-00480-00147563-00147837 But the first one we actually started
+odN92kI-rX0-00481-00147837-00148088 a little bit at OSPM summit in Pisa,
+odN92kI-rX0-00482-00148088-00148391 and Rafael and I discussed that it might be a good idea
+odN92kI-rX0-00483-00148391-00148609 to bring it up in this audience.
+odN92kI-rX0-00484-00148609-00149109 So the thought we've been having with regards
+odN92kI-rX0-00485-00149122-00149501 to thermal and, well most systems today,
+odN92kI-rX0-00486-00149612-00149820 when you run stuff on them, they get hot,
+odN92kI-rX0-00487-00149820-00150033 you can't sustain a high level of performance,
+odN92kI-rX0-00488-00150033-00150216 you need to scale things back.
+odN92kI-rX0-00489-00150216-00150521 And I think that applies both to mobile and to laptops
+odN92kI-rX0-00490-00150521-00150872 and well most systems today, you can't just assume
+odN92kI-rX0-00491-00150872-00151163 that a CPU is something that just runs flat out
+odN92kI-rX0-00492-00151163-00151358 and you can just put busy loops everywhere,
+odN92kI-rX0-00493-00151358-00151466 and it will just keep running
+odN92kI-rX0-00494-00151466-00151593 at a high level of performance.
+odN92kI-rX0-00495-00151593-00151808 So I think performance will be capped.
+odN92kI-rX0-00496-00152020-00152120 Is that a problem?
+odN92kI-rX0-00497-00152300-00152536 It depends on what you're doing I think.
+odN92kI-rX0-00498-00152792-00152964 We'll try this one.
+odN92kI-rX0-00499-00153135-00153312 So what you effectively have is
+odN92kI-rX0-00500-00153312-00153596 unpredictable compute bandwidth,
+odN92kI-rX0-00501-00153596-00153950 and that might be fine for some things but
+odN92kI-rX0-00502-00154143-00154574 user-space today has no idea that this capping is happening
+odN92kI-rX0-00503-00154574-00155005 and we have at least one thing in the kernel where we
+odN92kI-rX0-00504-00155005-00155401 imply a certain level of performance with what user-space
+odN92kI-rX0-00505-00155788-00155951 or we're having an interface where you can basically
+odN92kI-rX0-00506-00155951-00156126 request a certain performance level,
+odN92kI-rX0-00507-00156126-00156246 a notice SCHED_DEADLINE.
+odN92kI-rX0-00508-00156352-00156734 In SCHED_DEADLINE you can request a reservation
+odN92kI-rX0-00509-00156734-00157172 in terms of compute bandwidth, where you specify a period
+odN92kI-rX0-00510-00157172-00157585 and an amount of compute for each period, right.
+odN92kI-rX0-00511-00157585-00157875 And I think the implicit assumption at the moment is that
+odN92kI-rX0-00512-00157875-00158255 the, well the period this is not scaled
+odN92kI-rX0-00513-00158255-00158412 according to how fast you're running
+odN92kI-rX0-00514-00158412-00158685 but the amount of busy time is assumed
+odN92kI-rX0-00515-00158685-00158956 to be the busy time at the highest performance level.
+odN92kI-rX0-00516-00158956-00159105 Am I right, Yuri?
+odN92kI-rX0-00517-00159105-00159396 (speaking faintly)
+odN92kI-rX0-00518-00159793-00160005 - Yeah, you're right but then a run-time
+odN92kI-rX0-00519-00160005-00160187 it is actually scale considering.
+odN92kI-rX0-00520-00160187-00160456 - Yeah sure, so if you book a reservation
+odN92kI-rX0-00521-00160456-00160677 which gives you five milliseconds every ten
+odN92kI-rX0-00522-00160768-00160901 and you run at half the speed,
+odN92kI-rX0-00523-00160901-00161026 it basically means that you would be
+odN92kI-rX0-00524-00161026-00161374 running all the time, right, at that frequency.
+odN92kI-rX0-00525-00161374-00161457 - [Yuri] Yeah.
+odN92kI-rX0-00526-00161457-00161540 - Yeah.
+odN92kI-rX0-00527-00161652-00162042 So given that we have systems where
+odN92kI-rX0-00528-00162042-00162429 the budget you have to spend on CPUs
+odN92kI-rX0-00529-00162429-00162790 and running a certain OPP can change at any point in time.
+odN92kI-rX0-00530-00162882-00163080 Is it a good idea to have an interface
+odN92kI-rX0-00531-00163080-00163580 where you can sort of lure the user space into thinking,
+odN92kI-rX0-00532-00163587-00163839 "Oh I can make this reservation and it's all good,
+odN92kI-rX0-00533-00163839-00164179 and I'm going to get this performance level forever.
+odN92kI-rX0-00534-00164179-00164311 Is that a good idea?
+odN92kI-rX0-00535-00164311-00164726 Or should we start thinking about exposing
+odN92kI-rX0-00536-00164726-00165011 what level of performance user-space can actually expect?
+odN92kI-rX0-00537-00165180-00165348 I don't know if there are other examples
+odN92kI-rX0-00538-00165348-00165568 in the kernel where we make,
+odN92kI-rX0-00539-00165568-00165723 where there are these implicit,
+odN92kI-rX0-00540-00165897-00166159 or well, explicit requirements about how fast
+odN92kI-rX0-00541-00166159-00166476 what level of performance you'll be getting,
+odN92kI-rX0-00542-00166476-00166698 but at least for that line, we do have it.
+odN92kI-rX0-00543-00166698-00166893 I think we have it a little bit with util clamp
+odN92kI-rX0-00544-00166893-00167127 that Patrick just up streamed as well.
+odN92kI-rX0-00545-00167347-00167653 Your clamp level does have implications
+odN92kI-rX0-00546-00167653-00167841 for the OPP you are choosing
+odN92kI-rX0-00547-00167841-00168012 but it's not a hard guarantee.
+odN92kI-rX0-00548-00168012-00168272 I think it's like worse for SCHED_DEADLINE
+odN92kI-rX0-00549-00168272-00168398 where you might be thinking,
+odN92kI-rX0-00550-00168398-00168558 "Oh I can make this reservation,
+odN92kI-rX0-00551-00168558-00168823 "I'm always going to get this performance level."
+odN92kI-rX0-00552-00168964-00169200 So what do we do with SCHED_DEADLINE?
+odN92kI-rX0-00553-00169200-00169469 Is it fine that you make a reservation
+odN92kI-rX0-00554-00169469-00169695 and then if the system can't deliver it,
+odN92kI-rX0-00555-00169695-00169928 it put silent, it just ignore your reservation
+odN92kI-rX0-00556-00169928-00170149 and you don't get the performance level you need?
+odN92kI-rX0-00557-00170424-00170507 - So what's this notification,
+odN92kI-rX0-00558-00170507-00170835 whether you are constrained, right,
+odN92kI-rX0-00559-00170835-00170937 that is missing.
+odN92kI-rX0-00560-00170937-00171103 - [Morten] Yeah.
+odN92kI-rX0-00561-00171103-00171265 - So first of all, yeah, like you requested,
+odN92kI-rX0-00562-00171265-00171470 you may not get that performance,
+odN92kI-rX0-00563-00171470-00171603 so you need to notify, you need to know
+odN92kI-rX0-00564-00171603-00171738 that you are not getting it.
+odN92kI-rX0-00565-00171738-00171880 - [Morten] Exactly, and that's not there today.
+odN92kI-rX0-00566-00171880-00171963 - And that's not there.
+odN92kI-rX0-00567-00171963-00172209 We have in, you know, we expose it in,
+odN92kI-rX0-00568-00172343-00172589 we bring it to the kernel but we never use it.
+odN92kI-rX0-00569-00172589-00172750 At least in for X36.
+odN92kI-rX0-00570-00172877-00173151 - Yeah, and the way you use SCHED_DEADLINE today,
+odN92kI-rX0-00571-00173151-00173349 you make a reservation and then you assume
+odN92kI-rX0-00572-00173349-00173733 that it's there forever, or is that just a false assumption?
+odN92kI-rX0-00573-00173733-00173921 Should we just say that SCHED_DEADLINE,
+odN92kI-rX0-00574-00173921-00174177 you have a reservation until it can't be honored.
+odN92kI-rX0-00575-00174177-00174339 - Yeah once you get interrupter,
+odN92kI-rX0-00576-00174486-00174817 till you get constrained interrupt, you are done.
+odN92kI-rX0-00577-00174817-00175244 - What happens to SCHED_DEADLINE when we cap it,
+odN92kI-rX0-00578-00175244-00175379 for example due to thermos?
+odN92kI-rX0-00579-00175882-00176163 - Yeah currently we don't react to that.
+odN92kI-rX0-00580-00176163-00176632 My way of thinking about this is that if you know
+odN92kI-rX0-00581-00176632-00176826 that you can actually be capped,
+odN92kI-rX0-00582-00176826-00177076 you shouldn't meet stuff that you know
+odN92kI-rX0-00583-00177076-00177234 that you cannot guarantee, right.
+odN92kI-rX0-00584-00177234-00177533 So it's something that happens before.
+odN92kI-rX0-00585-00177533-00178008 Or if you know that the, if you can, I mean,
+odN92kI-rX0-00586-00178100-00178470 problem is that, as you say, once you emitted something
+odN92kI-rX0-00587-00178470-00178724 then there is this contract going on with user-space
+odN92kI-rX0-00588-00178724-00178956 and then, if your then your capped,
+odN92kI-rX0-00589-00178956-00179143 you're basically not respecting that.
+odN92kI-rX0-00590-00179143-00179362 So I guess what you can do if you want
+odN92kI-rX0-00591-00179362-00179727 to actually guarantee performance, is to admit less,
+odN92kI-rX0-00592-00179727-00180003 and with Dettol you can do it.
+odN92kI-rX0-00593-00180003-00180164 - That's what I'm getting at.
+odN92kI-rX0-00594-00180164-00180485 Do we want to have that?
+odN92kI-rX0-00595-00180485-00180713 Because at the moment it's not clear.
+odN92kI-rX0-00596-00180713-00180988 The SCHED_DEADLINE user can't be tricked
+odN92kI-rX0-00597-00180988-00181183 into thinking that it is actually a contract
+odN92kI-rX0-00598-00181183-00181401 with the kernel where it's not today.
+odN92kI-rX0-00599-00181401-00181501 So should we fix it?
+odN92kI-rX0-00600-00181642-00182012 - So on the one hand we need to reintroduce
+odN92kI-rX0-00601-00182012-00182378 the signal for for missing deadlines.
+odN92kI-rX0-00602-00182497-00182580 (faintly speaking)
+odN92kI-rX0-00603-00182580-00182663 Step back in?
+odN92kI-rX0-00604-00182663-00182825 Okay so that's at least one hint
+odN92kI-rX0-00605-00182825-00183066 that stuff has gone to bits.
+odN92kI-rX0-00606-00183563-00184063 We could add information to the signal that says why.
+odN92kI-rX0-00607-00184231-00184731 And then, I mean, for a variable system like that
+odN92kI-rX0-00608-00184983-00185447 you can of course a priori configure things
+odN92kI-rX0-00609-00185447-00185716 to not hand out the highest of the high.
+odN92kI-rX0-00610-00185716-00185937 I mean but of course if you put it,
+odN92kI-rX0-00611-00185937-00186293 if you put your phone in bright sunlight in your car,
+odN92kI-rX0-00612-00186293-00186662 it, yeah, at some point you have to just say, DM.
+odN92kI-rX0-00613-00187185-00187517 So when the thermals really kick in,
+odN92kI-rX0-00614-00187517-00187906 we should probably tell user-space about it.
+odN92kI-rX0-00615-00187995-00188275 (faintly speaking)
+odN92kI-rX0-00616-00188275-00188515 - Yeah I guess we can have probably both things.
+odN92kI-rX0-00617-00188617-00189005 So it also depends on how bad it is, if you are actually
+odN92kI-rX0-00618-00189005-00189361 breaking that this contract with some of your tasks.
+odN92kI-rX0-00619-00189361-00189742 So if you have a task that you cannot, I mean,
+odN92kI-rX0-00620-00189742-00190066 it has to run always, even when you're capped,
+odN92kI-rX0-00621-00190066-00190466 I think we for that you actually need to have,
+odN92kI-rX0-00622-00190466-00190854 I mean, the sustainable cap must be present
+odN92kI-rX0-00623-00190854-00191084 before you are meet that does to the system,
+odN92kI-rX0-00624-00191084-00191584 so that one it is in, you will always be guaranteed,
+odN92kI-rX0-00625-00191588-00191714 if that's possible.
+odN92kI-rX0-00626-00191714-00192002 And for the others that can actually maybe be stopped
+odN92kI-rX0-00627-00192002-00192368 or behave worse when the system is capped,
+odN92kI-rX0-00628-00192368-00192752 we can have a signal and then user-space can actually,
+odN92kI-rX0-00629-00192752-00193124 so for example if you have a video that is playing back,
+odN92kI-rX0-00630-00193124-00193371 and when you're capped you cannot try to reduce
+odN92kI-rX0-00631-00193371-00193827 the frame rate or anything but you can get the signal back.
+odN92kI-rX0-00632-00193916-00194217 - So are you proposing to have sort of two kinds of tasks,
+odN92kI-rX0-00633-00194217-00194537 ones that requires almost like a hard guarantee,
+odN92kI-rX0-00634-00194537-00194862 saying it will never ever take away your reservation,
+odN92kI-rX0-00635-00194862-00195113 and then you can allow other tasks to actually
+odN92kI-rX0-00636-00195113-00195467 oversubscribe the CPU, go above the sustained level
+odN92kI-rX0-00637-00195467-00195728 and then when it can't deliver that any more,
+odN92kI-rX0-00638-00195728-00195862 then just notified and say,
+odN92kI-rX0-00639-00195862-00196129 "Well your reservation is no longer on it."
+odN92kI-rX0-00640-00196129-00196604 - This just sounds like ooM and we're awfully at that.
+odN92kI-rX0-00641-00197015-00197371 I don't know, I'd start by just sending signals
+odN92kI-rX0-00642-00197371-00197753 when programs run and they completely fail,
+odN92kI-rX0-00643-00197899-00198271 mostly due to stuff changing and then
+odN92kI-rX0-00644-00198271-00198550 let user-space figure out what to do.
+odN92kI-rX0-00645-00198550-00198797 For Android that should probably be enough,
+odN92kI-rX0-00646-00198797-00199007 and then user-space can figure out which task
+odN92kI-rX0-00647-00199007-00199229 to knock on the head.
+odN92kI-rX0-00648-00199541-00199783 - [Morten] So we don't want to try to put in like a bar
+odN92kI-rX0-00649-00199783-00199982 saying, you should not make reservations
+odN92kI-rX0-00650-00199982-00200261 above, I don't know, 50%.
+odN92kI-rX0-00651-00200261-00200624 - We already have this bar and it's configurable.
+odN92kI-rX0-00652-00200624-00200798 - It's at 95% now, isn't it.
+odN92kI-rX0-00653-00200798-00200934 - But you can lower it.
+odN92kI-rX0-00654-00200934-00201044 - [Morten] True.
+odN92kI-rX0-00655-00201044-00201267 - And that's, it's the administrator.
+odN92kI-rX0-00656-00201267-00201569 I mean, we as kernel developers have no friggin clue.
+odN92kI-rX0-00657-00201770-00201936 It's the SOC.
+odN92kI-rX0-00658-00201936-00202104 - But is that entirely true?
+odN92kI-rX0-00659-00202104-00202352 You do have information in ACPI for example.
+odN92kI-rX0-00660-00202525-00203025 - Didn't I earlier, it was yesterday, right, yeah.
+odN92kI-rX0-00661-00203042-00203531 So if we have to rely on ACPI it's a sad, sad world.
+odN92kI-rX0-00662-00203531-00203703 (speaking faintly)
+odN92kI-rX0-00663-00203703-00204131 - So saying rely to ACPI is like saying rely on
+odN92kI-rX0-00664-00204131-00204411 English language, it's basically a communication means.
+odN92kI-rX0-00665-00204411-00204878 So what you are really saying is we can't rely
+odN92kI-rX0-00666-00204878-00205221 on firmware to provide actual information
+odN92kI-rX0-00667-00205221-00205488 which is very true, I mean, in practice.
+odN92kI-rX0-00668-00205586-00205880 (laughing)
+odN92kI-rX0-00669-00205880-00206010 - [Morten] Well we do rely on
+odN92kI-rX0-00670-00206010-00206225 some information coming from firmware.
+odN92kI-rX0-00671-00206225-00206398 - I mean if there is ACPI, we can of course
+odN92kI-rX0-00672-00206398-00206696 use it and set a number, it's just, I
+odN92kI-rX0-00673-00206784-00207040 let me say I am not full of faith that this number
+odN92kI-rX0-00674-00207040-00207404 will be as useful as it could be.
+odN92kI-rX0-00675-00207404-00207567 - But is it better or worse than nothing?
+odN92kI-rX0-00676-00207567-00207835 I mean the default is 95% which is above
+odN92kI-rX0-00677-00207835-00208335 your turbo arranged level on most of your parts at Intel.
+odN92kI-rX0-00678-00208473-00208777 And on arm I think that's probably true for many,
+odN92kI-rX0-00679-00208777-00208913 at least for the bigger cores that
+odN92kI-rX0-00680-00208913-00209129 you can't sustain a 95% frequency.
+odN92kI-rX0-00681-00209317-00209531 - So like once we had in thermal limits,
+odN92kI-rX0-00682-00209531-00209833 those guarantees don't hold, in none of
+odN92kI-rX0-00683-00209833-00210023 the guarantees hold, basically.
+odN92kI-rX0-00684-00210023-00210107 - That's how we do it.
+odN92kI-rX0-00685-00210107-00210394 - LFM, the the lowest frequency, even that's not guaranteed
+odN92kI-rX0-00686-00210394-00210586 because we can even have a duty cycle
+odN92kI-rX0-00687-00210586-00210796 eight steps inside that.
+odN92kI-rX0-00688-00210894-00211182 So 1800 megahertz divided by 8, 100 megahertz.
+odN92kI-rX0-00689-00211182-00211419 So 100 megahertz is the guarantee.
+odN92kI-rX0-00690-00211419-00211590 - So what you're is that SCHED_DEADLINE
+odN92kI-rX0-00691-00211590-00211796 and things like that is completely useless on a--
+odN92kI-rX0-00692-00211796-00211948 (speaking faintly)
+odN92kI-rX0-00693-00211948-00212106 - Yeah, it's very rare.
+odN92kI-rX0-00694-00212106-00212348 - It's not aware of any of those deadlines
+odN92kI-rX0-00695-00212348-00212746 and if you prioritize controlling the temperatures,
+odN92kI-rX0-00696-00212746-00213202 getting our way of the hot spots, as I was telling before.
+odN92kI-rX0-00697-00213202-00213421 That's how it's actually done today.
+odN92kI-rX0-00698-00213421-00213710 I mean if we can maybe have like from thermal,
+odN92kI-rX0-00699-00213710-00213871 from the thermal subsystem likes,
+odN92kI-rX0-00700-00213871-00214256 and is it even possible today for the system administrator
+odN92kI-rX0-00701-00214256-00214615 to actually define or the from AC guy to define
+odN92kI-rX0-00702-00214615-00214865 tree points where if the thermal subsystem
+odN92kI-rX0-00703-00214993-00215130 could notify the user-space, for example,
+odN92kI-rX0-00704-00215130-00215415 okay we are getting hot but we are not hammering it yet.
+odN92kI-rX0-00705-00215415-00215596 I mean yeah, that's possible you could do that
+odN92kI-rX0-00706-00215596-00215919 but you know, at the end of the day
+odN92kI-rX0-00707-00215919-00216284 when we get to the point where thermals are theme,
+odN92kI-rX0-00708-00216284-00216539 we need to reduce the performance.
+odN92kI-rX0-00709-00216539-00217039 - So yeah I'm all for changing the 95% in a sensible way
+odN92kI-rX0-00710-00217133-00217329 if we have sensible information.
+odN92kI-rX0-00711-00217329-00217433 That's not a problem.
+odN92kI-rX0-00712-00217433-00217684 - Yeah that's where I'm going, I mean, if we have
+odN92kI-rX0-00713-00217791-00217874 information from firmware, why don't we
+odN92kI-rX0-00714-00217874-00218144 use that to set a more sensible limit.
+odN92kI-rX0-00715-00218144-00218534 I totally agree that will be different for different devices
+odN92kI-rX0-00716-00218534-00218886 and each vendor might choose to have a different level
+odN92kI-rX0-00717-00218886-00219192 depending on what level of guarantees they need.
+odN92kI-rX0-00718-00219192-00219439 - What do we do in extreme thermal events,
+odN92kI-rX0-00719-00219439-00219606 that's another story.
+odN92kI-rX0-00720-00219606-00219703 - [Morten] I mean if you're at the point
+odN92kI-rX0-00721-00219703-00219816 where the only thing you can do
+odN92kI-rX0-00722-00219816-00220018 is to shut down, then you're screwed anyway.
+odN92kI-rX0-00723-00220018-00220101 - Right.
+odN92kI-rX0-00724-00220101-00220350 - Then everybody's screwed up, right, including
+odN92kI-rX0-00725-00220350-00220650 system integrators because they didn't foresee that working.
+odN92kI-rX0-00726-00220941-00221037 - True.
+odN92kI-rX0-00727-00221037-00221276 There's a question here or a comment.
+odN92kI-rX0-00728-00221685-00221923 - So I was wondering, is your assumption here
+odN92kI-rX0-00729-00221923-00222178 that you've gotten multiple tasks running SCHED_DEADLINE
+odN92kI-rX0-00730-00222178-00222509 or are you a single task scenario because you could,
+odN92kI-rX0-00731-00222509-00222716 if you're hitting that limit you're gonna get
+odN92kI-rX0-00732-00222716-00222984 like three signals if there's a signal there,
+odN92kI-rX0-00733-00222984-00223146 and you're just gonna get confused
+odN92kI-rX0-00734-00223146-00223242 with all that information.
+odN92kI-rX0-00735-00223242-00223398 I mean I don't know how that's necessarily
+odN92kI-rX0-00736-00223398-00223581 going to help you other than, you know,
+odN92kI-rX0-00737-00223581-00223844 the fire alarm that's going off, right, you know.
+odN92kI-rX0-00738-00223973-00224269 So some of it's more about the system's architecture too,
+odN92kI-rX0-00739-00224269-00224529 understanding your limits and understanding
+odN92kI-rX0-00740-00224529-00224757 what the predicted behavior of some of these workloads are
+odN92kI-rX0-00741-00224757-00225126 and how to distribute the workload across the course.
+odN92kI-rX0-00742-00225126-00225509 But I do agree, if we had meaningful information
+odN92kI-rX0-00743-00225509-00225682 than people trying to schedule these workloads
+odN92kI-rX0-00744-00225682-00225880 and meet some deadlines, you know,
+odN92kI-rX0-00745-00225880-00226159 maybe we could make better choices, right.
+odN92kI-rX0-00746-00226159-00226480 But I think that as you start to add more to the workload
+odN92kI-rX0-00747-00226480-00226764 and consolidate your workload on your multi-core systems,
+odN92kI-rX0-00748-00226764-00226973 I think then it's just noise at that point.
+odN92kI-rX0-00749-00226973-00227212 You've got all these fires going off because nobody's
+odN92kI-rX0-00750-00227212-00227523 hitting their deadlines on that core and then you've got it.
+odN92kI-rX0-00751-00227523-00227757 That gets back to just understanding,
+odN92kI-rX0-00752-00227757-00227969 what are my priority tasks, right.
+odN92kI-rX0-00753-00227969-00228234 And then, and in architecting maybe setting affinity
+odN92kI-rX0-00754-00228234-00228383 for specific things, you know.
+odN92kI-rX0-00755-00228383-00228566 I don't know what your use case is
+odN92kI-rX0-00756-00228566-00228786 and sometimes the use case will define this.
+odN92kI-rX0-00757-00228971-00229168 - But for us, it's getting the information
+odN92kI-rX0-00758-00229168-00229528 out to user-space.
+odN92kI-rX0-00759-00229528-00229985 - My worry is that if you want to architect everything now
+odN92kI-rX0-00760-00229985-00230196 you need to know how are the platform behaves.
+odN92kI-rX0-00761-00230196-00230459 I mean, if an end user picks up a dashboard or a device
+odN92kI-rX0-00762-00230459-00230701 and want your SCHED_DEADLINE on that,
+odN92kI-rX0-00763-00230701-00230834 then we're screwed, right.
+odN92kI-rX0-00764-00230834-00230987 - We'd like to have meaningful data
+odN92kI-rX0-00765-00230987-00231157 that we could provide the user-space.
+odN92kI-rX0-00766-00231157-00231240 - [Morten] Yeah.
+odN92kI-rX0-00767-00231240-00231323 - I agree.
+odN92kI-rX0-00768-00231494-00231674 - Yeah, I mean, unless you are foreseeing to say
+odN92kI-rX0-00769-00231674-00232154 play on the gray area where only one core is having
+odN92kI-rX0-00770-00232154-00232438 like a thermal event like a real fire alarm,
+odN92kI-rX0-00771-00232525-00232735 and then you want to move your workload out of it
+odN92kI-rX0-00772-00232735-00232992 so that it can cool off, but I mean,
+odN92kI-rX0-00773-00232992-00233288 it starts to be a little slightly bit complicated
+odN92kI-rX0-00774-00233288-00233398 but I don't know.
+odN92kI-rX0-00775-00233398-00233692 - Yeah that's probably not what I had in mind.
+odN92kI-rX0-00776-00233692-00233955 I was more after like providing
+odN92kI-rX0-00777-00233955-00234455 like some level of guarantee where if you use SCHED_DEADLINE
+odN92kI-rX0-00778-00234469-00234769 then you would be stopped before you can actually
+odN92kI-rX0-00779-00234769-00235092 make a reservation, that you would be almost sure
+odN92kI-rX0-00780-00235092-00235221 it would, wouldn't be met.
+odN92kI-rX0-00781-00235350-00235463 - Well the the thing is is that,
+odN92kI-rX0-00782-00235463-00235561 let's say you had some information
+odN92kI-rX0-00783-00235561-00235748 and you've got forecasts running on that core
+odN92kI-rX0-00784-00235748-00235861 and you thought you had everything
+odN92kI-rX0-00785-00235861-00236148 precisely managed as far as your deadlines.
+odN92kI-rX0-00786-00236148-00236373 But all of a sudden the fire alarms go off, right,
+odN92kI-rX0-00787-00236373-00236619 and so you do have that information to tweak it
+odN92kI-rX0-00788-00236619-00236892 from 95% to something a little more,
+odN92kI-rX0-00789-00236892-00237122 and fire alarms are still going off,
+odN92kI-rX0-00790-00237122-00237280 then do you just get the bucket
+odN92kI-rX0-00791-00237280-00237414 and then throw one of the tasks off
+odN92kI-rX0-00792-00237414-00237519 and move it to another core?
+odN92kI-rX0-00793-00237519-00237669 You know, yeah, you need more.
+odN92kI-rX0-00794-00237966-00238164 - The thermal is not the core, I think.
+odN92kI-rX0-00795-00238164-00238664 - Yeah well it depends on your time frame,
+odN92kI-rX0-00796-00238665-00238889 but I don't know if it's actually feasible
+odN92kI-rX0-00797-00238889-00239130 to try and to shift, I mean, the hot spot
+odN92kI-rX0-00798-00239130-00239304 from one core to the next or the whole thing
+odN92kI-rX0-00799-00239304-00239621 heats up so fast that it's just a global problem anyway.
+odN92kI-rX0-00800-00239621-00239898 - Yeah, if you have some idol CPUs, I don't know.
+odN92kI-rX0-00801-00240398-00240583 - Yeah so with deadline, you of course
+odN92kI-rX0-00802-00240583-00240931 have the period of the task and mostly
+odN92kI-rX0-00803-00240931-00241128 this would be like the 60 hertz
+odN92kI-rX0-00804-00241128-00241476 of your rendering pipeline or whatever is convenient
+odN92kI-rX0-00805-00241476-00241771 for your audio pipeline or something like that.
+odN92kI-rX0-00806-00241771-00242271 It's the core thermal versus packaged thermal
+odN92kI-rX0-00807-00242404-00242616 anywhere near that range of time?
+odN92kI-rX0-00808-00242719-00242953 I mean does it make sense to shift
+odN92kI-rX0-00809-00242953-00243146 to another core in those time frames?
+odN92kI-rX0-00810-00243444-00243847 - Yeah, it can happen fast.
+odN92kI-rX0-00811-00243847-00244156 - It can be microseconds.
+odN92kI-rX0-00812-00244156-00244592 - But you can't mic radio task that fast.
+odN92kI-rX0-00813-00244592-00244797 So you can't do anything about it anyway.
+odN92kI-rX0-00814-00244797-00245130 - And if they're in a BM, then, you know.
+odN92kI-rX0-00815-00245130-00245213 - Yeah.
+odN92kI-rX0-00816-00245652-00246008 - So if we plan to use the knobs that are available
+odN92kI-rX0-00817-00246008-00246423 to define the reservation we can use for deadline
+odN92kI-rX0-00818-00246423-00246606 with that note that we already have,
+odN92kI-rX0-00819-00246606-00247054 is not that knob affecting both deadline and RT tasks.
+odN92kI-rX0-00820-00247054-00247377 So does that mean that if we are under thermal pressure
+odN92kI-rX0-00821-00247377-00247639 we have only deadline tasks, we basically keel out
+odN92kI-rX0-00822-00247639-00248019 also the RT tasks in our system, so.
+odN92kI-rX0-00823-00248402-00248821 - So RT doesn't have admission control.
+odN92kI-rX0-00824-00248821-00248905 - No, exactly.
+odN92kI-rX0-00825-00248905-00249271 - RT is basically basically a failed scheduling policy.
+odN92kI-rX0-00826-00249582-00249858 There is nothing you can really do for them.
+odN92kI-rX0-00827-00249858-00249942 - So when it says that--
+odN92kI-rX0-00828-00249942-00250217 - Basically, so you only have then deadline,
+odN92kI-rX0-00829-00250217-00250567 all the RT will won't be scheduled because it's a total
+odN92kI-rX0-00830-00250567-00250737 when deadline basically finishes in doing its work.
+odN92kI-rX0-00831-00250737-00251025 So they are emitted they will never run.
+odN92kI-rX0-00832-00251188-00251363 - If it's worth at least to try to distinguish
+odN92kI-rX0-00833-00251363-00251863 what we can allocate with deadline and what is left for RT
+odN92kI-rX0-00834-00251887-00252212 still within the sustainable performance capacity.
+odN92kI-rX0-00835-00252494-00252719 Which goes back to what Morten was proposing
+odN92kI-rX0-00836-00252719-00253130 so we have to define a deadline granted capacity
+odN92kI-rX0-00837-00253130-00253423 that leave some space for RT for example
+odN92kI-rX0-00838-00253423-00253656 and other stuff under thermal conditions also.
+odN92kI-rX0-00839-00254084-00254375 - So this comes back to U clamp of course
+odN92kI-rX0-00840-00254375-00254831 because that's the only thing that RT has to.
+odN92kI-rX0-00841-00254925-00255258 - Yeah but U clamp is not really enforcing anything.
+odN92kI-rX0-00842-00255258-00255538 It has to be an extension on that eventually.
+odN92kI-rX0-00843-00255538-00255967 - So I'd hate to add bandwidth to RT, it's just.
+odN92kI-rX0-00844-00256143-00256494 - Well okay, so there are different categories
+odN92kI-rX0-00845-00256494-00256877 of thermal events, different time frames, and there are such
+odN92kI-rX0-00846-00256877-00257187 that you can't do anything about it because they,
+odN92kI-rX0-00847-00257187-00257501 it will just happen in a matter of microseconds
+odN92kI-rX0-00848-00257501-00257757 and then the hardware will decided.
+odN92kI-rX0-00849-00257757-00258078 You will get less power and that's it.
+odN92kI-rX0-00850-00258185-00258567 And you can't do anything except for getting,
+odN92kI-rX0-00851-00258567-00258791 possibly getting notified about that.
+odN92kI-rX0-00852-00258791-00259103 - Yeah but for those, the workload
+odN92kI-rX0-00853-00259103-00259329 should already take that into account.
+odN92kI-rX0-00854-00259329-00259673 Those will always happen and then, I mean,
+odN92kI-rX0-00855-00259673-00260151 if you have a work that is like four milliseconds long
+odN92kI-rX0-00856-00260315-00260517 and the event happens on microsecond scale,
+odN92kI-rX0-00857-00260604-00260901 it will always happen and therefore the workload
+odN92kI-rX0-00858-00260901-00261192 should take this into account anyway,
+odN92kI-rX0-00859-00261288-00261617 because it's just nothing we can do about it.
+odN92kI-rX0-00860-00261617-00262117 - Well so the work, there is a difference between a workload
+odN92kI-rX0-00861-00262129-00262629 and a, how the kernel allows to happen
+odN92kI-rX0-00862-00262886-00263052 because the workload is just, you know,
+odN92kI-rX0-00863-00263052-00263276 a bunch of tasks that will run,
+odN92kI-rX0-00864-00263276-00263536 and a kernel decides how fast they will run,
+odN92kI-rX0-00865-00263536-00263643 right, and that is--
+odN92kI-rX0-00866-00263643-00263748 - Sure, sure, but so,
+odN92kI-rX0-00867-00263988-00264142 this is for admission control
+odN92kI-rX0-00868-00264459-00264755 and we should allow so much work that it is possible
+odN92kI-rX0-00869-00265193-00265693 and suppose we do a while 1 loop
+odN92kI-rX0-00870-00265920-00266413 and load the CPU up and then it throws us down
+odN92kI-rX0-00871-00266413-00266747 and then we see we cannot ever load this machine up
+odN92kI-rX0-00872-00266747-00267247 more than now 60%, maybe 40% depending on the platform.
+odN92kI-rX0-00873-00267494-00267759 And that's where we should set our admission control.
+odN92kI-rX0-00874-00267891-00268314 That means that admission control should be
+odN92kI-rX0-00875-00268314-00268660 like tuned to the, to what's going on, right.
+odN92kI-rX0-00876-00268660-00268952 (faintly speaking)
+odN92kI-rX0-00877-00269215-00269462 - So then emission control control will disallow new tasks
+odN92kI-rX0-00878-00269462-00269799 and then the existing tasks, if you put it in the sun
+odN92kI-rX0-00879-00269799-00270040 or if something bad happens, your fan breaks
+odN92kI-rX0-00880-00270040-00270318 or thermal glue dried up, I don't know.
+odN92kI-rX0-00881-00270580-00270719 Weird stuff happens, I know.
+odN92kI-rX0-00882-00270719-00270953 - Yeah, I mean, I think just maybe circle back
+odN92kI-rX0-00883-00270953-00271128 on the same thing that you are saying,
+odN92kI-rX0-00884-00271220-00271445 two aspects here, right, if you are designing
+odN92kI-rX0-00885-00271533-00272017 your admission control assuming your maximum CPU frequency
+odN92kI-rX0-00886-00272017-00272456 or your turbo, you already started doing it wrong, right.
+odN92kI-rX0-00887-00272456-00272579 So there is that.
+odN92kI-rX0-00888-00272579-00272863 And the other thing is also that I think Peter mentioned
+odN92kI-rX0-00889-00272863-00273266 is like or maybe say in a different way is that
+odN92kI-rX0-00890-00273266-00273464 it's not only about thermals at least,
+odN92kI-rX0-00891-00273464-00273725 it's not only about the admission control
+odN92kI-rX0-00892-00273725-00273972 of a SCHED_DEADLINE doesn't know about that, right.
+odN92kI-rX0-00893-00273972-00274181 So if you have you as your high priority tasks
+odN92kI-rX0-00894-00274181-00274586 which had the endmies admitted on on a SCHED_DEADLINE,
+odN92kI-rX0-00895-00274586-00274846 and maybe you can even admit more,
+odN92kI-rX0-00896-00274846-00275145 but if you have like a noisy workload
+odN92kI-rX0-00897-00275145-00275645 which is consuming power like a while 1 over there,
+odN92kI-rX0-00898-00275663-00275870 it's also going to contribute to the thermals.
+odN92kI-rX0-00899-00275870-00276126 - It will.
+odN92kI-rX0-00900-00276126-00276468 - Right, it has nothing to do with the admission control
+odN92kI-rX0-00901-00276468-00276674 maybe, right, is not part of SCHED_DEADLINE,
+odN92kI-rX0-00902-00276674-00277013 it's just like a CFS task, but if you produce power
+odN92kI-rX0-00903-00277013-00277137 if you heat the--
+odN92kI-rX0-00904-00277137-00277487 - Yeah but its contribution will eventually be reduced to 5%
+odN92kI-rX0-00905-00277487-00277720 because if you start capping because of thermal,
+odN92kI-rX0-00906-00277720-00277903 then in the end, it's only your deadline task
+odN92kI-rX0-00907-00277903-00278111 that will end up running minus the 5%
+odN92kI-rX0-00908-00278111-00278271 that will reserve further stuff.
+odN92kI-rX0-00909-00278271-00278483 - And so the history of the 95,
+odN92kI-rX0-00910-00278483-00278964 the 95 was set way before we ever did any DVFS smarts.
+odN92kI-rX0-00911-00279260-00279651 The moment we started scaling work for SCHED_DEADLINE,
+odN92kI-rX0-00912-00280009-00280396 for DVFS, the whole 95% thing when went out the window,
+odN92kI-rX0-00913-00280396-00280726 and this is basically an overlooked thing.
+odN92kI-rX0-00914-00280726-00281125 So I'm absolutely okay with changing the 95
+odN92kI-rX0-00915-00281125-00281459 to a sensible number and even doing it
+odN92kI-rX0-00916-00281459-00281817 dynamically if we have the information.
+odN92kI-rX0-00917-00282065-00282370 - [Morten] Okay, fair enough.
+odN92kI-rX0-00918-00282370-00282571 - But maybe one point is also that
+odN92kI-rX0-00919-00282571-00282767 yes we can do ismar stuff here
+odN92kI-rX0-00920-00282767-00283020 but in the end of the day, is also part of the job
+odN92kI-rX0-00921-00283020-00283274 to design the system that fits
+odN92kI-rX0-00922-00283274-00283574 on your thermals envelope, right.
+odN92kI-rX0-00923-00283574-00283688 - Yeah.
+odN92kI-rX0-00924-00283688-00283828 - So we have one more minute for this topic,
+odN92kI-rX0-00925-00283828-00284063 so last thoughts and then we need to.
+odN92kI-rX0-00926-00284607-00284734 - I think I agree with your point
+odN92kI-rX0-00927-00284734-00284983 but I still think we can do slightly more
+odN92kI-rX0-00928-00284983-00285194 than just say it's up to the user
+odN92kI-rX0-00929-00285194-00285373 if we do have a little bit of information
+odN92kI-rX0-00930-00285373-00285720 to not let people put in deadlines
+odN92kI-rX0-00931-00285720-00286019 after use in 90% of the CPU capacity.
+odN92kI-rX0-00932-00286248-00286492 - So what kind of exact information are you looking for,
+odN92kI-rX0-00933-00286492-00286740 like from the thermal subsystem for example
+odN92kI-rX0-00934-00286740-00287148 or some notification through that, or?
+odN92kI-rX0-00935-00287148-00287648 - I think, well you can subscribe to notifications, can you,
+odN92kI-rX0-00936-00287681-00288167 I mean if you want to know when the system gets hot.
+odN92kI-rX0-00937-00288167-00288357 - Right, you already can do that.
+odN92kI-rX0-00938-00288357-00288440 - Yeah.
+odN92kI-rX0-00939-00288440-00288523 - From user-space.
+odN92kI-rX0-00940-00288523-00288606 - Yeah.
+odN92kI-rX0-00941-00288725-00289138 So yeah, maybe just lower the 95% to something derived
+odN92kI-rX0-00942-00289138-00289470 from whatever information will have the firmware table.
+odN92kI-rX0-00943-00289470-00289684 - Basically the sustained rate.
+odN92kI-rX0-00944-00289684-00289767 - Yeah.
+odN92kI-rX0-00945-00289767-00289979 - I think there's an ACPI thing or whatever we have,
+odN92kI-rX0-00946-00289979-00290245 but a sustained rate and set it to that.
+odN92kI-rX0-00947-00290245-00290390 - [Morten] And we can probably
+odN92kI-rX0-00948-00290390-00290473 find something similar for DT.
+odN92kI-rX0-00949-00290473-00290674 - But if you lower this dynamically then what do you do
+odN92kI-rX0-00950-00290674-00290925 with all the tasks which are already subscribed to.
+odN92kI-rX0-00951-00290925-00291217 - [Morten] It's not dynamic, it's a static table.
+odN92kI-rX0-00952-00291217-00291366 - They get a signal.
+odN92kI-rX0-00953-00291366-00291591 (laughing)
+odN92kI-rX0-00954-00291711-00292034 (faintly speaking)
+odN92kI-rX0-00955-00292034-00292264 - Or maybe you can also start on server 3.
+odN92kI-rX0-00956-00292264-00292583 So you first start with a really low cap
+odN92kI-rX0-00957-00292583-00292817 and see if you actually need to increase it
+odN92kI-rX0-00958-00292817-00293168 if you actually have more tasks that tries
+odN92kI-rX0-00959-00293168-00293519 to enter a system, I guess can monitor the thing.
+odN92kI-rX0-00960-00293519-00293729 - Right, okay, good.
+odN92kI-rX0-00961-00293953-00294086 I think that's it for this topic.
+odN92kI-rX0-00962-00294086-00294586 So I have one more topic which is sort of related.
+odN92kI-rX0-00963-00294619-00294843 Let me see if I can work this out again.
+odN92kI-rX0-00964-00294843-00295003 Oh no, I messed it up, didn't I.
+odN92kI-rX0-00965-00295729-00295997 I think we're here and can I realign it somehow.
+odN92kI-rX0-00966-00296601-00296773 Especially when they're not your own.
+odN92kI-rX0-00967-00296930-00297236 Right, so this is a little bit related
+odN92kI-rX0-00968-00297236-00297395 and it actually came out of the discussion
+odN92kI-rX0-00969-00297395-00297684 again from OSPM Summit in Pisa.
+odN92kI-rX0-00970-00297684-00298043 So the way we do with thermal management today in Linux
+odN92kI-rX0-00971-00298043-00298447 is very much focused on what devices do we have
+odN92kI-rX0-00972-00298447-00298596 and how can we control them.
+odN92kI-rX0-00973-00298596-00298805 And if we're in a thermal situation,
+odN92kI-rX0-00974-00298805-00299096 we decide a set of caps and we apply them to each device
+odN92kI-rX0-00975-00299096-00299306 and say you can't go any faster than this.
+odN92kI-rX0-00976-00299921-00300038 But when you have a thermal situation,
+odN92kI-rX0-00977-00300038-00300167 what you really need to do is to,
+odN92kI-rX0-00978-00300167-00300405 you need to lower the amount of work you need to do,
+odN92kI-rX0-00979-00300405-00300697 lower the amount of compute, you need to reduce it somehow.
+odN92kI-rX0-00980-00300697-00300978 And one way is just to slow everything down.
+odN92kI-rX0-00981-00300978-00301332 But what if the set of caps that our current
+odN92kI-rX0-00982-00301332-00301597 device centric thermal management governor
+odN92kI-rX0-00983-00301752-00302032 decides is not the most optimum one for your use case.
+odN92kI-rX0-00984-00302148-00302370 You could have a use case where
+odN92kI-rX0-00985-00302370-00302617 there is a task you might want to run really fast
+odN92kI-rX0-00986-00302617-00302977 because it's absolutely essential for the performance
+odN92kI-rX0-00987-00302977-00303222 and there are some other tasks you can sacrifice,
+odN92kI-rX0-00988-00303513-00303669 that you can't really handle today
+odN92kI-rX0-00989-00303669-00303869 in the way power management works.
+odN92kI-rX0-00990-00303955-00304455 So should we consider changing the way we look
+odN92kI-rX0-00991-00304507-00305007 at a thermal management, and make it task centric instead
+odN92kI-rX0-00992-00305152-00305652 because as we have today, we don't have any any information
+odN92kI-rX0-00993-00305792-00306035 about the importance of tasks,
+odN92kI-rX0-00994-00306035-00306362 we don't know how to best spent the thermal budget
+odN92kI-rX0-00995-00306362-00306610 and if we try to do clever things in the scheduler
+odN92kI-rX0-00996-00306610-00306847 we might actually make things worse.
+odN92kI-rX0-00997-00306847-00307131 I mean we can have task placement trying to escape
+odN92kI-rX0-00998-00307131-00307417 thermal caps, if you cap one set of CPUs,
+odN92kI-rX0-00999-00307417-00307655 the scheduler says, oh these CPUs are not capped as hard,
+odN92kI-rX0-01000-00307655-00307812 then you move more load over there
+odN92kI-rX0-01001-00307812-00308037 and then you have to cap them a moment later.
+odN92kI-rX0-01002-00308145-00308386 So it doesn't really work that well
+odN92kI-rX0-01003-00308386-00308526 I think, for some scenarios.
+odN92kI-rX0-01004-00308752-00309196 So in an ideal world, if you have a thermal situation,
+odN92kI-rX0-01005-00309196-00309425 you want to lower the amount of compute
+odN92kI-rX0-01006-00309425-00309879 but I think the application is actually better place
+odN92kI-rX0-01007-00309879-00310160 at deciding how to lower the amount of compute
+odN92kI-rX0-01008-00310160-00310387 rather than kernel firmware stepping units,
+odN92kI-rX0-01009-00310387-00310623 say you can't go any faster than this,
+odN92kI-rX0-01010-00310623-00310797 all your CPUs are capped at this level,
+odN92kI-rX0-01011-00310797-00310967 your GPU is capped at that level,
+odN92kI-rX0-01012-00310967-00311243 and you just have to live inside those constraints.
+odN92kI-rX0-01013-00311342-00311558 What if you had a workload, the new,
+odN92kI-rX0-01014-00311558-00311813 I actually want the few CPUs that run really fast
+odN92kI-rX0-01015-00311813-00311981 but I'm okay with the other ones
+odN92kI-rX0-01016-00311981-00312268 being being capped at very low OPP
+odN92kI-rX0-01017-00312268-00312593 or I can sacrifice my GPU completely.
+odN92kI-rX0-01018-00312593-00312809 There is no room to do that with the way
+odN92kI-rX0-01019-00312809-00312974 we currently do power management.
+odN92kI-rX0-01020-00313093-00313432 - Okay, thank you, so one thought about this.
+odN92kI-rX0-01021-00313432-00313850 So we still need to cut the kernel to be the last boundary,
+odN92kI-rX0-01022-00313977-00314248 like for you know, because we can we can be nice
+odN92kI-rX0-01023-00314248-00314738 to applications but then when they still don't, you know.
+odN92kI-rX0-01024-00314738-00314871 - Exactly.
+odN92kI-rX0-01025-00314871-00315086 - So this is not about just shifting
+odN92kI-rX0-01026-00315086-00315274 a whole problem up into the application
+odN92kI-rX0-01027-00315274-00315504 because we can't trust them in the end.
+odN92kI-rX0-01028-00315504-00315689 This is about enabling the applications
+odN92kI-rX0-01029-00315689-00315950 to do better if they behave.
+odN92kI-rX0-01030-00315950-00316116 And I don't think the kernel is going to
+odN92kI-rX0-01031-00316116-00316280 be the last line of defense anyway,
+odN92kI-rX0-01032-00316280-00316395 you have something in firmware.
+odN92kI-rX0-01033-00316395-00316571 So you sort of have a number of layers
+odN92kI-rX0-01034-00316571-00316798 starting from the bottom where you have firmware,
+odN92kI-rX0-01035-00316798-00317008 then you have kernel, you might have middleware
+odN92kI-rX0-01036-00317008-00317250 having an opinion in there, and then
+odN92kI-rX0-01037-00317250-00317571 at the very top you have the application.
+odN92kI-rX0-01038-00318302-00318486 - But these applications don't have deadlines
+odN92kI-rX0-01039-00318486-00318671 because they're kind of contradicting things here
+odN92kI-rX0-01040-00318671-00318867 is because if they have a deadline and you want
+odN92kI-rX0-01041-00318867-00319082 some sort of user space notification that
+odN92kI-rX0-01042-00319082-00319247 hey I should slow things down,
+odN92kI-rX0-01043-00319338-00319593 if the applications aware that it's deadline centered,
+odN92kI-rX0-01044-00319593-00319738 I guess it could reject that.
+odN92kI-rX0-01045-00319997-00320179 - Well the application can choose to ignore
+odN92kI-rX0-01046-00320179-00320429 whatever we tell it, and if that happens
+odN92kI-rX0-01047-00320429-00320616 and at some point you have you will,
+odN92kI-rX0-01048-00320616-00320942 it oversteps some boundaries, the kernel will step in
+odN92kI-rX0-01049-00320942-00321083 or the firmware will step in and say,
+odN92kI-rX0-01050-00321083-00321378 well you still exceed the power budget now and need to--
+odN92kI-rX0-01051-00321378-00321490 - Yeah, now I need to clamp.
+odN92kI-rX0-01052-00321490-00321867 - Boundaries, but this is around trying to enable
+odN92kI-rX0-01053-00321867-00322274 the application to stay away from those boundaries
+odN92kI-rX0-01054-00322274-00322454 so they actually don't get enforced.
+odN92kI-rX0-01055-00322551-00322931 So letting the application itself manage
+odN92kI-rX0-01056-00322931-00323151 how much compute it's requiring.
+odN92kI-rX0-01057-00323236-00323534 There are a few examples out there.
+odN92kI-rX0-01058-00323534-00323824 I came across a blog post where some graphics people
+odN92kI-rX0-01059-00323824-00324211 had worked with a specific phone vendor to try to
+odN92kI-rX0-01060-00324452-00324543 basically self-adapt the amount of detail
+odN92kI-rX0-01061-00324543-00324788 you were rendering in the graphics engine
+odN92kI-rX0-01062-00324788-00324985 depending on the thermal pressure.
+odN92kI-rX0-01063-00324985-00325203 So if you're overheating, they basically adapted
+odN92kI-rX0-01064-00325203-00325495 the workload to become less by reducing the complexity
+odN92kI-rX0-01065-00325495-00325656 of what you are rendering, and thereby you could
+odN92kI-rX0-01066-00325656-00325823 stay within the thermal budget, but you still
+odN92kI-rX0-01067-00325823-00326235 had a nice frame rate because the graphics really sucks
+odN92kI-rX0-01068-00326235-00326391 if you cap your system so hard
+odN92kI-rX0-01069-00326391-00326668 that you can't meet 30 or 60 FPS.
+odN92kI-rX0-01070-00326668-00326998 You might be better off only being able to see
+odN92kI-rX0-01071-00326998-00327239 half a mile ahead instead of one mile
+odN92kI-rX0-01072-00327385-00327639 in your render distance or whatever,
+odN92kI-rX0-01073-00327639-00327769 I'm not a graphics expert.
+odN92kI-rX0-01074-00327943-00328405 For applications like that, where you might be interested in
+odN92kI-rX0-01075-00328405-00328806 in investing into making them self-adapting.
+odN92kI-rX0-01076-00328806-00329106 Could we come up with a way to enable them to do that?
+odN92kI-rX0-01077-00329106-00329319 I know that in Android, I think they recently introduced
+odN92kI-rX0-01078-00329319-00329631 an interface where you can get notifications,
+odN92kI-rX0-01079-00329631-00329909 I think they have four different notifications going
+odN92kI-rX0-01080-00329909-00330368 from mildly critical up to severe for thermal.
+odN92kI-rX0-01081-00330487-00330646 But as far as I know, it doesn't use,
+odN92kI-rX0-01082-00330646-00330932 we don't have a standardized kernel interface
+odN92kI-rX0-01083-00330932-00331170 for user-space to subscribe to or read,
+odN92kI-rX0-01084-00331170-00331477 saying what is the current thermal situation.
+odN92kI-rX0-01085-00331477-00331657 And what should that interface actually be?
+odN92kI-rX0-01086-00331657-00331986 Can we come up with a metric that they can use to say
+odN92kI-rX0-01087-00331986-00332249 how far away am i from from hitting that boundary
+odN92kI-rX0-01088-00332249-00332616 where firmware will step in and hard cap the frequency?
+odN92kI-rX0-01089-00332616-00332895 - Right, so just like from a graphic side of things,
+odN92kI-rX0-01090-00332895-00333208 like you're saying, if I'm starting to see some,
+odN92kI-rX0-01091-00333208-00333696 I'm dropping frames, then I start to make some adjustments,
+odN92kI-rX0-01092-00333696-00333881 and then you had that feedback loop.
+odN92kI-rX0-01093-00333881-00334045 And I think what you're looking for is that,
+odN92kI-rX0-01094-00334045-00334302 also that feedback, it's not a one-time thing,
+odN92kI-rX0-01095-00334302-00334605 it's sort of like the application makes some adjustments,
+odN92kI-rX0-01096-00334605-00334761 gets additional feedback somehow.
+odN92kI-rX0-01097-00334761-00334875 - Right, yes.
+odN92kI-rX0-01098-00334875-00335242 And I want to enable the application to do the adjustments
+odN92kI-rX0-01099-00335242-00335532 before you have kernel firmware kicking in
+odN92kI-rX0-01100-00335532-00335802 and making hard caps, so you actually drop the frames.
+odN92kI-rX0-01101-00336200-00336380 Well you could of course see that you exceed
+odN92kI-rX0-01102-00336380-00336698 the thermal budget if you start to see frames dropping
+odN92kI-rX0-01103-00336698-00336880 and then you can start thinking about doing something.
+odN92kI-rX0-01104-00336880-00337157 But if we told it beforehand and now you're actually
+odN92kI-rX0-01105-00337157-00337639 approaching whatever limit you have in terms of thermal
+odN92kI-rX0-01106-00337639-00337904 and then do something about it proactively,
+odN92kI-rX0-01107-00337904-00338097 I think you can end up with a better user experience.
+odN92kI-rX0-01108-00338097-00338547 - So just to, maybe I mentioned that on the other topic
+odN92kI-rX0-01109-00338547-00338915 is the way that assist integrator would do it,
+odN92kI-rX0-01110-00338915-00339177 you would like create bands on the temperature domains,
+odN92kI-rX0-01111-00339177-00339433 right, and have tree points defined
+odN92kI-rX0-01112-00339433-00339746 on this temperature domain.
+odN92kI-rX0-01113-00339746-00340010 It's very common for people to say,
+odN92kI-rX0-01114-00340010-00340386 hey yeah, on this band we are fully okay,
+odN92kI-rX0-01115-00340386-00340650 and then on the next band we are in a warm zone
+odN92kI-rX0-01116-00340650-00341007 and then on the next band, we would be on a throttling zone,
+odN92kI-rX0-01117-00341007-00341144 and then the last band would be,
+odN92kI-rX0-01118-00341144-00341349 okay at this point, you are screwed,
+odN92kI-rX0-01119-00341349-00341528 you are gonna be shut down right,
+odN92kI-rX0-01120-00341528-00341732 that's a very typical design.
+odN92kI-rX0-01121-00341732-00342150 And the thermal subsystem actually sends notifications
+odN92kI-rX0-01122-00342150-00342370 to user-space when you cross the tree points,
+odN92kI-rX0-01123-00342370-00342505 when you cross up and down.
+odN92kI-rX0-01124-00342604-00342983 That would be one way of, obviously I mean,
+odN92kI-rX0-01125-00342983-00343292 out of tree policies that people do,
+odN92kI-rX0-01126-00343292-00343484 it's a whole different ballgame,
+odN92kI-rX0-01127-00343484-00343788 but that's at least how it is recommended today.
+odN92kI-rX0-01128-00343788-00343972 You don't need to necessarily throttle
+odN92kI-rX0-01129-00343972-00344185 on all the tree points, you can just define tree points
+odN92kI-rX0-01130-00344185-00344366 that are just for notification.
+odN92kI-rX0-01131-00344798-00344991 - So you're saying it is, somewhat there
+odN92kI-rX0-01132-00344991-00345186 already but it's out of tree.
+odN92kI-rX0-01133-00345186-00345398 - No, it's there, it's not out of tree.
+odN92kI-rX0-01134-00345547-00345639 - The notifications for that.
+odN92kI-rX0-01135-00345639-00346033 - It's probably not even used by auto tree systems,
+odN92kI-rX0-01136-00346033-00346204 but yeah, it's actually there.
+odN92kI-rX0-01137-00346440-00346603 So can you get those notifications?
+odN92kI-rX0-01138-00346603-00346861 Are they per thermal zone or?
+odN92kI-rX0-01139-00346861-00346974 - On the thermal zone.
+odN92kI-rX0-01140-00346974-00347057 - Okay.
+odN92kI-rX0-01141-00347186-00347345 So maybe they're--
+odN92kI-rX0-01142-00347345-00347623 - And you define those on the by-street as well for example.
+odN92kI-rX0-01143-00347623-00348071 There's no standardized way of saying what are the bands
+odN92kI-rX0-01144-00348071-00348262 these are left for the system integrator
+odN92kI-rX0-01145-00348262-00348558 because some people think, okay, two bands are fine,
+odN92kI-rX0-01146-00348558-00348671 three bands are fine, some people,
+odN92kI-rX0-01147-00348671-00348973 okay I want to have six bands.
+odN92kI-rX0-01148-00348973-00349227 - Yeah so but that would need to be
+odN92kI-rX0-01149-00349227-00349518 like defined by the system integrator.
+odN92kI-rX0-01150-00349518-00349601 - Right.
+odN92kI-rX0-01151-00349601-00349900 - Somehow like more, to add more trip points
+odN92kI-rX0-01152-00349900-00350112 for this kind of work.
+odN92kI-rX0-01153-00350112-00350612 So that would require some input from like the
+odN92kI-rX0-01154-00350618-00350916 firmware people as well as--
+odN92kI-rX0-01155-00350916-00351066 - I mean, yeah, if you have like something in ACPI
+odN92kI-rX0-01156-00351066-00351216 for example, then yeah.
+odN92kI-rX0-01157-00351216-00351409 - Well yeah that's what I mean.
+odN92kI-rX0-01158-00351409-00351863 So somebody needs to tell us what what are those,
+odN92kI-rX0-01159-00351863-00352117 what the bands are essentially.
+odN92kI-rX0-01160-00352335-00352503 - So to better understand, we can define tree points
+odN92kI-rX0-01161-00352503-00352776 already but we also have support to get signals
+odN92kI-rX0-01162-00352776-00352929 when we cross tree points?
+odN92kI-rX0-01163-00352929-00353215 - To the user-space we have Caesar fast notifications,
+odN92kI-rX0-01164-00353215-00353441 Caesar fast notifies that there are sent out.
+odN92kI-rX0-01165-00353745-00354031 If that's enough that's a whole different story, but yeah.
+odN92kI-rX0-01166-00354031-00354306 Yeah that would be easy if it's already there.
+odN92kI-rX0-01167-00354306-00354740 I was thinking maybe have some continuous signal
+odN92kI-rX0-01168-00354740-00354876 or something that you could use
+odN92kI-rX0-01169-00354876-00355216 if you want to basically create a feedback loop
+odN92kI-rX0-01170-00355216-00355558 in your application saying, okay I've got this much spare
+odN92kI-rX0-01171-00355558-00355904 of thermal, I mean, you could almost get there
+odN92kI-rX0-01172-00355904-00356142 if you define a lot of thread points I guess.
+odN92kI-rX0-01173-00356295-00356486 - It's like a concept of thermal capacity,
+odN92kI-rX0-01174-00356486-00356953 this thermal capacity we have crossing the next tree point.
+odN92kI-rX0-01175-00356953-00357124 Or maybe defining, okay I'm interested
+odN92kI-rX0-01176-00357124-00357445 to these specific level, can you notify me
+odN92kI-rX0-01177-00357445-00357558 when I'm crossing these levels,
+odN92kI-rX0-01178-00357558-00357741 some kind of programmable interface
+odN92kI-rX0-01179-00357741-00357918 where the application can actually require
+odN92kI-rX0-01180-00357918-00358285 to be notified on certain specific tree points.
+odN92kI-rX0-01181-00358469-00358646 - Yeah I think Shuima was at some point
+odN92kI-rX0-01182-00358646-00359146 was proposing something similar, right.
+odN92kI-rX0-01183-00359463-00359749 - Dell based notifications, to the,
+odN92kI-rX0-01184-00359749-00359874 the current is very slow.
+odN92kI-rX0-01185-00359998-00360191 Current mechanism is net link based
+odN92kI-rX0-01186-00360191-00360520 and it goes to Udev and then it distributes.
+odN92kI-rX0-01187-00360520-00361020 So it's a Dell based device and directing notification
+odN92kI-rX0-01188-00361051-00361134 that will be faster on that node.
+odN92kI-rX0-01189-00361308-00361484 - I mean that's the Dell nodes not existing,
+odN92kI-rX0-01190-00361484-00361640 he was proposing to have that.
+odN92kI-rX0-01191-00361640-00361980 Would that be more helpful?
+odN92kI-rX0-01192-00362120-00362219 - Potentially, yeah.
+odN92kI-rX0-01193-00362219-00362437 So what we did in turn is just a mock of things.
+odN92kI-rX0-01194-00362437-00362699 We basically took the power allocator governor
+odN92kI-rX0-01195-00362699-00362999 and it has like an error feedback in the feedback loop
+odN92kI-rX0-01196-00362999-00363226 it has, and we basically just exposed that
+odN92kI-rX0-01197-00363226-00363499 to user-space and trying to let user-space adapt
+odN92kI-rX0-01198-00363499-00363772 itself based on that information.
+odN92kI-rX0-01199-00363772-00363999 And that worked reasonably well.
+odN92kI-rX0-01200-00363999-00364083 - Yeah, I think he contrited that.
+odN92kI-rX0-01201-00364083-00364196 The other thing I was thinking about is
+odN92kI-rX0-01202-00364196-00364635 depends on how many gradients of information that you have
+odN92kI-rX0-01203-00364635-00364964 and like was mentioned earlier, a systems integrator
+odN92kI-rX0-01204-00364964-00365286 or something could actually tune it and if the gradients
+odN92kI-rX0-01205-00365286-00365580 are reliable and predictable, then the applications
+odN92kI-rX0-01206-00365580-00365805 could be tuned ahead of time that when they hit
+odN92kI-rX0-01207-00365805-00366129 a certain gradient, then the application goes into another
+odN92kI-rX0-01208-00366129-00366372 like a lower performance mode.
+odN92kI-rX0-01209-00367044-00367352 - There's a question at the far back.
+odN92kI-rX0-01210-00368124-00368355 - Does it need to be signals or could it just look like
+odN92kI-rX0-01211-00368355-00368694 the CPU freak interface where I've got a base temperature
+odN92kI-rX0-01212-00368694-00369100 and a min and a max that I can deal with and then
+odN92kI-rX0-01213-00369100-00369471 the applications can do whatever they wish at that point.
+odN92kI-rX0-01214-00369586-00369861 - Yeah, I'm not sure if it needs to be signals, I mean,
+odN92kI-rX0-01215-00370192-00370314 we probably can't react that fast anyway,
+odN92kI-rX0-01216-00370314-00370730 I mean, it's frame by frame so if it's just something
+odN92kI-rX0-01217-00370730-00371230 we could Paul or Reed, yeah, maybe that's sufficient.
+odN92kI-rX0-01218-00371251-00371412 - Yeah but maybe you don't want to be
+odN92kI-rX0-01219-00371412-00371720 like reacting frame by frame, maybe you wanted to find
+odN92kI-rX0-01220-00371720-00372131 a specific notification entry point where you just
+odN92kI-rX0-01221-00372131-00372370 do the switch from the application perspective, right.
+odN92kI-rX0-01222-00372370-00372804 So when we cross that tree point then
+odN92kI-rX0-01223-00372804-00373130 I would switch to another mode where I would
+odN92kI-rX0-01224-00373130-00373474 have less of a rendering requirement.
+odN92kI-rX0-01225-00373660-00373750 I mean it doesn't need to be frame by frame.
+odN92kI-rX0-01226-00373750-00373987 You would glitch a little bit when you cross it.
+odN92kI-rX0-01227-00373987-00374199 - You'll fixed it in a few frames.
+odN92kI-rX0-01228-00374291-00374381 - But you wouldn't be like
+odN92kI-rX0-01229-00374381-00374491 a frame by frame kind of adaptation.
+odN92kI-rX0-01230-00374789-00374959 - That's going to be a long throw.
+odN92kI-rX0-01231-00375664-00376110 - So but then we'll need user-space to subscribe to those.
+odN92kI-rX0-01232-00376266-00376349 - [Morten] Right.
+odN92kI-rX0-01233-00376349-00376582 - Not notification mechanisms, whatever they are,
+odN92kI-rX0-01234-00376582-00376823 so it will have to either open a dev
+odN92kI-rX0-01235-00376823-00377057 or listen to something, right.
+odN92kI-rX0-01236-00377379-00377570 - Yeah, they will need to listen to that.
+odN92kI-rX0-01237-00377570-00377736 - Yeah, okay.
+odN92kI-rX0-01238-00377736-00378026 - But that's fair enough, I mean, if I wants information
+odN92kI-rX0-01239-00378026-00378213 it needs to get it from somewhere.
+odN92kI-rX0-01240-00378309-00378392 - Yeah, okay.
+odN92kI-rX0-01241-00378739-00378907 - Fair enough.
+odN92kI-rX0-01242-00378907-00379093 - Where are we on time?
+odN92kI-rX0-01243-00379093-00379260 - We are fine, so we are five minutes
+odN92kI-rX0-01244-00379260-00379496 ahead of time, actually even ten minutes.
+odN92kI-rX0-01245-00379496-00379960 So are there any more thoughts on this particular topic
+odN92kI-rX0-01246-00379960-00380288 or we can switch over to the next one and then.
+odN92kI-rX0-01247-00380288-00380376 - [Morten] No I think that's it,
+odN92kI-rX0-01248-00380376-00380496 it sounds like we have something--
+odN92kI-rX0-01249-00380496-00380774 - I have one question related to this.
+odN92kI-rX0-01250-00380774-00381040 Last year they were talking about thermal capacity govern
+odN92kI-rX0-01251-00381040-00381188 and thermal capacity end scheduler.
+odN92kI-rX0-01252-00381188-00381453 Is that patches upstream or is it like?
+odN92kI-rX0-01253-00381453-00381673 - [Morten] You should ask Vason in the back.
+odN92kI-rX0-01254-00381871-00382053 - They're actually still working on that.
+odN92kI-rX0-01255-00382053-00382283 She is running some tests, she has been a bit sidetracked
+odN92kI-rX0-01256-00382283-00382538 on that activity, but it's ongoing.
+odN92kI-rX0-01257-00382538-00382855 We'll make sure that this will go outside in coming weeks.
+odN92kI-rX0-01258-00383426-00383651 - So I think for this topic this summer,
+odN92kI-rX0-01259-00383651-00383981 yes there is some Caesar fast notification going on
+odN92kI-rX0-01260-00383981-00384176 that you can already subscribe
+odN92kI-rX0-01261-00384176-00384363 and you would require the syst integrator
+odN92kI-rX0-01262-00384363-00384733 to define your bands but either way if you have
+odN92kI-rX0-01263-00384733-00384904 the syst integrator defining your bands,
+odN92kI-rX0-01264-00384904-00385259 reliable bands, I think there's still a question
+odN92kI-rX0-01265-00385259-00385568 to answer is if this is the Caesar fast notifications
+odN92kI-rX0-01266-00385568-00385896 is enough or we need a dev node for example
+odN92kI-rX0-01267-00385896-00386225 to be at the right time scale that you need.
+odN92kI-rX0-01268-00386225-00386455 - Yeah, okay that makes sense.
+odN92kI-rX0-01269-00386455-00386650 I'll have a look at that patch and see.
+odN92kI-rX0-01270-00386830-00386955 It's on the list already.
+odN92kI-rX0-01271-00387134-00387336 There was a patch for you in dev node.
+odN92kI-rX0-01272-00387336-00387591 - Is your dev node patches already on my list?
+odN92kI-rX0-01273-00387591-00387800 I think that we were supposed to send it.
+odN92kI-rX0-01274-00387800-00387916 - Yeah, I'll send a mailing list.
+odN92kI-rX0-01275-00387916-00387999 - Okay.
+odN92kI-rX0-01276-00387999-00388352 - Okay, cool, that's it, more questions?
+odN92kI-rX0-01277-00388624-00388723 No, next.
+odN92kI-rX0-01278-00388982-00389207 (applause)
+odN92kI-rX0-01279-00389628-00389816 - Sorry about that meddling stuff,
+odN92kI-rX0-01280-00389816-00390053 it was the latest thing when we put it in.
+odN92kI-rX0-01281-00392059-00392286 - So I have a problem but I don't
+odN92kI-rX0-01282-00392286-00392453 have a good solutions beforehand.
+odN92kI-rX0-01283-00392814-00393219 So we have Per Core P-States in Intel platform
+odN92kI-rX0-01284-00393306-00393722 for really long but previously the power was not
+odN92kI-rX0-01285-00393722-00393886 a consideration on this platform.
+odN92kI-rX0-01286-00393886-00394245 I mean, it was okay too but we have power sensitive
+odN92kI-rX0-01287-00394245-00394536 platforms now with Per Core P-States.
+odN92kI-rX0-01288-00394536-00395036 It's great for power, it significantly improves
+odN92kI-rX0-01289-00395086-00395301 the power savings and active power savings,
+odN92kI-rX0-01290-00395394-00395730 and but they will have some small performance loss
+odN92kI-rX0-01291-00395730-00396125 with some producer-consumer type workloads.
+odN92kI-rX0-01292-00396261-00396573 And good part, good news is that it's not that bad
+odN92kI-rX0-01293-00396573-00396856 because our hardware P algorithms
+odN92kI-rX0-01294-00396856-00397082 already has built-in optimization for that,
+odN92kI-rX0-01295-00397199-00397688 but there are still some cases where seams are required.
+odN92kI-rX0-01296-00397866-00398179 So it's very typical example in this is
+odN92kI-rX0-01297-00398179-00398380 one of the thread is continuously pushing
+odN92kI-rX0-01298-00398380-00398629 data and producing some new data,
+odN92kI-rX0-01299-00398629-00398985 and trying to ask some consumers stretch to do things,
+odN92kI-rX0-01300-00398985-00399215 like rendering or an audio playback
+odN92kI-rX0-01301-00399215-00399435 or some type of you know situations.
+odN92kI-rX0-01302-00399526-00399898 And in that case, the compared to the,
+odN92kI-rX0-01303-00399898-00400108 it's nothing to do with, I know it's in a core
+odN92kI-rX0-01304-00400108-00400294 but compared to the older generations,
+odN92kI-rX0-01305-00400294-00400633 we have some performance loss.
+odN92kI-rX0-01306-00400633-00401101 You can see simply do with schbench type workload,
+odN92kI-rX0-01307-00401101-00401436 with sleep option, like if you look at,
+odN92kI-rX0-01308-00401436-00401792 the PCPS means the old style and that's a new style
+odN92kI-rX0-01309-00401792-00401971 the Per Core P-States and SPD means
+odN92kI-rX0-01310-00402103-00402343 we don't have Per Core P-State here.
+odN92kI-rX0-01311-00402343-00402724 And if you look at this you know CPU in the case,
+odN92kI-rX0-01312-00402724-00403224 in the Per Core P case case, the CPU at two
+odN92kI-rX0-01313-00403303-00403803 which was in previously was planning at much higher
+odN92kI-rX0-01314-00403827-00404214 frequency because of the other core is running high
+odN92kI-rX0-01315-00404214-00404511 so it also gets a boost because of others.
+odN92kI-rX0-01316-00404511-00404797 But when you run Per Core P-State you won't get that boost.
+odN92kI-rX0-01317-00404797-00405134 It's good for power obviously but we don't get that boost
+odN92kI-rX0-01318-00405134-00405591 and then you lose some performance.
+odN92kI-rX0-01319-00405591-00405977 And it's not all workloads, it's very isolated workloads
+odN92kI-rX0-01320-00406069-00406278 but you won't see much impact but
+odN92kI-rX0-01321-00406278-00406523 we still need to solve this thing.
+odN92kI-rX0-01322-00406682-00406983 So I then I try to look at this,
+odN92kI-rX0-01323-00406983-00407243 the average CFS utilization.
+odN92kI-rX0-01324-00407243-00407507 So like in this that example which I was giving you
+odN92kI-rX0-01325-00407507-00407846 the one other CPU, one is basically is fully busy
+odN92kI-rX0-01326-00407846-00408172 with full capacity and the other CPU2
+odN92kI-rX0-01327-00408172-00408462 is the actual CFS utilization is almost
+odN92kI-rX0-01328-00408462-00408868 and very low because it sleeps most of the time, wakes up,
+odN92kI-rX0-01329-00408868-00409168 do something, again dump to some hardware again go to sleep.
+odN92kI-rX0-01330-00409168-00409313 So it doesn't ever build
+odN92kI-rX0-01331-00409313-00409571 a utilization enough to do anything.
+odN92kI-rX0-01332-00409826-00410143 And I see that you on the, had this similar problem
+odN92kI-rX0-01333-00410143-00410405 but they, it's not same problem but similar,
+odN92kI-rX0-01334-00410405-00410581 where do you have some utilization clamps
+odN92kI-rX0-01335-00410581-00410810 and other things, right, I think you did some
+odN92kI-rX0-01336-00410810-00411135 lot of work on because it's long sleeping tasks.
+odN92kI-rX0-01337-00411135-00411434 But it doesn't help because I know if we cannot really
+odN92kI-rX0-01338-00411434-00411681 define clamps, it's like x86 system,
+odN92kI-rX0-01339-00411681-00412041 anybody can run any workload, it's not tuned vertically,
+odN92kI-rX0-01340-00412041-00412513 so we don't have option.
+odN92kI-rX0-01341-00412782-00412865 So we don't have--
+odN92kI-rX0-01342-00412865-00413170 - There's a worst case, the worst case is when,
+odN92kI-rX0-01343-00413170-00413478 instead of, you've got 99% and 5%,
+odN92kI-rX0-01344-00413478-00413695 when they're 50/50 then they both run slow.
+odN92kI-rX0-01345-00413809-00414043 - Yeah, that is the worst.
+odN92kI-rX0-01346-00414356-00414825 So I did simple experiment, may not be the great one
+odN92kI-rX0-01347-00414825-00415071 but if I know that like in this case,
+odN92kI-rX0-01348-00415071-00415341 if the current task which is basically waking up
+odN92kI-rX0-01349-00415341-00415841 the consumer in this case and its utilization is
+odN92kI-rX0-01350-00415976-00416354 and the current one is like half of the full CPU capacity,
+odN92kI-rX0-01351-00416354-00416854 then you just give an indication to the CPU freq update
+odN92kI-rX0-01352-00417039-00417539 to sched util and current because to reduce the impact
+odN92kI-rX0-01353-00417554-00417880 the sched util governor can totally ignore that signal,
+odN92kI-rX0-01354-00417880-00418075 but like an Intel P-state case,
+odN92kI-rX0-01355-00418075-00418347 we have mechanism where we can somehow relate
+odN92kI-rX0-01356-00418347-00418621 between different processes, we know that they are related.
+odN92kI-rX0-01357-00418721-00418992 We can tell the hardware that they are related then it will
+odN92kI-rX0-01358-00418992-00419262 know that let me push this also at the same time.
+odN92kI-rX0-01359-00419396-00419691 But I don't know how how good is that,
+odN92kI-rX0-01360-00419691-00420018 what's the impact, so that's what I have questions.
+odN92kI-rX0-01361-00420018-00420274 What's that, what will happen if I put this change?
+odN92kI-rX0-01362-00420929-00421167 So we don't want back and forth
+odN92kI-rX0-01363-00421167-00421505 like didn't want a cyclic positive feedback loop,
+odN92kI-rX0-01364-00421505-00421756 it's gonna keep the frequency high, right,
+odN92kI-rX0-01365-00421756-00421952 if one is waking up another and then that's waking
+odN92kI-rX0-01366-00421952-00422303 it back up, I don't I mean it's just the core snippets,
+odN92kI-rX0-01367-00422303-00422464 I don't know how he implemented it,
+odN92kI-rX0-01368-00422464-00422664 but would that cause both CPUs to be stay
+odN92kI-rX0-01369-00422664-00423039 at a higher frequency because the other one needs you.
+odN92kI-rX0-01370-00423039-00423135 - Yeah, yeah.
+odN92kI-rX0-01371-00423135-00423285 - Is that a potential issue?
+odN92kI-rX0-01372-00423285-00423542 - Yeah, as long as both are related they will be high,
+odN92kI-rX0-01373-00423542-00423772 and that's what we want, if they're both are related,
+odN92kI-rX0-01374-00423772-00424122 If two CPUS are related, if two to CPS running a workload
+odN92kI-rX0-01375-00424122-00424368 which are related, they need to be high
+odN92kI-rX0-01376-00424368-00424635 otherwise it will cause the power distribution
+odN92kI-rX0-01377-00424635-00424809 an uneven distribution of the power.
+odN92kI-rX0-01378-00424809-00425096 - So if they're sharing a CPU, this happens naturally.
+odN92kI-rX0-01379-00425096-00425179 - Yeah.
+odN92kI-rX0-01380-00425179-00425276 - Yeah.
+odN92kI-rX0-01381-00425276-00425386 - But if their on separate CPUs
+odN92kI-rX0-01382-00425386-00425477 that's when we run into this.
+odN92kI-rX0-01383-00425477-00425560 - Yeah.
+odN92kI-rX0-01384-00425978-00426274 - So maybe I don't understand the case but it looks like
+odN92kI-rX0-01385-00426274-00426596 that if you know the tasks now with U clamp
+odN92kI-rX0-01386-00426596-00426977 you should be able to take those tasks and keep the
+odN92kI-rX0-01387-00426977-00427426 frequency to whatever minimum you define as a clamp value.
+odN92kI-rX0-01388-00427426-00427682 - So I need to know the task ahead.
+odN92kI-rX0-01389-00427682-00427977 - That's true, I mean, what we do for example
+odN92kI-rX0-01390-00427977-00428268 Android is the tasks are usually in C groups
+odN92kI-rX0-01391-00428268-00428506 and we know that those are the tasks most important
+odN92kI-rX0-01392-00428506-00428762 right now, so always keep the frequency at this level
+odN92kI-rX0-01393-00428762-00429064 whenever tasks are runnable independent
+odN92kI-rX0-01394-00429064-00429268 from their utility structure, if it is issue
+odN92kI-rX0-01395-00429268-00429483 is above the clamp value you follow the utilization
+odN92kI-rX0-01396-00429483-00429758 but if there are small tasks you still keep these.
+odN92kI-rX0-01397-00429758-00430258 - Yeah, that I understand, the problem is tuning
+odN92kI-rX0-01398-00430297-00430516 some waiting to clear, but if I like, if you are using
+odN92kI-rX0-01399-00430516-00431004 x86 system, like ten generation, and you don't know
+odN92kI-rX0-01400-00431004-00431289 unless you are expert user you don't nearly know
+odN92kI-rX0-01401-00431289-00431517 how to chain to C group and tie this.
+odN92kI-rX0-01402-00431517-00431805 He will just run a graphics work, something.
+odN92kI-rX0-01403-00431805-00432039 - There is per tasks API so with a Cisco
+odN92kI-rX0-01404-00432039-00432191 but you you need to know the task,
+odN92kI-rX0-01405-00432191-00432292 so if you need all these--
+odN92kI-rX0-01406-00432292-00432381 - Yeah, something which is--
+odN92kI-rX0-01407-00432381-00432572 - I get the point is ideally though it was configured out
+odN92kI-rX0-01408-00432572-00432791 itself and not force the end user to do it.
+odN92kI-rX0-01409-00432791-00433087 - Not every user can do that.
+odN92kI-rX0-01410-00433087-00433220 And very special because you know
+odN92kI-rX0-01411-00433220-00433527 somebody's tuning for you before, right.
+odN92kI-rX0-01412-00433527-00433760 - Is this a server workload or what kind of workload is it?
+odN92kI-rX0-01413-00433760-00433882 - It's a client workload.
+odN92kI-rX0-01414-00434170-00434258 (faintly speaking)
+odN92kI-rX0-01415-00434258-00434341 No, a client.
+odN92kI-rX0-01416-00434627-00434989 - I'm wondering if this, in this kind of workload
+odN92kI-rX0-01417-00434989-00435439 you could identify as sort of wake or wakey pattern
+odN92kI-rX0-01418-00435439-00435939 and it happens that one of those two process wakes up
+odN92kI-rX0-01419-00436105-00436491 there's a wakey wakey, in that case shouldn't the scheduler
+odN92kI-rX0-01420-00436491-00436764 already realize that there is this relationship
+odN92kI-rX0-01421-00436764-00437264 and put the two tasks close together so on the same core?
+odN92kI-rX0-01422-00437366-00437763 When you describe this problem, I would say,
+odN92kI-rX0-01423-00437763-00437967 doesn't the scheduler already solve this?
+odN92kI-rX0-01424-00437967-00438050 - No, no.
+odN92kI-rX0-01425-00438050-00438214 - Because they recognized that they are waking each other up
+odN92kI-rX0-01426-00438214-00438351 and put them on the same core.
+odN92kI-rX0-01427-00438351-00438651 I don't know if this is the case, I'm just asking it.
+odN92kI-rX0-01428-00438651-00439010 - It won't happen because the other core is hopefully busy,
+odN92kI-rX0-01429-00439010-00439143 it's running continuously, right,
+odN92kI-rX0-01430-00439143-00439506 it's occupying full hundred capacity of the CPU, it's not.
+odN92kI-rX0-01431-00439714-00440214 - So the thing you just mentioned just a fine wake up
+odN92kI-rX0-01432-00440218-00440718 and that only works for if there's idle time.
+odN92kI-rX0-01433-00441076-00441316 - Another question is, another question I had is
+odN92kI-rX0-01434-00441316-00441555 in this case are each of the CPUs running
+odN92kI-rX0-01435-00441555-00441728 in a completely independent power supply?
+odN92kI-rX0-01436-00441728-00441847 - Yes, yeah.
+odN92kI-rX0-01437-00441847-00441999 - They are, okay.
+odN92kI-rX0-01438-00441999-00442083 (laughing)
+odN92kI-rX0-01439-00442083-00442230 - Because sometimes the frequency, you can get the power
+odN92kI-rX0-01440-00442230-00442528 sings by frequency savings and I've done some analysis
+odN92kI-rX0-01441-00442528-00442858 on that before and it was like not worth it.
+odN92kI-rX0-01442-00442858-00442945 - Yeah yeah yeah.
+odN92kI-rX0-01443-00443042-00443265 - SPD stands for single power.
+odN92kI-rX0-01444-00443265-00443402 (speaking faintly)
+odN92kI-rX0-01445-00443402-00443485 - A power, okay, not P-state.
+odN92kI-rX0-01446-00443842-00444147 - I might have missed this earlier on, so you mentioned
+odN92kI-rX0-01447-00444147-00444427 this is actually running slower but do you have any kind
+odN92kI-rX0-01448-00444427-00444766 of data to show how much slower the total task is running?
+odN92kI-rX0-01449-00444766-00445024 It sounds like it's kind of working as intended
+odN92kI-rX0-01450-00445024-00445328 based on my interpretation because the other--
+odN92kI-rX0-01451-00445328-00445566 - It is working as intended but the--
+odN92kI-rX0-01452-00445566-00445893 (speaking faintly)
+odN92kI-rX0-01453-00445893-00446118 (laughing)
+odN92kI-rX0-01454-00446278-00446371 - I guess what, because--
+odN92kI-rX0-01455-00446371-00446821 - No, this is working as designed, I would say,
+odN92kI-rX0-01456-00446821-00447237 but it has it has the some disadvantage
+odN92kI-rX0-01457-00447237-00447566 if you compare with the old style with the new style, right.
+odN92kI-rX0-01458-00447566-00447675 - Right.
+odN92kI-rX0-01459-00447675-00447769 - What's the performance of all time?
+odN92kI-rX0-01460-00447769-00447983 - I mean you're at 100% on one CPU guaranteed,
+odN92kI-rX0-01461-00447983-00448189 the other CPU you're at 5%, right,
+odN92kI-rX0-01462-00448189-00448560 so if you run that second core faster
+odN92kI-rX0-01463-00448560-00448781 then maybe you get down to like 2%.
+odN92kI-rX0-01464-00448781-00448897 Is that the?
+odN92kI-rX0-01465-00448897-00449239 - No, you won't get because of whenever we try
+odN92kI-rX0-01466-00449239-00449493 to associate tasks we have overhead to some
+odN92kI-rX0-01467-00449493-00449812 in firm hardware, so that's not going to be hundred equal
+odN92kI-rX0-01468-00449812-00450118 but it will be like with so base this change,
+odN92kI-rX0-01469-00450118-00450317 I could all get performance very similar.
+odN92kI-rX0-01470-00450317-00450472 You'll never get similar, same.
+odN92kI-rX0-01471-00450700-00451041 - So it's look like you want the two CPU
+odN92kI-rX0-01472-00451041-00451245 to share the same frequency domain.
+odN92kI-rX0-01473-00451245-00451328 - [Lecturer] Yeah.
+odN92kI-rX0-01474-00451328-00451616 - So why don't you, I mean, instead because they have some
+odN92kI-rX0-01475-00451616-00451816 dedicated frequency control that you
+odN92kI-rX0-01476-00451816-00452068 can't have a shared frequency domain.
+odN92kI-rX0-01477-00452068-00452364 - No, they have dedicated strict control on each,
+odN92kI-rX0-01478-00452364-00452584 so hardware does doesn't know each other.
+odN92kI-rX0-01479-00452584-00452730 - Yeah but from a software point of view
+odN92kI-rX0-01480-00452730-00452885 we already have something like that.
+odN92kI-rX0-01481-00452885-00453229 - No, we cannot tell hardware that they are,
+odN92kI-rX0-01482-00453229-00453372 we have to tell them they are together.
+odN92kI-rX0-01483-00453372-00453655 - Not from an another but in an
+odN92kI-rX0-01484-00453821-00454131 ARM platform we have one frequency domain
+odN92kI-rX0-01485-00454131-00454328 for all the CPU and we are looking,
+odN92kI-rX0-01486-00454328-00454591 we are selecting the highest utilization
+odN92kI-rX0-01487-00454591-00454953 of all the CPU in the in the frequency domain.
+odN92kI-rX0-01488-00454953-00455138 So can't you create something similar?
+odN92kI-rX0-01489-00455138-00455221 - Yeah but--
+odN92kI-rX0-01490-00455221-00455385 - Looking at the utilization of all the CPU.
+odN92kI-rX0-01491-00455385-00455670 - Yeah but then I did defeat the purpose of
+odN92kI-rX0-01492-00455670-00455859 Per Core P-States if I do it.
+odN92kI-rX0-01493-00455859-00455954 (speaking faintly)
+odN92kI-rX0-01494-00455954-00456197 - Yeah it's like, it's very, that's what I said initially,
+odN92kI-rX0-01495-00456197-00456477 it's very isolated case, it's not a panic situation,
+odN92kI-rX0-01496-00456708-00457027 very, yeah, it's outliers, not really a mainstream.
+odN92kI-rX0-01497-00457503-00457586 - So IOA is exactly the same except we have to hack for IOA.
+odN92kI-rX0-01498-00457982-00458273 (speaking faintly)
+odN92kI-rX0-01499-00458778-00459075 - Can't we repurpose the IOA 'cause this,
+odN92kI-rX0-01500-00459075-00459218 so what I said just for the record,
+odN92kI-rX0-01501-00459218-00459459 this is the exact same problem we have with IOA.
+odN92kI-rX0-01502-00459553-00459941 IOA it is for block devices, this just happens to be
+odN92kI-rX0-01503-00459941-00460263 not a block device but it's the exact same problem.
+odN92kI-rX0-01504-00460437-00460580 - Yes, yes, it is the exact same problem,
+odN92kI-rX0-01505-00460580-00460831 I just didn't want to because IOA is set
+odN92kI-rX0-01506-00460831-00461179 by the drivers in the only thing is.
+odN92kI-rX0-01507-00461334-00461626 (faintly speaking)
+odN92kI-rX0-01508-00461885-00462043 - So in this case you would have the consumer
+odN92kI-rX0-01509-00462043-00462297 do the IOA on the supplier?
+odN92kI-rX0-01510-00462297-00462539 - Yeah but I need to identify consumer, how do I.
+odN92kI-rX0-01511-00462539-00462733 - I know, it's a question for Peter.
+odN92kI-rX0-01512-00462733-00463154 So in this case you would have the consumer do the IOA
+odN92kI-rX0-01513-00463154-00463470 on the producer, is that what you would want?
+odN92kI-rX0-01514-00463470-00463675 - So it's the consumer that is waiting,
+odN92kI-rX0-01515-00463675-00463839 it's waiting on a device that just happens
+odN92kI-rX0-01516-00463839-00463964 to not be a block device.
+odN92kI-rX0-01517-00464058-00464266 - Well it's not, okay, it might be talking to device
+odN92kI-rX0-01518-00464266-00464467 but I think his problem was not that,
+odN92kI-rX0-01519-00464467-00464677 waiting on a device right.
+odN92kI-rX0-01520-00464677-00464987 He was in a sense is waiting on the producer.
+odN92kI-rX0-01521-00464987-00465271 - It's waiting on Hutex basically yet in this example.
+odN92kI-rX0-01522-00465271-00465640 We are in a futex on that one.
+odN92kI-rX0-01523-00465640-00466099 So we have to simulate IOA, like I'm simulating
+odN92kI-rX0-01524-00466099-00466413 a remote wake, it's basically nothing.
+odN92kI-rX0-01525-00466413-00466739 - So you kinda want to include the producers utilization
+odN92kI-rX0-01526-00466739-00467011 as part of your utilization so that you can run fast.
+odN92kI-rX0-01527-00467011-00467094 - [Lecturer] Yeah.
+odN92kI-rX0-01528-00467094-00467301 - In a sense, I'm not sure that's the right thing to do
+odN92kI-rX0-01529-00467301-00467454 but that would solve your issue.
+odN92kI-rX0-01530-00467454-00467693 - [Lecturer] Yeah that's why question is.
+odN92kI-rX0-01531-00467826-00467988 - I was also thinking of a similar scenario
+odN92kI-rX0-01532-00467988-00468266 with Chromebooks, and how we'd have a scenario
+odN92kI-rX0-01533-00468266-00468467 where we would have a shared handle and the handle
+odN92kI-rX0-01534-00468467-00468722 would be open, right, but when the handle was open
+odN92kI-rX0-01535-00468722-00468939 for the display what would happen is
+odN92kI-rX0-01536-00468939-00469249 is the GPU processor would be pegged at that frequency.
+odN92kI-rX0-01537-00469249-00469475 It was actually a bug because we needed to let go
+odN92kI-rX0-01538-00469475-00469622 of the handle at a better time
+odN92kI-rX0-01539-00469622-00469799 and it was a power management issue.
+odN92kI-rX0-01540-00469799-00469997 But if there is some way like with IOA
+odN92kI-rX0-01541-00469997-00470343 or some other mechanism, they could signal or notify
+odN92kI-rX0-01542-00470343-00470817 that while this is held, keep it up, right,
+odN92kI-rX0-01543-00470817-00471118 you wouldn't necessarily need to know specifically,
+odN92kI-rX0-01544-00471118-00471507 that the consumer producer wouldn't have to know who is who.
+odN92kI-rX0-01545-00471507-00471838 (speaking faintly)
+odN92kI-rX0-01546-00471838-00472338 - This kind of resembles this thing that Yuri
+odN92kI-rX0-01547-00472343-00472646 was talking about yesterday in a scheduler
+odN92kI-rX0-01548-00472646-00472821 proxy execution problem, right.
+odN92kI-rX0-01549-00473064-00473435 Pretty much it is very similar to that one
+odN92kI-rX0-01550-00473435-00473935 because this is just pretty much like
+odN92kI-rX0-01551-00474033-00474213 the same problem in different terms.
+odN92kI-rX0-01552-00474764-00475076 - So with proxy there is a clear and unambiguous
+odN92kI-rX0-01553-00475076-00475576 blocked on relation and this might not have that.
+odN92kI-rX0-01554-00475917-00476260 - Right, that's correct, but maybe the,
+odN92kI-rX0-01555-00476260-00476746 so my point is that maybe somebody should know that
+odN92kI-rX0-01556-00476746-00476943 those two things are related, right.
+odN92kI-rX0-01557-00477029-00477223 (speaking faintly)
+odN92kI-rX0-01558-00477223-00477723 - Yeah, yes.
+odN92kI-rX0-01559-00477855-00478164 - So if you use a mute, a few tags in two threads
+odN92kI-rX0-01560-00478164-00478664 in application that they obviously are correlated.
+odN92kI-rX0-01561-00479055-00479138 - Maybe.
+odN92kI-rX0-01562-00479138-00479333 - Well no, they are because they use
+odN92kI-rX0-01563-00479333-00479628 the same lock so there is at least
+odN92kI-rX0-01564-00479628-00479846 one memory location they want to access.
+odN92kI-rX0-01565-00480116-00480429 - Yeah but so the few tags,
+odN92kI-rX0-01566-00480429-00480624 you don't know who will wake the futex.
+odN92kI-rX0-01567-00480725-00480947 The futex is just a wait operation.
+odN92kI-rX0-01568-00480947-00481211 It's futex wait and it'll sit there until
+odN92kI-rX0-01569-00481211-00481520 somebody in that address space flips the bit
+odN92kI-rX0-01570-00481520-00481752 and tells it, hey wake up now.
+odN92kI-rX0-01571-00482035-00482231 So you don't know who of the many
+odN92kI-rX0-01572-00482231-00482437 possible tasks that might be.
+odN92kI-rX0-01573-00482526-00482820 - No but my point was that if you are
+odN92kI-rX0-01574-00482820-00483272 a user-space programmer and if you put that few code
+odN92kI-rX0-01575-00483272-00483446 then you kind of know that your threads
+odN92kI-rX0-01576-00483446-00483660 are going to be working together.
+odN92kI-rX0-01577-00483760-00483954 So you in principle, you could also
+odN92kI-rX0-01578-00483954-00484199 do something else and say, hey these two threads
+odN92kI-rX0-01579-00484199-00484587 are really correlated because XYZ, right.
+odN92kI-rX0-01580-00485014-00485321 - So then you have a multi-threaded program
+odN92kI-rX0-01581-00485405-00485672 that has a gazillion threads that do absolutely
+odN92kI-rX0-01582-00485672-00486123 nothing important and two that actually do something.
+odN92kI-rX0-01583-00486123-00486348 (laughing)
+odN92kI-rX0-01584-00486629-00486821 - And I kind of had two similar issues in the
+odN92kI-rX0-01585-00486821-00486983 embedded space, when you're talking about
+odN92kI-rX0-01586-00486983-00487299 say display pipeline, we may have two threads working on it
+odN92kI-rX0-01587-00487299-00487518 but if you are double buffering, you don't want
+odN92kI-rX0-01588-00487518-00487763 to treat them as one combined load
+odN92kI-rX0-01589-00487763-00488022 because they're not going to really delay each other.
+odN92kI-rX0-01590-00488022-00488154 And then there are sometimes you're not doing
+odN92kI-rX0-01591-00488154-00488325 double buffering, in which case you do
+odN92kI-rX0-01592-00488325-00488547 want to consider them as one workload.
+odN92kI-rX0-01593-00488547-00488734 This is going by wake or wakey doesn't
+odN92kI-rX0-01594-00488734-00488899 really give you that information.
+odN92kI-rX0-01595-00488997-00489253 So if you're single buffered, you want to run
+odN92kI-rX0-01596-00489253-00489635 both of the CPUs really fast so that you can meet your needs
+odN92kI-rX0-01597-00489635-00490070 but if you're double buffered you can take your own time.
+odN92kI-rX0-01598-00490070-00490238 So it's pretty complicated.
+odN92kI-rX0-01599-00490238-00490514 - Yes I know that, that's why there's no good solution,
+odN92kI-rX0-01600-00490514-00490800 so that's, this is the best I could think about
+odN92kI-rX0-01601-00490800-00491271 but I know it may impact some of those mobile.
+odN92kI-rX0-01602-00491412-00491572 - I guess, I mean, would it be right to say
+odN92kI-rX0-01603-00491572-00491889 we need some amount of input from those two threads
+odN92kI-rX0-01604-00491889-00492084 I think to make useful decision.
+odN92kI-rX0-01605-00492084-00492306 - Yeah but then somebody--
+odN92kI-rX0-01606-00492306-00492516 - That is not forthcoming.
+odN92kI-rX0-01607-00492516-00492599 - There's not be tuning or util clamping
+odN92kI-rX0-01608-00492599-00492784 on a per app or per basis but.
+odN92kI-rX0-01609-00492909-00493101 - That's not going to happen.
+odN92kI-rX0-01610-00493101-00493271 (speaking faintly)
+odN92kI-rX0-01611-00493271-00493578 Okay so if you know if you have better suggestion,
+odN92kI-rX0-01612-00493578-00493999 let me know, or I'll just send a spanner patch,
+odN92kI-rX0-01613-00493999-00494229 you can just do it as a record and just maybe comment
+odN92kI-rX0-01614-00494229-00494484 and leave it if there's a solution you can suggest.
+odN92kI-rX0-01615-00494720-00495099 - One more consideration, so if you can identify the tasks
+odN92kI-rX0-01616-00495099-00495345 from user-space so without changing the application
+odN92kI-rX0-01617-00495345-00495581 but you can still have some kind of monitoring mechanism
+odN92kI-rX0-01618-00495581-00495970 to identify those misbehaviors, then using the pair task
+odN92kI-rX0-01619-00495970-00496153 API can still set the output use.
+odN92kI-rX0-01620-00496153-00496422 - Yes, if I have an user-space which is smart enough
+odN92kI-rX0-01621-00496422-00496709 to identify the relationship, you can do it, yes.
+odN92kI-rX0-01622-00496709-00496859 - Without changing the application, I mean,
+odN92kI-rX0-01623-00496859-00497016 it can be a third party.
+odN92kI-rX0-01624-00497016-00497102 - Yes, without changing the application.
+odN92kI-rX0-01625-00497281-00497722 - Question, which Intel generation has this
+odN92kI-rX0-01626-00497722-00498060 per core capability, normag Xeon like Skylake,
+odN92kI-rX0-01627-00498060-00498266 Broadwell, do they have it?
+odN92kI-rX0-01628-00498266-00498349 - Yes.
+odN92kI-rX0-01629-00498349-00498478 - Okay so it's not like a latest
+odN92kI-rX0-01630-00498478-00498699 I mean it was a while, okay.
+odN92kI-rX0-01631-00498699-00498968 So in order to reproduce we send that if it's something
+odN92kI-rX0-01632-00498968-00499139 that we don't have the other four.
+odN92kI-rX0-01633-00499139-00499222 - (Lecturer) Yeah you need ten generation
+odN92kI-rX0-01634-00499222-00499351 now you probably.
+odN92kI-rX0-01635-00499351-00499488 - What do I need?
+odN92kI-rX0-01636-00499488-00499613 - Tenth generation cores.
+odN92kI-rX0-01637-00499835-00500055 - So for a client it is like what?
+odN92kI-rX0-01638-00500055-00500203 - The Ice Lake.
+odN92kI-rX0-01639-00500203-00500287 - The latest one, Ice Lake, yeah.
+odN92kI-rX0-01640-00500287-00500551 So for the client, in the client space you need the latest.
+odN92kI-rX0-01641-00500551-00500688 - But you can reproduce this on
+odN92kI-rX0-01642-00500688-00500993 the servers, in any of the servers, yes.
+odN92kI-rX0-01643-00500993-00501296 Yeah just run Schedule schedbench with sleep,
+odN92kI-rX0-01644-00501392-00501475 you will see it.
+odN92kI-rX0-01645-00501475-00501778 - Yeah and it's just that the ranges get bigger every.
+odN92kI-rX0-01646-00501778-00502233 - Yeah, fine, sure.
+odN92kI-rX0-01647-00502233-00502510 - Can you say again what you do in case
+odN92kI-rX0-01648-00502510-00502846 your governor receives a shed CPU freq wake remote?
+odN92kI-rX0-01649-00502846-00503346 - I, we have the, we have a boost, we know that
+odN92kI-rX0-01650-00503607-00503909 we just increase the, we have something called EPP knob
+odN92kI-rX0-01651-00503909-00504256 which all the minimum frequency, we can just increase it.
+odN92kI-rX0-01652-00504256-00504368 So it it will--
+odN92kI-rX0-01653-00504476-00504631 - The knob is expensive, right.
+odN92kI-rX0-01654-00504747-00505031 - In Ice Lake, not, yeah, it's the that,
+odN92kI-rX0-01655-00505031-00505201 it takes hundred cycles, it's not.
+odN92kI-rX0-01656-00505414-00505587 - Too bright not to take affect.
+odN92kI-rX0-01657-00505587-00505751 - At the ah sure.
+odN92kI-rX0-01658-00505751-00505835 - Yeah.
+odN92kI-rX0-01659-00505835-00505963 - As long as we don't have to wait.
+odN92kI-rX0-01660-00505963-00506242 - Yeah, no, you don't wait, yeah.
+odN92kI-rX0-01661-00506242-00506560 Okay so yeah if commented, let's test it
+odN92kI-rX0-01662-00506560-00506927 and if you, I'm sure there are better approaches
+odN92kI-rX0-01663-00506927-00507171 and one approach is I was thinking about doing in
+odN92kI-rX0-01664-00507171-00507339 user-space write some smart app
+odN92kI-rX0-01665-00507339-00507799 and to identify this relationship but it's not that,
+odN92kI-rX0-01666-00507799-00507882 it's not immediately.
+odN92kI-rX0-01667-00507882-00507966 - It's not easy.
+odN92kI-rX0-01668-00508071-00508334 - It's not easy because that's has to be very smart, yeah.
+odN92kI-rX0-01669-00508947-00509341 No question.
+odN92kI-rX0-01670-00509517-00509822 - We are we are pretty much in a break time for, yeah,
+odN92kI-rX0-01671-00509990-00510172 for the official break, so let's just do a break
+odN92kI-rX0-01672-00510172-00510389 and then we will resume after it.
+odN92kI-rX0-01673-00510508-00510733 (applause)
+odN92kI-rX0-01674-00510827-00510927 - Thank you.
+odN92kI-rX0-01675-00511096-00511395 - Hi, I'm Sudeep from Linux kernel team at ARM.
+odN92kI-rX0-01676-00511541-00511760 So today I'm going to talk about
+odN92kI-rX0-01677-00511760-00512068 device power management based on platform firmware.
+odN92kI-rX0-01678-00512068-00512493 It's not just device, could be anything in the system.
+odN92kI-rX0-01679-00512493-00512847 So basically the idea is using platform firmware
+odN92kI-rX0-01680-00512847-00513160 instead of driving everything from OS.
+odN92kI-rX0-01681-00513247-00513747 Lots of people are not happy with that idea,
+odN92kI-rX0-01682-00513754-00513964 everybody wants this, can't also do it.
+odN92kI-rX0-01683-00513964-00514332 Has been controversial topic but there are use cases
+odN92kI-rX0-01684-00514332-00514625 and requirements that we just can't address,
+odN92kI-rX0-01685-00514625-00514989 having everything all the control in Linux.
+odN92kI-rX0-01686-00514989-00515489 So the idea here is, I'm just trying to standardize
+odN92kI-rX0-01687-00515738-00516238 the interface and does achieve to some extent and like
+odN92kI-rX0-01688-00516539-00516825 this is mostly getting used on quite complex systems
+odN92kI-rX0-01689-00516825-00517184 like the high end mobile phones you would see
+odN92kI-rX0-01690-00517184-00517479 which have a dedicated controller to do
+odN92kI-rX0-01691-00517479-00517702 all the power management on your system,
+odN92kI-rX0-01692-00517702-00518075 and every vendor trying to invent
+odN92kI-rX0-01693-00518075-00518338 their own way to talk to this controller.
+odN92kI-rX0-01694-00518338-00518797 So this interface is about standardizing that
+odN92kI-rX0-01695-00518797-00519258 and we did start initially standardizing everything
+odN92kI-rX0-01696-00519258-00519625 just to keep the changes minimal.
+odN92kI-rX0-01697-00519625-00519904 The way we approach the solution is like,
+odN92kI-rX0-01698-00519904-00520264 okay we have device, it needs some power domain,
+odN92kI-rX0-01699-00520264-00520617 some performance domain or reset domain,
+odN92kI-rX0-01700-00520617-00520814 clock domains, just hook them.
+odN92kI-rX0-01701-00520814-00521177 And the sharing information is today
+odN92kI-rX0-01702-00521177-00521375 given through the device tree,
+odN92kI-rX0-01703-00521375-00521633 so you need to know for the device X,
+odN92kI-rX0-01704-00521633-00521883 what do my power domain it's attached to
+odN92kI-rX0-01705-00521979-00522337 say the domain Z power, performance domain Y.
+odN92kI-rX0-01706-00522337-00522591 So all this information is needed.
+odN92kI-rX0-01707-00522591-00522875 So we are just thinking about how we can
+odN92kI-rX0-01708-00523291-00523467 move to like making a device centric
+odN92kI-rX0-01709-00523467-00523795 and see, like it's most likely the way
+odN92kI-rX0-01710-00523795-00524088 ACPI works today, like it's all device centric.
+odN92kI-rX0-01711-00524088-00524342 Turn on the device, set to this state,
+odN92kI-rX0-01712-00524342-00524652 it doesn't bother about what domains
+odN92kI-rX0-01713-00524652-00524894 or where it is in the system.
+odN92kI-rX0-01714-00524894-00525247 So the approach is more like that.
+odN92kI-rX0-01715-00525247-00525574 So just to give a background on what
+odN92kI-rX0-01716-00525574-00526032 most of the immediate systems had in the past,
+odN92kI-rX0-01717-00526032-00526314 this shows how everything is in the operating system
+odN92kI-rX0-01718-00526314-00526617 trying to deal with all the clocks, reset
+odN92kI-rX0-01719-00526617-00526913 power domains, everything in the OS.
+odN92kI-rX0-01720-00526913-00527413 So now it's migrating to this with some standard interface
+odN92kI-rX0-01721-00527708-00528110 talking a standard protocol over anything
+odN92kI-rX0-01722-00528110-00528294 can be the transport but you have
+odN92kI-rX0-01723-00528419-00528539 a dedicated controller that talks.
+odN92kI-rX0-01724-00528539-00528826 But you need not have a dedicated controller
+odN92kI-rX0-01725-00528826-00529149 on your system, it could be just a service
+odN92kI-rX0-01726-00529149-00529387 which is running somewhere remotely,
+odN92kI-rX0-01727-00529387-00529617 it could be as a secure service on your
+odN92kI-rX0-01728-00529768-00530045 like application processor itself.
+odN92kI-rX0-01729-00530045-00530362 It's just abstracted and moved away
+odN92kI-rX0-01730-00530669-00530786 so that the operating system need not know
+odN92kI-rX0-01731-00530786-00531183 all those details about where which register to book
+odN92kI-rX0-01732-00531183-00531530 and how to change for each and everything.
+odN92kI-rX0-01733-00532494-00532759 - If you are saying that it could just be a service running
+odN92kI-rX0-01734-00532759-00533064 either entrusted side or some other core processor
+odN92kI-rX0-01735-00533174-00533637 and then you're also trying to standardize the protocol,
+odN92kI-rX0-01736-00533637-00533765 wouldn't it be just simpler to just have
+odN92kI-rX0-01737-00533765-00533953 a standardized ops and let them implement it?
+odN92kI-rX0-01738-00533953-00534206 Like what is the benefit of doing this?
+odN92kI-rX0-01739-00534206-00534347 Like there's already an abstraction layer
+odN92kI-rX0-01740-00534347-00534564 by allowing people to implement ops
+odN92kI-rX0-01741-00534564-00534739 and drivers can plug in and implement ops.
+odN92kI-rX0-01742-00534739-00535239 - So I agree, I just gave that as the initial motivation
+odN92kI-rX0-01743-00535304-00535652 with which we started this, but the requirements,
+odN92kI-rX0-01744-00535652-00535877 as more and more requirements come,
+odN92kI-rX0-01745-00535877-00536043 so we see the real lean.
+odN92kI-rX0-01746-00536043-00536437 So just to get into the next requirement,
+odN92kI-rX0-01747-00536437-00536840 so how would you do if you had two virtual machines
+odN92kI-rX0-01748-00536840-00537019 one probably running some other OS
+odN92kI-rX0-01749-00537019-00537181 and one running Linux?
+odN92kI-rX0-01750-00537181-00537508 So do you go and implement custom interfaces
+odN92kI-rX0-01751-00537508-00537890 in each of those and try to solve the problem
+odN92kI-rX0-01752-00537890-00538174 or just have this standard interface
+odN92kI-rX0-01753-00538174-00538674 and each virtual machines talk over the same interface.
+odN92kI-rX0-01754-00538710-00539210 So this is becoming increasingly like high requirement,
+odN92kI-rX0-01755-00539263-00539629 even in mobile platforms people want to run
+odN92kI-rX0-01756-00539629-00539886 in the virtualized environment isolating
+odN92kI-rX0-01757-00539886-00540194 the like partitioning the system
+odN92kI-rX0-01758-00540194-00540616 and giving some parts isolated to a virtual machine
+odN92kI-rX0-01759-00540616-00541100 and it deals with all the controls and power management
+odN92kI-rX0-01760-00541100-00541407 or whatnot for that part of the, so.
+odN92kI-rX0-01761-00541407-00541724 So that's one of like another motivation,
+odN92kI-rX0-01762-00541724-00542068 why we need this more than ever.
+odN92kI-rX0-01763-00542315-00542694 So what we have today is, as I said like
+odN92kI-rX0-01764-00542694-00543058 when we started initially, it was all like
+odN92kI-rX0-01765-00543058-00543252 just keep the changes minimal,
+odN92kI-rX0-01766-00543252-00543623 let's just hook into existing frameworks
+odN92kI-rX0-01767-00543623-00544041 we have in kernel, like we have devfs for performance,
+odN92kI-rX0-01768-00544041-00544541 reset clock power domains, sensors, all those things.
+odN92kI-rX0-01769-00544596-00544960 So it's all fine, it hooks up perfectly well
+odN92kI-rX0-01770-00544960-00545431 and it's all fine but as I said,
+odN92kI-rX0-01771-00545431-00545581 one of the motivation for this
+odN92kI-rX0-01772-00545581-00545814 is also the virtualized environment.
+odN92kI-rX0-01773-00545814-00546211 Do we want the virtual machine configuration manager
+odN92kI-rX0-01774-00546211-00546434 to have each of these information
+odN92kI-rX0-01775-00546434-00546697 like it has to know which clock domain
+odN92kI-rX0-01776-00546697-00547001 it belongs to reset, so on and so.
+odN92kI-rX0-01777-00547001-00547501 So the idea is can we make everything device centric.
+odN92kI-rX0-01778-00547570-00548018 We just say, set this device to something
+odN92kI-rX0-01779-00548018-00548514 and all the arbitration happens in the platform firmware.
+odN92kI-rX0-01780-00548757-00549118 And also we have with this interface
+odN92kI-rX0-01781-00549118-00549407 as we are dealing with virtual machines
+odN92kI-rX0-01782-00549407-00549715 like the device isolation and giving permission
+odN92kI-rX0-01783-00549715-00550102 for each of these devices and the controls
+odN92kI-rX0-01784-00550102-00550482 they can have with this.
+odN92kI-rX0-01785-00550760-00551159 So as I was telling, so this is a use case
+odN92kI-rX0-01786-00551159-00551467 where we have these two resources here
+odN92kI-rX0-01787-00551467-00551753 can be one of these power clock.
+odN92kI-rX0-01788-00551846-00552325 So we are just saying this is one VM which is isolated
+odN92kI-rX0-01789-00552325-00552825 on its own, it talks over this transport and controls this.
+odN92kI-rX0-01790-00552825-00553084 Whereas another virtual machine can be having
+odN92kI-rX0-01791-00553084-00553386 and your mean virtual machine
+odN92kI-rX0-01792-00553386-00553858 manager can control what each of those domain
+odN92kI-rX0-01793-00553858-00554161 can actually have control over.
+odN92kI-rX0-01794-00554161-00554541 So the permissions and are all set here.
+odN92kI-rX0-01795-00554541-00555041 So that's the idea but again like if we have to describe
+odN92kI-rX0-01796-00555503-00556003 the entire topology, like what clock what power
+odN92kI-rX0-01797-00556040-00556404 what performance, that is not going to scale
+odN92kI-rX0-01798-00556404-00556791 because there is also requirements say that have to change.
+odN92kI-rX0-01799-00557121-00557372 - Don't you need to describe that somewhere anyway?
+odN92kI-rX0-01800-00557518-00557708 - Yeah, the idea what I am trying to say is like,
+odN92kI-rX0-01801-00557708-00558037 do we need that or do we let platform, deal with it,
+odN92kI-rX0-01802-00558037-00558298 because that's never going to end,
+odN92kI-rX0-01803-00558298-00558577 as I was about to say like GPIO pin marks.
+odN92kI-rX0-01804-00558577-00558896 So for this device to work, I have to do all these
+odN92kI-rX0-01805-00558896-00559256 and to keep increasing this controls.
+odN92kI-rX0-01806-00559802-00559946 - So basically you are saying that
+odN92kI-rX0-01807-00560040-00560455 that it could in principle be done in the OS
+odN92kI-rX0-01808-00560455-00560955 but that would require more and more complicated code
+odN92kI-rX0-01809-00561283-00561767 in going forward, and which may not be scalable enough.
+odN92kI-rX0-01810-00561767-00562116 And then it will be hard to address the case
+odN92kI-rX0-01811-00562116-00562435 in which different OS is run in different VMs
+odN92kI-rX0-01812-00562435-00562745 and have to control the same resources, right.
+odN92kI-rX0-01813-00562745-00563065 - Yeah, it may be possible but not always true.
+odN92kI-rX0-01814-00563065-00563367 So because of, as I said, like they want to isolate
+odN92kI-rX0-01815-00563367-00563730 the system, so yeah, it's possible
+odN92kI-rX0-01816-00563730-00564171 but if there is another VM running and it's sharing,
+odN92kI-rX0-01817-00564171-00564528 if you have Linux running, taking control over all,
+odN92kI-rX0-01818-00564636-00564812 that's just like stepping on each other.
+odN92kI-rX0-01819-00564812-00565282 So yeah you could probably do but we want it to be isolated
+odN92kI-rX0-01820-00565282-00565782 and put, left complete to the platform to solve this
+odN92kI-rX0-01821-00565832-00566143 for you because you just don't know what
+odN92kI-rX0-01822-00566143-00566311 other domains are actually running
+odN92kI-rX0-01823-00566311-00566673 or VMs are running and what they are managing.
+odN92kI-rX0-01824-00567390-00567592 - So one thing I see here, and like understand
+odN92kI-rX0-01825-00567592-00567856 the value of what you're bringing forward.
+odN92kI-rX0-01826-00567856-00568118 On a flip side, it pushes the complexity
+odN92kI-rX0-01827-00568118-00568455 down into the firmware, and I fear that
+odN92kI-rX0-01828-00568455-00568665 the complexity that we currently have
+odN92kI-rX0-01829-00568665-00568898 and have to manage in any operating system,
+odN92kI-rX0-01830-00568993-00569275 are just going to be coming back in a few years time
+odN92kI-rX0-01831-00569275-00569544 when we see that there's more dependencies
+odN92kI-rX0-01832-00569544-00569702 that cannot be accommodated in the firmware.
+odN92kI-rX0-01833-00569702-00569979 So we're gonna, in my, I fear that we will
+odN92kI-rX0-01834-00569979-00570426 have to revisit this again in the future.
+odN92kI-rX0-01835-00570426-00570641 The complexity is just going to creep up again
+odN92kI-rX0-01836-00570641-00570785 because we are going to see that there are things
+odN92kI-rX0-01837-00570785-00570902 that we can't do in firmware
+odN92kI-rX0-01838-00570902-00571094 that we're going to have to start doing
+odN92kI-rX0-01839-00571094-00571248 in the operating system again.
+odN92kI-rX0-01840-00571361-00571575 (faintly speaking)
+odN92kI-rX0-01841-00571575-00571775 And then there will be battles between firmware
+odN92kI-rX0-01842-00571775-00572117 and operating systems, and it will beget very complex.
+odN92kI-rX0-01843-00572117-00572348 - Yep but like if you have a solution
+odN92kI-rX0-01844-00572348-00572645 that we can solve this problem, you know is today,
+odN92kI-rX0-01845-00572645-00572907 yeah like happy, but do we have
+odN92kI-rX0-01846-00572907-00573330 any other better solution today to deal with this?
+odN92kI-rX0-01847-00573330-00573830 - So you were saying something about you didn't want to,
+odN92kI-rX0-01848-00573993-00574206 I completely didn't understand one of points
+odN92kI-rX0-01849-00574206-00574443 you were trying to make about letting,
+odN92kI-rX0-01850-00574443-00574669 doing this on a device level instead of a resource level.
+odN92kI-rX0-01851-00574669-00574982 I'm not sure what you're, can you explain that more?
+odN92kI-rX0-01852-00574982-00575285 - Yeah, just building up the case why we need.
+odN92kI-rX0-01853-00575285-00575439 I'll just come to that soon.
+odN92kI-rX0-01854-00575618-00575958 So this is just just two different,
+odN92kI-rX0-01855-00575958-00576304 like it could be your application processor
+odN92kI-rX0-01856-00576304-00576593 modem controlling shared resources.
+odN92kI-rX0-01857-00576593-00577001 So what I was trying to tell is like,
+odN92kI-rX0-01858-00577001-00577266 instead of having all this information,
+odN92kI-rX0-01859-00577266-00577542 can we just make it device centric
+odN92kI-rX0-01860-00577542-00577879 and say, I operate everything over the device.
+odN92kI-rX0-01861-00577879-00578260 Like you just get a device ID and you just say,
+odN92kI-rX0-01862-00578260-00578649 okay, I want to power on or whatever state,
+odN92kI-rX0-01863-00578649-00579089 I want this performance level or I just want to reset.
+odN92kI-rX0-01864-00579089-00579532 So just drive everything because the resource
+odN92kI-rX0-01865-00579532-00579938 in the previous slide as I showed, the VMs have to
+odN92kI-rX0-01866-00580033-00580355 like they just isolate the devices
+odN92kI-rX0-01867-00580355-00580761 and having to give all the information around that device
+odN92kI-rX0-01868-00580761-00581015 to each, is adds more complexity
+odN92kI-rX0-01869-00581015-00581339 than just make it everything device centric and then say
+odN92kI-rX0-01870-00581339-00581616 this is isolated and then deal with it, and if--
+odN92kI-rX0-01871-00581616-00581890 - So okay, one second, when you're saying device
+odN92kI-rX0-01872-00581890-00582177 here, in what context are you defining that?
+odN92kI-rX0-01873-00582177-00582423 Are you talking about it as like a device
+odN92kI-rX0-01874-00582423-00582635 as in the driver from driver core,
+odN92kI-rX0-01875-00582635-00582887 we're talking about a device that?
+odN92kI-rX0-01876-00582887-00583006 - Yes--
+odN92kI-rX0-01877-00583006-00583132 - Because for everything listed here
+odN92kI-rX0-01878-00583132-00583331 you have a corresponding thing on the right side,
+odN92kI-rX0-01879-00583331-00583480 so I'm like not sure what the point
+odN92kI-rX0-01880-00583480-00583687 you're trying to simplify here, I don't see it.
+odN92kI-rX0-01881-00583687-00583924 - This device as I said, it could be even a CPU
+odN92kI-rX0-01882-00583924-00584424 or your, some IP block, which it could be
+odN92kI-rX0-01883-00584458-00584628 a security engine or it could be
+odN92kI-rX0-01884-00584628-00584978 a random number generating, whatever it could be.
+odN92kI-rX0-01885-00584978-00585164 - Okay maybe let me rephrase it
+odN92kI-rX0-01886-00585164-00585331 and tell me if this is what you're trying to say.
+odN92kI-rX0-01887-00585331-00585674 Are you trying to say instead of SCMI exporting,
+odN92kI-rX0-01888-00585674-00585918 here are the ten clocks that you could use
+odN92kI-rX0-01889-00585918-00586165 and here are the ten research states you could use.
+odN92kI-rX0-01890-00586165-00586396 Are trying to say, here are the 10 devices
+odN92kI-rX0-01891-00586396-00586559 you could use and out of these 10 devices,
+odN92kI-rX0-01892-00586559-00586824 some of them have clocks, some of them have control.
+odN92kI-rX0-01893-00586824-00586920 Is that what you are trying to say, or?
+odN92kI-rX0-01894-00586920-00587038 - Yeah, those are--
+odN92kI-rX0-01895-00587137-00587321 - Are you trying to like encapsulate the clocks
+odN92kI-rX0-01896-00587321-00587654 and performance like as a subset of each device?
+odN92kI-rX0-01897-00587763-00587847 Is that what you're trying to say?
+odN92kI-rX0-01898-00587847-00587947 - Not each device, it's like,
+odN92kI-rX0-01899-00588051-00588349 instead of having all this operate on the device
+odN92kI-rX0-01900-00588349-00588849 and let the firmware or the platform deal with what is
+odN92kI-rX0-01901-00588999-00589240 the clock it's connected to power domain connected to--
+odN92kI-rX0-01902-00589240-00589461 - But it still won't ask the frequency for the device.
+odN92kI-rX0-01903-00589461-00589544 - Yes.
+odN92kI-rX0-01904-00589747-00590023 - So the idea is that for each device
+odN92kI-rX0-01905-00590253-00590389 you have a certain
+odN92kI-rX0-01906-00590389-00590716 number of resources related to it,
+odN92kI-rX0-01907-00590716-00591162 and then you can either control all these resources directly
+odN92kI-rX0-01908-00591162-00591573 or you can define power states say for the device,
+odN92kI-rX0-01909-00591573-00592025 and define those power States in terms of those resources.
+odN92kI-rX0-01910-00592025-00592207 So you say you say for example
+odN92kI-rX0-01911-00592207-00592538 power state one is this clock is gated.
+odN92kI-rX0-01912-00592659-00592987 Power state two is this clock is gated
+odN92kI-rX0-01913-00592987-00593476 and that regulator is in this state, right.
+odN92kI-rX0-01914-00593564-00593648 - Is it what he's trying to say then?
+odN92kI-rX0-01915-00593648-00593782 - Yeah I think so.
+odN92kI-rX0-01916-00593782-00594282 - So making it device specific or device centric--
+odN92kI-rX0-01917-00594355-00594795 - So the Linux or any hose running under the VM
+odN92kI-rX0-01918-00594795-00594969 would never actually specifically try
+odN92kI-rX0-01919-00594969-00595252 to change a clock or a power domain.
+odN92kI-rX0-01920-00595252-00595650 - They may not have that information if I'm trying to,
+odN92kI-rX0-01921-00595650-00595875 the question I have is that like,
+odN92kI-rX0-01922-00595875-00596165 in Linux how do we drive this?
+odN92kI-rX0-01923-00596165-00596505 So this is how we would like to drive
+odN92kI-rX0-01924-00596505-00596842 the specification towards because this removes
+odN92kI-rX0-01925-00596842-00597282 a lot of complexity but how do we drive this inland access?
+odN92kI-rX0-01926-00597282-00597572 My open question here, like do we just you reuse
+odN92kI-rX0-01927-00597572-00597777 the power gen pd power domain?
+odN92kI-rX0-01928-00597777-00598064 And since it already have this performance level
+odN92kI-rX0-01929-00598064-00598327 do we extend it or do we come up with some
+odN92kI-rX0-01930-00598327-00598639 different solution to all together?
+odN92kI-rX0-01931-00598639-00598914 I'm just opening that.
+odN92kI-rX0-01932-00598914-00599399 - Doesn't it just hook in, right in, into the structure?
+odN92kI-rX0-01933-00599399-00599578 All you have to do is specify the domain
+odN92kI-rX0-01934-00599578-00599964 and then you can do whatever you want on the back end part.
+odN92kI-rX0-01935-00599964-00600268 - Yeah, I'm just asking the question here,
+odN92kI-rX0-01936-00600268-00600651 so I'm fine it before we get into there,
+odN92kI-rX0-01937-00600651-00600957 I just want to get the opinion of how would we look this,
+odN92kI-rX0-01938-00600957-00601192 how do we want to solve this in Linux.
+odN92kI-rX0-01939-00601192-00601375 - So essentially, depending on the granularity
+odN92kI-rX0-01940-00601375-00601875 you want to kind of control things.
+odN92kI-rX0-01941-00602045-00602420 You can either define a PM domain like ACPI does,
+odN92kI-rX0-01942-00602420-00602920 with a set of operations in there
+odN92kI-rX0-01943-00602922-00603422 and that could be called like ACMI PM domain,
+odN92kI-rX0-01944-00603426-00603696 and then work pretty much the same way
+odN92kI-rX0-01945-00603696-00603962 as the ACPI PM domain does.
+odN92kI-rX0-01946-00603962-00604322 Or you hook up with Gen PD at a kind of lower level
+odN92kI-rX0-01947-00604322-00604822 but then you have to expose the power domain hierarchy
+odN92kI-rX0-01948-00604972-00605449 through the device tree, right, for that to work actually.
+odN92kI-rX0-01949-00605573-00605913 - In the idealistic world we want just the leaf node
+odN92kI-rX0-01950-00605913-00606203 not the hierarchy, leave the hierarchy to the firmware.
+odN92kI-rX0-01951-00606306-00606404 - Well yeah, something.
+odN92kI-rX0-01952-00606404-00606591 - But something to see if, I agree.
+odN92kI-rX0-01953-00606591-00606740 - A part of it, right, so it depends on
+odN92kI-rX0-01954-00606740-00607163 that you want to, where you want
+odN92kI-rX0-01955-00607163-00607663 to have your control plane essentially, right.
+odN92kI-rX0-01956-00608177-00608443 - So the idea is to have something either connect to
+odN92kI-rX0-01957-00608443-00608865 Gen PD or create this IC on my power domain
+odN92kI-rX0-01958-00608865-00609100 and create one for each device.
+odN92kI-rX0-01959-00609100-00609268 - Have you tried anything?
+odN92kI-rX0-01960-00609268-00609597 It's already-- - No, it's just new idea,
+odN92kI-rX0-01961-00609597-00609756 so I'm just trying to.
+odN92kI-rX0-01962-00609756-00610256 - So I guess, okay, so I guess my advice would be
+odN92kI-rX0-01963-00610364-00610675 to try to do it in this way and that way
+odN92kI-rX0-01964-00610675-00610900 and compare and then have an idea
+odN92kI-rX0-01965-00610900-00611107 what's simpler, what works for you
+odN92kI-rX0-01966-00611107-00611408 because at this point it's like we,
+odN92kI-rX0-01967-00611408-00611550 yeah, you can do you can do this,
+odN92kI-rX0-01968-00611550-00611857 you can do that, you can do something else.
+odN92kI-rX0-01969-00611857-00612301 It would be good to have like an example to look at, right.
+odN92kI-rX0-01970-00612463-00612913 - So another question that is kind of like my pet peeve
+odN92kI-rX0-01971-00612913-00613143 with the CMI, I just want to call it out here.
+odN92kI-rX0-01972-00613371-00613730 If ACMI is going to be adopted his arms position
+odN92kI-rX0-01973-00613730-00613914 or upstream position gonna be,
+odN92kI-rX0-01974-00613914-00614198 every former needs to follow ACMI going forward?
+odN92kI-rX0-01975-00614198-00614594 - Like at least for the solution space
+odN92kI-rX0-01976-00614594-00614809 we are addressing here, they should not come up
+odN92kI-rX0-01977-00614809-00615162 with something which is just a,
+odN92kI-rX0-01978-00615162-00615418 which makes no sense but no add on,
+odN92kI-rX0-01979-00615418-00615614 so that would, what I would say like--
+odN92kI-rX0-01980-00615614-00615757 - Because I see that the answer is--
+odN92kI-rX0-01981-00615862-00615997 - If it's just another protocol trying
+odN92kI-rX0-01982-00615997-00616162 to solve the same problem, I would say no,
+odN92kI-rX0-01983-00616162-00616499 but if it's addressing something out of there--
+odN92kI-rX0-01984-00616499-00616723 - Then I'm, okay, that's an okay pushing tab
+odN92kI-rX0-01985-00616723-00616993 but just one thing that I noticed was that
+odN92kI-rX0-01986-00616993-00617232 with all of these things ASCMI effectively makes
+odN92kI-rX0-01987-00617232-00617490 fast switching impossible on ARM for CPU frequency scaling.
+odN92kI-rX0-01988-00617490-00617737 - We do have that today, or in the latest.
+odN92kI-rX0-01989-00617957-00618232 - So we are having a separate protocol for CP frequency?
+odN92kI-rX0-01990-00618232-00618682 - No, we discovered that this has a fast transport
+odN92kI-rX0-01991-00618767-00619021 where it just says, this is the registers,
+odN92kI-rX0-01992-00619021-00619235 you can just write and it's--
+odN92kI-rX0-01993-00619235-00619512 - Okay, part of the pet peeve is gone, thank you.
+odN92kI-rX0-01994-00619512-00619595 - Yeah.
+odN92kI-rX0-01995-00619974-00620074 Any other questions?
+odN92kI-rX0-01996-00620469-00620581 No, nothing, okay.
+odN92kI-rX0-01997-00620994-00621219 (applause)
+odN92kI-rX0-01998-00624522-00624651 - Yeah, yeah it's a Windows laptop,
+odN92kI-rX0-01999-00624651-00624776 boo hiss, get used to it.
+odN92kI-rX0-02000-00625215-00625453 See if it wakes up, that would be funny, wouldn't it?
+odN92kI-rX0-02001-00625569-00625784 We're about to do a talk on suspend resume.
+odN92kI-rX0-02002-00625899-00626225 See how the other guys do.
+odN92kI-rX0-02003-00626225-00626573 It may not wake up, that would be freaking hysterical.
+odN92kI-rX0-02004-00627011-00627153 All right.
+odN92kI-rX0-02005-00627153-00627236 - It's waking up.
+odN92kI-rX0-02006-00627236-00627356 - Yeah, I can hear it.
+odN92kI-rX0-02007-00627356-00627568 (laughing)
+odN92kI-rX0-02008-00627568-00627767 - Yeah, I have a bunch of data on here that
+odN92kI-rX0-02009-00627960-00628110 can't really get to otherwise.
+odN92kI-rX0-02010-00628199-00628350 So my name's Lynn Brown.
+odN92kI-rX0-02011-00628350-00628554 It's not this, I don't really have slides,
+odN92kI-rX0-02012-00628554-00628730 I'm just going to show you some stuff.
+odN92kI-rX0-02013-00628849-00629131 I work at Intel Open Source Technology Center
+odN92kI-rX0-02014-00629131-00629281 along with a couple of guys in this room
+odN92kI-rX0-02015-00629281-00629620 if you don't know me, and today we're going to talk
+odN92kI-rX0-02016-00629620-00629921 about suspend resume quality in Linux.
+odN92kI-rX0-02017-00630372-00630678 So five years ago, who was at Linux plumbers in Seattle?
+odN92kI-rX0-02018-00631101-00631284 Okay, well then you may recognize some of this
+odN92kI-rX0-02019-00631284-00631539 because that's when we introduced this tool called,
+odN92kI-rX0-02020-00631662-00631905 at that time it was called Analyze Suspend.
+odN92kI-rX0-02021-00631905-00632185 And now the way we ship it is in
+odN92kI-rX0-02022-00632185-00632460 a part of this project called PM-Graph.
+odN92kI-rX0-02023-00632581-00632732 It also has bootgraph in it but today we're going
+odN92kI-rX0-02024-00632732-00633153 to talk just about the suspend graph part.
+odN92kI-rX0-02025-00633305-00633507 So what I'm going to do right now is actually go to
+odN92kI-rX0-02026-00633507-00633841 a web page that you can get to, it's 01.org PM-graph,
+odN92kI-rX0-02027-00633841-00634097 and I'm just going to highlight some stuff
+odN92kI-rX0-02028-00634097-00634287 from the getting started for those of you
+odN92kI-rX0-02029-00634287-00634576 that weren't in that session 5 years ago.
+odN92kI-rX0-02030-00634785-00635214 Very simple, oh good, you can actually see.
+odN92kI-rX0-02031-00635214-00635699 You can clone this, it's open source, make install run it,
+odN92kI-rX0-02032-00635699-00635899 run it as root, it's really that simple.
+odN92kI-rX0-02033-00636007-00636182 Does have some, this tool does have
+odN92kI-rX0-02034-00636309-00636725 some kernel dependencies, most distros have
+odN92kI-rX0-02035-00636725-00636967 all of these in there already,
+odN92kI-rX0-02036-00637097-00637241 and if you don't have all of them
+odN92kI-rX0-02037-00637241-00637499 and some pieces of functionality work
+odN92kI-rX0-02038-00637499-00637679 and some do not, I'll talk about that
+odN92kI-rX0-02039-00637679-00637858 a little bit more when we go through
+odN92kI-rX0-02040-00637858-00638013 examples of how the tool works.
+odN92kI-rX0-02041-00638311-00638787 And then basic usage, it'd be sort of like this,
+odN92kI-rX0-02042-00638953-00639235 you say sleepgraph - m mem would be for a suspend to mem.
+odN92kI-rX0-02043-00639235-00639648 If you do a freeze it would do a suspend to idle.
+odN92kI-rX0-02044-00639648-00639853 RTC wake says, after you've been,
+odN92kI-rX0-02045-00639959-00640066 actually right before you suspend ARM
+odN92kI-rX0-02046-00640066-00640279 the RTC to wake up in 15 seconds,
+odN92kI-rX0-02047-00640279-00640428 it can be whatever number you like,
+odN92kI-rX0-02048-00640428-00640617 that's how we usually run it.
+odN92kI-rX0-02049-00640617-00640838 You could run it without that for testing
+odN92kI-rX0-02050-00640838-00641308 other wake up sources, but for automated testing
+odN92kI-rX0-02051-00641308-00641496 that's what we use because we don't have
+odN92kI-rX0-02052-00641496-00641661 somebody there to press a button.
+odN92kI-rX0-02053-00641835-00641939 There's a bunch of options,
+odN92kI-rX0-02054-00641939-00642125 you can stick them into config files,
+odN92kI-rX0-02055-00642236-00642530 and I think now I'm going to show you some,
+odN92kI-rX0-02056-00642530-00642761 oh a couple of things we can do with it,
+odN92kI-rX0-02057-00642761-00642989 this is an example output, but you know what,
+odN92kI-rX0-02058-00642989-00643212 I'm going to go to a real webpage.
+odN92kI-rX0-02059-00643212-00643377 So it's a HTML page like this.
+odN92kI-rX0-02060-00643630-00643942 On the top it says, you know what it is,
+odN92kI-rX0-02061-00643942-00644221 this is run on a 9360 which there's probably some
+odN92kI-rX0-02062-00644221-00644615 in this room, it's a pretty good Linux laptop from Dell.
+odN92kI-rX0-02063-00644615-00645115 And suspend time was 247 milliseconds,
+odN92kI-rX0-02064-00645231-00645504 resume time was 3/4 of a second.
+odN92kI-rX0-02065-00645670-00645897 In purple it reports, it self-reports
+odN92kI-rX0-02066-00645897-00646108 how long it was in firmware.
+odN92kI-rX0-02067-00646108-00646325 I don't believe what it's reporting here, it's saying
+odN92kI-rX0-02068-00646325-00646679 one millisecond, but some of them are actually accurate.
+odN92kI-rX0-02069-00646907-00647070 So that's sort of informational.
+odN92kI-rX0-02070-00647070-00647241 And then what you're seeing on the screen
+odN92kI-rX0-02071-00647241-00647595 is a graph from left to right in time.
+odN92kI-rX0-02072-00647595-00647800 We initiated the suspend here about
+odN92kI-rX0-02073-00647800-00648051 a quarter second before the actual suspend,
+odN92kI-rX0-02074-00648051-00648340 SR this is the line between suspend resume.
+odN92kI-rX0-02075-00648340-00648584 And then all of this to the right
+odN92kI-rX0-02076-00648584-00648805 is how long it took us to wake up,
+odN92kI-rX0-02077-00648805-00649095 and this basically says where the time went, okay.
+odN92kI-rX0-02078-00649237-00649532 This is a suspend to mem.
+odN92kI-rX0-02079-00649679-00650154 In mem, which commonly is a set up to do an ACPI S3
+odN92kI-rX0-02080-00650154-00650444 but it could be a suspend to idle on another machine,
+odN92kI-rX0-02081-00650444-00650597 but what you'll see on the suspend to mem
+odN92kI-rX0-02082-00650597-00650924 is this blue and red, is the low level suspend
+odN92kI-rX0-02083-00650924-00651082 and resume, that you don't see was to see
+odN92kI-rX0-02084-00651082-00651273 with suspend to mem where you're offlining
+odN92kI-rX0-02085-00651273-00651418 your processors and so forth.
+odN92kI-rX0-02086-00651595-00651885 This is actually a pretty common scenario
+odN92kI-rX0-02087-00651885-00652385 right here, where you're big guy might be
+odN92kI-rX0-02088-00652402-00652902 nvme on suspend side, and USB is our nemesis
+odN92kI-rX0-02089-00653013-00653320 on the resume side for a typical system.
+odN92kI-rX0-02090-00653596-00654090 HD audio, not as much of a problem as it used to be,
+odN92kI-rX0-02091-00654090-00654340 and often graphics, I don't know if
+odN92kI-rX0-02092-00654340-00654777 this example right here has graphics on or off.
+odN92kI-rX0-02093-00654777-00654949 If the graphics was off when you suspended
+odN92kI-rX0-02094-00654949-00655089 then it will look different.
+odN92kI-rX0-02095-00655226-00655497 You can zoom in, you can zoom out, you can,
+odN92kI-rX0-02096-00655497-00655929 if we wanted to go look at our RS CPUs off lines,
+odN92kI-rX0-02097-00655929-00656079 they'd be in here for example.
+odN92kI-rX0-02098-00656417-00656672 Here's our machine suspend, machine resume,
+odN92kI-rX0-02099-00656672-00656900 calls into the BIOS, it's all here.
+odN92kI-rX0-02100-00657131-00657298 There's just a couple of other things
+odN92kI-rX0-02101-00657298-00657434 that I'd like to point out with this tool,
+odN92kI-rX0-02102-00657434-00657570 and then I'm going to talk about
+odN92kI-rX0-02103-00657570-00657814 why I'm talking about this tool today,
+odN92kI-rX0-02104-00657814-00658028 which is what we want to do with it.
+odN92kI-rX0-02105-00658119-00658291 Actually there's, I think there's an example
+odN92kI-rX0-02106-00658291-00658611 on the web page, at one point we were very concerned
+odN92kI-rX0-02107-00658611-00658962 about back to back suspend resume, so then an X two.
+odN92kI-rX0-02108-00659066-00659506 Sometimes there, to make suspend go faster, some things
+odN92kI-rX0-02109-00659506-00659876 got deferred to the right side of that yellow bar,
+odN92kI-rX0-02110-00659876-00660031 and you can't resume until those things
+odN92kI-rX0-02111-00660031-00660245 have finished, you can't suspend again til
+odN92kI-rX0-02112-00660245-00660373 those things have finished resuming.
+odN92kI-rX0-02113-00660373-00660861 So this can be interesting, particularly in the case
+odN92kI-rX0-02114-00660861-00661123 where you have a low latency, you wake up
+odN92kI-rX0-02115-00661123-00661557 from Network, you want to wake up, wake up, wake up,
+odN92kI-rX0-02116-00661557-00661799 well if you're still, you can't come back to sleep,
+odN92kI-rX0-02117-00661799-00661994 you can't, if you're spending all your time
+odN92kI-rX0-02118-00661994-00662381 suspend resuming, then you can't get any work done,
+odN92kI-rX0-02119-00662381-00662525 you're basically on all the time.
+odN92kI-rX0-02120-00662525-00662953 So we need fast suspend and resume and nothing deferred
+odN92kI-rX0-02121-00662953-00663199 in between them that delays the next one.
+odN92kI-rX0-02122-00663411-00663601 That's pretty important on some machines.
+odN92kI-rX0-02123-00663601-00663954 On a laptop, if something takes under a second
+odN92kI-rX0-02124-00663954-00664099 most people are pretty happy.
+odN92kI-rX0-02125-00664255-00664510 But I'm showing you sort of the success cases here.
+odN92kI-rX0-02126-00664607-00664789 There are a lot of non-success cases.
+odN92kI-rX0-02127-00664789-00665061 And with that, I'm going to go to a summary.
+odN92kI-rX0-02128-00665061-00665561 So first I have to apologize, we usually put our summary
+odN92kI-rX0-02129-00665679-00665911 into Google sheet and this has been exported
+odN92kI-rX0-02130-00665911-00666101 so that I could show it on this kind of laptop
+odN92kI-rX0-02131-00666101-00666352 and I'm really pretty bad at Windows.
+odN92kI-rX0-02132-00666547-00666776 But what this is, is a list of machines
+odN92kI-rX0-02133-00667066-00667389 and it's the output of a number of tests.
+odN92kI-rX0-02134-00667389-00667624 So say there's a Dell 9380 at the top,
+odN92kI-rX0-02135-00667853-00668193 and line 2 is, we ran freeze for 24 hours
+odN92kI-rX0-02136-00668303-00668803 and it suspended and resumed 2500 times,
+odN92kI-rX0-02137-00668825-00669192 similarly with mem, 2700 times.
+odN92kI-rX0-02138-00669192-00669446 This particular test setup we run for 24 hours
+odN92kI-rX0-02139-00669446-00669631 and however many we fit in there.
+odN92kI-rX0-02140-00669631-00669914 Another test lab we have we run
+odN92kI-rX0-02141-00669914-00670093 generally around 2,000 cycles.
+odN92kI-rX0-02142-00670406-00670802 Let's see, this health, we have a summary where we
+odN92kI-rX0-02143-00670922-00671127 sort of wait different things, like if it crashed,
+odN92kI-rX0-02144-00671127-00671310 that would be that would really
+odN92kI-rX0-02145-00671310-00671545 impact this health, so we could sort by health.
+odN92kI-rX0-02146-00671677-00671881 I do have machines which are 100% healthy,
+odN92kI-rX0-02147-00671881-00672009 they're just not on here.
+odN92kI-rX0-02148-00672449-00672753 And let's see, pass fail and hang,
+odN92kI-rX0-02149-00672753-00673153 so one way you can run the tool is
+odN92kI-rX0-02150-00673153-00673459 when we run the tool is, we would run it say
+odN92kI-rX0-02151-00673459-00673959 for 24 hours on a system which has no state, and so on disk,
+odN92kI-rX0-02152-00673998-00674272 and so when it resumes then we pull the data off.
+odN92kI-rX0-02153-00674272-00674656 Well if it failed to resume, then we've lost the results.
+odN92kI-rX0-02154-00674789-00675114 And so from our point of view, that's a hang
+odN92kI-rX0-02155-00675114-00675314 and there's no result to look at,
+odN92kI-rX0-02156-00675314-00675614 which is pretty annoying whereas on the ones we have
+odN92kI-rX0-02157-00675614-00675922 with the disk, you can go back, you can reboot the machine,
+odN92kI-rX0-02158-00675922-00676272 you still have your your system results there.
+odN92kI-rX0-02159-00676454-00676736 Back of C 10, is this LPIR, so some details.
+odN92kI-rX0-02160-00676736-00676973 This, probably a graphic is better for this.
+odN92kI-rX0-02161-00676973-00677415 So this is suspend max medium and min time.
+odN92kI-rX0-02162-00677415-00677884 So this first one, the maximum suspend time was 5.8 seconds,
+odN92kI-rX0-02163-00677884-00678034 these are all in milliseconds.
+odN92kI-rX0-02164-00678129-00678629 Resume, 7 seconds, usually 2 seconds and a 1/2 a second
+odN92kI-rX0-02165-00678728-00678883 for resume, for that particular machine.
+odN92kI-rX0-02166-00678883-00679237 It's sort of a new machine which still has some quirks.
+odN92kI-rX0-02167-00679385-00679638 All right, so what do we do with this?
+odN92kI-rX0-02168-00679638-00679853 We have this whole huge list of machines,
+odN92kI-rX0-02169-00679853-00680089 actually we have a list a lot longer than this,
+odN92kI-rX0-02170-00680089-00680260 I only have public machines here.
+odN92kI-rX0-02171-00680380-00680634 So there's a, we happen to have in our lab,
+odN92kI-rX0-02172-00680634-00680923 this is an organ lab we, have a couple of Dells,
+odN92kI-rX0-02173-00680923-00681423 thank you Dell, and I think a Galaxy Book is a Samsung,
+odN92kI-rX0-02174-00681463-00681950 thanks Samsung, and some HPs, there's a MacBook in there,
+odN92kI-rX0-02175-00682150-00682360 and then a couple of Lenovo machines.
+odN92kI-rX0-02176-00682763-00682977 And we've scrubbed all of these results
+odN92kI-rX0-02177-00682977-00683206 which have been running for 24 hours a day.
+odN92kI-rX0-02178-00683206-00683657 And in priority order, the first thing we do is
+odN92kI-rX0-02179-00683657-00683910 grep for bad things in D message,
+odN92kI-rX0-02180-00683910-00684327 and so say on this 9360, this DMAR,
+odN92kI-rX0-02181-00684436-00684862 this happened 2800 times out of 28 tests, so.
+odN92kI-rX0-02182-00684996-00685253 What we're trying to do when we are casting this net
+odN92kI-rX0-02183-00685253-00685504 and looking for failures, is to find out
+odN92kI-rX0-02184-00685504-00685739 what happens all the time, what's intermittent.
+odN92kI-rX0-02185-00685865-00686135 If I wanted to bug something, where do I go to bug it.
+odN92kI-rX0-02186-00686325-00686553 If I ran more tests on this machine
+odN92kI-rX0-02187-00686553-00686723 would I learn any more, or is it better
+odN92kI-rX0-02188-00686723-00686964 to run on that machine over there, okay.
+odN92kI-rX0-02189-00687052-00687536 And what we're learning is that after around 2000 tests
+odN92kI-rX0-02190-00687536-00687659 you pretty much got a good idea
+odN92kI-rX0-02191-00687659-00687888 what this machine is going to do.
+odN92kI-rX0-02192-00687888-00688283 I remember the day when I was,
+odN92kI-rX0-02193-00688283-00688463 actually it was wasn't too long ago,
+odN92kI-rX0-02194-00688463-00688694 where doing two suspend resumes on Linux
+odN92kI-rX0-02195-00688694-00688931 was a big freaking deal, like if you could do it
+odN92kI-rX0-02196-00688931-00689219 and then you did it again, you're like yes.
+odN92kI-rX0-02197-00689219-00689567 Now it's like 2000 and oh, somebody else
+odN92kI-rX0-02198-00689567-00689890 wants the machine for something else, 'cause.
+odN92kI-rX0-02199-00689890-00690194 So really what we want to do now is
+odN92kI-rX0-02200-00690194-00690369 find more faults from more machines.
+odN92kI-rX0-02201-00690369-00690761 But let me scroll down, so these happen 100% of the time,
+odN92kI-rX0-02202-00690761-00691195 but down here you'll get, sorry,
+odN92kI-rX0-02203-00691195-00691341 like I said I'm bad at this,
+odN92kI-rX0-02204-00691531-00691819 you'll get a lot of things that out of say
+odN92kI-rX0-02205-00692165-00692574 1.08% of the time, we got some error message
+odN92kI-rX0-02206-00692574-00693033 on say this Dell, okay, so the BIOS basically freaked out.
+odN92kI-rX0-02207-00693033-00693222 We have no idea, it looks like a fan
+odN92kI-rX0-02208-00693222-00693486 tried to turn itself off at some point,
+odN92kI-rX0-02209-00693486-00693986 but it only happened 0.08% of the time, and that's
+odN92kI-rX0-02210-00694043-00694469 a tough one, that's a tough path to follow to debug that.
+odN92kI-rX0-02211-00694469-00694969 So we cast this net and we filed a bunch of Bugzillas.
+odN92kI-rX0-02212-00695072-00695213 This is an example of one.
+odN92kI-rX0-02213-00695213-00695310 By the way, I should say about
+odN92kI-rX0-02214-00695310-00695510 all the stuff in this spreadsheet, Paul?
+odN92kI-rX0-02215-00695663-00695963 - I'm just curious, is there some way
+odN92kI-rX0-02216-00695963-00696119 you to accelerate the failure rate?
+odN92kI-rX0-02217-00696119-00696276 For example, you mentioned the fan.
+odN92kI-rX0-02218-00696276-00696530 Is it possible just to force a fan on and off at odd times
+odN92kI-rX0-02219-00696530-00696674 while you're suspending or resuming?
+odN92kI-rX0-02220-00696674-00696811 - Yeah, so that's a good question.
+odN92kI-rX0-02221-00696811-00697120 So when you have a bug that's difficult to reproduce,
+odN92kI-rX0-02222-00697120-00697421 how do you force it to happen more often?
+odN92kI-rX0-02223-00697421-00697540 (speaking faintly)
+odN92kI-rX0-02224-00697540-00697779 In that case, yeah, in that case, so for the fan
+odN92kI-rX0-02225-00697779-00697930 if you were debugging that one in particular,
+odN92kI-rX0-02226-00697930-00698201 which we do have some work going on in that,
+odN92kI-rX0-02227-00698201-00698376 on that particular thing right now.
+odN92kI-rX0-02228-00698583-00698832 Yeah, we could perhaps provoke that failure
+odN92kI-rX0-02229-00698832-00698986 without even suspending and resuming,
+odN92kI-rX0-02230-00698986-00699145 just by turning the device on and off,
+odN92kI-rX0-02231-00699145-00699490 and with runtime device power management we can do that.
+odN92kI-rX0-02232-00699619-00699979 This is one that actually, I like this example
+odN92kI-rX0-02233-00699979-00700184 the best because it's something that's,
+odN92kI-rX0-02234-00700184-00700420 it's on a bunch of different machines.
+odN92kI-rX0-02235-00700420-00700585 So basically you may see in some of our bugzillas
+odN92kI-rX0-02236-00700585-00700896 we suck a cookie, basically we have a script,
+odN92kI-rX0-02237-00700896-00701108 it goes up to bugzilla and says,
+odN92kI-rX0-02238-00701108-00701485 oh do I see that bug in this test results, okay.
+odN92kI-rX0-02239-00701485-00701715 And so say I were looking for this bug I'd say,
+odN92kI-rX0-02240-00701715-00702065 well on this Dell 9370 which is a cab u leik refresh
+odN92kI-rX0-02241-00702065-00702424 I saw it 400 times, which is 17% of the time,
+odN92kI-rX0-02242-00702424-00702753 that's a good machine to debug this problem, okay.
+odN92kI-rX0-02243-00702753-00702979 But it's, this is the sound one,
+odN92kI-rX0-02244-00703081-00703335 very intermittent but you see all these other machines,
+odN92kI-rX0-02245-00703335-00703457 it's like whoa, okay this thing
+odN92kI-rX0-02246-00703457-00703825 is happening .07 times on this, on say the 9360.
+odN92kI-rX0-02247-00704062-00704472 And we have, we really had no idea.
+odN92kI-rX0-02248-00704472-00704659 This one is actually a 5.0 result,
+odN92kI-rX0-02249-00704659-00704865 we've actually fixed this.
+odN92kI-rX0-02250-00704865-00705205 But then there's other bugs where it disappeared,
+odN92kI-rX0-02251-00705205-00705364 like on these particular machines,
+odN92kI-rX0-02252-00705364-00705654 this bug we found none of the test results,
+odN92kI-rX0-02253-00705654-00705923 all 0% of the time found this failure.
+odN92kI-rX0-02254-00706111-00706341 And these are just what we have happened to have documented,
+odN92kI-rX0-02255-00706341-00706841 and then over here, this PCEI port device bug,
+odN92kI-rX0-02256-00706847-00707146 this happens 100% of the time on the 9380.
+odN92kI-rX0-02257-00707547-00707773 So one thing that we can do is
+odN92kI-rX0-02258-00707773-00708111 we can look and see, okay if I want to,
+odN92kI-rX0-02259-00708111-00708522 that's the kernel issue, so we've got,
+odN92kI-rX0-02260-00708522-00708693 so we're scraping for kernel issues,
+odN92kI-rX0-02261-00708693-00708966 we're scraping for file bugs, and now
+odN92kI-rX0-02262-00708966-00709166 we're looking for slow devices.
+odN92kI-rX0-02263-00709166-00709450 I didn't really show except for USB
+odN92kI-rX0-02264-00709450-00709641 that some devices sometimes are very slow.
+odN92kI-rX0-02265-00709641-00709860 I think there's an example in here
+odN92kI-rX0-02266-00709860-00710055 where something took 50 seconds, in that amount of time
+odN92kI-rX0-02267-00710055-00710272 somebody's going to press the button,
+odN92kI-rX0-02268-00710272-00710427 right, so catastrophic failure.
+odN92kI-rX0-02269-00710525-00710848 And here in fact we have sorted the worst suspend devices.
+odN92kI-rX0-02270-00710848-00711222 Here's one, 32 seconds for this device
+odN92kI-rX0-02271-00711222-00711581 on a MacBook Pro and it's the nouveau driver,
+odN92kI-rX0-02272-00711906-00712043 so not sure what to say about that.
+odN92kI-rX0-02273-00712043-00712543 But the interesting thing is that the average time,
+odN92kI-rX0-02274-00712664-00712865 so the average time was one second
+odN92kI-rX0-02275-00712865-00713118 but one, one time, it took 32 seconds
+odN92kI-rX0-02276-00713203-00713554 out of 6,000 times that it was the worst device.
+odN92kI-rX0-02277-00713638-00713807 And you can go all the way down.
+odN92kI-rX0-02278-00713807-00714071 This is, we can sort all of our devices
+odN92kI-rX0-02279-00714071-00714389 to I think, where we'd do the cut off at very low
+odN92kI-rX0-02280-00714641-00715012 for some of these other, like CPU on and off,
+odN92kI-rX0-02281-00715012-00715208 for some reason they're very slow on the Mac.
+odN92kI-rX0-02282-00715208-00715523 I think they've got their MTRS set up wrong.
+odN92kI-rX0-02283-00715523-00715956 Or this is a Browsewell, that's an old slow machine.
+odN92kI-rX0-02284-00715956-00716401 So we can say, okay from the population of failures
+odN92kI-rX0-02285-00716401-00716583 which ones are the most dramatic.
+odN92kI-rX0-02286-00716736-00716922 We can see how often they happen.
+odN92kI-rX0-02287-00717083-00717282 And say I go to worst resume devices,
+odN92kI-rX0-02288-00717282-00717576 and then we can say, here's the 915,
+odN92kI-rX0-02289-00717576-00717852 I could say, well for all the results I have,
+odN92kI-rX0-02290-00717852-00718071 for the million results I have,
+odN92kI-rX0-02291-00718071-00718351 on what machines is the 915 the slowest device.
+odN92kI-rX0-02292-00718351-00718851 Oh okay, so I can scroll over and say, oh well it's great.
+odN92kI-rX0-02293-00718878-00719201 This one right here is the machine I want to debug
+odN92kI-rX0-02294-00719201-00719701 to 915 on because basically 3,000 times
+odN92kI-rX0-02295-00719746-00719870 that's the slowest device.
+odN92kI-rX0-02296-00719870-00720336 So if I want to work on that device, it's easy to sort them.
+odN92kI-rX0-02297-00720336-00720785 So anyway, so what we have is we have this capability.
+odN92kI-rX0-02298-00720785-00720954 A, you should be running it, particularly
+odN92kI-rX0-02299-00720954-00721254 if you're concerned about suspend resume performance,
+odN92kI-rX0-02300-00721254-00721445 particularly if you're concerned about it
+odN92kI-rX0-02301-00721445-00721643 on a particular laptop you may have.
+odN92kI-rX0-02302-00721643-00721793 B, you should be filing bugs.
+odN92kI-rX0-02303-00722169-00722380 C, we are doing this automatically
+odN92kI-rX0-02304-00722380-00722708 and we are we have moved from the,
+odN92kI-rX0-02305-00722708-00723014 does it work and to how fast it is
+odN92kI-rX0-02306-00723014-00723313 into the how reliable is it stage.
+odN92kI-rX0-02307-00723313-00723538 We still have problems where it fails,
+odN92kI-rX0-02308-00723538-00723792 we still have problems where it's slow,
+odN92kI-rX0-02309-00723792-00724140 but now we're, gee does it run all night stage.
+odN92kI-rX0-02310-00724140-00724311 And for the most part it does.
+odN92kI-rX0-02311-00724496-00724878 And now the thing that we've discovered is
+odN92kI-rX0-02312-00724878-00725084 we have 30 machines but what we need
+odN92kI-rX0-02313-00725084-00725583 is 3500 machines, and my lab has a budget
+odN92kI-rX0-02314-00725583-00725907 but it's not going to have 3500 machines.
+odN92kI-rX0-02315-00725907-00726280 So what I'd really like is, I'd really like this test
+odN92kI-rX0-02316-00726280-00726594 run by say Ubuntu, like opt in,
+odN92kI-rX0-02317-00726594-00726916 send interesting information back to Ubuntu
+odN92kI-rX0-02318-00726916-00727270 and have these results squirreled away
+odN92kI-rX0-02319-00727270-00727475 when I, when your random Ubuntu user
+odN92kI-rX0-02320-00727475-00727745 does their a suspend resume somewhere
+odN92kI-rX0-02321-00727745-00728029 where somebody can debug them and have their finger
+odN92kI-rX0-02322-00728029-00728300 on the pulse of how is Linux suspend resume
+odN92kI-rX0-02323-00728300-00728428 working out in the wild.
+odN92kI-rX0-02324-00728529-00728693 We can observe that and not only that,
+odN92kI-rX0-02325-00728693-00728830 maybe even have enough information
+odN92kI-rX0-02326-00728830-00729171 that we can go debug it and attack it.
+odN92kI-rX0-02327-00729171-00729364 So I'm interested to know if anybody
+odN92kI-rX0-02328-00729364-00729670 has any insight into public telemetry
+odN92kI-rX0-02329-00729799-00729939 because that's what I really want.
+odN92kI-rX0-02330-00729939-00730154 We're really good on the private telemetry.
+odN92kI-rX0-02331-00730265-00730583 - Are you publishing this data already?
+odN92kI-rX0-02332-00730583-00730830 - Not until today am i publishing this data.
+odN92kI-rX0-02333-00730830-00731330 So say like if I were looking at this,
+odN92kI-rX0-02334-00731699-00731972 say this bug report, right, so yeah.
+odN92kI-rX0-02335-00731972-00732383 We let people know that we have this
+odN92kI-rX0-02336-00732383-00732811 and particularly in the bug reports that we find,
+odN92kI-rX0-02337-00732811-00733065 yes, I'm seeing this on the following 18 machines
+odN92kI-rX0-02338-00733065-00733335 or I'm seeing this on this one weird machine over there.
+odN92kI-rX0-02339-00733555-00734009 But it's not in some public place where say
+odN92kI-rX0-02340-00734009-00734217 I got run over by a bus, somebody else could say,
+odN92kI-rX0-02341-00734217-00734457 oh yeah and let's go work on the number one bug.
+odN92kI-rX0-02342-00734677-00734822 If we weren't pushing this, I don't know if we,
+odN92kI-rX0-02343-00734822-00734986 we don't really have a lot of community,
+odN92kI-rX0-02344-00734986-00735221 we have a bunch, a couple of people at Intel working
+odN92kI-rX0-02345-00735221-00735668 at this but I'd like to to be broader than that.
+odN92kI-rX0-02346-00735668-00735873 I'd like to leverage the diversity
+odN92kI-rX0-02347-00735873-00736276 of the community for a number of reasons, one is
+odN92kI-rX0-02348-00736276-00736549 because the community has a lot more laptops than I have.
+odN92kI-rX0-02349-00736753-00736973 Even if all the community did was report bugs,
+odN92kI-rX0-02350-00736973-00737158 that would be a huge boon to us
+odN92kI-rX0-02351-00737158-00737387 because we could say, oh my gosh, look at
+odN92kI-rX0-02352-00737387-00737677 all these bugs that are appearing on this class of machine.
+odN92kI-rX0-02353-00737916-00738110 So any ideas, thoughts?
+odN92kI-rX0-02354-00738110-00738307 - There is the kernel CI thing, right,
+odN92kI-rX0-02355-00738307-00738690 that they have like a distributed way of doing tests.
+odN92kI-rX0-02356-00738690-00739133 I've seen that, I mean it's not power management specific,
+odN92kI-rX0-02357-00739133-00739484 but they have some sort of testing.
+odN92kI-rX0-02358-00739484-00739857 - So I mean I've talked about this with a couple of people,
+odN92kI-rX0-02359-00739857-00740296 and say in Seattle, I said, hey we have this tool
+odN92kI-rX0-02360-00740296-00740462 and the guys in this room, the people that care
+odN92kI-rX0-02361-00740462-00740669 about power management, go run it.
+odN92kI-rX0-02362-00740669-00740752 And some people said, sure.
+odN92kI-rX0-02363-00740752-00740913 I mean you could all have downloaded it and run it
+odN92kI-rX0-02364-00740913-00741157 in the time I've been talking, nobody has.
+odN92kI-rX0-02365-00741315-00741698 And very few people have sent the results back to me
+odN92kI-rX0-02366-00741698-00741996 or anybody that can debug them.
+odN92kI-rX0-02367-00742152-00742487 So I think we're now past the stage where,
+odN92kI-rX0-02368-00742487-00742717 hey we'll get the community, I'm going to
+odN92kI-rX0-02369-00742717-00742800 volunteer my system for testing.
+odN92kI-rX0-02370-00742800-00743110 No, what we really want is, during the day
+odN92kI-rX0-02371-00743110-00743333 I've suspended and resumed eight times today,
+odN92kI-rX0-02372-00743333-00743546 I could have sent eight results back, right.
+odN92kI-rX0-02373-00743546-00743680 - Oh I see.
+odN92kI-rX0-02374-00743680-00743944 - Yeah, automatically by opting in, right,
+odN92kI-rX0-02375-00743944-00744174 by opting in, or maybe 1/8 of them
+odN92kI-rX0-02376-00744174-00744590 or some sampling, some number larger than zero
+odN92kI-rX0-02377-00744590-00744782 and that would be really useful
+odN92kI-rX0-02378-00744782-00745081 because this is just a table of the 15 machines
+odN92kI-rX0-02379-00745081-00745260 that we have right here and we've found
+odN92kI-rX0-02380-00745260-00745634 some really interesting issues
+odN92kI-rX0-02381-00745634-00745815 that you would never, you'd say,
+odN92kI-rX0-02382-00745921-00746201 in your own laptop, if it didn't work 100% of the time,
+odN92kI-rX0-02383-00746201-00746344 you're like, I must have done something wrong
+odN92kI-rX0-02384-00746344-00746628 or maybe I sat on my laptop, whatever, reset it, but no.
+odN92kI-rX0-02385-00746628-00747004 If it recorded it, we'll see one out of 2,000,
+odN92kI-rX0-02386-00747004-00747319 we'll see that, oh wait, that's a real failure.
+odN92kI-rX0-02387-00747319-00747502 And if we can nail that one out of 2,000
+odN92kI-rX0-02388-00747502-00747733 then people will really start to trust
+odN92kI-rX0-02389-00747825-00747959 Suspend Resume even more.
+odN92kI-rX0-02390-00748167-00748437 So I want to have more machines with a 100% health
+odN92kI-rX0-02391-00748437-00748637 for 2,000 tests, that's my goal.
+odN92kI-rX0-02392-00748777-00749107 I have some but it's a very short list.
+odN92kI-rX0-02393-00749107-00749358 Everyone has some blemish of some minute,
+odN92kI-rX0-02394-00749491-00749813 100% perfect is a high bar.
+odN92kI-rX0-02395-00749813-00750073 So yeah, I'm thinking opt in,
+odN92kI-rX0-02396-00750337-00750470 and where does the results get stored,
+odN92kI-rX0-02397-00750470-00750701 I should say that the results are like about,
+odN92kI-rX0-02398-00750701-00750973 it's about 100K for one of these webpages.
+odN92kI-rX0-02399-00750973-00751363 Yes, it's compressible, we can save the data
+odN92kI-rX0-02400-00751363-00751627 without generating HTML to squeeze it down.
+odN92kI-rX0-02401-00751725-00751972 Or we could just send the summaries.
+odN92kI-rX0-02402-00751972-00752185 We have a script that will generate
+odN92kI-rX0-02403-00752185-00752513 a Google Doc of a Google sheet of a summary
+odN92kI-rX0-02404-00752513-00752902 of an entire directory of thousands of results.
+odN92kI-rX0-02405-00752902-00753171 So if we could even just upload them somewhere
+odN92kI-rX0-02406-00753171-00753365 and then you could run your favorite scraping tool
+odN92kI-rX0-02407-00753365-00753552 to go look for problems, that would work,
+odN92kI-rX0-02408-00753552-00753764 but we need a repository and we need
+odN92kI-rX0-02409-00753764-00754193 community uploading those results.
+odN92kI-rX0-02410-00754193-00754401 - I mean, I think it's a great idea,
+odN92kI-rX0-02411-00754401-00754769 it's just that, how do you think
+odN92kI-rX0-02412-00754877-00755079 you're gonna solve the problem
+odN92kI-rX0-02413-00755079-00755486 having tests done on distro kernels?
+odN92kI-rX0-02414-00755603-00755873 - On distro kernels, this will run on a distro kernel.
+odN92kI-rX0-02415-00755997-00756265 - No, I meant like, you may see a problem
+odN92kI-rX0-02416-00756265-00756493 that's only reproducible on kernels.
+odN92kI-rX0-02417-00756596-00756796 - Okay, yeah, so most of the time it doesn't matter.
+odN92kI-rX0-02418-00756796-00757010 I mean at this point, there was a time
+odN92kI-rX0-02419-00757010-00757291 when it was pretty frantic, getting suspend resume running,
+odN92kI-rX0-02420-00757291-00757791 now it's usually a device, like there's a couple of,
+odN92kI-rX0-02421-00757877-00758041 I won't make fun of them, but there's a couple of
+odN92kI-rX0-02422-00758041-00758281 choice devices that our OEMs choose
+odN92kI-rX0-02423-00758281-00758751 to save like 10 cents, and yeah.
+odN92kI-rX0-02424-00758879-00759300 So usually it's a device, and that's a device driver,
+odN92kI-rX0-02425-00759300-00759598 and the distros, the same driver upstream has, so yeah.
+odN92kI-rX0-02426-00759598-00759800 It's more information.
+odN92kI-rX0-02427-00759800-00760018 Did I just turn myself off?
+odN92kI-rX0-02428-00760018-00760101 - You did.
+odN92kI-rX0-02429-00760188-00760688 - How did I do that, sorry.
+odN92kI-rX0-02430-00760706-00761069 (speaking faintly)
+odN92kI-rX0-02431-00761069-00761259 - She is coming.
+odN92kI-rX0-02432-00761259-00761382 - My phone just talked to me too.
+odN92kI-rX0-02433-00761382-00761641 Everything's malfunctioning up here.
+odN92kI-rX0-02434-00761641-00761961 (speaking faintly)
+odN92kI-rX0-02435-00761961-00762051 - Is that on, yes.
+odN92kI-rX0-02436-00762207-00762301 - It's always the battery, it's the first.
+odN92kI-rX0-02437-00762301-00762593 - I wanted to comment, this is inspiring,
+odN92kI-rX0-02438-00762593-00762918 that's a good, I'm impressed by this project.
+odN92kI-rX0-02439-00762918-00763227 I didn't think about this and it's a great idea.
+odN92kI-rX0-02440-00763227-00763566 You obviously may want to talk to distro people,
+odN92kI-rX0-02441-00763566-00763917 this has to go into the default images
+odN92kI-rX0-02442-00763917-00764360 of the most popular distribution.
+odN92kI-rX0-02443-00764360-00764619 This is one thing that I can think of.
+odN92kI-rX0-02444-00764619-00765035 Obviously there is this, should I send back
+odN92kI-rX0-02445-00765035-00765281 the bug report, to somebody is more sensitive
+odN92kI-rX0-02446-00765281-00765450 to this than others, but I don't know,
+odN92kI-rX0-02447-00765450-00765875 Firefox sent on telemetry back home and people accept that.
+odN92kI-rX0-02448-00765875-00766339 I was thinking that a privileged venu
+odN92kI-rX0-02449-00766339-00766748 to advertise this kind of work is the FOSDEM Conference
+odN92kI-rX0-02450-00766748-00767248 in Brussels in Europe, it's a, stands for
+odN92kI-rX0-02451-00767250-00767550 Free and Open Source Developer European Meeting,
+odN92kI-rX0-02452-00767550-00767746 and it's a place where this kind of stuff
+odN92kI-rX0-02453-00767746-00767999 is really taken to heart,
+odN92kI-rX0-02454-00767999-00768201 like you would have people cheering for you
+odN92kI-rX0-02455-00768201-00768355 if you show them this because they say,
+odN92kI-rX0-02456-00768355-00768677 yes this is something I can really contribute to just,
+odN92kI-rX0-02457-00768677-00769030 and the suspend resume affects everybody
+odN92kI-rX0-02458-00769030-00769122 and everybody hates it.
+odN92kI-rX0-02459-00769122-00769309 Like when I don't see my laptop coming back,
+odN92kI-rX0-02460-00769309-00769504 it's not something like,
+odN92kI-rX0-02461-00769504-00769727 I don't know, my network doesn't do IPV6,
+odN92kI-rX0-02462-00769727-00769955 yeah I mean, I am I on the--
+odN92kI-rX0-02463-00769955-00770150 - [Lynn] Very visible, right, some people think
+odN92kI-rX0-02464-00770150-00770335 suspend resume and PM are synonymous.
+odN92kI-rX0-02465-00770461-00770547 - But suspend resume, it's really
+odN92kI-rX0-02466-00770547-00770687 something like, it's mission critical.
+odN92kI-rX0-02467-00770687-00770860 So people are going to be enthusiastic about this
+odN92kI-rX0-02468-00770860-00771064 if you or some of your collaborators
+odN92kI-rX0-02469-00771064-00771243 can go and present there and have,
+odN92kI-rX0-02470-00771243-00771578 and also you will meet a lot of distro people there.
+odN92kI-rX0-02471-00771578-00771938 Like so you will have people approaching you saying,
+odN92kI-rX0-02472-00772025-00772211 we should do this for fedora,
+odN92kI-rX0-02473-00772211-00772358 we should do this for openSUSE,
+odN92kI-rX0-02474-00772358-00772519 we should do this for Ubuntu.
+odN92kI-rX0-02475-00772667-00772882 - Basdom, okay, good suggestion, thank you.
+odN92kI-rX0-02476-00773067-00773399 Yeah so if any, does any, first of all I should,
+odN92kI-rX0-02477-00773506-00773756 does anybody have any questions about the tool itself
+odN92kI-rX0-02478-00773756-00774062 or its results, because I just sort of
+odN92kI-rX0-02479-00774062-00774237 skipped through that a little bit,
+odN92kI-rX0-02480-00774237-00774438 but you can all look at that,
+odN92kI-rX0-02481-00774438-00774612 you can run it and look at your own results.
+odN92kI-rX0-02482-00774612-00774846 It creates an HTML file, you open the file
+odN92kI-rX0-02483-00774846-00775035 and you have what's run the screen right, here so.
+odN92kI-rX0-02484-00775035-00775156 - I think you mentioned already
+odN92kI-rX0-02485-00775156-00775447 but does it have an option to output
+odN92kI-rX0-02486-00775447-00775752 a computer readable format?
+odN92kI-rX0-02487-00775885-00776038 - Well it's--
+odN92kI-rX0-02488-00776038-00776184 - Or is it only HTML?
+odN92kI-rX0-02489-00776184-00776684 - It creates, it saves the D message in an F trace log
+odN92kI-rX0-02490-00776741-00777101 and a tool specific log, and then a separate program
+odN92kI-rX0-02491-00777101-00777266 comes and renders this from that.
+odN92kI-rX0-02492-00777369-00777598 So you could grep through it, if you wanted
+odN92kI-rX0-02493-00777598-00777808 another tool, you could do that too, yeah.
+odN92kI-rX0-02494-00778204-00778369 A couple, get the mic back there.
+odN92kI-rX0-02495-00779123-00779357 - I think that there's some privacy implications
+odN92kI-rX0-02496-00779357-00779597 for this of course, so have you thought about
+odN92kI-rX0-02497-00779597-00779772 what sort of things you'd be filtering
+odN92kI-rX0-02498-00779897-00780101 from D message that would be sent up with this yet?
+odN92kI-rX0-02499-00780101-00780601 - So yes, so like if, like we save the D message here,
+odN92kI-rX0-02500-00780693-00780903 I guess maybe, how can I make that bigger,
+odN92kI-rX0-02501-00781044-00781275 we save the D message from when you start to suspend
+odN92kI-rX0-02502-00781275-00781572 to when your resume to see if something jumps out.
+odN92kI-rX0-02503-00781572-00781853 We have some other logs involved,
+odN92kI-rX0-02504-00781853-00782229 a tool specific log that grabs some stuff
+odN92kI-rX0-02505-00782229-00782379 from the system that's useful.
+odN92kI-rX0-02506-00782517-00782875 There is an option to trace processes
+odN92kI-rX0-02507-00782875-00783084 that are to show what's running
+odN92kI-rX0-02508-00783084-00783374 because sometimes a process will be doing something
+odN92kI-rX0-02509-00783374-00783558 and we'll hang up a device and then
+odN92kI-rX0-02510-00783558-00783877 the device hangs up suspend, like you're, I don't know,
+odN92kI-rX0-02511-00783877-00784219 you're reading something out on Thunderbolt or something.
+odN92kI-rX0-02512-00784376-00784712 And so, but that's an option, that's not on by default.
+odN92kI-rX0-02513-00784955-00785129 This bit here, like where we can see,
+odN92kI-rX0-02514-00785129-00785263 actually the one in front of us this,
+odN92kI-rX0-02515-00785263-00785567 is an M sleep, this is, yeah.
+odN92kI-rX0-02516-00785567-00785703 So a couple vendors have this,
+odN92kI-rX0-02517-00785703-00785834 I think it's a bug workaround,
+odN92kI-rX0-02518-00785834-00786017 that's hard-coded into the BIOS.
+odN92kI-rX0-02519-00786017-00786328 It's sort of embarrassing but that's a, yeah,
+odN92kI-rX0-02520-00786328-00786536 it is what it looks like,
+odN92kI-rX0-02521-00786536-00786750 it's a two seconds sleep in the BIOS.
+odN92kI-rX0-02522-00786750-00787009 There's really nothing we can do about it, you're like--
+odN92kI-rX0-02523-00787009-00787167 (speaking faintly)
+odN92kI-rX0-02524-00787167-00787355 - Yeah it's, I think it's an off method
+odN92kI-rX0-02525-00787355-00787600 for maybe it's a fan or something, I forgot what.
+odN92kI-rX0-02526-00787738-00788222 Yeah, so it's, this isn't the, HP has this one too.
+odN92kI-rX0-02527-00788222-00788586 - Yeah so there are systems in which
+odN92kI-rX0-02528-00788586-00788825 there are things like that in AML,
+odN92kI-rX0-02529-00788825-00789309 essentially they wake, they wait for a certain fixed time.
+odN92kI-rX0-02530-00789309-00789443 - [Lynn] Yes, it's bad.
+odN92kI-rX0-02531-00789443-00789581 - Because to work around some issues.
+odN92kI-rX0-02532-00789581-00789797 - And I believe this one, the workaround
+odN92kI-rX0-02533-00789797-00790009 is only for resume, so the suspense side
+odN92kI-rX0-02534-00790009-00790271 is completely bogus but yeah.
+odN92kI-rX0-02535-00790271-00790498 I mean, you can really, closer you look at a BIOS
+odN92kI-rX0-02536-00790498-00790628 the more you can embarrass a BIOS guy
+odN92kI-rX0-02537-00790628-00790766 and we could do that all day long
+odN92kI-rX0-02538-00790766-00790990 but that's really not what I want to talk about today.
+odN92kI-rX0-02539-00791089-00791172 - That's not the main goal here.
+odN92kI-rX0-02540-00791172-00791372 - Yeah, so we have, for me--
+odN92kI-rX0-02541-00791372-00791476 - I should show a different one,
+odN92kI-rX0-02542-00791476-00791581 I'll show a happier one.
+odN92kI-rX0-02543-00791581-00791814 - Lynn, so we have two minutes left for this topic, so.
+odN92kI-rX0-02544-00791814-00791945 - Well I'm just asking questions
+odN92kI-rX0-02545-00791945-00792328 so I really had nothing to, I think one.
+odN92kI-rX0-02546-00792491-00792645 - So there are two things there.
+odN92kI-rX0-02547-00792645-00793003 Actually the packaging that stuff was easy
+odN92kI-rX0-02548-00793003-00793224 and I actually did for the openSUSE
+odN92kI-rX0-02549-00793224-00793511 and I want to try to push up to the factory
+odN92kI-rX0-02550-00793511-00793878 and but the question is, how to,
+odN92kI-rX0-02551-00793878-00794072 so where to gather the data?
+odN92kI-rX0-02552-00794072-00794224 So where to upload?
+odN92kI-rX0-02553-00794224-00794379 That's not defined yet, right.
+odN92kI-rX0-02554-00794379-00794593 - Right, right, so if, say Google stepped forward
+odN92kI-rX0-02555-00794593-00794828 and said hey, you can stick this on our Google drive
+odN92kI-rX0-02556-00794828-00795099 with some credentials, we'd all be really happy about that
+odN92kI-rX0-02557-00795099-00795336 because we know how to do, how to get to that.
+odN92kI-rX0-02558-00795336-00795619 - Yeah that's, I'm not sure how this will work.
+odN92kI-rX0-02559-00795619-00795976 - There's, so this spreadsheet here had a,
+odN92kI-rX0-02560-00795976-00796305 say I ran 2000, we save all 2,000 results
+odN92kI-rX0-02561-00796305-00796640 but we also say, we can also just reduce
+odN92kI-rX0-02562-00796640-00796902 this down to save, like say, these.
+odN92kI-rX0-02563-00796902-00797071 Like this, each one of these is a link
+odN92kI-rX0-02564-00797071-00797376 to the fastest slowest and median, and it's a real result.
+odN92kI-rX0-02565-00797376-00797581 I can open any of these, at least,
+odN92kI-rX0-02566-00797581-00797814 I should be able to open any of those.
+odN92kI-rX0-02567-00797814-00798056 And that's the median for this machine.
+odN92kI-rX0-02568-00798056-00798307 So instead of all 2000, there's like
+odN92kI-rX0-02569-00798307-00798544 about a dozen interesting ones, like the slowest device,
+odN92kI-rX0-02570-00798544-00798910 the fastest device, anything with an oops
+odN92kI-rX0-02571-00798910-00799334 or a kernel message in it, we grab those.
+odN92kI-rX0-02572-00799420-00799576 And then a lot of them are just repeats,
+odN92kI-rX0-02573-00799576-00799721 so we don't really have to store that.
+odN92kI-rX0-02574-00799721-00800221 But yeah, I'd ballpark about 100K per result.
+odN92kI-rX0-02575-00800249-00800561 We have to store that someplace, yeah.
+odN92kI-rX0-02576-00800561-00800853 - Yeah and a weighted question is,
+odN92kI-rX0-02577-00800853-00801048 so suppose that we get a data
+odN92kI-rX0-02578-00801048-00801297 and we had a bug report for that
+odN92kI-rX0-02579-00801297-00801680 and we fixed that, but how to verify the fix?
+odN92kI-rX0-02580-00801815-00802122 - Well I mean, we run this every RC actually,
+odN92kI-rX0-02581-00802122-00802334 and so the 5.0, of the bug I showed you
+odN92kI-rX0-02582-00802440-00802600 with the audio is gone.
+odN92kI-rX0-02583-00802600-00802781 - Yeah in that case, you guys have the machines
+odN92kI-rX0-02584-00802781-00803031 but in the case the supposed access,
+odN92kI-rX0-02585-00803031-00803216 we get we get a data from the pub,
+odN92kI-rX0-02586-00803216-00803676 someone else publicly, and how to verify that?
+odN92kI-rX0-02587-00803765-00803961 - I don't know, like any other bug,
+odN92kI-rX0-02588-00803961-00804044 you debug it, you make it go away
+odN92kI-rX0-02589-00804044-00804247 on what's in front of you and hopefully
+odN92kI-rX0-02590-00804247-00804560 it goes away for everybody else, sometimes that works.
+odN92kI-rX0-02591-00804560-00804849 But if they run it again and we get a result
+odN92kI-rX0-02592-00804849-00805051 for that machine and it's clean, then we know it worked.
+odN92kI-rX0-02593-00805051-00805460 So this also happens for regular bugs, like bugzilla,
+odN92kI-rX0-02594-00805460-00805867 somebody files the bug, you see it,
+odN92kI-rX0-02595-00805867-00806125 you fix it and then there's no response.
+odN92kI-rX0-02596-00806125-00806443 You have to assume that it has been fixed, right, kind of.
+odN92kI-rX0-02597-00806443-00806909 - The case of bugzilla is clear that there is a bug reporter
+odN92kI-rX0-02598-00806909-00807293 but in that saboratory we get, we gather the data
+odN92kI-rX0-02599-00807293-00807793 for improvement and, but we yeah,
+odN92kI-rX0-02600-00807815-00808096 don't know exactly who submitted that data
+odN92kI-rX0-02601-00808096-00808344 and we can verify that and improve it.
+odN92kI-rX0-02602-00808535-00808673 - So say for this one, what we did was,
+odN92kI-rX0-02603-00808786-00808909 we put a little attachment in the bug report,
+odN92kI-rX0-02604-00808909-00809235 which sort of has a profile, this issue.def
+odN92kI-rX0-02605-00809235-00809584 so that our tool will go out and say, okay, here's a result.
+odN92kI-rX0-02606-00809584-00809958 And then we could see that that thing I showed you,
+odN92kI-rX0-02607-00809958-00810294 with the audio bug, that was all done automatically
+odN92kI-rX0-02608-00810294-00810794 by comparing this bug report with those test results
+odN92kI-rX0-02609-00810808-00811129 and we could see that it was on all these machines in 5.0.
+odN92kI-rX0-02610-00811129-00811570 It's gone and 5.0 RC, whatever it was, 5.1 RC6.
+odN92kI-rX0-02611-00811694-00811895 So yeah, Bugzilla is good, if we can get
+odN92kI-rX0-02612-00811895-00812075 into Bugzilla, we're in pretty good shape.
+odN92kI-rX0-02613-00812075-00812280 - So what, just to clarify, what's your idea right now?
+odN92kI-rX0-02614-00812280-00812488 Do you want people to send the reports to you,
+odN92kI-rX0-02615-00812488-00812597 or do you want--
+odN92kI-rX0-02616-00812597-00812887 - First of all, filing bugs we're all good with that,
+odN92kI-rX0-02617-00812887-00813065 but yeah, what we really want,
+odN92kI-rX0-02618-00813065-00813554 what the grand strategy is, we want to get data
+odN92kI-rX0-02619-00813554-00813718 stuff running on all kinds of machines,
+odN92kI-rX0-02620-00813718-00814023 that's what we really want is, we want to cast a broad net.
+odN92kI-rX0-02621-00814023-00814275 We can run 24 hours a day on a machine
+odN92kI-rX0-02622-00814275-00814701 we have in front of us, but I need 3,000 machines.
+odN92kI-rX0-02623-00814701-00814842 That's really the message.
+odN92kI-rX0-02624-00814842-00814934 I think this one last.
+odN92kI-rX0-02625-00814934-00815421 - Yep, hello, I'm Victor Ruiz, I work in Red Hat.
+odN92kI-rX0-02626-00815421-00815611 I own suspend resume, I work on
+odN92kI-rX0-02627-00815611-00815880 the current hardware team QE.
+odN92kI-rX0-02628-00815880-00816255 You have a pull request for me on Github
+odN92kI-rX0-02629-00816255-00816601 from two years ago, so I'm very interested in in your work.
+odN92kI-rX0-02630-00816777-00817180 I have a question about the tool itself,
+odN92kI-rX0-02631-00817180-00817586 whether you are thinking about doing a client,
+odN92kI-rX0-02632-00817753-00818104 client-server architecture to be able to get that
+odN92kI-rX0-02633-00818104-00818511 when something growth run because if the machine
+odN92kI-rX0-02634-00818511-00818982 doesn't resume then you are losing data.
+odN92kI-rX0-02635-00818982-00819107 - Yeah that's a good question,
+odN92kI-rX0-02636-00819107-00819414 so we basically have been running it
+odN92kI-rX0-02637-00819414-00819687 via SSH on it when we have that situation.
+odN92kI-rX0-02638-00819870-00820315 But that hasn't been a priority yet anyway.
+odN92kI-rX0-02639-00820315-00820456 - Okay but let's talk.
+odN92kI-rX0-02640-00820569-00820798 - Yeah okay, thank you, thanks.
+odN92kI-rX0-02641-00820798-00820911 - [Moderator] All right, thank you.
+odN92kI-rX0-02642-00821158-00821383 (applause)
+odN92kI-rX0-02643-00821585-00821877 (faintly speaking)
+odN92kI-rX0-02644-00823885-00824027 - Just press it to turn it on.
+odN92kI-rX0-02645-00824027-00824318 (faintly speaking)
+odN92kI-rX0-02646-00827002-00827282 - Okay, hi everyone thanks for having me.
+odN92kI-rX0-02647-00827383-00827831 My name is Artem I worked for Intel from Finland.
+odN92kI-rX0-02648-00827831-00828009 The subject of my talk is the
+odN92kI-rX0-02649-00828009-00828231 C-state latency measurement infrastructure.
+odN92kI-rX0-02650-00828391-00828741 Brief introduction, so in Intel
+odN92kI-rX0-02651-00828741-00829058 we've developed a tool for measuring C-state latency,
+odN92kI-rX0-02652-00829058-00829495 we call it WULT which stands for Wake Up Latency Tracer.
+odN92kI-rX0-02653-00829726-00829839 Right now we are going through
+odN92kI-rX0-02654-00829839-00830167 the process of open sourcing it.
+odN92kI-rX0-02655-00830167-00830376 Unfortunately we are not done yet
+odN92kI-rX0-02656-00830476-00830839 but as soon as we get the company approval
+odN92kI-rX0-02657-00830839-00831187 we will publish it on Github under the GPL license.
+odN92kI-rX0-02658-00831327-00831748 This tool has user space components and a kernel driver,
+odN92kI-rX0-02659-00831929-00832186 and the plan is to upstream the kernel driver.
+odN92kI-rX0-02660-00832329-00832531 So the reason I'm here is to get
+odN92kI-rX0-02661-00832531-00832841 early feedback on the WULT kernel drivers
+odN92kI-rX0-02662-00832841-00833150 and how do we transform them into something up streamable.
+odN92kI-rX0-02663-00833320-00833559 Okay so let's talk about what exactly
+odN92kI-rX0-02664-00833559-00833745 we measure and how we do this.
+odN92kI-rX0-02665-00833977-00834383 I'll walk you through a simplified version
+odN92kI-rX0-02666-00834383-00834643 of one single measurement cycle.
+odN92kI-rX0-02667-00834744-00835007 So we start with Arm in a delayed interrupt.
+odN92kI-rX0-02668-00835116-00835320 I'll talk a little bit more about what it is
+odN92kI-rX0-02669-00835320-00835651 but just assume for now that the measured platform
+odN92kI-rX0-02670-00835651-00835985 has a capability of generating an interrupt
+odN92kI-rX0-02671-00835985-00836254 at a precise pre-programmed time in the future,
+odN92kI-rX0-02672-00836254-00836561 sort of like a timer but there are some nuances there.
+odN92kI-rX0-02673-00836712-00836894 So we armed the delayed interrupt
+odN92kI-rX0-02674-00836894-00837168 so that it should happen at launch time in the future
+odN92kI-rX0-02675-00837314-00837620 and then since we are measuring an idle system,
+odN92kI-rX0-02676-00837620-00837816 it will at some point go to a C-state.
+odN92kI-rX0-02677-00837923-00838423 On Intel platforms the CPU instruction that we execute
+odN92kI-rX0-02678-00838424-00838595 to enter a C-state is mwait.
+odN92kI-rX0-02679-00838691-00838900 I think on arms it's something like wfi.
+odN92kI-rX0-02680-00838994-00839111 - Wfi, yeah.
+odN92kI-rX0-02681-00839111-00839487 - Yeah, so okay, so the measured CPU executes mwait,
+odN92kI-rX0-02682-00839487-00839690 it stops executing further instructions.
+odN92kI-rX0-02683-00839795-00840202 It enters the C-state, sits there saving power
+odN92kI-rX0-02684-00840202-00840498 and waiting for an event like an interrupt to happen.
+odN92kI-rX0-02685-00840679-00840977 Then when it's launch time the delayed interrupt
+odN92kI-rX0-02686-00840977-00841316 actually happens, it gets delivered to the mirrored CPU
+odN92kI-rX0-02687-00841316-00841593 and kicks off the process of exiting the C-state.
+odN92kI-rX0-02688-00841783-00842025 This process will take some time.
+odN92kI-rX0-02689-00842025-00842252 When it's finished, the CPU will
+odN92kI-rX0-02690-00842252-00842462 start executing the instructions
+odN92kI-rX0-02691-00842462-00842696 after the end wait, and that's the point
+odN92kI-rX0-02692-00842696-00842936 where we take the time after idle timestamp.
+odN92kI-rX0-02693-00843223-00843608 So the C-state exit latency is the period of time
+odN92kI-rX0-02694-00843608-00843966 between launch time and time after idle.
+odN92kI-rX0-02695-00843966-00844370 In other words, it's a time between the moment we
+odN92kI-rX0-02696-00844370-00844641 the delayed interrupt happens until the moment
+odN92kI-rX0-02697-00844641-00844831 the CPU starts executing instructions.
+odN92kI-rX0-02698-00845024-00845367 So the C-state latency is this difference.
+odN92kI-rX0-02699-00845462-00845645 And that's basically what we measure
+odN92kI-rX0-02700-00845645-00846045 and how we do it in a in a nutshell, the principle.
+odN92kI-rX0-02701-00846312-00846437 Any questions so far?
+odN92kI-rX0-02702-00846717-00846890 I assume that if there are questions,
+odN92kI-rX0-02703-00846890-00847348 you can just interrupt me, I mean, you don't need to wait.
+odN92kI-rX0-02704-00847348-00847518 Okay, few words, why is it useful.
+odN92kI-rX0-02705-00847790-00848086 Okay, so the C-state latency information
+odN92kI-rX0-02706-00848086-00848358 is today, is the foundation for
+odN92kI-rX0-02707-00848453-00848654 the power management quality of service subsystem.
+odN92kI-rX0-02708-00848654-00848979 On Intel platforms, we get this information
+odN92kI-rX0-02709-00848979-00849354 either from the firmware through the ACPI tables
+odN92kI-rX0-02710-00849354-00849754 or from the driver from intel_idle driver,
+odN92kI-rX0-02711-00849754-00849920 they are just hard coded there.
+odN92kI-rX0-02712-00849920-00850112 Now the truth is that these numbers
+odN92kI-rX0-02713-00850112-00850301 are not always 100% correct.
+odN92kI-rX0-02714-00850528-00850804 And they also may depend on platform configuration.
+odN92kI-rX0-02715-00850804-00851304 So if we have a tool that is easy to use
+odN92kI-rX0-02716-00851337-00851534 and that can measure these latencies,
+odN92kI-rX0-02717-00851534-00851719 we will open lots of opportunities
+odN92kI-rX0-02718-00851719-00851854 for improving this subsystem.
+odN92kI-rX0-02719-00851854-00852118 For example we can improve numbers in intel_driver.
+odN92kI-rX0-02720-00852268-00852485 Other point is that if you are someone
+odN92kI-rX0-02721-00852485-00852769 who is building real time-ish product
+odN92kI-rX0-02722-00852769-00853058 and for some reasons you are going to use C-states,
+odN92kI-rX0-02723-00853179-00853379 maybe some of them, then measuring their latency
+odN92kI-rX0-02724-00853379-00853565 is probably crucial for your product.
+odN92kI-rX0-02725-00853704-00853879 And with this tool you wouldn't need
+odN92kI-rX0-02726-00853879-00854112 expensive lab equipment to do so.
+odN92kI-rX0-02727-00854372-00854456 - So can you just repeat again,
+odN92kI-rX0-02728-00854456-00854864 what were the requirements from your hardware
+odN92kI-rX0-02729-00854864-00855039 to use your tool or your driver?
+odN92kI-rX0-02730-00855039-00855341 - Yeah so the question is, what are the requirements
+odN92kI-rX0-02731-00855341-00855496 from the hardware to use the tool?
+odN92kI-rX0-02732-00855496-00855712 So the requirement is to be able to
+odN92kI-rX0-02733-00855812-00855984 generate a delay interrupt and I said
+odN92kI-rX0-02734-00855984-00856154 I would talk a little bit later
+odN92kI-rX0-02735-00856154-00856289 and I will very soon.
+odN92kI-rX0-02736-00856289-00856399 - Okay.
+odN92kI-rX0-02737-00856399-00856752 - Yeah, but before I will talk about that,
+odN92kI-rX0-02738-00856752-00856970 I'll just very briefly show a quick,
+odN92kI-rX0-02739-00856970-00857131 a visual example, just you know.
+odN92kI-rX0-02740-00857131-00857223 - Sure.
+odN92kI-rX0-02741-00857223-00857640 - So this talk is not about talking about the results
+odN92kI-rX0-02742-00857640-00857956 and all that and user-space, it's about drivers,
+odN92kI-rX0-02743-00857956-00858240 but I thought just showing an example of,
+odN92kI-rX0-02744-00858240-00858585 this example will help understanding what we are doing.
+odN92kI-rX0-02745-00858585-00858978 So this is a snapshot from a report,
+odN92kI-rX0-02746-00858978-00859184 in this case it's the scatterplot,
+odN92kI-rX0-02747-00859184-00859479 every dot is one single measurement,
+odN92kI-rX0-02748-00859479-00859723 on the X we have the amount of time
+odN92kI-rX0-02749-00859723-00859998 was spent in the end wait before the interrupt happened.
+odN92kI-rX0-02750-00859998-00860228 We call it silent time, it's in milliseconds.
+odN92kI-rX0-02751-00860228-00860537 On the y-axis we have the C-state latency,
+odN92kI-rX0-02752-00860537-00860701 it's in microseconds.
+odN92kI-rX0-02753-00860701-00861070 Now so in HTML, you can hover your mouse over a dot
+odN92kI-rX0-02754-00861070-00861262 and then you get this little pop-up menu
+odN92kI-rX0-02755-00861262-00861512 that tells details about the C-state.
+odN92kI-rX0-02756-00861512-00861895 In this specific case, what do we have here,
+odN92kI-rX0-02757-00861895-00862384 it says that we spend 2.5 milliseconds in end wait
+odN92kI-rX0-02758-00862384-00862540 before the interrupt happened,
+odN92kI-rX0-02759-00862540-00862911 and it took us 94 microseconds to wake up.
+odN92kI-rX0-02760-00862911-00863187 And then the Linux request, it says six,
+odN92kI-rX0-02761-00863187-00863437 but you know the C-state counters tells us
+odN92kI-rX0-02762-00863437-00863768 that we were around 1% in CC zero course E0,
+odN92kI-rX0-02763-00863768-00864081 like 13% in course E1 and so on.
+odN92kI-rX0-02764-00864347-00864609 We had 73% in PC6 that's, yeah.
+odN92kI-rX0-02765-00864929-00865170 So yeah, it's not it's not my goal to talk
+odN92kI-rX0-02766-00865170-00865644 about the results, it's just a quick example.
+odN92kI-rX0-02767-00865644-00866000 Okay, now let's talk about implementation details.
+odN92kI-rX0-02768-00866000-00866384 Now if you have like rotten tomatoes prepared,
+odN92kI-rX0-02769-00866384-00866787 don't throw it let me immediately, do hesitate.
+odN92kI-rX0-02770-00866787-00867066 It's how it's implemented, right,
+odN92kI-rX0-02771-00867066-00867467 it's not necessarily what we need to do in upstream, okay.
+odN92kI-rX0-02772-00867467-00867854 Okay, delayed interrupts, so on your question.
+odN92kI-rX0-02773-00867854-00868091 So today, in how we implemented it now,
+odN92kI-rX0-02774-00868091-00868545 we use the I210 network card to generate delayed interrupts.
+odN92kI-rX0-02775-00868648-00869127 So it's a Gigabit Ethernet card PCI Express device
+odN92kI-rX0-02776-00869127-00869367 that you can buy in many shops around the globe.
+odN92kI-rX0-02777-00869525-00869879 Now the nice thing about it, it has a built-in clock,
+odN92kI-rX0-02778-00869983-00870260 high precision clock that host can read,
+odN92kI-rX0-02779-00870260-00870475 and it also can generate an interrupt,
+odN92kI-rX0-02780-00870475-00870686 a delayed interrupt for you.
+odN92kI-rX0-02781-00870686-00870845 So basically what we do, we read the clock from the card,
+odN92kI-rX0-02782-00870933-00871079 add a delay, write it back, tell the card
+odN92kI-rX0-02783-00871079-00871227 generate interrupt at that point,
+odN92kI-rX0-02784-00871227-00871453 and then we can go in into end wait.
+odN92kI-rX0-02785-00871453-00871822 Now is this the only device that can do this?
+odN92kI-rX0-02786-00871822-00872118 (faintly speaking)
+odN92kI-rX0-02787-00872118-00872250 Yes, the question is,
+odN92kI-rX0-02788-00872250-00872489 do we affinitize the interrupt to CPU, yes.
+odN92kI-rX0-02789-00872489-00872670 So we select a CPU that we measure,
+odN92kI-rX0-02790-00872670-00872886 affinitize the interrupted to that CPU, that's right.
+odN92kI-rX0-02791-00872886-00873149 Now, yeah, it's not the only device
+odN92kI-rX0-02792-00873149-00873379 that can do this, is what we use.
+odN92kI-rX0-02793-00873379-00873712 I know other NICs can do this and there are
+odN92kI-rX0-02794-00873712-00873982 other types of devices that can do this.
+odN92kI-rX0-02795-00873982-00874245 For example, you can use display controller for that.
+odN92kI-rX0-02796-00874346-00874493 Okay, so now the question is,
+odN92kI-rX0-02797-00874493-00874650 why not using the timers, right,
+odN92kI-rX0-02798-00874650-00874893 they are a perfect source of delayed interrupt,
+odN92kI-rX0-02799-00874893-00875023 they're designed for that.
+odN92kI-rX0-02800-00875112-00875418 The answer is that on Intel platforms they are too perfect,
+odN92kI-rX0-02801-00875510-00875684 they are so perfect that they
+odN92kI-rX0-02802-00875684-00875924 make C-state latency invisible.
+odN92kI-rX0-02803-00875924-00876187 So the problem is that the P you need the microchip
+odN92kI-rX0-02804-00876187-00876687 inside this Intel chip or PMC, it is aware of timers.
+odN92kI-rX0-02805-00876728-00877062 So it will pre-wake the system, the core,
+odN92kI-rX0-02806-00877062-00877417 and hide this latency and make it
+odN92kI-rX0-02807-00877417-00877645 disappear or drop substantially.
+odN92kI-rX0-02808-00877645-00877995 So in Intel platforms, we need to use something external,
+odN92kI-rX0-02809-00877995-00878181 that's what why we use the network card.
+odN92kI-rX0-02810-00878181-00878364 But probably there are platforms out there
+odN92kI-rX0-02811-00878364-00878504 where timer's could be used.
+odN92kI-rX0-02812-00878720-00878911 So yeah, that's on delayed interrupts.
+odN92kI-rX0-02813-00878995-00879271 Now, high-level design, how we did it,
+odN92kI-rX0-02814-00879271-00879403 how we implemented this.
+odN92kI-rX0-02815-00879487-00879975 Okay so today we have two kernel components,
+odN92kI-rX0-02816-00879975-00880228 and I will be talking just about the kernel.
+odN92kI-rX0-02817-00880228-00880504 Everything I show is are kernel components.
+odN92kI-rX0-02818-00880504-00880798 So we have two models, we implement this as models.
+odN92kI-rX0-02819-00880798-00881109 One is called wult, one is called, the other is wult_igb.
+odN92kI-rX0-02820-00881212-00881574 So the main logic of measurement is implemented in wult.
+odN92kI-rX0-02821-00881686-00882004 And wult_igb is a simple driver
+odN92kI-rX0-02822-00882004-00882385 that knows how to talk to the network card, okay.
+odN92kI-rX0-02823-00882385-00882836 So wult implements in API and wult defines
+odN92kI-rX0-02824-00882969-00883120 the API wult_igb implements it.
+odN92kI-rX0-02825-00883120-00883287 So from wult point of view, wult_igb
+odN92kI-rX0-02826-00883287-00883489 is just a delayed interrupts provider.
+odN92kI-rX0-02827-00883489-00883736 It could be some other device, something else,
+odN92kI-rX0-02828-00883736-00883879 so it's a small layer of abstraction.
+odN92kI-rX0-02829-00884041-00884258 Now in Linux today, there is a standard driver
+odN92kI-rX0-02830-00884258-00884584 for these network cards, it's called igb, Intel gigabit.
+odN92kI-rX0-02831-00884823-00885323 So what we do, we unbind the NIC from this driver,
+odN92kI-rX0-02832-00885371-00885553 and then bind it to wult.
+odN92kI-rX0-02833-00885553-00885770 So igb doesn't own it anymore.
+odN92kI-rX0-02834-00885867-00886187 Wult_igb driver is very small comparing to igb.
+odN92kI-rX0-02835-00886187-00886653 All we do, reset the card, we do few instilization steps
+odN92kI-rX0-02836-00886653-00886896 and then it's ready to be used as delayed interrupt source.
+odN92kI-rX0-02837-00886896-00887229 We don't initialize the fi, we don't need to allocate rings
+odN92kI-rX0-02838-00887229-00887561 and deal with these descriptors, no complexity.
+odN92kI-rX0-02839-00887561-00887720 It's very tiny comparing to igb.
+odN92kI-rX0-02840-00887934-00888323 Okay, now one of the components in the wult driver
+odN92kI-rX0-02841-00888323-00888614 is what we call Armer kernel thread.
+odN92kI-rX0-02842-00888614-00888892 It's a kernel thread and the mission of this thread
+odN92kI-rX0-02843-00888892-00889169 is to continuously arm delayed interrupts.
+odN92kI-rX0-02844-00889169-00889414 So all it does, it will pick a delay
+odN92kI-rX0-02845-00889414-00889860 arm the interrupt, wait for it to happen,
+odN92kI-rX0-02846-00889860-00890108 and continue this again and again and again.
+odN92kI-rX0-02847-00890108-00890236 That's all it does.
+odN92kI-rX0-02848-00890337-00890617 Then we have a debugging, debug fs interface
+odN92kI-rX0-02849-00890617-00890815 for configuring this stuff, for example
+odN92kI-rX0-02850-00890815-00891043 you can enable disable measurements,
+odN92kI-rX0-02851-00891043-00891289 you have this delay min and max knobs
+odN92kI-rX0-02852-00891289-00891685 where you can set the interrupt delay range.
+odN92kI-rX0-02853-00891685-00892017 So the thread will pick a random number in this range
+odN92kI-rX0-02854-00892017-00892359 and arm the interrupt that far away.
+odN92kI-rX0-02855-00892554-00892668 Okay I'm just zooming the out components,
+odN92kI-rX0-02856-00892668-00892865 so just to recap what we have,
+odN92kI-rX0-02857-00892865-00893008 is the driver with a thread,
+odN92kI-rX0-02858-00893008-00893270 it arms interrupts and waits for them to happen,
+odN92kI-rX0-02859-00893270-00893442 otherwise the system is idle.
+odN92kI-rX0-02860-00893442-00893623 The mirrored CPU has plenty of time
+odN92kI-rX0-02861-00893623-00893892 to go to a C-state, so what happens, it will go
+odN92kI-rX0-02862-00893892-00894285 to in a C-state, the interrupt will wake it up.
+odN92kI-rX0-02863-00894285-00894507 Then it goes into a C-state again and we repeat this.
+odN92kI-rX0-02864-00894507-00894750 So now how do we actually measure the latency?
+odN92kI-rX0-02865-00894910-00895331 Okay, let's briefly talk about the current Intel idle flow.
+odN92kI-rX0-02866-00895331-00895685 So we have the idle task, we have the CPU idle subsystem.
+odN92kI-rX0-02867-00895685-00895994 So the subsystem will choose a C-state
+odN92kI-rX0-02868-00895994-00896362 and then will enter it by calling the inter call back.
+odN92kI-rX0-02869-00896362-00896634 This will go into the idle driver,
+odN92kI-rX0-02870-00896634-00897034 in our case it's Intel idle driver, and Intel idle driver
+odN92kI-rX0-02871-00897034-00897332 will ultimately call the end wait instruction.
+odN92kI-rX0-02872-00897332-00897583 So here I assume that the mirrored CPU
+odN92kI-rX0-02873-00897583-00897756 CPU is zero, but it can be anything else.
+odN92kI-rX0-02874-00897756-00898145 And yes, the delayed interrupts are affinities to the CPU.
+odN92kI-rX0-02875-00898296-00898631 Okay, so we have another component in wult,
+odN92kI-rX0-02876-00898631-00898774 we call it interposer.
+odN92kI-rX0-02877-00899053-00899253 So it basically puts itself right there.
+odN92kI-rX0-02878-00899541-00899631 Right there, just.
+odN92kI-rX0-02879-00899878-00900148 Now with interposer, the flow looks like this.
+odN92kI-rX0-02880-00900245-00900579 So CPU idle tries to call the idle driver
+odN92kI-rX0-02881-00900579-00900882 but it actually goes to the interposer
+odN92kI-rX0-02882-00900882-00901299 so we have chance to take some stuff like time before idle.
+odN92kI-rX0-02883-00901299-00901607 Then we go back to intel_idle, go to end wait.
+odN92kI-rX0-02884-00901607-00901775 From end wait we go to Intel idle
+odN92kI-rX0-02885-00901864-00902167 and back to interposer, we can measure stuff after idle,
+odN92kI-rX0-02886-00902167-00902346 and then it will return to CPU idle.
+odN92kI-rX0-02887-00902785-00902954 So now a few points here, important points,
+odN92kI-rX0-02888-00902954-00903358 one important point is that CPU idle disables interrupts
+odN92kI-rX0-02889-00903358-00903640 and preemption right before calling to idle driver.
+odN92kI-rX0-02890-00903640-00903908 So in interposer we have this disabled.
+odN92kI-rX0-02891-00904054-00904306 Other point is that in current implementation,
+odN92kI-rX0-02892-00904306-00904589 we don't modify CPU idle, we don't modify Intel idle.
+odN92kI-rX0-02893-00904589-00905058 What we do is we use Kiel sims to find the address
+odN92kI-rX0-02894-00905058-00905336 of the function and then we use F trace
+odN92kI-rX0-02895-00905336-00905637 to hook to it and interpose it.
+odN92kI-rX0-02896-00905637-00905845 So no need to modify these subsystems.
+odN92kI-rX0-02897-00906062-00906337 Okay, few words about what we do in interposer.
+odN92kI-rX0-02898-00906337-00906837 So basically when CPU idle calls enter, we take,
+odN92kI-rX0-02899-00906858-00907200 we can remember launch time, we take time before idle,
+odN92kI-rX0-02900-00907200-00907517 we take snapshots of C-state counters.
+odN92kI-rX0-02901-00907517-00907841 Then we go run until idle which goes to end weight.
+odN92kI-rX0-02902-00907841-00908185 We sleep, we wake up, we go back to interposer.
+odN92kI-rX0-02903-00908185-00908451 We take statistics after idle,
+odN92kI-rX0-02904-00908451-00908708 like time after idle, C-states counter after idle.
+odN92kI-rX0-02905-00908805-00908947 Linus?
+odN92kI-rX0-02906-00908947-00909030 - So why don't you just call end wait.
+odN92kI-rX0-02907-00909142-00909433 (speaking faintly)
+odN92kI-rX0-02908-00909536-00909775 There must be some reason obviously.
+odN92kI-rX0-02909-00909775-00910074 - Why don't we just call end wait our cells?
+odN92kI-rX0-02910-00910074-00910286 - Intel idle does a suggestion.
+odN92kI-rX0-02911-00910286-00910585 (speaking faintly)
+odN92kI-rX0-02912-00910585-00910790 - Yes but I mean, like you would to
+odN92kI-rX0-02913-00910790-00911177 iterate over all the suggestions,
+odN92kI-rX0-02914-00911177-00911677 and iterate over all the different distances,
+odN92kI-rX0-02915-00911720-00912206 time distances and get a scatter plot of that one, maybe.
+odN92kI-rX0-02916-00912206-00912363 - Yeah we could do something like this,
+odN92kI-rX0-02917-00912363-00912784 it's just, there is lots of knowledge
+odN92kI-rX0-02918-00912784-00913214 about C-states in end wait, oh sorry, in the intel idle.
+odN92kI-rX0-02919-00913214-00913573 We didn't want to duplicate it,
+odN92kI-rX0-02920-00913573-00913759 but that's a right point, we want to be as close
+odN92kI-rX0-02921-00913759-00913978 to this way as possible from both ends.
+odN92kI-rX0-02922-00913978-00914270 And right now we are we're a little bit further away.
+odN92kI-rX0-02923-00914449-00914621 I think it doesn't really matter.
+odN92kI-rX0-02924-00914621-00914783 So we are talking about like hundreds of
+odN92kI-rX0-02925-00914783-00915276 nano seconds versus microseconds here, but we could do this.
+odN92kI-rX0-02926-00915276-00915602 The other point is that, with this sort of architecture
+odN92kI-rX0-02927-00915602-00916085 we could actually also hook to ACPI idle.
+odN92kI-rX0-02928-00916085-00916359 But yeah, that's an option, yeah.
+odN92kI-rX0-02929-00916494-00916704 I don't know if this answers the question.
+odN92kI-rX0-02930-00916837-00917069 Yeah so the other thing we do,
+odN92kI-rX0-02931-00917069-00917209 I just wanted to mention is that
+odN92kI-rX0-02932-00917209-00917382 we try to check the wake reason.
+odN92kI-rX0-02933-00917382-00917586 So if we think we woke up because of something else
+odN92kI-rX0-02934-00917586-00917862 not our interrupt, we just ignore all this data.
+odN92kI-rX0-02935-00917862-00918057 But if we do think we woke up from interrupt,
+odN92kI-rX0-02936-00918057-00918261 we send the data to user-space
+odN92kI-rX0-02937-00918261-00918496 and how we do this, we use trace print K.
+odN92kI-rX0-02938-00918590-00919037 So that's where I hope rotten tomatoes don't fly at me.
+odN92kI-rX0-02939-00919037-00919374 I know that this is like not a way to go in upstream
+odN92kI-rX0-02940-00919374-00919631 but that was a quick shortcut.
+odN92kI-rX0-02941-00919631-00919861 So it ends up in the trace buffer,
+odN92kI-rX0-02942-00919861-00920210 so user-space picks it from the trace of S.
+odN92kI-rX0-02943-00920210-00920507 And yeah, that's basically the current design,
+odN92kI-rX0-02944-00920507-00920614 that's how it works.
+odN92kI-rX0-02945-00920879-00921281 So I have a list of challenges that I see.
+odN92kI-rX0-02946-00921419-00921649 I can go to them and unless someone
+odN92kI-rX0-02947-00921649-00922004 has comments or questions or suggestions.
+odN92kI-rX0-02948-00922276-00922768 Okay so again, now I need high-level feedback,
+odN92kI-rX0-02949-00922768-00922898 there are lots of details.
+odN92kI-rX0-02950-00923025-00923186 Let's see, on high level first,
+odN92kI-rX0-02951-00923186-00923441 kind of first thing is, where,
+odN92kI-rX0-02952-00923441-00923696 if we to upstream this stuff, where would it leave?
+odN92kI-rX0-02953-00923696-00924085 I drivers, idle wult, something like that.
+odN92kI-rX0-02954-00924217-00924514 That's what comes to mind.
+odN92kI-rX0-02955-00924514-00924721 Any other ideas?
+odN92kI-rX0-02956-00924819-00924998 - [Lynn] All those in favor, say aye.
+odN92kI-rX0-02957-00924998-00925213 - [Group] Aye.
+odN92kI-rX0-02958-00925213-00925330 (laughing)
+odN92kI-rX0-02959-00925330-00925569 - I mean, yeah, if no one comments
+odN92kI-rX0-02960-00925569-00925812 then probably it's this or we'll see.
+odN92kI-rX0-02961-00925965-00926208 Okay the other thing is the measurement data.
+odN92kI-rX0-02962-00926303-00926644 So what do we send now to upstream?
+odN92kI-rX0-02963-00926644-00926845 Well first of all we send the launch time,
+odN92kI-rX0-02964-00926845-00927091 the time before time are after idle,
+odN92kI-rX0-02965-00927091-00927534 then we calculate delta TSC, delta APERF MPERF,
+odN92kI-rX0-02966-00927534-00927789 and then we get statistics before and after idle
+odN92kI-rX0-02967-00927789-00928132 for C-states, and we send those, the deltas, to upstream.
+odN92kI-rX0-02968-00928132-00928487 So is basically ends up right now being in one CSV line,
+odN92kI-rX0-02969-00928582-00928871 comma separated line with of numbers.
+odN92kI-rX0-02970-00928871-00929175 And but the for a very first line will be CSV header
+odN92kI-rX0-02971-00929175-00929423 with a short name for those things.
+odN92kI-rX0-02972-00929423-00929631 So what are the issues here?
+odN92kI-rX0-02973-00929631-00929951 First of all, obviously is architecture dependent, right,
+odN92kI-rX0-02974-00929951-00930398 but more than that, it also depends
+odN92kI-rX0-02975-00930398-00930695 on which specific Intel platform we measure.
+odN92kI-rX0-02976-00930695-00931148 C-states are different on add amps, on Kleins, on Xeons.
+odN92kI-rX0-02977-00931332-00931751 Also, even MSR registers to read those C-states
+odN92kI-rX0-02978-00931751-00931936 are different on different platforms,
+odN92kI-rX0-02979-00932029-00932381 and we need to somehow read them.
+odN92kI-rX0-02980-00932381-00932633 So what we do today, we just read all the possible MSRs
+odN92kI-rX0-02981-00932633-00932845 for all the possible C-states we know
+odN92kI-rX0-02982-00932845-00933134 for all the Intel platforms, right, and send lots of data.
+odN92kI-rX0-02983-00933134-00933634 Now luckily, in most cases if we read C-state counter
+odN92kI-rX0-02984-00933676-00933867 for a C-state that doesn't exist on this platform,
+odN92kI-rX0-02985-00933867-00934096 we just zeros and we send zeros.
+odN92kI-rX0-02986-00934096-00934341 In other cases we have an exception
+odN92kI-rX0-02987-00934341-00934445 and then we just handle it,
+odN92kI-rX0-02988-00934445-00934585 we remember we had an exception,
+odN92kI-rX0-02989-00934585-00934886 never read again, and just always put zeros.
+odN92kI-rX0-02990-00934886-00935168 So the user-space recognizes that,
+odN92kI-rX0-02991-00935168-00935491 okay, for this C-state there are always zeros
+odN92kI-rX0-02992-00935491-00935645 then it either it doesn't exist
+odN92kI-rX0-02993-00935753-00935943 or it just disabled, then just ignore it.
+odN92kI-rX0-02994-00935943-00936103 That's how we do it but obviously
+odN92kI-rX0-02995-00936103-00936386 it's not like nice and cool.
+odN92kI-rX0-02996-00936386-00936777 Ideally we need to put knowledge to kernel
+odN92kI-rX0-02997-00936777-00937249 about hardware C-states on all the Intel platforms.
+odN92kI-rX0-02998-00937249-00937525 Now we have all this knowledge in the Turbostar tool,
+odN92kI-rX0-02999-00937525-00937825 which is part of the kernel tree but it's a user-space tool.
+odN92kI-rX0-03000-00937945-00938174 It's there, so it would be cool
+odN92kI-rX0-03001-00938174-00938504 to have this knowledge inside the kernel instead.
+odN92kI-rX0-03002-00938504-00938804 Now, so Lynn Brown is here, I talked to him,
+odN92kI-rX0-03003-00938804-00939084 he says that he would also like to have this too
+odN92kI-rX0-03004-00939084-00939366 in Turbostat because it's a problem for turbostat.
+odN92kI-rX0-03005-00939366-00939564 Now Turbostat uses, they have MSR
+odN92kI-rX0-03006-00939564-00939803 and it reads these counters like one after another
+odN92kI-rX0-03007-00939803-00940034 so there is a big time gap between them.
+odN92kI-rX0-03008-00940034-00940401 Also the security people don't like dev MSR, they want
+odN92kI-rX0-03009-00940401-00940729 to disable it, so Turbostat would also benefit from this.
+odN92kI-rX0-03010-00940729-00941070 So probably we need something, some sort of function
+odN92kI-rX0-03011-00941070-00941277 that would disable interrupts preemption
+odN92kI-rX0-03012-00941277-00941531 and read all those counters, it would know
+odN92kI-rX0-03013-00941531-00941818 all those counters for this platform, read them it
+odN92kI-rX0-03014-00941818-00942284 in one go and then kernel consumers
+odN92kI-rX0-03015-00942284-00942720 could run this function and somehow we could use,
+odN92kI-rX0-03016-00942720-00943073 we could send it to user-space if Turbostat wants it.
+odN92kI-rX0-03017-00943073-00943520 - What exactly do you mean by more C-state in the kernel?
+odN92kI-rX0-03018-00943620-00943979 - Ah okay, so yeah when I say C-state syllable confusion
+odN92kI-rX0-03019-00943979-00944245 there are, so I like to, this terminology,
+odN92kI-rX0-03020-00944245-00944532 there are requested C-states, that's those,
+odN92kI-rX0-03021-00944532-00944819 those that see that Linux can request,
+odN92kI-rX0-03022-00944819-00945153 like arguments of the mwait instruction, right,
+odN92kI-rX0-03023-00945153-00945426 but under the hood in the hardware,
+odN92kI-rX0-03024-00945426-00945627 there are more C-states than this.
+odN92kI-rX0-03025-00945627-00945875 I call them hardware C-states.
+odN92kI-rX0-03026-00945875-00946195 So now Linux is aware of those requested C-states
+odN92kI-rX0-03027-00946195-00946589 but Linux kernel is not aware of the hardware C-states.
+odN92kI-rX0-03028-00946589-00946970 - Oh I see, so you are pretty much talking about,
+odN92kI-rX0-03029-00946970-00947231 about the residency counters.
+odN92kI-rX0-03030-00947231-00947314 - Exactly, yeah.
+odN92kI-rX0-03031-00947314-00947469 - Yeah, okay.
+odN92kI-rX0-03032-00947469-00947751 - So all this knowledge is in Turbostat today.
+odN92kI-rX0-03033-00947931-00948383 Yeah so, any ideas, suggestions
+odN92kI-rX0-03034-00948383-00948607 how to do this sort of stuff?
+odN92kI-rX0-03035-00948701-00949059 - I guess maybe it's a question too,
+odN92kI-rX0-03036-00949059-00949423 why the hardware C-states are not in the kernel yet?
+odN92kI-rX0-03037-00949596-00949776 - [Moderator] You mean residency counters?
+odN92kI-rX0-03038-00949990-00950073 - Or to Lynn.
+odN92kI-rX0-03039-00950271-00950766 - So yeah, so I've shown a couple people this patch
+odN92kI-rX0-03040-00950766-00950961 but what we're thinking of doing
+odN92kI-rX0-03041-00950961-00951179 is moving the counters into perf
+odN92kI-rX0-03042-00951179-00951485 where they'll be explicitly advertised by name,
+odN92kI-rX0-03043-00951485-00951760 and so that a tool could look at them by name.
+odN92kI-rX0-03044-00951760-00951984 What Artem wants them, I want them in user-space,
+odN92kI-rX0-03045-00951984-00952198 he wants them in the kernel space.
+odN92kI-rX0-03046-00952198-00952404 So right now we're using just an offset
+odN92kI-rX0-03047-00952404-00952667 that we know from the data book into dev MSR
+odN92kI-rX0-03048-00952667-00953025 and that's convenient but it has downsides,
+odN92kI-rX0-03049-00953025-00953302 multiple system calls is one of them.
+odN92kI-rX0-03050-00953302-00953445 So we want to be able to, we have to update perf
+odN92kI-rX0-03051-00953445-00953667 to do this because you can't read
+odN92kI-rX0-03052-00953667-00953824 multiple counters in one system call now.
+odN92kI-rX0-03053-00953908-00954158 And we want to be able to name them with a vector.
+odN92kI-rX0-03054-00954282-00954591 So that's I think the solution of that problem.
+odN92kI-rX0-03055-00954591-00954992 - So maybe, to circle back on the question,
+odN92kI-rX0-03056-00954992-00955439 the kernel cannot request for those specific states then,
+odN92kI-rX0-03057-00955439-00955602 so that's why it doesn't make sense to have it.
+odN92kI-rX0-03058-00955602-00955836 - [Lynn] And it just doesn't know.
+odN92kI-rX0-03059-00955836-00955919 (speaking faintly)
+odN92kI-rX0-03060-00955919-00956419 - So I think it hasn't been needed in the kernel so far,
+odN92kI-rX0-03061-00956469-00956649 okay, so that's why it's not there.
+odN92kI-rX0-03062-00956778-00956970 - So I have a different question
+odN92kI-rX0-03063-00956970-00957139 from an architectural point of view,
+odN92kI-rX0-03064-00957271-00957429 I say this with no love for BPF
+odN92kI-rX0-03065-00957651-00957734 but this seems actually like quite a natural thing
+odN92kI-rX0-03066-00957734-00957999 to try and solve with BPF because we could have
+odN92kI-rX0-03067-00957999-00958241 arch dependent compilation with the right MSRs.
+odN92kI-rX0-03068-00958346-00958529 You can do a much better export path
+odN92kI-rX0-03069-00958529-00958807 relative versus trace print K.
+odN92kI-rX0-03070-00958807-00959124 You could, there is BPF and puff interaction.
+odN92kI-rX0-03071-00959304-00959487 It seems to solve a lot of problems.
+odN92kI-rX0-03072-00959487-00959661 Has it been considered?
+odN92kI-rX0-03073-00959661-00959789 - [Lynn] Can we access the read MSR?
+odN92kI-rX0-03074-00959789-00960009 - You would have to extend the BPF program
+odN92kI-rX0-03075-00960009-00960256 to give you the access to do that, right.
+odN92kI-rX0-03076-00960256-00960498 There are a few things you could do that.
+odN92kI-rX0-03077-00960498-00960592 You could either have it be something
+odN92kI-rX0-03078-00960712-00960874 that side loads from like,
+odN92kI-rX0-03079-00960874-00961037 I'm not saying it works out the box,
+odN92kI-rX0-03080-00961037-00961277 I'm just saying whether you extended
+odN92kI-rX0-03081-00961277-00961578 the BPF instruction set or you have side load
+odN92kI-rX0-03082-00961578-00962004 prior to the BPF program in CPU idle, right,
+odN92kI-rX0-03083-00962004-00962175 both of those things would work.
+odN92kI-rX0-03084-00962175-00962360 But both of those things seem much simpler
+odN92kI-rX0-03085-00962360-00962678 than what's been outlined with respect to
+odN92kI-rX0-03086-00962678-00962979 both the arc concerns and getting data in and out
+odN92kI-rX0-03087-00962979-00963135 and doing some of the analysis.
+odN92kI-rX0-03088-00963527-00963695 So that's a kind of a general question.
+odN92kI-rX0-03089-00963695-00963946 - [Lynn] Yeah no, we hadn't thought of that, thanks.
+odN92kI-rX0-03090-00963946-00964217 - Yeah thanks, that's interesting suggestion.
+odN92kI-rX0-03091-00964720-00965178 Okay another thing set of challenges
+odN92kI-rX0-03092-00965178-00965346 are user-space interfaces.
+odN92kI-rX0-03093-00965454-00965654 Let's see, so first of all, debugfs
+odN92kI-rX0-03094-00965654-00966062 I feel like for upstream we, if we were in upstream
+odN92kI-rX0-03095-00966062-00966436 it would need to be cisFS rather than the bugfs
+odN92kI-rX0-03096-00966436-00966583 and I would imagine it would be
+odN92kI-rX0-03097-00966583-00966745 something like sys class wult
+odN92kI-rX0-03098-00966840-00967137 or any suggestions here or?
+odN92kI-rX0-03099-00967241-00967496 - [Attendant] But if you, yeah, if you go BPF.
+odN92kI-rX0-03100-00967496-00967801 - Yeah, if we go BPF, if I guess.
+odN92kI-rX0-03101-00967801-00968119 Okay, so for, so these are more like
+odN92kI-rX0-03102-00968119-00968279 configure the measurement themselves,
+odN92kI-rX0-03103-00968279-00968451 like what are those delays.
+odN92kI-rX0-03104-00968451-00968662 Oh they would be in the BPF program, you mean.
+odN92kI-rX0-03105-00968662-00968987 - [Attendant] Right, it would simplify things.
+odN92kI-rX0-03106-00969244-00969443 - Yeah but also we would need enough
+odN92kI-rX0-03107-00969443-00969601 to just switch it on and off.
+odN92kI-rX0-03108-00969695-00969811 - [Attendant] Load and unload.
+odN92kI-rX0-03109-00969900-00970400 - Okay but if it's not BPF then something like this
+odN92kI-rX0-03110-00970665-00970824 would probably make sense to start with.
+odN92kI-rX0-03111-00971039-00971229 - [Attendant] So just trying BPF.
+odN92kI-rX0-03112-00971229-00971585 - Yeah yeah, I understand but let's assume we tried
+odN92kI-rX0-03113-00971585-00971854 and something didn't work, let's just assume.
+odN92kI-rX0-03114-00971854-00972137 - [Attendant] Then you get worried.
+odN92kI-rX0-03115-00972137-00972378 - I would make the meta suggestion
+odN92kI-rX0-03116-00972378-00972523 building on the meta suggestion,
+odN92kI-rX0-03117-00972523-00972745 that if you tried BPF and it didn't work,
+odN92kI-rX0-03118-00972745-00972984 what didn't work in BPF might give you guidance
+odN92kI-rX0-03119-00972984-00973260 as to what this interface should look like.
+odN92kI-rX0-03120-00973260-00973503 - Right.
+odN92kI-rX0-03121-00973503-00973867 - So in BPF, how would it look like?
+odN92kI-rX0-03122-00973867-00974269 Is it like I write a program in some,
+odN92kI-rX0-03123-00974269-00974579 I don't know much about, in some like language,
+odN92kI-rX0-03124-00974579-00974786 special language and then I put it to the kernel
+odN92kI-rX0-03125-00974786-00975118 and it's compiled and kernel execute it.
+odN92kI-rX0-03126-00975118-00975353 - It compiles to native code.
+odN92kI-rX0-03127-00975949-00976090 It compiles to native code but obviously you have some
+odN92kI-rX0-03128-00976090-00976459 specific like read MSR dependency for your arch bits here.
+odN92kI-rX0-03129-00976459-00976669 So it would be a combination of that
+odN92kI-rX0-03130-00976669-00976929 plus either some extension to your program
+odN92kI-rX0-03131-00976929-00977114 to allow it to make those reads,
+odN92kI-rX0-03132-00977114-00977332 or an extension to read and hand it off
+odN92kI-rX0-03133-00977332-00977554 to the BPF program if it's activated.
+odN92kI-rX0-03134-00977554-00977895 But both of those, I'm not suggesting
+odN92kI-rX0-03135-00977895-00978036 either of those is the answer, I'm saying
+odN92kI-rX0-03136-00978036-00978477 those with two possibilities, yeah.
+odN92kI-rX0-03137-00978677-00979177 - Okay, then yeah, this slide was
+odN92kI-rX0-03138-00979231-00979684 about trace print K and my thinking about this.
+odN92kI-rX0-03139-00979684-00980054 Yeah so I, what I want to, my vision here is that
+odN92kI-rX0-03140-00980054-00980292 first of all the, by nature this data
+odN92kI-rX0-03141-00980292-00980542 that we sent to upstream kind of belongs to tracing,
+odN92kI-rX0-03142-00980542-00980931 it just feels like tracing is right, is the right fit.
+odN92kI-rX0-03143-00981058-00981270 Now of course we shouldn't use trace print K.
+odN92kI-rX0-03144-00981270-00981738 So I was thinking that if we,
+odN92kI-rX0-03145-00981829-00982329 if we can use F trace anyway, so we can,
+odN92kI-rX0-03146-00982445-00982825 CPU idle in CPU idle, when we call the enter function,
+odN92kI-rX0-03147-00982825-00983170 we already have a pair of trace events,
+odN92kI-rX0-03148-00983170-00983486 trace points around the enter invocation.
+odN92kI-rX0-03149-00983486-00983791 So if we could hook to them so we can
+odN92kI-rX0-03150-00983791-00984181 take measurements before in idle and after idle, that's one.
+odN92kI-rX0-03151-00984181-00984476 And then if we could create a dynamically from the model,
+odN92kI-rX0-03152-00984476-00984869 created a trace point, trace event that we can trigger,
+odN92kI-rX0-03153-00984869-00985135 right, because the problem now is that
+odN92kI-rX0-03154-00985135-00985275 trace points are, they are kind of static.
+odN92kI-rX0-03155-00985275-00985474 You need to define the arguments
+odN92kI-rX0-03156-00985474-00985731 in advance and like in compile time.
+odN92kI-rX0-03157-00985731-00985940 In this case if we don't know in advance
+odN92kI-rX0-03158-00985940-00986221 what are our C-states, what going to be our arguments,
+odN92kI-rX0-03159-00986221-00986541 we need to create a trace point,
+odN92kI-rX0-03160-00986541-00986821 trace event on the fly, right,
+odN92kI-rX0-03161-00986821-00987082 and like dynamically from within the model.
+odN92kI-rX0-03162-00987082-00987502 So if we had a way to dynamically create trace points.
+odN92kI-rX0-03163-00987502-00987875 Okay Steven taught us yesterday the trace points
+odN92kI-rX0-03164-00987875-00988001 and trace events are different things,
+odN92kI-rX0-03165-00988001-00988309 that's why I kind of get confusing, but yeah.
+odN92kI-rX0-03166-00988309-00988596 So if we could dynamically define a trace event
+odN92kI-rX0-03167-00988596-00988861 from within the model, that would help us.
+odN92kI-rX0-03168-00988861-00988990 We could take the measurement,
+odN92kI-rX0-03169-00988990-00989354 trigger this trace event, it goes to the F trace
+odN92kI-rX0-03170-00989354-00989570 and then perf can also consume it.
+odN92kI-rX0-03171-00989570-00989829 I have patches in my mailbox from Thomson Lucy.
+odN92kI-rX0-03172-00989829-00990032 I'm not sure if he sent them upstream or not
+odN92kI-rX0-03173-00990032-00990281 but these patches implement exactly that
+odN92kI-rX0-03174-00990281-00990616 you can create a trace event from within the model
+odN92kI-rX0-03175-00990616-00991012 by calling several functions within the kernel.
+odN92kI-rX0-03176-00991012-00991350 So that looks very promising, I didn't try them.
+odN92kI-rX0-03177-00991350-00991495 So yeah that's my thinking.
+odN92kI-rX0-03178-00991495-00991749 Any suggestions, comments here?
+odN92kI-rX0-03179-00992321-00992766 - So with BPF you can hook up to existing trace points
+odN92kI-rX0-03180-00992766-00993101 or you can use new trace points by,
+odN92kI-rX0-03181-00993263-00993563 you could add trace points for that
+odN92kI-rX0-03182-00993563-00993993 and hook up to them from the BPF program,
+odN92kI-rX0-03183-00993993-00994349 and then you can define what information
+odN92kI-rX0-03184-00994349-00994618 you will export from that.
+odN92kI-rX0-03185-00994618-00994939 - Yeah right, so there, we have two tasks, right.
+odN92kI-rX0-03186-00994939-00995211 One is to hook to before an after idle,
+odN92kI-rX0-03187-00995211-00995329 that's task number one.
+odN92kI-rX0-03188-00995329-00995492 So we have those trace points.
+odN92kI-rX0-03189-00995492-00995736 - Yeah, so you just add trace points as you,
+odN92kI-rX0-03190-00995736-00996008 defined as places to hook up to.
+odN92kI-rX0-03191-00996008-00996475 - Right, and now we hook up, we collect data,
+odN92kI-rX0-03192-00996475-00996662 now we have something to deliver to user-space.
+odN92kI-rX0-03193-00996662-00996960 How do we deliver through which interface?
+odN92kI-rX0-03194-00996960-00997152 So my thinking was to deliver,
+odN92kI-rX0-03195-00997152-00997592 we dynamically create another trace event,
+odN92kI-rX0-03196-00997592-00997856 let's call it wult trace event, sorry.
+odN92kI-rX0-03197-00997964-00998048 - [Moderator] Well so I don't really
+odN92kI-rX0-03198-00998048-00998208 have an answer ready for that,
+odN92kI-rX0-03199-00998208-00998440 so we, let's just talk about it later.
+odN92kI-rX0-03200-00998440-00998940 - For the eBPF, it can hook to the trace point
+odN92kI-rX0-03201-00998961-00999407 like the idle entry or idle exit.
+odN92kI-rX0-03202-00999519-00999902 This had been existed in the coder side.
+odN92kI-rX0-03203-00999902-01000120 - Right right, I hooked, I have this hooked.
+odN92kI-rX0-03204-01000120-01000415 I run my interposer it before and after, fine.
+odN92kI-rX0-03205-01000415-01000573 I have data now.
+odN92kI-rX0-03206-01000573-01000830 How do I deliver this data to user-space?
+odN92kI-rX0-03207-01000830-01001050 - The eBPF where they have the map
+odN92kI-rX0-03208-01001050-01001348 so they can communicate the data from the color side
+odN92kI-rX0-03209-01001348-01001619 to the user side, so you can.
+odN92kI-rX0-03210-01001619-01001862 But for the eBPF, I follow up one question
+odN92kI-rX0-03211-01002033-01002256 that difficult to see in that,
+odN92kI-rX0-03212-01002256-01002756 how you can create the map or to create eBPF program
+odN92kI-rX0-03213-01002976-01003196 based on the different the CPU topology
+odN92kI-rX0-03214-01003196-01003577 and the CPU idle states.
+odN92kI-rX0-03215-01003577-01004040 So before that I write to the program
+odN92kI-rX0-03216-01004040-01004160 like a hot code to that.
+odN92kI-rX0-03217-01004252-01004508 I'm going to show how how to do that.
+odN92kI-rX0-03218-01004508-01004986 I just to bring up maybe you need to resolve this.
+odN92kI-rX0-03219-01005204-01005288 So you if you have that
+odN92kI-rX0-03220-01005288-01005562 and you need to know the CPUs,
+odN92kI-rX0-03221-01005562-01006039 how many CPU numbers and know what's the CPU idle states,
+odN92kI-rX0-03222-01006168-01006428 then you need to create the eBPF of the program.
+odN92kI-rX0-03223-01006744-01006929 This me might be issue.
+odN92kI-rX0-03224-01006929-01007129 - Okay, so basically what I hear is that
+odN92kI-rX0-03225-01007234-01007600 BPF has mechanisms, so you go figure out
+odN92kI-rX0-03226-01007600-01007837 how to do it through BPF, okay.
+odN92kI-rX0-03227-01007837-01007965 (faintly speaking)
+odN92kI-rX0-03228-01007965-01008060 - Right, okay.
+odN92kI-rX0-03229-01008312-01008529 - [Basdom] You said you want to hook
+odN92kI-rX0-03230-01008643-01008762 before and after something, right?
+odN92kI-rX0-03231-01008762-01008846 - [Artem] Yeah.
+odN92kI-rX0-03232-01008951-01009133 - I am not sure I understand that part
+odN92kI-rX0-03233-01009133-01009492 cause this event that you have is something that
+odN92kI-rX0-03234-01009492-01009850 after you woke up from an idle residency
+odN92kI-rX0-03235-01009850-01010092 you have a number which is the exit latency, right?
+odN92kI-rX0-03236-01010092-01010175 - Yes.
+odN92kI-rX0-03237-01010175-01010425 - So why do you want to hook up before and after?
+odN92kI-rX0-03238-01010425-01010779 I mean you have the number, you just want to be after.
+odN92kI-rX0-03239-01010779-01010896 - [Artem] Yes.
+odN92kI-rX0-03240-01010896-01011140 - Like when you're after the residency completed
+odN92kI-rX0-03241-01011140-01011458 then you have a number which is the exit latency.
+odN92kI-rX0-03242-01011458-01011609 - [Artem] Exactly, yeah.
+odN92kI-rX0-03243-01011609-01011761 - But I understand this being before and after something,
+odN92kI-rX0-03244-01011761-01012029 like if you want to compute a difference, you have
+odN92kI-rX0-03245-01012029-01012324 the difference, it's the knick who is computing that.
+odN92kI-rX0-03246-01012324-01012607 So there's no before and after anything, just after.
+odN92kI-rX0-03247-01012607-01013045 - So if for every data point, we need only a single number
+odN92kI-rX0-03248-01013045-01013293 and this is the latency, that that's enough.
+odN92kI-rX0-03249-01013293-01013562 Now we need more data than that.
+odN92kI-rX0-03250-01013562-01013734 We need several other things.
+odN92kI-rX0-03251-01013734-01014093 First of all is the C-state residency.
+odN92kI-rX0-03252-01014093-01014470 So in x86 we have counters, that just count
+odN92kI-rX0-03253-01014470-01014744 amount of cycles you spent in this C-state.
+odN92kI-rX0-03254-01014744-01014957 I'm talking about hardware C-states now.
+odN92kI-rX0-03255-01014957-01015149 In these hardware C-states, like let's say
+odN92kI-rX0-03256-01015149-01015507 core C1 core C6 package C2 package C6,
+odN92kI-rX0-03257-01015507-01015676 these are absolute numbers.
+odN92kI-rX0-03258-01015676-01015953 So we need to take snapshot of them before idle,
+odN92kI-rX0-03259-01015953-01016275 before M wait and after M wait, and then have the delta.
+odN92kI-rX0-03260-01016374-01016689 Then we know that out of 100%,
+odN92kI-rX0-03261-01016689-01016941 how much we spent in this and that and that
+odN92kI-rX0-03262-01016941-01017098 because that is very helpful when
+odN92kI-rX0-03263-01017098-01017270 you actually start analyzing the data
+odN92kI-rX0-03264-01017270-01017558 and figuring out why it goes down here always
+odN92kI-rX0-03265-01017558-01017917 because you don't have package C6 here.
+odN92kI-rX0-03266-01017917-01018087 And there are there are so many things
+odN92kI-rX0-03267-01018247-01018452 actually you can learn from that, that's why.
+odN92kI-rX0-03268-01018452-01018552 - I have a question.
+odN92kI-rX0-03269-01018748-01018912 So when you imagine the new latency,
+odN92kI-rX0-03270-01018912-01019210 how do you change the system to use a new number?
+odN92kI-rX0-03271-01019477-01019773 - Okay, so right now we didn't change anything.
+odN92kI-rX0-03272-01019773-01020083 So I didn't even finish open source in this tool
+odN92kI-rX0-03273-01020083-01020244 so it's too early to say but
+odN92kI-rX0-03274-01020485-01020765 yeah that's a good for one forward-looking question.
+odN92kI-rX0-03275-01020765-01021088 I don't know if I have good answers right now.
+odN92kI-rX0-03276-01021088-01021439 But I would imagine that others could also try this
+odN92kI-rX0-03277-01021439-01021858 and we could figure it out, yeah.
+odN92kI-rX0-03278-01022193-01022537 - So we have like two minutes to go and then
+odN92kI-rX0-03279-01022537-01022779 the session recording is going to stop at that point.
+odN92kI-rX0-03280-01022779-01023154 So if you want your question recorded, ask it now
+odN92kI-rX0-03281-01023256-01023756 and then yeah, there is, give the mic to them.
+odN92kI-rX0-03282-01024137-01024344 - Yeah it's something related.
+odN92kI-rX0-03283-01024344-01024758 So you mentioned there's a pre wake up CPU timers
+odN92kI-rX0-03284-01024856-01025154 and I'm not sure if manual governor has already been fixed
+odN92kI-rX0-03285-01025260-01025592 because there's a lot, there's a few heuristics
+odN92kI-rX0-03286-01025592-01026016 based on the timer so they can decide to inter C6 or not.
+odN92kI-rX0-03287-01026184-01026427 So if you have this feature, you can just
+odN92kI-rX0-03288-01026514-01026654 for any platform supporting this feature,
+odN92kI-rX0-03289-01026654-01026974 we need to turn off this in the menu governor.
+odN92kI-rX0-03290-01026974-01027194 - Yeah I don't think they are related.
+odN92kI-rX0-03291-01027194-01027458 So menu governor, when you select the C-state to enter,
+odN92kI-rX0-03292-01027458-01027820 right, it's a different thing,
+odN92kI-rX0-03293-01027820-01027937 it's not see state latency.
+odN92kI-rX0-03294-01027937-01028037 It's the break even.
+odN92kI-rX0-03295-01028037-01028203 - It's not orse thing but it's something
+odN92kI-rX0-03296-01028203-01028446 you might need to be fixed in menu governor
+odN92kI-rX0-03297-01028446-01028629 because you have this feature.
+odN92kI-rX0-03298-01028629-01028876 - No, I don't think they are related.
+odN92kI-rX0-03299-01028876-01029053 In menu governor when you select the C-state
+odN92kI-rX0-03300-01029053-01029286 and you know you have to be in this C-state
+odN92kI-rX0-03301-01029286-01029629 at least let's say 200 microseconds
+odN92kI-rX0-03302-01029629-01029871 for it to make sense to save any power
+odN92kI-rX0-03303-01029871-01030335 and you know the next interrupt is 100 microseconds away,
+odN92kI-rX0-03304-01030335-01030544 doesn't make sense to request these deep C-state,
+odN92kI-rX0-03305-01030544-01030674 you request more shallow.
+odN92kI-rX0-03306-01030674-01030849 So it's about power break even,
+odN92kI-rX0-03307-01030849-01031066 it's not about C-state exit latency.
+odN92kI-rX0-03308-01031066-01031228 Although, okay.
+odN92kI-rX0-03309-01031228-01031532 - It's all based on the estimation of future recaps
+odN92kI-rX0-03310-01031532-01031848 but one of the input of how long it's going to
+odN92kI-rX0-03311-01031848-01032098 wake up is actual timer expiration.
+odN92kI-rX0-03312-01032098-01032365 So actually with only that, if you have already have.
+odN92kI-rX0-03313-01032598-01032810 - Okay yeah, we don't have time I think
+odN92kI-rX0-03314-01032905-01033206 but I suggest to talk about this to Rafael.
+odN92kI-rX0-03315-01033206-01033370 He knows everything about that.
+odN92kI-rX0-03316-01033454-01033687 Yeah but my personal opinion, these are not,
+odN92kI-rX0-03317-01033687-01033867 these are kind of orthogonal things.
+odN92kI-rX0-03318-01033867-01034127 - So the short answer is, this is a different timer.
+odN92kI-rX0-03319-01034429-01034662 This is a different timer, it's not the timer he's using.
+odN92kI-rX0-03320-01034662-01034826 It's a different timer.
+odN92kI-rX0-03321-01034826-01034954 - [Attendant] Yeah I'm not talking about different,
+odN92kI-rX0-03322-01034954-01035402 maybe computated, totally independent now his work
+odN92kI-rX0-03323-01035402-01035673 but something like somewhat related.
+odN92kI-rX0-03324-01035673-01035912 - Okay, let's talk about that later.
+odN92kI-rX0-03325-01035912-01036237 Now we need to finish the session at this point.
+odN92kI-rX0-03326-01036393-01036517 Of course we can continue the discussion.
+odN92kI-rX0-03327-01036517-01036849 We have the room for the rest of the evening today.
+odN92kI-rX0-03328-01036849-01037144 But the recording as I said is going to finish
+odN92kI-rX0-03329-01037144-01037450 pretty much right now, so thanks a lot everybody for.
+odN92kI-rX0-03330-01037450-01037641 - [Lynn] Everything else is off the record.
+odN92kI-rX0-03331-01037641-01037909 - Everything else is off the, record, yeah.
+odN92kI-rX0-03332-01037909-01038126 (laughing)
+odN92kI-rX0-03333-01038126-01038309 So yeah, so thanks a lot for participating
+odN92kI-rX0-03334-01038309-01038464 and thanks a lot to the--
+ofgd2kYPNLE-00000-00000000-00000498 The Alaskan Malamute is a large breed of dog that is highly prized for its strength,
+ofgd2kYPNLE-00001-00000498-00001068 endurance, and loyalty. Originally bred by the Inuit people of Alaska to pull sleds and hunt,
+ofgd2kYPNLE-00002-00001068-00001530 the Alaskan Malamute is now a popular family pet and working dog around the world.
+ofgd2kYPNLE-00003-00001530-00002076 The Alaskan Malamute is a strikingly beautiful dog, with a thick double coat that is designed
+ofgd2kYPNLE-00004-00002076-00002580 to keep it warm in even the harshest Arctic conditions. The coat is typically a combination of
+ofgd2kYPNLE-00005-00002580-00003174 white and gray, with some dogs having black or red markings. The Malamute's ears are small and erect,
+ofgd2kYPNLE-00006-00003174-00003600 and its tail is carried over its back. One of the most distinctive features of
+ofgd2kYPNLE-00007-00003600-00004206 the Alaskan Malamute is its size. Adult males can weigh up to 85 pounds and stand as tall as
+ofgd2kYPNLE-00008-00004206-00004788 25 inches at the shoulder, while females are slightly smaller at 75 pounds and 23
+ofgd2kYPNLE-00009-00004788-00005358 inches tall. Despite their large size, Alaskan Malamutes are known for their gentle and friendly
+ofgd2kYPNLE-00010-00005358-00005856 temperament, making them great family pets. While Alaskan Malamutes make great companions,
+ofgd2kYPNLE-00011-00005856-00006431 they are also highly valued as working dogs. The breed is still used for sled racing and
+ofgd2kYPNLE-00012-00006431-00006894 other types of work in cold weather environments, and they are known for their exceptional strength
+ofgd2kYPNLE-00013-00006894-00007409 and endurance. Alaskan Malamutes are also used for search and rescue, therapy work,
+ofgd2kYPNLE-00014-00007409-00007956 and as service dogs for people with disabilities. Because of their intelligence and energy levels,
+ofgd2kYPNLE-00015-00007956-00008478 Alaskan Malamutes require a lot of exercise and stimulation. They are not recommended
+ofgd2kYPNLE-00016-00008478-00008922 for people who live in small apartments or who are not able to provide them with plenty
+ofgd2kYPNLE-00017-00008922-00009462 of exercise and mental stimulation. Alaskan Malamutes thrive on activities like hiking,
+ofgd2kYPNLE-00018-00009462-00009972 running, and other outdoor adventures, and they love to play and interact with their owners.
+ofgd2kYPNLE-00019-00009972-00010446 When it comes to training, Alaskan Malamutes can be challenging for inexperienced dog
+ofgd2kYPNLE-00020-00010446-00010926 owners. They are intelligent and independent, which can make them stubborn and difficult to
+ofgd2kYPNLE-00021-00010926-00011472 train. It's important to establish clear rules and boundaries early on, and to use positive
+ofgd2kYPNLE-00022-00011472-00011958 reinforcement methods to encourage good behavior. One thing to keep in mind when considering an
+ofgd2kYPNLE-00023-00011958-00012324 Alaskan Malamute as a pet is that they can be prone to certain health
+ofgd2kYPNLE-00024-00012324-00012816 issues. Like all large breeds, Alaskan Malamutes are at risk for hip dysplasia,
+ofgd2kYPNLE-00025-00012816-00013380 a genetic condition that affects the joints and can lead to arthritis and other problems. They are
+ofgd2kYPNLE-00026-00013380-00013860 also at risk for certain eye conditions, such as cataracts and progressive retinal atrophy.
+ofgd2kYPNLE-00027-00013860-00014418 Overall, the Alaskan Malamute is a loyal, intelligent, and powerful breed that makes
+ofgd2kYPNLE-00028-00014418-00014808 a great companion for the right owner. They are happiest when they have a job to do,
+ofgd2kYPNLE-00029-00014808-00015282 whether that's pulling a sled or going on a long hike, and they thrive on love and
+ofgd2kYPNLE-00030-00015282-00015810 attention from their owners. If you're looking for a friendly, energetic, and loyal dog that can
+ofgd2kYPNLE-00031-00015810-00016278 keep up with your active lifestyle, the Alaskan Malamute might be the perfect breed for you.
+ok-HZVtppmU-00000-00001628-00002316 Welcome, Welcome! I'm so excited! This is such a special occasion
+ok-HZVtppmU-00001-00002480-00003215 The producers are very pleased, the show has been a great success!
+ok-HZVtppmU-00002-00003620-00004312 And we've been getting a lot of mail from you saying "You, Rantala, are the best!"
+ok-HZVtppmU-00003-00004552-00004984 So the suits asked me "Rantala, would you like to create your own episode?"
+ok-HZVtppmU-00004-00005084-00005616 So I told them "Hmmm... OF COURSE!" So this time, I'm calling the shots!
+ok-HZVtppmU-00005-00005748-00006156 So to start things off, we are going to a sex club! I'm gonna do some fucking
+ok-HZVtppmU-00006-00007152-00007523 -How are you? -I'm fine
+ok-HZVtppmU-00007-00007608-00008059 -Nice propeller -It goes round and round
+ok-HZVtppmU-00008-00008164-00008724 -Are you blushing? -No, I've been here many times
+ok-HZVtppmU-00009-00012780-00013132 Get out fucker!
+ok-HZVtppmU-00010-00013291-00013772 -Up yours scumbag! -It's on now motherfucker
+ok-HZVtppmU-00011-00016136-00016508 -My sister can hit better than that! -You asked for it
+ok-HZVtppmU-00012-00018144-00018776 That didn't work out too well but that's ok. I'm gonna fuck somewhere's else!
+ok-HZVtppmU-00013-00019652-00020480 We have no more money to go to sexclubs. So I'll just go fuck The Party Boys!
+ok-HZVtppmU-00014-00026995-00027392 -You guys wanna hear a joke? -Yeah sure
+ok-HZVtppmU-00015-00027456-00028008 A man was walking in the woods when he saw a sign "Beware of flying cocks"
+ok-HZVtppmU-00016-00028227-00028564 -That was a good one -Yeah just wait a second
+ok-HZVtppmU-00017-00028612-00029164 -So the guy starts laughing "HahahGuarggh!" -Hehehe, that was pretty good
+ok-HZVtppmU-00018-00029232-00029700 -I thought that was silly -Yeah it sucked. I can top that, check it
+ok-HZVtppmU-00019-00029880-00030327 A guy walked in to the woods and saw a sign "Beware of flying cocks"
+ok-HZVtppmU-00020-00030444-00031227 And another sign, "Beware of shitty fucking assholes flying around"
+ok-HZVtppmU-00021-00031395-00032180 So then the cock flew into the asshole! So there was a cock in the ass!
+ok-HZVtppmU-00022-00032320-00033140 Well? Go ahead and laugh. I told you a funny joke. Laugh motherfuckers!
+ok-HZVtppmU-00023-00033484-00033936 You guys suck ass. You just don't get good humour. Rantala would have laughed
+ok-HZVtppmU-00024-00034108-00034696 -You're right. We're nothing compared to him -He is stone groove, my man
+ok-HZVtppmU-00025-00034764-00035260 -And he is so smart too -And above all: such a great stud!
+ok-HZVtppmU-00026-00035760-00036476 -Where do you think that lovely hunk is? -I don't know, I'm starting to miss him
+ok-HZVtppmU-00027-00036712-00037540 -What if he's lying in the gutter? -Shut up Alpo! That can't happen
+ok-HZVtppmU-00028-00037632-00038048 -If he doesn't come soon, I'm going to end it all
+ok-HZVtppmU-00029-00038944-00039300 I want to open!
+ok-HZVtppmU-00030-00041448-00042028 -You with the dope again -We don't want weed, we just want Rantala
+ok-HZVtppmU-00031-00042156-00042716 -There is no high like Rantala! -Damn straight!
+ok-HZVtppmU-00032-00042812-00043368 -You miss him too Janitor, don't you? -Don't worry, we all do
+ok-HZVtppmU-00033-00043496-00044300 -Hands off when he gets here, he's my darling -That is not true! He is my sweetiepie
+ok-HZVtppmU-00034-00044400-00044944 I don't think he wants a nerd with no teeth and no eyes!
+ok-HZVtppmU-00035-00045044-00045544 -Sure I do! Hellohellohello! -RANTALA!!!
+ok-HZVtppmU-00036-00045616-00046096 Here me now: If his teeth are kicked in, there's room for your friend's cock too!
+ok-HZVtppmU-00037-00046164-00046772 And with no eyes, there's two more holes to fuck, with a total of three fuckholes!
+ok-HZVtppmU-00038-00046988-00047360 -Where have you been Rantala?! -We have missed you so much!
+ok-HZVtppmU-00039-00047428-00047824 -We can't stand living without you! -Every day is agony when you're gone!
+ok-HZVtppmU-00040-00047900-00048276 -I was on a barnyard adventure! -Loving it! Give us the hole lowdown
+ok-HZVtppmU-00041-00048372-00048936 What a trip, que no! There were pigs, hens, horses, a cat and a dog!
+ok-HZVtppmU-00042-00049000-00049620 And the icing on the cake: a sow weighing a 1000 kilos! And they were all in heat!
+ok-HZVtppmU-00043-00049800-00050324 -Rantala, you are the fucking master! -Don't I know it. Jealous much?
+ok-HZVtppmU-00044-00050392-00051064 -Oh yes! But I'm hungry for you now... -Rantala, can I have a little taste of you?
+ok-HZVtppmU-00045-00051160-00051864 -Hands off cracker! Nobody touches him but me -Don't be a silly Sally! Everyone gets a piece
+ok-HZVtppmU-00046-00051967-00052640 -But this time: Alpo gets to go first -Really? I get the honour?
+ok-HZVtppmU-00047-00052952-00053408 -Thank you god! -I didn't even get a shower after the pig
+ok-HZVtppmU-00048-00053655-00054079 Calm down, everybody's gonna get theirs Now you know what I like Pekkuli
+ok-HZVtppmU-00049-00054132-00054488 I  know what you like baby!
+ok-HZVtppmU-00050-00055188-00055628 -Well, that was nice, but the other thing -Oh right, I know!
+ok-HZVtppmU-00051-00055691-00056404 Kipe, gimme some tongueplay! Jammu, lick them balls like I taught you!
+ok-HZVtppmU-00052-00056452-00056804 Yes dear!
+ok-HZVtppmU-00053-00057767-00058179 -A marvellous drink! Next time, a tad more giraffe's sperm and you've nailed it
+ok-HZVtppmU-00054-00058408-00059016 -Boys, you want some mustard on that sausage? -Yes love! Of course!
+ok-HZVtppmU-00055-00059064-00059420 Pekkuli, take a shit
+ok-HZVtppmU-00056-00059520-00059972 Just a moment
+ok-HZVtppmU-00057-00060279-00060708 You don't understand! It's art! Art belongs to the people!
+ok-HZVtppmU-00058-00060916-00061744 We had some technical difficulties, but The show must go on, like my idol said
+ok-HZVtppmU-00059-00061876-00062288 So, without further ado, we go now to our viewer feedback
+ok-HZVtppmU-00060-00062367-00062748 VIEWER FEEDBACK!
+ok-HZVtppmU-00061-00062844-00063420 Time for our viewer feedback! Let's see what we have here
+ok-HZVtppmU-00062-00063491-00064140 A guy from Lappeenranta says: Rantala, you are qute! Stop it, I'm blushing
+ok-HZVtppmU-00063-00064276-00065055 Next up... A grandmother from Espoo says
+ok-HZVtppmU-00064-00065128-00065632 She wants more fucking! Grannys in silkstockings getting pounded in the night
+ok-HZVtppmU-00065-00065740-00066160 Send us your address, love!
+ok-HZVtppmU-00066-00066544-00067172 Next Rauli from Jyväskylä says he wants to see more violence!
+ok-HZVtppmU-00067-00067436-00067812 I guess I'll have to fuck with Jammu some more. No problemo!
+ok-HZVtppmU-00068-00067972-00068772 Ulla from Hervanta wants more history, geography, statistics and religion
+ok-HZVtppmU-00069-00068844-00069248 Ulla, my dear, I believe our next insert is right up your alley!
+ok-HZVtppmU-00070-00069348-00069776 RANTALA'S GOURMET HEAVEN
+ok-HZVtppmU-00071-00069860-00070640 Hellohellohello! This is Rantala's gourmet heaven and this is mr Perdus!
+ok-HZVtppmU-00072-00070748-00071176 Good evening Finland, how are you!
+ok-HZVtppmU-00073-00071240-00071972 -We have some delicious treats for you! -Today, we make a the pizza!
+ok-HZVtppmU-00074-00072048-00072628 But first, we give you an organic recipe, and the best thing about these is:
+ok-HZVtppmU-00075-00072676-00073024 THE PRICE!!
+ok-HZVtppmU-00076-00073076-00073692 That's right, because the ingredients are free, so why waste your money!
+ok-HZVtppmU-00077-00073796-00074428 I have produced all these ingredients out of my body yesterday
+ok-HZVtppmU-00078-00074660-00075236 We need 300 grams of faeces, and half a litre of urine
+ok-HZVtppmU-00079-00075340-00076020 Two desilitres of seamen, and one desilitre of sweat
+ok-HZVtppmU-00080-00076220-00076764 A litre of vomit and to spice it up, a fistful of buggers
+ok-HZVtppmU-00081-00076908-00077284 Quite lovely! And what will you cook for us from these delicatessens?
+ok-HZVtppmU-00082-00077500-00078288 Merde á la sauce de la pissé, in other words shit in piss-sauce
+ok-HZVtppmU-00083-00078356-00078708 SHIT IN PISS-SAUCE
+ok-HZVtppmU-00084-00078868-00079476 I have started by frying shit, five minutes will do the trick
+ok-HZVtppmU-00085-00079596-00079940 -Is that a fact? -Oh absolutely
+ok-HZVtppmU-00086-00080296-00080812 -Next, we add the pissé -Goodness gracious!
+ok-HZVtppmU-00087-00081168-00081908 To get the best aroma, the piss must have alcohol, distilled through a bad liver
+ok-HZVtppmU-00088-00082024-00082416 -Merci, Pekkuli -You're welcome, monsieur
+ok-HZVtppmU-00089-00083100-00083576 -The bottle is empty! -Pekkuli drank the secret dip!
+ok-HZVtppmU-00090-00083828-00084620 -Everything is ruined! -Maybe not. I'll handle this
+ok-HZVtppmU-00091-00084724-00085104 The first ingredient was sweat, right?
+ok-HZVtppmU-00092-00085440-00086064 And next, vomit. Looky here Pekkuli, time for lunch... A thank you.
+ok-HZVtppmU-00093-00086288-00086644 Then, the sperm
+ok-HZVtppmU-00094-00087264-00087948 This is not the way to make dip! All the ingredients must be carefully balanced
+ok-HZVtppmU-00095-00088176-00088904 It must be chilled over night in exactly 6,5 degrees celsius! All is lost
+ok-HZVtppmU-00096-00089700-00090108 -Where's the chow Rantala? -I'm hungry as a goddamn motherfucker!
+ok-HZVtppmU-00097-00090332-00090804 -You fucked it up huh? -We wouldn't have eaten that shit anyway
+ok-HZVtppmU-00098-00091004-00091452 Now you listen to me. I'm the boss here, I run this episode!
+ok-HZVtppmU-00099-00091572-00092124 Why don't you go fuck yourself. We ain't getting paid enough for this shit
+ok-HZVtppmU-00100-00092192-00092760 And this gay business has been going on for days! It has to stop or I'm out
+ok-HZVtppmU-00101-00092836-00093664 Me too! I can play the nerd and have fucko here punch on me but no more gay stuff!
+ok-HZVtppmU-00102-00093908-00094280 And how is a man to drink when my mouth is always full of something else
+ok-HZVtppmU-00103-00095108-00095688 We have to get pizza for the guys, they're all cranky and tired, how will they fuck!
+ok-HZVtppmU-00104-00096028-00096772 -Yes, I have the situation under control -What? This is a cookingshow not a pizza-ad
+ok-HZVtppmU-00105-00096928-00097304 You said this was an emergency!
+ok-HZVtppmU-00106-00097636-00098380 -Pizza, pay sixty euros -We're not giving you shit
+ok-HZVtppmU-00107-00098500-00098928 -I don't want shit but sixty euros -Blow me
+ok-HZVtppmU-00108-00098948-00099516 -I knew this you Party bastards! -Ask Rantala for the money
+ok-HZVtppmU-00109-00099584-00099956 -Yeah, this is his show -Yeah, his show
+ok-HZVtppmU-00110-00100028-00100652 -Rantala! -Save me monsieur Perdus!
+ok-HZVtppmU-00111-00100752-00101112 Yes my darling!
+ok-HZVtppmU-00112-00101404-00101784 I want sixty euros right now you vile piece of shit
+ok-HZVtppmU-00113-00101996-00102404 I don't have the money right now but of course I will pay you
+ok-HZVtppmU-00114-00102456-00102832 -How? -Right here baby!
+ok-HZVtppmU-00115-00103256-00103628 -Rantala is dead -Rantala is dead!!!!!
+ok-HZVtppmU-00116-00103835-00104444 -Rantala, my darling! -Rantala is dead now!!!
+ok-HZVtppmU-00117-00104908-00105411 -Why??!! -I love this shit
+ok-HZVtppmU-00118-00105520-00106000 Dear viewers, we are experiencing minor technical difficulties...
+ok-HZVtppmU-00119-00106148-00106535 Out of air-time? You can't do this to me!
+ok-HZVtppmU-00120-00106680-00107180 You listen to me now! I will not work with that fag for another second!
+ok-HZVtppmU-00121-00107248-00108035 You will hear from our lawyers! Cause this is totally bullshit!
+ok-HZVtppmU-00122-00108135-00108576 -And I'm turning you in to the cops! -And I'm going to buy more booze!
+ok-HZVtppmU-00123-00109468-00109972 What are you guys whispering? Talk to me!
+ok-HZVtppmU-00124-00110083-00110708 -That sounds pretty good! -What sounds pretty good?
+ok-HZVtppmU-00125-00110816-00111388 You can't do shit without me! This is my show! I run shit here!
+ok-HZVtppmU-00126-00111459-00111900 Give it here! I want to go first. Ok faggot, now you get your comeuppance
+ok-HZVtppmU-00127-00112816-00113220 -What is this?! Stop it Jammu! -I'm gonna loop this!
+ok-HZVtppmU-00128-00113372-00113928 -This is my show and you must obey me! -Give it here Jammu, my turn
+ok-HZVtppmU-00129-00114444-00114856 -Show mercy you guys! -No mercy for you you freaking pervert!
+ok-HZVtppmU-00130-00114920-00115296 -You try too Pekkuli! -I can dooo thhis
+ok-HZVtppmU-00131-00115708-00116076 Ok, prisonrules. Give me that remote!
+ok-HZVtppmU-00132-00116624-00117232 This is a very nice scene here, loop this and take out the batteries...
+ok-HZVtppmU-00133-00117283-00117944 Give me those motherfucking batteries! Puke them out you sick bastard!
+ok-HZVtppmU-00134-00117996-00118711 Kipe, you can tear them out of my rectum tomorrow, but now enjoy the best scene ever!
+okWcZse_QA8-00000-00000000-00000450 Richard is one of the first legendary commanders that players can get their hands on here in Rise
+okWcZse_QA8-00001-00000450-00000953 of Kingdoms and we haven't talked about him in a long time and seeing as how his Relic recently
+okWcZse_QA8-00002-00000953-00001488 did get a buff today we're gonna be going over Richard in 2023 and whether or not he's worth
+okWcZse_QA8-00003-00001488-00002028 investing in and if you should be using him but first what's going on guys cheers some of you
+okWcZse_QA8-00004-00002028-00002544 have been commenting on my soda drinking this is sparkling water this is Lacroix okay anyway
+okWcZse_QA8-00005-00002544-00003018 let's take an updated look at Richard here in Rise of Kingdoms and I just want to let you guys know
+okWcZse_QA8-00006-00003018-00003432 that this video is mainly for newer players if you've been playing the game since Richard
+okWcZse_QA8-00007-00003432-00004098 came out then you you probably already know a lot about Richard but Richard is an infantry Garrison
+okWcZse_QA8-00008-00004098-00004692 and defense Commander but don't be confused because the Garrison tree does not mean that
+okWcZse_QA8-00009-00004692-00005226 he is a good Garrison Commander or that he should even be used in a Garrison and as a matter of fact
+okWcZse_QA8-00010-00005226-00005880 outside of game modes like maybe Ark of Osiris you pretty much should never use Richard in a Garrison
+okWcZse_QA8-00011-00005880-00006264 and the reason for that is his his active skill and this is the perfect time where we're going
+okWcZse_QA8-00012-00006264-00006786 to go over all the skills on Richard so we know exactly what he's doing his primary skill and the
+okWcZse_QA8-00013-00006786-00007290 main reason that people use Richard to this day maybe not for fighting but for many other things
+okWcZse_QA8-00014-00007290-00007944 is because of his healing Factor so he has a 1400 healing factor and debuffs up to five enemies in
+okWcZse_QA8-00015-00007944-00008610 a fan-shaped area reducing their damage by 30 in March speed by 15 for 2 seconds so very powerful
+okWcZse_QA8-00016-00008610-00009216 debuff relatively short duration unfortunately but the healing Factor will ensure that Richard stays
+okWcZse_QA8-00017-00009216-00009690 in the open field longer the problem with this in a Garrison is that you are taking slightly
+okWcZse_QA8-00018-00009690-00010242 wounded units and putting them back as healthy and then they're attacked again and then a portion
+okWcZse_QA8-00019-00010242-00010794 of those units will become severely wounded so effectively you are converting slightly wounded
+okWcZse_QA8-00020-00010794-00011322 units into severely wounded units you do not want to do that in a Garrison because well because some
+okWcZse_QA8-00021-00011322-00011820 of those troops actually die when he's a Garrison okay you can actually just straight up get deaths
+okWcZse_QA8-00022-00011820-00012330 in a Garrison scenario and you Don don't want a Garrison that will increase the number of deaths
+okWcZse_QA8-00023-00012330-00012834 that you get like that's literally the opposite of of the point of a Garrison okay his second skill
+okWcZse_QA8-00024-00012834-00013446 here says troops led by this Commander gained 10 increased counter attack damage and take 15 less
+okWcZse_QA8-00025-00013446-00014100 damage that's nice that's all damage takes 15 less damage across the board very tanky and worthy of
+okWcZse_QA8-00026-00014100-00014706 the defense talent tree third skill here says infantry units gain 15 attack and defense very
+okWcZse_QA8-00027-00014706-00015318 vanilla very plain very boring third skill fourth skill here says troops the best man receive 30
+okWcZse_QA8-00028-00015318-00015780 increased healing effects and take 30 less damage from watchtowers for those of you who don't know
+okWcZse_QA8-00029-00015780-00016326 watchtowers are only on your wall they're on your city wall if I can go back I will show you I don't
+okWcZse_QA8-00030-00016326-00016776 know why that's not working here you can see this is your Watchtower uh they don't deal any damage
+okWcZse_QA8-00031-00016776-00017304 anyway so that part of the skill is literally nothing maybe at the very beginning of the game if
+okWcZse_QA8-00032-00017304-00017766 you're less than City Hall 25 maybe the Watchtower damage has some influence on you but at the end
+okWcZse_QA8-00033-00017766-00018186 of the day it's not gonna matter at all okay and healing infect enhancement obviously is just good
+okWcZse_QA8-00034-00018186-00018516 for his first skill but really this four skill is not that great and then if you take a look
+okWcZse_QA8-00035-00018516-00018936 at his expertise troops led by this Commander take five percent less damage from all sources
+okWcZse_QA8-00036-00018936-00019524 so that basically bumps this up to 20 damage taken reduction and all the Infantry units led by this
+okWcZse_QA8-00037-00019524-00020004 Commander deal two percent increase damage to Calvary units every 10 seconds the target's
+okWcZse_QA8-00038-00020004-00020670 March speed is reduced by 50 for 5 seconds that last part of his expertise is really the best
+okWcZse_QA8-00039-00020670-00021186 part of his expertise two percent more damage to Cavs is irrelevant because Richard is not dealing
+okWcZse_QA8-00040-00021186-00021546 very much damage anyway and he's not gonna be able to catch the Cavalry in the open field because he
+okWcZse_QA8-00041-00021546-00021990 has no March speed at all now he is slowing down targets but in order for that to work you have to
+okWcZse_QA8-00042-00021990-00022404 actually catch them in the first place next let's take a look at some Talent builds now remember
+okWcZse_QA8-00043-00022404-00022956 talents only apply to the primary commander and there are some instances where you may want to use
+okWcZse_QA8-00044-00022956-00023580 Richard as a primary Commander this is the talent build that I have for my Richard is it very good
+okWcZse_QA8-00045-00023580-00024132 I mean not really but at the end of the day the problem with Richard is that he's not going to
+okWcZse_QA8-00046-00024132-00024726 deal that much damage regardless this talent build I have specifically for battling Barbarians in the
+okWcZse_QA8-00047-00024726-00025224 open field and that is at this point in the game the primary use that I have for Richard now there
+okWcZse_QA8-00048-00025224-00025698 is another use or a few other uses for Richard that we'll talk about later in the video so stay
+okWcZse_QA8-00049-00025698-00026052 tuned until then and you've made it this far into the video hopefully you'll drop a thumbs up on it
+okWcZse_QA8-00050-00026052-00026412 it really helps out a ton but taking a look at the talent trees here we go all the way up to hold the
+okWcZse_QA8-00051-00026412-00026772 line here I'm a huge fan of this Talent it says when this Commander's troops have only infantry
+okWcZse_QA8-00052-00026772-00027227 units 10 chance for 20 damage taking reduction of two seconds it's a short timer but that's almost
+okWcZse_QA8-00053-00027227-00027630 like a skill right there that's really good we get the six percent healthier one percent of Defense
+okWcZse_QA8-00054-00027630-00028092 we also get the rage engine on undying Fury we actually skipped out on iron sphere again because
+okWcZse_QA8-00055-00028092-00028566 this is really for killing barbs but also again you're not really going to catch the Cavalry and
+okWcZse_QA8-00056-00028566-00029094 if you're gonna use Richard for PVP he's almost definitely not going to be a primary Commander
+okWcZse_QA8-00057-00029094-00029520 then we came all the way here we grabbed balance which reduces damage taken by three percent but
+okWcZse_QA8-00058-00029520-00029939 also reduces damage dealt remember he's not really dealing damage anyway so there's no downside to
+okWcZse_QA8-00059-00029939-00030389 that one and then loose formation reduces skill damage taken by nine percent totally fine we came
+okWcZse_QA8-00060-00030389-00030695 all the way up to desperate elegy and you could see that we actually didn't have enough points
+okWcZse_QA8-00061-00030695-00031086 to go all the way in here it's possible that you could argue you want to take this point away and
+okWcZse_QA8-00062-00031086-00031512 put in it to desperate elegy perhaps that that that's the way to go but essentially when this
+okWcZse_QA8-00063-00031512-00031980 troop is reduced to 30 of units remaining they gain an extra 20 raids from each normal attack
+okWcZse_QA8-00064-00031980-00032538 that they take that's at four and then at five it's 25 rage so that's really nice I only got it
+okWcZse_QA8-00065-00032538-00032892 to three unfortunately but the other use that I have for Richard and probably one of the reasons
+okWcZse_QA8-00066-00032892-00033390 that you're going to be using Richard is also in Sunset Canyon he's actually incredibly good in
+okWcZse_QA8-00067-00033390-00033966 Sunset Canyon because of his healing and in Sunset Canyon you always fight to the death and if that's
+okWcZse_QA8-00068-00033966-00034470 the case desperate elegy will come into play every single time which is very good but beyond that we
+okWcZse_QA8-00069-00034470-00034872 actually grab medicinal supplies which gives you even more healing factors so essentially
+okWcZse_QA8-00070-00034872-00035466 his healing factor is 1400 with this it's 1700 and his fourth skill gives him plus 30 if you
+okWcZse_QA8-00071-00035466-00035832 have all three points there which is great we also grabbed testudo formation on the way there which
+okWcZse_QA8-00072-00035832-00036342 is a very good talent as well oh overall a very tanky build that is not going to deal damage but
+okWcZse_QA8-00073-00036342-00036900 it will last a very long time in the open field now if I was going to use Richard for PVP for some
+okWcZse_QA8-00074-00036900-00037398 reason maybe because I'm using him as a buffing March or something like that or maybe it's like
+okWcZse_QA8-00075-00037398-00037896 kvk1 and you want to use them as primary I would probably do something like this where you grab
+okWcZse_QA8-00076-00037896-00038358 balance and loose formation over here you grab all the other talents on that row and then you
+okWcZse_QA8-00077-00038358-00038952 just go all in on the Infantry tree now one thing that you could do is actually take some points
+okWcZse_QA8-00078-00038952-00039330 away and put them in Fleet of foot if you want to move a little bit faster because you're going to
+okWcZse_QA8-00079-00039330-00039864 be very frustrated with how slow Richard actually is so if you wanted to remove maybe talents from
+okWcZse_QA8-00080-00039864-00040236 balance or maybe these little points off on the side here just to gain some more March speed
+okWcZse_QA8-00081-00040236-00040728 that's something that you could do it's up to you and you know if your kvk1 a lot of players aren't
+okWcZse_QA8-00082-00040728-00041136 going to have Max Tech a lot of players are going to be relatively slow regardless so you know if
+okWcZse_QA8-00083-00041136-00041634 you're using him in kvk1 you're probably at least a dolphin or a well and this might be a good build
+okWcZse_QA8-00084-00041634-00042066 for you to go for now I'm worried about putting this talent build in the video but I figured
+okWcZse_QA8-00085-00042066-00042498 if you were going to use him as the Garrison Commander you could consider doing something
+okWcZse_QA8-00086-00042498-00043002 like this now my assumption is that you're using this as a city defense which again you probably
+okWcZse_QA8-00087-00043002-00043416 don't want to you just don't use Richard please use like Martel Sun Tzu or something like that or
+okWcZse_QA8-00088-00043416-00043890 Martel ysg something like that in the early game don't use Richard okay but if you are gonna use
+okWcZse_QA8-00089-00043890-00044328 him guess something like this would be good uh you could see I avoided medicinal supplies because you
+okWcZse_QA8-00090-00044328-00044844 don't want more healing on your wall again nowhere to turn is good go for all the basically all the
+okWcZse_QA8-00091-00044844-00045252 front line of that of that Garrison tree you can come up here and get the front line of all the
+okWcZse_QA8-00092-00045252-00045654 inventory because all this stuff does not have really anything to do with infantry except for
+okWcZse_QA8-00093-00045654-00046122 this but you know realistically if your city's getting hit is probably Cavalry I mean Attila
+okWcZse_QA8-00094-00046122-00046554 Takeda hello I know that's not kbk1 but regardless we have an extra couple extra points I threw it in
+okWcZse_QA8-00095-00046554-00047046 infantry Health even though it might be mixed if it's a city we grabbed balance over here you could
+okWcZse_QA8-00096-00047046-00047370 rearrange some of these extra talents I didn't know what to do with them so I put them where
+okWcZse_QA8-00097-00047370-00047832 I thought it mattered the most and that's what I came up with but again don't use Richard primary
+okWcZse_QA8-00098-00047832-00048282 for Garrison okay okay so now that you have an idea as to what Richard's stats actually look
+okWcZse_QA8-00099-00048282-00048660 like let's talk about what the pros of Richard actually are the first one and this is the most
+okWcZse_QA8-00100-00048660-00049182 obvious one and that is chaining Barbarians in the open field okay and we're gonna talk about PVP so
+okWcZse_QA8-00101-00049182-00049536 don't don't leave the video just yet we're going to talk about PVP in a moment but first Richard
+okWcZse_QA8-00102-00049536-00050088 is my number one barbarian killer and you might be thinking Omniarch that makes no sense he doesn't
+okWcZse_QA8-00103-00050088-00050544 have the peacekeeping tree and that's correct however the thing about Richard is that with his
+okWcZse_QA8-00104-00050544-00051102 healing factor and the fact that he is so tanky he literally can last forever in the open field
+okWcZse_QA8-00105-00051102-00051702 when fighting against barbarians assuming you have decent enough Tech and tier 4 units you you could
+okWcZse_QA8-00106-00051702-00052152 use tier five but like that's expensive and of course his favorite pair is Yi Seong Gye for this
+okWcZse_QA8-00107-00052152-00052626 uh the whole point of Barbarian chaining if you're new to the game is that when Richard is paired
+okWcZse_QA8-00108-00052626-00053196 with Yi Seong Gye Yi Seong Gye's circular AOE is going to trigger nearby barbarians to attack you
+okWcZse_QA8-00109-00053196-00053748 and that means Richard's going to attack them for free and the fact that he's so tanky means that
+okWcZse_QA8-00110-00053748-00054282 he's gonna be able to take these punches like a king uh and then your other peacekeeping marches
+okWcZse_QA8-00111-00054282-00054767 can swarm down that Barbarian meanwhile Richard gets to hit it for free you also could just have
+okWcZse_QA8-00112-00054767-00055248 Richard ysg no other marches and just chain free kills over and over again that's something you
+okWcZse_QA8-00113-00055248-00055638 could do if you have the patience to do that the other thing that's amazing about Richard that a
+okWcZse_QA8-00114-00055638-00056105 lot of people don't talk about is all the other events that you can use Richard in for example
+okWcZse_QA8-00115-00056105-00056670 the golden Kingdom okay this is effectively a PVE gay mode it's an event that comes around that's
+okWcZse_QA8-00116-00056670-00057402 essentially you versus 20 floors of enemies and basically a Sunset Canyon Style game mode and
+okWcZse_QA8-00117-00057402-00058032 Richard actually shines in Sunset Canyon remember he survives a long time and he heals himself in a
+okWcZse_QA8-00118-00058032-00058541 game mode where you can't refresh your troops the only way to get or troops in Sunset Canyon
+okWcZse_QA8-00119-00058541-00059112 is to just bring more with commanders that have an increased troop capacity or by healing them
+okWcZse_QA8-00120-00059112-00059616 and Richard is the king of healing so he's very good in game modes like Sunset Canyon like lost
+okWcZse_QA8-00121-00059616-00060102 Cannon also game modes like golden Kingdom but even beyond that you can use him in events like
+okWcZse_QA8-00122-00060102-00060594 karuak ceremony for example because he's going to stay alive for a long time as I mentioned
+okWcZse_QA8-00123-00060594-00061062 there are some uses for him in Arc of Osiris as well but in general you're mainly going to use
+okWcZse_QA8-00124-00061062-00061746 him for tanking barbs and tanking during events that are really difficult for you to take down
+okWcZse_QA8-00125-00061746-00062279 the enemies now I don't use Richard in my Sunset Canyon lineup right now I've been actually testing
+okWcZse_QA8-00126-00062279-00062748 a couple of different things you can see here I have a Tark with a Herald over here in my offline
+okWcZse_QA8-00127-00062748-00063155 which is not not very common but I'm just testing it out I'm seeing how it's doing and it's actually
+okWcZse_QA8-00128-00063155-00063564 it's actually better than you might think but a couple things that you could do with Richard in
+okWcZse_QA8-00129-00063564-00063929 your Sunset Canyon if you wanted to use him is one you could basically use him just like
+okWcZse_QA8-00130-00063929-00064458 I'm using my constant Constantine here Guan with Richard is something you can use in Sunset Canyon
+okWcZse_QA8-00131-00064458-00065052 in Sunset Canyon you are effectively making guan yu extremely tanky which means his AOE is going to
+okWcZse_QA8-00132-00065052-00065550 pop off for a long time and he's gonna be healing he's gonna stick around for a while and this 30
+okWcZse_QA8-00133-00065550-00066096 increase damage applies to all the enemies in Sunset Canyon so essentially you're debuffing
+okWcZse_QA8-00134-00066096-00066606 the entire enemy's lineup just by having Richard over here in this corner this location is very
+okWcZse_QA8-00135-00066606-00067146 important okay he either wants to be here or he wants to be like this okay just that though
+okWcZse_QA8-00136-00067146-00067530 if you're that that's it okay you want Richard there whether it depends on if you're attacking
+okWcZse_QA8-00137-00067530-00067932 or defending and it also depends on the enemy's lineup if you're attacking so just keep that in
+okWcZse_QA8-00138-00067932-00068388 mind but uh you want to put him here and put them behind Guan the other thing you could do is you
+okWcZse_QA8-00139-00068388-00068976 could put them in your off Lane with a very tanky pairing such as uh Richard primary with Charles
+okWcZse_QA8-00140-00068976-00069582 Martel secondary this is a very common off lane and the reason for this is because again it's
+okWcZse_QA8-00141-00069582-00070212 very tanky so your enemy will have uh two choices either they can put their four armies like this to
+okWcZse_QA8-00142-00070212-00070686 attack the Richard Martel which is Gonna Last a very long time and it's gonna really cause their
+okWcZse_QA8-00143-00070686-00071100 offline to deal a lot of damage to your marches but the other thing is that if they don't attack
+okWcZse_QA8-00144-00071100-00071622 it it's just gonna last forever and kind of hold up one of your armies and that's basically the
+okWcZse_QA8-00145-00071622-00072132 point of that off lane and honestly you have to get pretty far into season of Conquest before you
+okWcZse_QA8-00146-00072132-00072708 start to consider replacing that like seriously Richard Martel is slept on it's very good in
+okWcZse_QA8-00147-00072708-00073266 Sunset Canyon for a very long time because it's just so tanky and there's so much healing it's
+okWcZse_QA8-00148-00073266-00073740 very good if I had better gear for my Richard I would consider just using it here as well but I
+okWcZse_QA8-00149-00073740-00074268 don't I do want to talk real quick about his Relic okay because we we talked about some of the pros
+okWcZse_QA8-00150-00074268-00074640 of him and we should at least mention that he does have a relic and a lot of commanders in the game
+okWcZse_QA8-00151-00074640-00075042 right now do not have a relic and essentially this in the season of Conquest will give him
+okWcZse_QA8-00152-00075042-00075660 that March speed that he desperately needed and you can even upgrade it to 15 now here you get a
+okWcZse_QA8-00153-00075660-00076260 little bit more counter attack damage but really I mean it's not enough damage and the mark speed
+okWcZse_QA8-00154-00076260-00076878 yeah it's nice but I don't know man I don't know it's just this upgraded Relic is like putting
+okWcZse_QA8-00155-00076878-00077340 makeup on a pig sure it looks a little bit better if you're getting pretty desperate if you know
+okWcZse_QA8-00156-00077340-00077766 what I'm saying but at the end of the day it's still a pick you know he really needed something
+okWcZse_QA8-00157-00077766-00078408 exceptional here uh and unfortunately he didn't now if his second star Relic was like 30 percent
+okWcZse_QA8-00158-00078408-00079128 March speed and like 10 counter attack damage he's still probably wouldn't be used for PVP very much
+okWcZse_QA8-00159-00079128-00079638 except for one usage which we'll talk about in a moment but as far as relics go at the end of
+okWcZse_QA8-00160-00079638-00080082 the day do I recommend upgrading as well like no I don't you can see here that I I haven't done
+okWcZse_QA8-00161-00080082-00080574 it okay I unlocked his Relic a long time ago for canyon and I did not upgrade the Relic it's just
+okWcZse_QA8-00162-00080574-00081216 expensive and these coins are very premium now so Richard is just not in it's not worth getting for
+okWcZse_QA8-00163-00081216-00081894 basically a PVE Commander now let's talk about the cons of Richard okay the cons of Richard are
+okWcZse_QA8-00164-00081894-00082566 very numerous okay if you look at his active skill it's a heel and it's a debuff there is no damage
+okWcZse_QA8-00165-00082566-00083106 here there's no instant proc damage on any of his other skills there's no normal attack damage buff
+okWcZse_QA8-00166-00083106-00083628 there's a small counter attack damage buff but if you look at somebody like Martel for example
+okWcZse_QA8-00167-00083628-00084246 he has a 30 counter attack damage buff whereas Richard only has 10. so really Richard is a very
+okWcZse_QA8-00168-00084246-00084924 supportive tank uh and he cannot deal damage back he is effectively a punching bag in the open field
+okWcZse_QA8-00169-00084924-00085326 because he has the healing Factor yes he's gonna last a long time but your trades are going to be
+okWcZse_QA8-00170-00085326-00085866 really bad with Richard okay Richard in the open field is basically feeding your enemy free kills
+okWcZse_QA8-00171-00085866-00086280 now the benefit of Richard in the open field is that if somebody's targeting him he's gonna
+okWcZse_QA8-00172-00086280-00086718 last longer than they might expect it's actually it's actually quite annoying to hit a Richard in
+okWcZse_QA8-00173-00086718-00087096 the open field because you just want to kill him and move on to the next Target but he just stays
+okWcZse_QA8-00174-00087096-00087534 there for a while and you would think in season of Conquest surely you can melt him quickly yeah kind
+okWcZse_QA8-00175-00087534-00088068 of but it's it's still annoying so there's really not a great use for PVP for him and we will talk
+okWcZse_QA8-00176-00088068-00088524 about a good use in just a moment when we talk about Commander pairs um but the other con here
+okWcZse_QA8-00177-00088524-00089154 is March speed there's no March speed here on Richard and he's infantry so he's already very
+okWcZse_QA8-00178-00089154-00089550 slow because he's infantry and there's nothing you can really do about it he's just gonna be
+okWcZse_QA8-00179-00089550-00090150 a slow punching bag okay good luck debuffing the enemy when he you get when the enemy's gone okay
+okWcZse_QA8-00180-00090150-00090672 by the time you get there he's like the Internet Explorer of commanders okay he's he's done he's
+okWcZse_QA8-00181-00090672-00091284 not he's not he's still in 2022. that's how slow Richard is okay very slow no damage those are the
+okWcZse_QA8-00182-00091284-00091740 biggest downsides to him now let's talk about who should get Richard like who should use him
+okWcZse_QA8-00183-00091740-00092172 right I think pretty much every player in the game should get him uh and I think that pretty
+okWcZse_QA8-00184-00092172-00092772 much every player in the game should have a 5-1-1 Richard that's pretty much where you want to stop
+okWcZse_QA8-00185-00092772-00093324 if you want to go a little bit further you could do five five one one and that's pretty good okay
+okWcZse_QA8-00186-00093324-00093906 you could pretty much leave him there and and then you're Gucci if you're a whale then uh you want
+okWcZse_QA8-00187-00093906-00094476 to expertise him in kvk1 or you know you can leave him at five five five one and you're really golden
+okWcZse_QA8-00188-00094476-00094950 there but the wells are the only people that are really should be getting this this expertise
+okWcZse_QA8-00189-00094950-00095472 because this fourth skill is not very good and the expertise really is only okay and even this third
+okWcZse_QA8-00190-00095472-00095994 skill I mean you get 20 more stats by getting it from one to five and twenty percent of stats is
+okWcZse_QA8-00191-00095994-00096486 like nevsky starts with 20 of stats right so we're talking about like legendary Commander sculptures
+okWcZse_QA8-00192-00096486-00096906 going into Richard or going into a newer Commander just you just don't want to use them on Richard
+okWcZse_QA8-00193-00096906-00097314 it's just a waste all right let's talk about some Commander pairs for Richard now if you're in the
+okWcZse_QA8-00194-00097314-00097788 early game this is really the only time you're gonna really consider using Richard I would say
+okWcZse_QA8-00195-00097788-00098124 and if you're using Richard primary there's a couple of options that you have obviously
+okWcZse_QA8-00196-00098124-00098706 in the early game right kvk1 uh you could use him with somebody like Martel obviously that's a good
+okWcZse_QA8-00197-00098706-00099270 choice it's a very tanky Commander pairing the healing combined with the shielding is nuts the
+okWcZse_QA8-00198-00099270-00099750 thing about having Richard debuff the target the healing combined with the shielding here is very
+okWcZse_QA8-00199-00099750-00100338 good you're also gaining a ton of infantry stats and there is some March speed on an expertise to
+okWcZse_QA8-00200-00100338-00100836 Martel keep that in mind it's not on base Martel only expertise unfortunately which makes it very
+okWcZse_QA8-00201-00100836-00101214 unlikely that you'll have this in kbk1 but if you do you get the extra extra large speed here
+okWcZse_QA8-00202-00101214-00101736 okay which Richard really needs the counter attack damage is nice here this is just a solid extremely
+okWcZse_QA8-00203-00101736-00102282 tanky March realistically though in k avk1 you're probably only going to have somebody like Sun Tzu
+okWcZse_QA8-00204-00102282-00102768 and you know this is fine uh the thing about this is that Sun Tzu actually deals damage okay he
+okWcZse_QA8-00205-00102768-00103374 has a five Target AOE with the rage regeneration very good okay very good he's also giving you 10
+okWcZse_QA8-00206-00103374-00103908 less damage which hey that's great Richard is already taking 15 less here but he also gives
+okWcZse_QA8-00207-00103908-00104340 you 10 infantry Health which is very good and unfortunately the skill damage is only gonna be
+okWcZse_QA8-00208-00104340-00104928 good for Sun Tzu but realistically this is a solid kvk1 pairing I really like that I think Martel Sun
+okWcZse_QA8-00209-00104928-00105461 Tzu is a better pairing but you could definitely do something like this moving past kvk1 you could
+okWcZse_QA8-00210-00105461-00105888 start to look at something like an ethyl flood back here and kvk1 you typically don't have very
+okWcZse_QA8-00211-00105888-00106326 much progress on Ethel fled if you have her at all but the thing about this is that this is actually
+okWcZse_QA8-00212-00106326-00106698 a very supportive March okay we talked before about how Richard doesn't really do anything
+okWcZse_QA8-00213-00106698-00107220 besides just be tanky and supportive and Ethel foot hiding behind that Richard is gonna be really
+okWcZse_QA8-00214-00107220-00107700 good because Richard's not really a threat in the open field right like people yeah they want to hit
+okWcZse_QA8-00215-00107700-00108078 him if they're greedy for kills but like really you shouldn't B hitting a Richard in the open
+okWcZse_QA8-00216-00108078-00108402 field there's no point it's not like he's gonna hurt you like what's he gonna do he's not gonna
+okWcZse_QA8-00217-00108402-00108858 do anything right so if people leave you Richard alone which they should be doing then your Ethel
+okWcZse_QA8-00218-00108858-00109361 foot behind him is going to actually do a really solid debuff to five targets has a pretty nice
+okWcZse_QA8-00219-00109361-00109956 AOE damage factor and you also deal more damage to enemies that are slowed and remember you have
+okWcZse_QA8-00220-00109956-00110435 a few different ways of slowing the target Ethel fled herself deals a little bit of a Slowdown here
+okWcZse_QA8-00221-00110435-00110940 on her second skill with a 10 chance but Richard's active skill actually deals a Slowdown here as
+okWcZse_QA8-00222-00110940-00111468 well and his expertise deals a really powerful slowdown if you did go ahead and bring him all
+okWcZse_QA8-00223-00111468-00111966 the way there so you're gonna get a lot of benefit out of this 20 bonus damage from Ethel fled by
+okWcZse_QA8-00224-00111966-00112368 putting her behind Richard now again this Army's still not really going to deal that much damage
+okWcZse_QA8-00225-00112368-00112842 because Ethel flood's AOE is relatively weak and Richard himself is not dealing much damage but
+okWcZse_QA8-00226-00112842-00113298 the fact that you just get that elevation pretty frequently is nice and this is a pretty good way
+okWcZse_QA8-00227-00113298-00113735 to get your Ethel flood out in the open field if you want to do a more supportive March as well
+okWcZse_QA8-00228-00113735-00114078 you could do something like Alexander the Great and I would actually recommend probably Alexander
+okWcZse_QA8-00229-00114078-00114606 the Great primary uh and Richard's secondary it depends if you want more tank go with Richard and
+okWcZse_QA8-00230-00114606-00114996 if you want to hide your Alex go with Richard but if you want more damage the attack tree is
+okWcZse_QA8-00231-00114996-00115440 gonna be it for Alexander the Great and this is just also more support in the open field
+okWcZse_QA8-00232-00115440-00115883 okay you're not only going to Shield yourself but you'll pass around some Shields as well and three
+okWcZse_QA8-00233-00115883-00116376 enemies in the circular AOE will take 30 increased damage which is great you're not gonna be the one
+okWcZse_QA8-00234-00116376-00116916 dealing that damage but at least your other armies or nearby allies will be which is pretty good and
+okWcZse_QA8-00235-00116916-00117474 he's giving you a ton of bonus stats here mainly infantry attack and March feed which again you
+okWcZse_QA8-00236-00117474-00117900 don't have a lot of March speed you have none on Richard and the extra attack is just nice in
+okWcZse_QA8-00237-00117900-00118278 general plus Alexander the Great's not that tanky so having Richard behind him makes him a little
+okWcZse_QA8-00238-00118278-00118716 bit more tanky which he really needs now we could talk about Richard ysg but I already mentioned him
+okWcZse_QA8-00239-00118716-00119100 before and really this is your this is your Barb killing machine this is what you're gonna use
+okWcZse_QA8-00240-00119100-00119711 to kill barbs you're really not gonna use this in PvP because the damage just isn't there and
+okWcZse_QA8-00241-00119711-00120233 you're wasting this 50 increase skill damage on ysg by putting him with a Richard in PvP so really
+okWcZse_QA8-00242-00120233-00120630 not a great PVP March so we're really not going to talk about that but the hidden pairing that
+okWcZse_QA8-00243-00120630-00121361 you could do starting in kvk2 is Saladin primary with Richard secondary and this is a very unique
+okWcZse_QA8-00244-00121361-00122064 combination but the benefit of this pairing is that you are going to slow down targets so much
+okWcZse_QA8-00245-00122064-00122628 it's actually insane now of course if you do this it's going to be all Cavalry okay because Saladin
+okWcZse_QA8-00246-00122628-00123030 is the primary and he has the Cavalry treat so it's gonna be full Cavalry you're gonna miss out
+okWcZse_QA8-00247-00123030-00123761 on these stats but remember if you're most players you have a 5511 or a 5-1-1 Richard and this these
+okWcZse_QA8-00248-00123761-00124194 stats it's irrelevant and anyway okay the first two skills here it doesn't matter what doesn't
+okWcZse_QA8-00249-00124194-00124644 matter what troop type he's actually paired with so if you look at the Synergy here it's actually
+okWcZse_QA8-00250-00124644-00125280 kind of insane because Saladin being Cavalry is actually gonna make Richard very quick okay his
+okWcZse_QA8-00251-00125280-00125759 active skill is going to deal single Target damage not that impressive but it will reduce the March
+okWcZse_QA8-00252-00125759-00126438 speed and healing for 5 seconds that's very that's a very long debuff 30 reduced March speed is huge
+okWcZse_QA8-00253-00126438-00126858 and once you've slowed them down with Saladin who actually has a chance of catching them because
+okWcZse_QA8-00254-00126858-00127380 he's Cavalry then you're gonna hit them with the debuff on Richard's skill here which is not only
+okWcZse_QA8-00255-00127380-00127914 going to heal you but also you're gonna reduce the enemy's March speed again now these overlap
+okWcZse_QA8-00256-00127914-00128418 unfortunately but only for the Target that you're actually hitting this also debuffs remember up to
+okWcZse_QA8-00257-00128418-00128933 four other nearby Targets so this is an AOE slowdown and this is gonna make your Saladin
+okWcZse_QA8-00258-00128933-00129420 a bit more tanky okay Saladin is already pretty tanky because he takes 30 less skill damage and
+okWcZse_QA8-00259-00129420-00129852 20 less counter attack damage now you're adding even more less counter attack damage and all
+okWcZse_QA8-00260-00129852-00130356 damage reduction which is huge but the best part is actually if your Richard is expertise so this
+okWcZse_QA8-00261-00130356-00130938 is a very well combination okay most layers cannot do this but if you're Richard is expertise then
+okWcZse_QA8-00262-00130938-00131532 this 50 March speed reduction every 10 seconds is gonna be huge okay because again Saladin as
+okWcZse_QA8-00263-00131532-00131964 a Cavalry commander is going to help you catch up to those targets as long as you stay connect acted
+okWcZse_QA8-00264-00131964-00132570 for 10 seconds it's like an instant snare okay it instantly applies a 5 Second 50 March speed
+okWcZse_QA8-00265-00132570-00133074 reduction which is a very powerful March speed reduction and if you use this combination right
+okWcZse_QA8-00266-00133074-00133644 here your other armies so your other three or four armies that you're using are really going to be
+okWcZse_QA8-00267-00133644-00134160 able to swarm down that Target that just can't get away it can't get away it's slowed down by Saladin
+okWcZse_QA8-00268-00134160-00134616 it's slowed down by Richard it's slowed down by Richard's expertise there's so much slow down
+okWcZse_QA8-00269-00134616-00135174 here it's ridiculous this is like the ultimate slow down Army but again it's unfortunate kind
+okWcZse_QA8-00270-00135174-00135720 of only a whale Army and really it's not gonna deal that much damage this is a very Niche use
+okWcZse_QA8-00271-00135720-00136398 case uh but you could definitely try it out now some players have tried replacing Saladin uh with
+okWcZse_QA8-00272-00136398-00136998 somebody like guan yu right now guan yu definitely misses a lot of the things that Saladin has like
+okWcZse_QA8-00273-00136998-00137490 being Cavalry and having his own slowdown but if you're just trying to get this expertise in
+okWcZse_QA8-00274-00137490-00137910 the open field you could definitely hide your Richard behind your Guan you if that's all you
+okWcZse_QA8-00275-00137910-00138342 care about are you going to get good trades not with this March okay maybe your other marches are
+okWcZse_QA8-00276-00138342-00138804 gonna get good trades because Richard is snaring them in one in one spot but yeah this is for those
+okWcZse_QA8-00277-00138804-00139260 uh really Advanced players who just have plenty of commanders that and they just don't know what to
+okWcZse_QA8-00278-00139260-00139704 do with them you could give this one a try and that's pretty much it unfortunately Richard's
+okWcZse_QA8-00279-00139704-00140328 only good in the late game for PVE content doing really good tanky stuff for events and Canyon and
+okWcZse_QA8-00280-00140328-00141048 for those whales he can kind of be a really good snare unit besides that PVP wise he's pretty much
+okWcZse_QA8-00281-00141048-00141534 in the garbage and no you should not get his Relic with that being said guys if you found this video
+okWcZse_QA8-00282-00141534-00141954 entertaining or informative drop a thumbs up on it it really helps out the channel a ton of helps get
+okWcZse_QA8-00283-00141954-00142308 this video out into the YouTube algorithm so other Rise of Kingdom Slayers might see it if you're new
+okWcZse_QA8-00284-00142308-00142668 here consider subscribing and clicking the Bell to be notified the next time I upload a rise of
+okWcZse_QA8-00285-00142668-00143088 Kingdom's video comment down below your thoughts on Richard did you actually get him when you first
+okWcZse_QA8-00286-00143088-00143538 started the game or did you completely pass and do you regret it I would love to know with that being
+okWcZse_QA8-00287-00143538-00143856 said guys thank you so much for watching this has been Omni Arc I will talk to you guys again soon
+p0y81P03i8o-00000-00000856-00001592 do
+p0y81P03i8o-00001-00002920-00004384 um
+p0y81P03i8o-00002-00004808-00005808 hello friends welcome back. i'm sitting here being surprised again by another gretsch guitar.
+p0y81P03i8o-00003-00005808-00006504 everyone that comes in regardless of the price point has something unique
+p0y81P03i8o-00004-00006600-00007552 surprising and every price range just delivers value that i'm really not seeing from
+p0y81P03i8o-00005-00007640-00008464 many other brands. this is the i have to look at my notes here the g2655T-P90
+p0y81P03i8o-00006-00008752-00010000 streamliner center block jr double cut p90 with bixby. now gretch i love you i love your guitars
+p0y81P03i8o-00007-00010000-00010528 but can we do something about the names can you just call it fred or something
+p0y81P03i8o-00008-00010648-00011384 um okay well let's get into this this is another guitar in their mid-price line
+p0y81P03i8o-00009-00011448-00012384 it is a mahogany body laminate beautiful binding front and back around the body it has a nato neck
+p0y81P03i8o-00010-00012512-00013200 which we talked about before in one of the gretsch videos it's referred to by some
+p0y81P03i8o-00011-00013200-00014184 people as eastern mahogany it's a very stable wood feels great laurel fretboard on here very
+p0y81P03i8o-00012-00014319-00014863 well finished i have several rosewood several ebony fretboard guitars this
+p0y81P03i8o-00013-00015047-00015672 this will stand up to the feel of any of them on the fretboard medium jumbo frets on a 12
+p0y81P03i8o-00014-00015672-00016456 inch radius and again very nicely finished polished no fret ends that are an issue
+p0y81P03i8o-00015-00016552-00017144 they're just putting great quality into these guitars and i know this is right around a 600
+p0y81P03i8o-00016-00017144-00018072 dollar price guitar it has also a spruce center block in here so it is not a full hollow body but
+p0y81P03i8o-00017-00018072-00018592 it's got a lot of the acoustic properties when we get into the clean sounds you'll see what
+p0y81P03i8o-00018-00018592-00019432 i'm talking about a licensed bigsby b-50 tremolo system on here the adjust-o-matic bridge very nice
+p0y81P03i8o-00019-00019496-00020280 just everything on this guitar to look at it physically the spring on the bigsby is set just
+p0y81P03i8o-00020-00020280-00021024 right it's not too loose it's not too tight other than that it's got the typical gretch control
+p0y81P03i8o-00021-00021024-00021584 configuration on here that it took me a little bit to get used to it but now i'm really liking it
+p0y81P03i8o-00022-00021584-00022344 you have a volume control for the neck pickup the bridge pickup a master tone control and then the
+p0y81P03i8o-00023-00022344-00023024 master volume control that's up here on the horn and i really like the concept behind it because
+p0y81P03i8o-00024-00023024-00023984 when you're in the middle position with both pickups active you can dial in the exact amount
+p0y81P03i8o-00025-00024296-00024608 of the two pickups that you would like to hear adjust the tone
+p0y81P03i8o-00026-00024864-00025927 and then have a master volume that controls that volume it's really very simple and very
+p0y81P03i8o-00027-00025927-00026527 easy to get used to and it's just a very cool implementation if you're playing styles and music
+p0y81P03i8o-00028-00026600-00027272 where this middle position is something that you'll go to on these guitars but i'm digging it
+p0y81P03i8o-00029-00027272-00028368 the tuners on here they appear to be just you know these stock gretch tuners i they feel very nice
+p0y81P03i8o-00030-00028368-00028792 haven't had any issues with this i've been playing on it for a while really liking it
+p0y81P03i8o-00031-00029152-00030016 but let me cut to the treats the reason i really like this guitar so much is the pickups on here
+p0y81P03i8o-00032-00030016-00030776 even though in the name it just says p90 this is a new gretch conception called the
+p0y81P03i8o-00033-00030776-00031695 fidela sonic 90. and it is a single coil large bobbin you know bar magnet underneath the coil
+p0y81P03i8o-00034-00031816-00033784 very much constructed like a p90 but you just take a listen to this
+p0y81P03i8o-00035-00034720-00035184 that's a bridge pickup with the tone full up on here it's
+p0y81P03i8o-00036-00035720-00036584 it has a great attack quality to it and the highs are not overbearing
+p0y81P03i8o-00037-00036704-00037296 and you know i'll give you a shot of the controls i've got back here basically i've got everything
+p0y81P03i8o-00038-00037296-00037984 set on our mesa film or between 12 and 1 o'clock so i'm not hyping the sound back
+p0y81P03i8o-00039-00038240-00038696 if you like a classic p90 for the aggressive
+p0y81P03i8o-00040-00038768-00039408 sound that you can get out of it this takes it in a different direction it's a very unique sounding
+p0y81P03i8o-00041-00039408-00040176 pickup and define something that is this different and this usable in a guitar of this price point
+p0y81P03i8o-00042-00040664-00041264 because i am not one of those guys who wants every guitar to sound like a les paul or a strat
+p0y81P03i8o-00043-00041264-00042184 i like variation in what i've got and this is extremely appealing to me
+p0y81P03i8o-00044-00044208-00044496 it's just such a well-balanced
+p0y81P03i8o-00045-00044584-00045432 and smooth sound for a bridge p90 uh the base is nice and round on it and the treble is nice and
+p0y81P03i8o-00046-00045432-00046112 round on it it's extremely clear it doesn't get fuzzy like a lot of some brands p90s do
+p0y81P03i8o-00047-00046320-00046856 just really dig it i'll go through the clean tones here really quickly for you
+p0y81P03i8o-00048-00046856-00047280 um and then we'll get into some heavy stuff and it's a little bit surprising there too
+p0y81P03i8o-00049-00047336-00047784 okay um you were hearing a little bit on the bridge pickup here
+p0y81P03i8o-00050-00048352-00048952 that's my one touch of the bigsby for the video that is just a gorgeous sound
+p0y81P03i8o-00051-00049080-00050584 go up to the neck pickup on here
+p0y81P03i8o-00052-00051591-00051984 beep
+p0y81P03i8o-00053-00052640-00052840 oops i did it twice but
+p0y81P03i8o-00054-00053255-00053688 it's got the sonic qualities that i like in p90s
+p0y81P03i8o-00055-00053744-00054784 but doesn't have anything any of the qualities i don't like about p90
+p0y81P03i8o-00056-00055240-00055440 and of course in the middle the blend of both of
+p0y81P03i8o-00057-00055440-00056184 them i'm just going to keep everything up full on here
+p0y81P03i8o-00058-00059167-00060264 that's really digging this thing okay now p90s you know if you're associating it more with a heavier
+p0y81P03i8o-00059-00060264-00061136 type of sound it's going to give you that too but very different than what you generally get
+p0y81P03i8o-00060-00061136-00061704 out of a p90 okay i'm going to put a little bit of gain through the fillmore back here
+p0y81P03i8o-00061-00061704-00062144 and i'm just on the crunch channel i'm not maxing it out and by the way these are
+p0y81P03i8o-00062-00062384-00063191 i would say i think we're looking at about an 8 km dc resistance on the bridge pickup
+p0y81P03i8o-00063-00063191-00063976 and about seven and a half on the neck so you're in the range of just basic output
+p0y81P03i8o-00064-00064032-00064632 of a vintage style paf humbucker so as far as the way it's going to hit the front of
+p0y81P03i8o-00065-00064632-00065984 the amp it's going to be very similar to that but let's go to the bridge pickup here
+p0y81P03i8o-00066-00067368-00068784 foreign
+p0y81P03i8o-00067-00068856-00069224 it's full and it's smooth and it's got some
+p0y81P03i8o-00068-00069224-00070184 rippage to it if you know what i'm talking about you can dig in
+p0y81P03i8o-00069-00070632-00071184 it's got the aggression you're looking for but it is not like stabbing yourself in the eye it's
+p0y81P03i8o-00070-00071432-00072096 the the highs and lows are both smooth and tight i just can't get over that enough
+p0y81P03i8o-00071-00072168-00072984 now um up to the neck pickup
+p0y81P03i8o-00072-00075008-00075432 it's gorgeous okay now i i haven't done this yet
+p0y81P03i8o-00073-00075432-00076104 quickly just because i'm going on about this guitar let's see what it does on some high high
+p0y81P03i8o-00074-00076168-00077184 gain here i'm going up to the highest gain channel on here i'm going to stay on the bridge pickup
+p0y81P03i8o-00075-00078848-00079984 let's push it a little bit more
+p0y81P03i8o-00076-00081216-00081384 look
+p0y81P03i8o-00077-00081584-00081880 we are i'll tell you exactly
+p0y81P03i8o-00078-00082248-00082488 599.99
+p0y81P03i8o-00079-00082656-00083632 and with the scaled down body 14 inch wide body on here this is just a fun guitar and
+p0y81P03i8o-00080-00083760-00084464 you heard the tones out of it it's a no-brainer you should have this in your collection
+p0y81P03i8o-00081-00084592-00085280 and it's available here at moorguitars.com and more music in evansville indiana so if you've
+p0y81P03i8o-00082-00085280-00086064 got questions about this guitar or any guitar bass amp pedals give the folks here at moreguitars.com
+p0y81P03i8o-00083-00086064-00086704 a call or come by our store visit us more music in evansville indiana we would love to see you
+p0y81P03i8o-00084-00086768-00087392 the folks here are experts in every brand they sell you're not going to find this kind of
+p0y81P03i8o-00085-00087392-00088296 knowledge about gear anywhere else give them a call and get this into your hand the grech 2655
+p0y81P03i8o-00086-00088296-00089592 tp90 streamliner center block jr double cut p90 with bixby or just ask for fred see you next time
+p2j76RiSyJg-00002-00015596-00016879 Beat x Offset
+p2j76RiSyJg-00003-00016879-00019257 •Alexx Melo•
+p3QIk20Z7CQ-00000-00000000-00000200 COOKING RECIPE
+p6EJg1XtIQY-00000-00000126-00000580 Thank you Mountain View Dodge for my Jeep Cherokee
+p6EJg1XtIQY-00001-00000580-00000882 The service provided by Patsy was exceptional
+p6EJg1XtIQY-00002-00000882-00001296 I will be back for oil changes, tire services and rotations
+p6EJg1XtIQY-00003-00001392-00001712 Thank you Geraldine!
+p97YMwvxJjQ-00000-00000000-00000112 Hello everyone
+p97YMwvxJjQ-00001-00000112-00000288 welcome on my channel
+p97YMwvxJjQ-00002-00000940-00001140 We're not going to do that
+p97YMwvxJjQ-00003-00001140-00001252 no
+p97YMwvxJjQ-00004-00002332-00002472 I'm going to make Dutch videos from now on
+p97YMwvxJjQ-00005-00002472-00002536 Sorry Trisha
+p97YMwvxJjQ-00006-00002536-00002696 I'm going to spend more time on you guys
+p97YMwvxJjQ-00007-00002696-00002856 trying to ignore you guys less
+p97YMwvxJjQ-00008-00002856-00003206 and have more of a dialog, to talk, and do stuff for you
+p97YMwvxJjQ-00009-00003261-00003524 I'm also going to try to make a video every week
+p97YMwvxJjQ-00010-00003524-00003715 That's not a promise
+p97YMwvxJjQ-00011-00003715-00003846 People on the internet can still lie
+p97YMwvxJjQ-00012-00003846-00004184 They do that a lot, especially in front of a camera
+p97YMwvxJjQ-00013-00004184-00004436 Someone also just asked to help them with a short film
+p97YMwvxJjQ-00014-00004436-00004636 And that always takes a lot of time
+p97YMwvxJjQ-00015-00004636-00005024 and I don't like to have a lot of frustration and stress
+p97YMwvxJjQ-00016-00005024-00005230 because I need to make something else like a video
+p97YMwvxJjQ-00017-00005230-00005448 whilst helping to make a short film
+p97YMwvxJjQ-00018-00005494-00005622 So those things are not promises
+p97YMwvxJjQ-00019-00005622-00005740 they're also not lies
+p97YMwvxJjQ-00020-00005740-00006038 They're just three things
+p97YMwvxJjQ-00021-00006742-00006942 I feel as if I'm making a new channel
+p97YMwvxJjQ-00022-00006942-00007116 just because now I'm talking Dutch
+p97YMwvxJjQ-00023-00007116-00007298 even if I'm making video's for three years on this channel
+p97YMwvxJjQ-00024-00007340-00007431 And maybe it is kind of like that
+p97YMwvxJjQ-00025-00007431-00007631 Maybe there's people who couldn't understand me
+p97YMwvxJjQ-00026-00007631-00007814 and now they can
+p97YMwvxJjQ-00027-00007814-00008068 and now they're like: "Yes' I can subscribe to this hotness!
+p97YMwvxJjQ-00028-00008068-00008150 so boys and girls
+p97YMwvxJjQ-00029-00008202-00008268 welcome
+p97YMwvxJjQ-00030-00008282-00008420 I think we need an introduction
+p97YMwvxJjQ-00031-00008420-00008552 Yes. I feel we do
+p97YMwvxJjQ-00032-00008552-00008678 We're going to do that. Introduction!
+p97YMwvxJjQ-00033-00008944-00009134 Hello, I'm Anton, I'm 24 years old
+p97YMwvxJjQ-00034-00009134-00009238 that's like... super old
+p97YMwvxJjQ-00035-00009258-00009464 And if you think I'm over exaggerating
+p97YMwvxJjQ-00036-00009494-00009588 I'm almost 25
+p97YMwvxJjQ-00037-00009636-00009866 And 25 is 1/4th of your life if you live to be 100
+p97YMwvxJjQ-00038-00009958-00010038 So that's pretty close
+p97YMwvxJjQ-00039-00010038-00010212 Even more. If you only get to be 30
+p97YMwvxJjQ-00040-00010212-00010318 that's even closer
+p97YMwvxJjQ-00041-00010318-00010530 and if you only get to be 25 I die next year
+p97YMwvxJjQ-00042-00010830-00010936 I'm a big nerd
+p97YMwvxJjQ-00043-00010936-00011100 I like to read books
+p97YMwvxJjQ-00044-00011100-00011278 I'm a very big fan of Harry Potter
+p97YMwvxJjQ-00045-00011278-00011514 I very much like to play Nintendo games
+p97YMwvxJjQ-00046-00011768-00011866 what was that?
+p97YMwvxJjQ-00047-00011990-00012098 did you guys... no?
+p97YMwvxJjQ-00048-00012144-00012262 I'm also big film fan
+p97YMwvxJjQ-00049-00012262-00012430 I even did 4 years of film school
+p97YMwvxJjQ-00050-00012430-00012740 So I have an official degree that says:
+p97YMwvxJjQ-00051-00012740-00012844 Film Director
+p97YMwvxJjQ-00052-00012894-00013091 Well... I don't have that degree
+p97YMwvxJjQ-00053-00013091-00013284 I graduated about 2 years ago
+p97YMwvxJjQ-00054-00013284-00013424 And I never went to get my degree...
+p97YMwvxJjQ-00055-00013506-00013612 I hope they don't throw it out
+p97YMwvxJjQ-00056-00013636-00013712 Imagine that
+p97YMwvxJjQ-00057-00013728-00013872 That they do throw it out
+p97YMwvxJjQ-00058-00013944-00014130 Ow shit. Maybe I should go and get it...
+p97YMwvxJjQ-00059-00014130-00014330 What else is there to say? I live in Antwerp
+p97YMwvxJjQ-00060-00014382-00014484 In an apartment
+p97YMwvxJjQ-00061-00014512-00014684 I work in Ghent at Mediaraven
+p97YMwvxJjQ-00062-00014684-00014858 which I use the train to get to
+p97YMwvxJjQ-00063-00014878-00015022 I make videos over there
+p97YMwvxJjQ-00064-00015022-00015188 Over there I get payed to make videos
+p97YMwvxJjQ-00065-00015188-00015424 And sometimes I come home and I make more videos
+p97YMwvxJjQ-00066-00015424-00015606 for which I don't get payed
+p97YMwvxJjQ-00067-00015694-00015797 But those are for you guys!
+p97YMwvxJjQ-00068-00015838-00016006 So actually I get payed with the love
+p97YMwvxJjQ-00069-00016006-00016240 you guys sent me through the internet
+p97YMwvxJjQ-00070-00016310-00016386 Thank you
+p97YMwvxJjQ-00071-00016420-00016594 I would like to finish by saying
+p97YMwvxJjQ-00072-00016594-00016768 that I've been watching videos for a long time
+p97YMwvxJjQ-00073-00016768-00017000 Since I was 17, which is 8 years, which I find very long
+p97YMwvxJjQ-00074-00017000-00017122 Especially since I'm only 24
+p97YMwvxJjQ-00075-00017122-00017282 That means Ive been watching YouTube for
+p97YMwvxJjQ-00076-00017282-00017420 1/3th of my life
+p97YMwvxJjQ-00077-00017450-00017650 And naturally, when I started film school
+p97YMwvxJjQ-00078-00017650-00017844 I wanted to make videos myself
+p97YMwvxJjQ-00079-00017896-00018040 And I did
+p97YMwvxJjQ-00080-00018146-00018346 But I just wanted to make stuff
+p97YMwvxJjQ-00081-00018346-00018618 And I put it online without much thought
+p97YMwvxJjQ-00082-00018618-00018732 So I didn't really mean it
+p97YMwvxJjQ-00083-00018732-00018906 But now I want do it serieusly
+p97YMwvxJjQ-00084-00018944-00019144 I feel like the time has come
+p97YMwvxJjQ-00085-00019144-00019406 for me to actually do something with this online stuff
+p97YMwvxJjQ-00086-00019456-00019656 And I think that'll be a lot easier by speaking Dutch
+p97YMwvxJjQ-00087-00019752-00019888 Because I live in Belgium
+p97YMwvxJjQ-00088-00019888-00020000 In Flanders
+p97YMwvxJjQ-00089-00020000-00020112 And we speak Dutch here
+p97YMwvxJjQ-00090-00020112-00020310 Oh my god... That sounded so conservative...
+p97YMwvxJjQ-00091-00020310-00020434 Like the Vlaams-Belang...
+p97YMwvxJjQ-00092-00020460-00020510 Or even N-VA...
+p97YMwvxJjQ-00093-00020550-00020776 Ok. So I hope you'll all keep watching
+p97YMwvxJjQ-00094-00020778-00021134 That you'll all want to be part of this new adventure
+p97YMwvxJjQ-00095-00021220-00021398 I'm excited for this
+p97YMwvxJjQ-00096-00021398-00021530 I'm very excited for this
+p97YMwvxJjQ-00097-00021538-00021668 I hope you guys too
+p97YMwvxJjQ-00098-00021668-00021874 I'd like to say subscribe...
+p97YMwvxJjQ-00099-00021874-00022168 but that's a bridge to far for me
+p97YMwvxJjQ-00100-00022216-00022266 ok
+p97YMwvxJjQ-00101-00022266-00022316 yes
+p97YMwvxJjQ-00102-00022384-00022544 thanks for watching
+p97YMwvxJjQ-00103-00022546-00022638 Let's do this
+p97YMwvxJjQ-00104-00022654-00022922 And apparently till next time!
+p97YMwvxJjQ-00105-00022922-00023032 Yow!
+p9rc0vn3iXc-00000-00000248-00000335 [Announcer] You have entered
+p9rc0vn3iXc-00001-00000335-00000533 the plumbing business success zone.
+p9rc0vn3iXc-00002-00000533-00000809 Teaching entrepreneurs the business of plumbing.
+p9rc0vn3iXc-00003-00000809-00000892 This is "Potty Talk Live."
+p9rc0vn3iXc-00004-00000892-00000986 Recorded live every week.
+p9rc0vn3iXc-00005-00001195-00001462 Turn on notifications to never miss an episode.
+p9rc0vn3iXc-00006-00001462-00001693 Now your host, Richard Behney,
+p9rc0vn3iXc-00007-00001693-00001892 the Million Dollar Plumber.
+p9rc0vn3iXc-00008-00002141-00002275 Hey, hey, hey, plumbing pros,
+p9rc0vn3iXc-00009-00002275-00002704 welcome to 310th episode of "Potty Talk Live!"
+p9rc0vn3iXc-00010-00002704-00002840 Hey there, I'm your host, Richard Behney,
+p9rc0vn3iXc-00011-00002840-00002989 the Million Dollar Plumber.
+p9rc0vn3iXc-00012-00002989-00003277 Along with my co-host, the lovely Laura.
+p9rc0vn3iXc-00013-00003277-00003360 Hello.
+p9rc0vn3iXc-00014-00003360-00003617 Where we're talking the business of plumbing,
+p9rc0vn3iXc-00015-00003617-00004075 and boy, do we have a great one in store for you tonight.
+p9rc0vn3iXc-00016-00004075-00004160 I can't wait to get to this one, all right?
+p9rc0vn3iXc-00017-00004160-00004244 Hey coach.
+p9rc0vn3iXc-00018-00004244-00004492 It's the first of a four-part series, all right?
+p9rc0vn3iXc-00019-00004492-00004576 Hey Bo.
+p9rc0vn3iXc-00020-00004576-00004704 The MDP mindset. MDP mindset.
+p9rc0vn3iXc-00021-00004704-00004841 All right, we're gonna bring it.
+p9rc0vn3iXc-00022-00004841-00004943 Hey, great to have you here.
+p9rc0vn3iXc-00023-00004943-00005102 Great to have you here, baby.
+p9rc0vn3iXc-00024-00005102-00005190 Thank you honey, hey Frankie.
+p9rc0vn3iXc-00025-00005190-00005429 All right, hey, everybody's checking in.
+p9rc0vn3iXc-00026-00005429-00005512 'Cause you know what to do.
+p9rc0vn3iXc-00027-00005512-00005680 All right, in fact, we have our title here.
+p9rc0vn3iXc-00028-00005680-00006045 Tonight's episode is "How to Be a Plumbing CEO."
+p9rc0vn3iXc-00029-00006045-00006128 Okay,
+p9rc0vn3iXc-00030-00006128-00006348 so we're gonna get to it. We're going to bring, all right?
+p9rc0vn3iXc-00031-00006348-00006681 We'll be talking about how to be a plumbing business CEO,
+p9rc0vn3iXc-00032-00006681-00006843 but as everyone's checking in-
+p9rc0vn3iXc-00033-00006843-00007000 Thank you, Darrell.
+p9rc0vn3iXc-00034-00007000-00007151 DW got busted. You made it.
+p9rc0vn3iXc-00035-00007151-00007273 He was doing a little thing on the side,
+p9rc0vn3iXc-00036-00007273-00007648 or whatever it was, had a little meeting or whatever.
+p9rc0vn3iXc-00037-00007648-00007839 Thank you, Darrell.
+p9rc0vn3iXc-00038-00007839-00007934 If you're watching live-
+p9rc0vn3iXc-00039-00007934-00008017 Hey Ricky Bobby.
+p9rc0vn3iXc-00040-00008017-00008232 Let us know that you're watching live with us.
+p9rc0vn3iXc-00041-00008232-00008356 So you're part of the live social audience,
+p9rc0vn3iXc-00042-00008356-00008539 type live in the comments. Yes please, hey Aaron.
+p9rc0vn3iXc-00043-00008539-00008774 If you're watching the replay, please type replay,
+p9rc0vn3iXc-00044-00008774-00008863 let us know that you caught it.
+p9rc0vn3iXc-00045-00008863-00008946 Hey Chris.
+p9rc0vn3iXc-00046-00008946-00009084 And if you're listening to the podcast,
+p9rc0vn3iXc-00047-00009084-00009184 you know what to do.
+p9rc0vn3iXc-00048-00009184-00009421 Just keep on listening.
+p9rc0vn3iXc-00049-00009421-00009535 Kathy, Chris and Kathy both.
+p9rc0vn3iXc-00050-00009535-00009637 All right, Kathy is joining us live.
+p9rc0vn3iXc-00051-00009637-00009820 All right, Chris, all right, good to have you.
+p9rc0vn3iXc-00052-00009820-00010096 Yes, and also, we wanna make sure you guys share this,
+p9rc0vn3iXc-00053-00010096-00010216 share "Potty Talk Live."
+p9rc0vn3iXc-00054-00010216-00010420 So right now, just hit the little share button
+p9rc0vn3iXc-00055-00010420-00010551 at the bottom.
+p9rc0vn3iXc-00056-00010551-00010740 So you can share with all your friends,
+p9rc0vn3iXc-00057-00010740-00010977 'cause everybody would wanna watch "Potty Talk Live."
+p9rc0vn3iXc-00058-00010977-00011215 Hey, you miss an episode of "Potty Talk Live-"
+p9rc0vn3iXc-00059-00011215-00011358 You gonna die.
+p9rc0vn3iXc-00060-00011479-00011729 The Babe is live, all right.
+p9rc0vn3iXc-00061-00011729-00012011 Hey Jeff.
+p9rc0vn3iXc-00062-00012011-00012094 All right, Bay Plumbing.
+p9rc0vn3iXc-00063-00012094-00012224 Hey Joseph.
+p9rc0vn3iXc-00064-00012405-00012548 Happy Monday, baby.
+p9rc0vn3iXc-00065-00012548-00012684 Happy Monday, I know.
+p9rc0vn3iXc-00066-00012684-00012847 You ever had one of those days
+p9rc0vn3iXc-00067-00012847-00012994 where you're just going like a hundred mile an hour?
+p9rc0vn3iXc-00068-00012994-00013196 I can't pronounce that, I'm gonna say live to that person.
+p9rc0vn3iXc-00069-00013196-00013328 Hey GB, welcome, good to see you, brother.
+p9rc0vn3iXc-00070-00013328-00013440 I know he's from South Africa,
+p9rc0vn3iXc-00071-00013440-00013566 but I don't know how to pronounce his name,
+p9rc0vn3iXc-00072-00013566-00013650 so, hello to South Africa.
+p9rc0vn3iXc-00073-00013650-00013806 All our South African mates here,
+p9rc0vn3iXc-00074-00013806-00014081 all right, good to have you here.
+p9rc0vn3iXc-00075-00014302-00014385 Angel!
+p9rc0vn3iXc-00076-00014586-00014819 Hey Miss Blevins.
+p9rc0vn3iXc-00077-00014819-00015126 Just call me angel of the morning, baby.
+p9rc0vn3iXc-00078-00015126-00015346 Oh my goodness, he's got a song for everything.
+p9rc0vn3iXc-00079-00015346-00015589 But I feel like I haven't done anything today.
+p9rc0vn3iXc-00080-00015589-00015672 Hey Christie.
+p9rc0vn3iXc-00081-00015672-00015774 You know, it's like, I look back and it's like,
+p9rc0vn3iXc-00082-00015774-00015896 what'd I do today?
+p9rc0vn3iXc-00083-00015896-00016013 We did, we got a lot of nothing done.
+p9rc0vn3iXc-00084-00016013-00016209 And know, we did a lot, but I know,
+p9rc0vn3iXc-00085-00016209-00016368 but it's like, what do we do?
+p9rc0vn3iXc-00086-00016368-00016485 Oh, there's Danny!
+p9rc0vn3iXc-00087-00016485-00016709 Danny, oh, we just had, Danny-
+p9rc0vn3iXc-00088-00016794-00016955 Danny and Taies were in town,
+p9rc0vn3iXc-00089-00016955-00017116 as well as Ted and Mickey Thompson.
+p9rc0vn3iXc-00090-00017116-00017297 Aw, had a great time with Ted and Mickey too.
+p9rc0vn3iXc-00091-00017297-00017570 Yes, and Elizabeth Fox, with Fox Plumbing.
+p9rc0vn3iXc-00092-00017570-00017671 Danny and Taies headed back today.
+p9rc0vn3iXc-00093-00017671-00017826 And their daughter, yes.
+p9rc0vn3iXc-00094-00017826-00017966 Back to New Era Plumbing.
+p9rc0vn3iXc-00095-00017966-00018100 Yes. A lot of Dracut, Dracut,
+p9rc0vn3iXc-00096-00018100-00018293 is that right? Loved hanging out.
+p9rc0vn3iXc-00097-00018293-00018499 Yes, loved hanging out with you guys, so fun.
+p9rc0vn3iXc-00098-00018499-00018647 They're probably just getting ready for the plane,
+p9rc0vn3iXc-00099-00018647-00018774 I'm assuming.
+p9rc0vn3iXc-00100-00018774-00018947 Oh yeah, 'cause they had later flight.
+p9rc0vn3iXc-00101-00018947-00019142 Yes, but we had a great time with them this week,
+p9rc0vn3iXc-00102-00019142-00019254 very fun.
+p9rc0vn3iXc-00103-00019254-00019372 We are lucky.
+p9rc0vn3iXc-00104-00019372-00019479 We are very blessed.
+p9rc0vn3iXc-00105-00019479-00019562 We have a good time.
+p9rc0vn3iXc-00106-00019562-00019757 Sign up to get text messages right now.
+p9rc0vn3iXc-00107-00019757-00019964 All you have to do is type text in the comments
+p9rc0vn3iXc-00108-00019964-00020240 to get a notification when we go live.
+p9rc0vn3iXc-00109-00020240-00020323 Yes.
+p9rc0vn3iXc-00110-00020323-00020443 So you never miss an episode of "Potty Talk Live,"
+p9rc0vn3iXc-00111-00020443-00020664 just type text, and magically,
+p9rc0vn3iXc-00112-00020664-00020857 a little text message will appear.
+p9rc0vn3iXc-00113-00020857-00020940 You get notified,
+p9rc0vn3iXc-00114-00020940-00021089 there's nothing else worth that's watching.
+p9rc0vn3iXc-00115-00021089-00021201 On Monday night, there's nothing.
+p9rc0vn3iXc-00116-00021201-00021361 Right, okay? Ain't no thing.
+p9rc0vn3iXc-00117-00021361-00021501 So join us here if you're serious
+p9rc0vn3iXc-00118-00021501-00021672 about growing your plumbing business.
+p9rc0vn3iXc-00119-00021672-00021874 PPS Live for the third time in South Africa.
+p9rc0vn3iXc-00120-00021874-00022078 All right, good to have you here, brother.
+p9rc0vn3iXc-00121-00022078-00022185 PPS, that's what we'll call you now.
+p9rc0vn3iXc-00122-00022185-00022412 Glad you're here.
+p9rc0vn3iXc-00123-00022412-00022616 Oh, yes, he did get the dishes done, thanks, Chris.
+p9rc0vn3iXc-00124-00022616-00022753 Did you bust me doing?
+p9rc0vn3iXc-00125-00022753-00022912 I didn't bust you, I like, that's exactly what I-
+p9rc0vn3iXc-00126-00022912-00023057 I mean, I don't know if the Million Dollar Plumber-
+p9rc0vn3iXc-00127-00023057-00023141 It's like my love language.
+p9rc0vn3iXc-00128-00023141-00023386 Should be seen doing the dishes.
+p9rc0vn3iXc-00129-00023386-00023634 Absolutely, because that's the million,
+p9rc0vn3iXc-00130-00023634-00023827 Mrs. Million Dollar Plumber's love language, so.
+p9rc0vn3iXc-00131-00023827-00024093 The lovely Laura beats me when I don't.
+p9rc0vn3iXc-00132-00024093-00024176 You wish.
+p9rc0vn3iXc-00133-00024176-00024336 Oh, I see, Ricky said, sorry I've been missing-
+p9rc0vn3iXc-00134-00024336-00024530 I like when she ties me up.
+p9rc0vn3iXc-00135-00024530-00024644 Missing in action, we are finally starting to slow down-
+p9rc0vn3iXc-00136-00024644-00024727 You didn't hear that.
+p9rc0vn3iXc-00137-00024727-00024863 No, I didn't, and I'm gonna ignore you.
+p9rc0vn3iXc-00138-00024863-00025006 Finally started to slow down in Oklahoma
+p9rc0vn3iXc-00139-00025006-00025089 after the big freeze.
+p9rc0vn3iXc-00140-00025089-00025172 Isn't that crazy?
+p9rc0vn3iXc-00141-00025172-00025283 That Oklahoma and Texas both had this big freeze.
+p9rc0vn3iXc-00142-00025283-00025407 I know, crazy.
+p9rc0vn3iXc-00143-00025407-00025648 So yes, I'm sure you guys were crazy,
+p9rc0vn3iXc-00144-00025648-00026039 first with frozen pipes, and then burst.
+p9rc0vn3iXc-00145-00026130-00026377 DW said he loves him some Million Dollar Plumber.
+p9rc0vn3iXc-00146-00026377-00026647 I love you too, brother.
+p9rc0vn3iXc-00147-00026647-00026850 So anyway, we had a great visit this week
+p9rc0vn3iXc-00148-00026850-00026975 with clients in town.
+p9rc0vn3iXc-00149-00026975-00027226 We got to go out on the boat. Yeah, had a great time.
+p9rc0vn3iXc-00150-00027226-00027311 Do all kinds of fun stuff with them.
+p9rc0vn3iXc-00151-00027311-00027485 Hung out, you know, it's just, we love those-
+p9rc0vn3iXc-00152-00027485-00027587 Ate everything in sight.
+p9rc0vn3iXc-00153-00027587-00027797 This is the problem with clients coming into town,
+p9rc0vn3iXc-00154-00027797-00027965 'cause I mean, they're on vacation.
+p9rc0vn3iXc-00155-00027965-00028049 It's killer.
+p9rc0vn3iXc-00156-00028049-00028132 So they wanna eat like they're on vacation,
+p9rc0vn3iXc-00157-00028132-00028264 but man oh man.
+p9rc0vn3iXc-00158-00028264-00028495 Man, I'm not gonna get to that naughty body
+p9rc0vn3iXc-00159-00028495-00028733 if I'm eating everything naughty.
+p9rc0vn3iXc-00160-00028733-00028988 I know, so we just can't eat again until clients come.
+p9rc0vn3iXc-00161-00028988-00029079 Until the next client.
+p9rc0vn3iXc-00162-00029079-00029206 That's right, we're gonna have water, that's it.
+p9rc0vn3iXc-00163-00029206-00029317 Just water. That's all we're having,
+p9rc0vn3iXc-00164-00029317-00029401 and bread here.
+p9rc0vn3iXc-00165-00029401-00029484 No, no bread, no bread, just water.
+p9rc0vn3iXc-00166-00029484-00029665 No bread, that's bad.
+p9rc0vn3iXc-00167-00029665-00029905 Okay, so it was good, but we had a great time.
+p9rc0vn3iXc-00168-00029905-00029988 We did have a really great time.
+p9rc0vn3iXc-00169-00029988-00030072 I love, it's, you know,
+p9rc0vn3iXc-00170-00030072-00030190 it's like you're a little, you know,
+p9rc0vn3iXc-00171-00030190-00030358 a little sad when things go.
+p9rc0vn3iXc-00172-00030358-00030441 And when they leave. Yeah.
+p9rc0vn3iXc-00173-00030441-00030631 When they leave us 'cause it's like family.
+p9rc0vn3iXc-00174-00030631-00030862 So, all right, I took out the trash, does that count?
+p9rc0vn3iXc-00175-00030862-00030946 It is weird that way.
+p9rc0vn3iXc-00176-00030946-00031227 Yes, of course, that's Kimberly's love language.
+p9rc0vn3iXc-00177-00031227-00031311 If that's her love language.
+p9rc0vn3iXc-00178-00031311-00031631 If that's her love language, then it counts.
+p9rc0vn3iXc-00179-00031631-00031963 Matt Delnay can't hear over the crying babies.
+p9rc0vn3iXc-00180-00031963-00032113 Right, yeah.
+p9rc0vn3iXc-00181-00032113-00032202 People, stop having babies,
+p9rc0vn3iXc-00182-00032202-00032373 there is an easy solution to that.
+p9rc0vn3iXc-00183-00032586-00032726 Just the season, man, it's just the season.
+p9rc0vn3iXc-00184-00032726-00032858 That's right, that's right.
+p9rc0vn3iXc-00185-00032858-00033091 Yeah, the romance starts in the kitchen.
+p9rc0vn3iXc-00186-00033091-00033174 I learned that.
+p9rc0vn3iXc-00187-00033174-00033307 That's right.
+p9rc0vn3iXc-00188-00033307-00033502 The romance starts in the kitchen.
+p9rc0vn3iXc-00189-00033502-00033610 That's right.
+p9rc0vn3iXc-00190-00034149-00034267 All right, what are we talking about tonight?
+p9rc0vn3iXc-00191-00034267-00034479 All right, well here, well, we're gonna talk about-
+p9rc0vn3iXc-00192-00034479-00034562 Oh, we have a question, that's right.
+p9rc0vn3iXc-00193-00034562-00034646 Yeah, but we're gonna talk about
+p9rc0vn3iXc-00194-00034646-00034816 how to be a plumbing business CEO.
+p9rc0vn3iXc-00195-00034816-00034899 Okay.
+p9rc0vn3iXc-00196-00034899-00035049 But the question here that we have
+p9rc0vn3iXc-00197-00035049-00035319 is why did you start your plumbing business?
+p9rc0vn3iXc-00198-00035319-00035402 Why? Listen up-
+p9rc0vn3iXc-00199-00035402-00035539 Why, why, why? Why, why, why?
+p9rc0vn3iXc-00200-00035539-00035745 Why did you start your plumbing business?
+p9rc0vn3iXc-00201-00035745-00035840 Okay?
+p9rc0vn3iXc-00202-00035840-00035930 So let us know.
+p9rc0vn3iXc-00203-00035930-00036114 Why did you start your plumbing business?
+p9rc0vn3iXc-00204-00036114-00036390 And it may be even different from what your why is now,
+p9rc0vn3iXc-00205-00036390-00036501 but what was the main reason?
+p9rc0vn3iXc-00206-00036501-00036916 What was the precipice to starting your plumbing business?
+p9rc0vn3iXc-00207-00036916-00037045 I think mine was I got fired,
+p9rc0vn3iXc-00208-00037045-00037276 so I had to come up with some kind of plan.
+p9rc0vn3iXc-00209-00037276-00037386 Actually, they know it's true.
+p9rc0vn3iXc-00210-00037386-00037549 I think, but I got, right.
+p9rc0vn3iXc-00211-00037549-00037749 I'm that guy, always got, always made-
+p9rc0vn3iXc-00212-00037749-00037896 Oh, you always waves.
+p9rc0vn3iXc-00213-00037896-00037979 Always made, yeah.
+p9rc0vn3iXc-00214-00037979-00038062 And made people nervous.
+p9rc0vn3iXc-00215-00038062-00038228 I'm just not meant to work for anyone.
+p9rc0vn3iXc-00216-00038228-00038323 No, you're not.
+p9rc0vn3iXc-00217-00038323-00038421 And I think I did, I think I got fired,
+p9rc0vn3iXc-00218-00038421-00038609 and I said, well, I gotta-
+p9rc0vn3iXc-00219-00038609-00038807 I'll show him. I'll show him, all right.
+p9rc0vn3iXc-00220-00038965-00039082 That's right. How hard can it be?
+p9rc0vn3iXc-00221-00039082-00039166 Yeah, right.
+p9rc0vn3iXc-00222-00039166-00039484 How hard can this plumbing business thing be?
+p9rc0vn3iXc-00223-00039484-00039791 And Joseph said, I was laid off and desperate!
+p9rc0vn3iXc-00224-00039791-00039874 Yes.
+p9rc0vn3iXc-00225-00039874-00040274 Desperation sometimes is a key motivator, right?
+p9rc0vn3iXc-00226-00040370-00040545 Thanks, GB.
+p9rc0vn3iXc-00227-00040545-00041045 Oh, are you agreeing? I think GB's wise, I think he's agreeing that the romance starts in the kitchen.
+p9rc0vn3iXc-00228-00041048-00041131 Oh yes, that's what he was agreeing to.
+p9rc0vn3iXc-00229-00041131-00041334 I think 'cause I know that GB gets that stuff, you know?
+p9rc0vn3iXc-00230-00041334-00041593 Yes, he is a wise man.
+p9rc0vn3iXc-00231-00041593-00041749 So, you start your plumbing business, guys?
+p9rc0vn3iXc-00232-00041749-00041838 Let's hear it.
+p9rc0vn3iXc-00233-00041838-00041921 Yeah, why did you start your plumbing business?
+p9rc0vn3iXc-00234-00041921-00042004 What was the reason?
+p9rc0vn3iXc-00235-00042004-00042091 Hey Lenny. Hey Lenny,
+p9rc0vn3iXc-00236-00042091-00042198 good to have you here live, brother.
+p9rc0vn3iXc-00237-00042198-00042295 Nice and live.
+p9rc0vn3iXc-00238-00042451-00042601 Yeah, most, a lot of guys will start
+p9rc0vn3iXc-00239-00042601-00042807 their plumbing business, for one,
+p9rc0vn3iXc-00240-00042807-00043060 because they wanna make the money themselves, right?
+p9rc0vn3iXc-00241-00043060-00043227 Right, and that's a biggie we hear,
+p9rc0vn3iXc-00242-00043227-00043484 wanna make the, you know, I can do a better job.
+p9rc0vn3iXc-00243-00043484-00043574 Right. Right.
+p9rc0vn3iXc-00244-00043574-00043721 And freedom. Right, freedom.
+p9rc0vn3iXc-00245-00043721-00043831 Freedom's always a big one, right?
+p9rc0vn3iXc-00246-00043831-00043914 So, why don't we have?
+p9rc0vn3iXc-00247-00043914-00044040 Did anyone type in yet, why don't we have?
+p9rc0vn3iXc-00248-00044040-00044123 Yeah, what's, come on, guys,
+p9rc0vn3iXc-00249-00044123-00044206 I'm looking for comments here.
+p9rc0vn3iXc-00250-00044206-00044351 Why did you start your plumbing business?
+p9rc0vn3iXc-00251-00044351-00044674 What was the main reason that you started it?
+p9rc0vn3iXc-00252-00044674-00045021 I think, yeah, Joseph was pretty transparent there.
+p9rc0vn3iXc-00253-00045021-00045104 That was really good. Laid off and desperate.
+p9rc0vn3iXc-00254-00045104-00045263 You know, and that's a good reason.
+p9rc0vn3iXc-00255-00045263-00045453 I think that's who I was, I mean, we'd just had,
+p9rc0vn3iXc-00256-00045453-00045931 Mallory was just born, and I got fired from my job
+p9rc0vn3iXc-00257-00045931-00046297 for standing up for, yeah, not wearing the mask.
+p9rc0vn3iXc-00258-00046297-00046520 Ricky said, so that I can retire.
+p9rc0vn3iXc-00259-00046520-00046614 So you could retire.
+p9rc0vn3iXc-00260-00046614-00046697 Okay, so, for-
+p9rc0vn3iXc-00261-00046697-00046781 How's that working out for you, man?
+p9rc0vn3iXc-00262-00046781-00047161 Yeah, he looks young, he really, really young, so.
+p9rc0vn3iXc-00263-00047161-00047453 Boss's kid syndrome, loved my old boss,
+p9rc0vn3iXc-00264-00047453-00047637 couldn't stand his kids, oh.
+p9rc0vn3iXc-00265-00047637-00047991 That's a thing, a new leadership comes into play,
+p9rc0vn3iXc-00266-00047991-00048112 and it's like, I'm outta here.
+p9rc0vn3iXc-00267-00048112-00048196 Right.
+p9rc0vn3iXc-00268-00048196-00048477 I wanted to be free to do bigger things, Bo, yeah.
+p9rc0vn3iXc-00269-00048477-00048584 Okay, we hear that with a lot.
+p9rc0vn3iXc-00270-00048584-00048711 Mm, freedom.
+p9rc0vn3iXc-00271-00048711-00048794 Okay, right, freedom.
+p9rc0vn3iXc-00272-00048794-00049084 To create your own destiny.
+p9rc0vn3iXc-00273-00049084-00049202 Well, that's an important reason
+p9rc0vn3iXc-00274-00049202-00049326 why you started your plumbing business.
+p9rc0vn3iXc-00275-00049326-00049430 Yes.
+p9rc0vn3iXc-00276-00049558-00049690 Right, the housing crash.
+p9rc0vn3iXc-00277-00049690-00049815 Oh, Mikey.
+p9rc0vn3iXc-00278-00049815-00049954 Wow, you started a plumbing business
+p9rc0vn3iXc-00279-00049954-00050078 during the housing crash.
+p9rc0vn3iXc-00280-00050078-00050172 Ugh. Whoa.
+p9rc0vn3iXc-00281-00050259-00050413 He is a brave guy.
+p9rc0vn3iXc-00282-00050413-00050524 I wanted to be my own boss.
+p9rc0vn3iXc-00283-00050524-00050832 That is a biggie.
+p9rc0vn3iXc-00284-00050832-00050977 You need an entrepreneur mindset,
+p9rc0vn3iXc-00285-00050977-00051277 one, you know, we wanna be in control,
+p9rc0vn3iXc-00286-00051277-00051387 but more control over your life.
+p9rc0vn3iXc-00287-00051387-00051796 'Cause some guys, like you, just cannot work for someone.
+p9rc0vn3iXc-00288-00051796-00051879 Yeah, I mean, I'm a good worker.
+p9rc0vn3iXc-00289-00051879-00051962 I mean, it wasn't a bunch of drama,
+p9rc0vn3iXc-00290-00052321-00052544 but I was always interested in everything.
+p9rc0vn3iXc-00291-00052544-00052628 Right.
+p9rc0vn3iXc-00292-00052628-00052787 And so, you get a guy that's holding onto the jobs
+p9rc0vn3iXc-00293-00052787-00053159 or the other leadership, they get nervous.
+p9rc0vn3iXc-00294-00053159-00053242 You made people nervous.
+p9rc0vn3iXc-00295-00053242-00053407 I made managers nervous, like I was gonna take their job
+p9rc0vn3iXc-00296-00053407-00053588 or something, because I looked good, you know,
+p9rc0vn3iXc-00297-00053588-00053829 doing the work, and I was always asking questions.
+p9rc0vn3iXc-00298-00053829-00054176 And so, okay. I heard someone say,
+p9rc0vn3iXc-00299-00054176-00054290 you can't fly with the eagles
+p9rc0vn3iXc-00300-00054290-00054432 if you're hanging out with turkeys.
+p9rc0vn3iXc-00301-00054432-00054550 That's right, Jeff!
+p9rc0vn3iXc-00302-00054550-00055050 And if you own your own plumbing business, you are an eagle.
+p9rc0vn3iXc-00303-00055113-00055388 The company to worked for for 28 years locked their doors,
+p9rc0vn3iXc-00304-00055388-00055611 so I had a shot to start up.
+p9rc0vn3iXc-00305-00055611-00055812 With oh, with a nice client list, because yeah.
+p9rc0vn3iXc-00306-00055812-00055969 All right, yeah. There we go.
+p9rc0vn3iXc-00307-00055969-00056088 Way to go, Jeff, okay.
+p9rc0vn3iXc-00308-00056088-00056307 So you got that, so you took the client list,
+p9rc0vn3iXc-00309-00056307-00056502 and you went to town.
+p9rc0vn3iXc-00310-00056502-00056642 Yeah, because they gotta
+p9rc0vn3iXc-00311-00056642-00056844 go somewhere, right?
+p9rc0vn3iXc-00312-00056844-00057135 So, got laid off to tell everyone, and nine years later,
+p9rc0vn3iXc-00313-00057135-00057218 I'm in business.
+p9rc0vn3iXc-00314-00057218-00057330 We're still in business. Got laid off.
+p9rc0vn3iXc-00315-00057330-00057423 Way to go, Chris.
+p9rc0vn3iXc-00316-00057423-00057576 Oh, nice.
+p9rc0vn3iXc-00317-00057576-00057674 All right, good for you.
+p9rc0vn3iXc-00318-00057674-00057776 I wanted to work.
+p9rc0vn3iXc-00319-00057776-00058094 I wanted to work 80-plus hours a week, not really, I know.
+p9rc0vn3iXc-00320-00058094-00058356 That's the thing we're getting to, and that is the-
+p9rc0vn3iXc-00321-00058356-00058439 You're funny, Bo.
+p9rc0vn3iXc-00322-00058439-00058680 But thank you, Bo, that is the rude awakening.
+p9rc0vn3iXc-00323-00058680-00058854 You know, we get into, same kinda thing.
+p9rc0vn3iXc-00324-00058854-00058990 Well, I'll just do it myself.
+p9rc0vn3iXc-00325-00058990-00059073 Freedom!
+p9rc0vn3iXc-00326-00059073-00059156 Right, and we get into it,
+p9rc0vn3iXc-00327-00059156-00059398 and we find out there's a whole lot more going on.
+p9rc0vn3iXc-00328-00059398-00059483 Like business.
+p9rc0vn3iXc-00329-00059483-00059647 Like business, right, exactly.
+p9rc0vn3iXc-00330-00059647-00060042 So, this is the first part
+p9rc0vn3iXc-00331-00060042-00060197 of a four-part series, we're gonna talk
+p9rc0vn3iXc-00332-00060197-00060417 about the MDP mindset.
+p9rc0vn3iXc-00333-00060417-00060635 I think we're gonna do a four-part sermon series
+p9rc0vn3iXc-00334-00060635-00060780 on that one.
+p9rc0vn3iXc-00335-00060780-00061105 This one is "How to Be a Plumbing CEO."
+p9rc0vn3iXc-00336-00061197-00061281 So it's the MDP mindset, that's the series.
+p9rc0vn3iXc-00337-00061281-00061438 The MDP mindset. Yes.
+p9rc0vn3iXc-00338-00061438-00061682 You know, we talk about this, you know, all the time,
+p9rc0vn3iXc-00339-00061682-00061879 that really, you know, like, you know,
+p9rc0vn3iXc-00340-00061879-00062021 with our academy, and you know,
+p9rc0vn3iXc-00341-00062021-00062160 our clients, and that kind of thing.
+p9rc0vn3iXc-00342-00062160-00062504 And the fact that it's good that we get to spend time
+p9rc0vn3iXc-00343-00062504-00062696 with our clients, our private clients,
+p9rc0vn3iXc-00344-00062696-00062911 it's not so much talking about the business stuff,
+p9rc0vn3iXc-00345-00062911-00063102 really about, the successful business,
+p9rc0vn3iXc-00346-00063102-00063382 the things you need to do is really about 20% of it.
+p9rc0vn3iXc-00347-00063382-00063572 The 80% is what goes on up here.
+p9rc0vn3iXc-00348-00063572-00063716 It's a mindset.
+p9rc0vn3iXc-00349-00063716-00063962 That's where most fail, is the mindset.
+p9rc0vn3iXc-00350-00063962-00064301 Not that they aren't good plumbers, or-
+p9rc0vn3iXc-00351-00064301-00064391 Just don't know the business.
+p9rc0vn3iXc-00352-00064391-00064564 You know, don't care, don't write, that kind of a thing.
+p9rc0vn3iXc-00353-00064564-00064868 It's the mindset on how we think of ourselves, okay?
+p9rc0vn3iXc-00354-00064868-00065019 And that kind of a thing.
+p9rc0vn3iXc-00355-00065019-00065513 So, that MDP, Million Dollar Plumber, mindset.
+p9rc0vn3iXc-00356-00065513-00065886 So, we're gonna cover that here in the next four episodes.
+p9rc0vn3iXc-00357-00065886-00065969 But tonight, the first part, we're talking
+p9rc0vn3iXc-00358-00065969-00066469 about how to be a plumbing business CEO.
+p9rc0vn3iXc-00359-00066567-00066831 Which leads to thank you, Sparky.
+p9rc0vn3iXc-00360-00066831-00066914 Moving us along.
+p9rc0vn3iXc-00361-00066914-00067038 Sparky's moving me along.
+p9rc0vn3iXc-00362-00067038-00067325 Not only am I being managed by the lovely Laura.
+p9rc0vn3iXc-00363-00067325-00067646 Now I got this young punk.
+p9rc0vn3iXc-00364-00067646-00067760 He's a punk?
+p9rc0vn3iXc-00365-00067760-00067843 I don't think Sparky's a punk.
+p9rc0vn3iXc-00366-00067843-00068084 Sparky, I love you, I love my little brother.
+p9rc0vn3iXc-00367-00068205-00068646 But number one, I know, he's pushing me along.
+p9rc0vn3iXc-00368-00068646-00068865 Is to speak it, okay?
+p9rc0vn3iXc-00369-00068865-00069043 How to be a plumbing business CEO?
+p9rc0vn3iXc-00370-00069043-00069165 You need to speak it.
+p9rc0vn3iXc-00371-00069165-00069352 I like how you,
+p9rc0vn3iXc-00372-00069352-00069703 you kind of whittled it down, okay, simplified it.
+p9rc0vn3iXc-00373-00069703-00069856 I like what, you know, one the first things
+p9rc0vn3iXc-00374-00069856-00070067 you'll ask our clients, what do you ask?
+p9rc0vn3iXc-00375-00070067-00070150 What do you ask them?
+p9rc0vn3iXc-00376-00070150-00070278 Oh, sure, when we meet with our clients,
+p9rc0vn3iXc-00377-00070278-00070553 and I think we even say this in the academy,
+p9rc0vn3iXc-00378-00070553-00070781 when you meet someone for the first time,
+p9rc0vn3iXc-00379-00070781-00071001 and most of the time, when you say, what's your name,
+p9rc0vn3iXc-00380-00071001-00071195 or whatever, and people usually ask,
+p9rc0vn3iXc-00381-00071195-00071372 so, what do you do for a living, right?
+p9rc0vn3iXc-00382-00071372-00071534 Right.
+p9rc0vn3iXc-00383-00071534-00071768 We find that most of our clients, when they come to us,
+p9rc0vn3iXc-00384-00071768-00071987 the answer to that question is they'll say,
+p9rc0vn3iXc-00385-00071987-00072125 well, I'm a plumber.
+p9rc0vn3iXc-00386-00072125-00072348 And we're like, you own your own plumbing business.
+p9rc0vn3iXc-00387-00072348-00072646 You're not a plumber anymore, you are a CEO.
+p9rc0vn3iXc-00388-00072646-00072729 Right.
+p9rc0vn3iXc-00389-00072729-00072877 And so, speaking that out, so we really challenge
+p9rc0vn3iXc-00390-00072877-00073121 our clients that from that moment on,
+p9rc0vn3iXc-00391-00073121-00073418 anytime someone asks you, what do you do for a living?
+p9rc0vn3iXc-00392-00073418-00073597 You say, I'm a CEO.
+p9rc0vn3iXc-00393-00073597-00073716 I'm a CEO, all right?
+p9rc0vn3iXc-00394-00073716-00073799 Right, and you know what?
+p9rc0vn3iXc-00395-00073799-00073922 And then they'll probably ask you
+p9rc0vn3iXc-00396-00073922-00074029 what kind of business it is.
+p9rc0vn3iXc-00397-00074029-00074209 You can say, I own my own plumbing business.
+p9rc0vn3iXc-00398-00074209-00074419 And then you don't follow it up by saying,
+p9rc0vn3iXc-00399-00074419-00074628 but you know, it's just a small business.
+p9rc0vn3iXc-00400-00074628-00074721 It's just me, and I have another guy.
+p9rc0vn3iXc-00401-00074721-00074915 It's just me and a helper and a truck.
+p9rc0vn3iXc-00402-00075019-00075102 And that kind of thing.
+p9rc0vn3iXc-00403-00075102-00075185 Again we're getting back to mindset.
+p9rc0vn3iXc-00404-00075185-00075269 Right, yes.
+p9rc0vn3iXc-00405-00075269-00075460 Okay. all right. Yeah, speak it.
+p9rc0vn3iXc-00406-00075460-00075571 Right.
+p9rc0vn3iXc-00407-00075571-00075695 Into existence, right?
+p9rc0vn3iXc-00408-00075695-00075779 Exactly.
+p9rc0vn3iXc-00409-00075779-00075884 And then live that.
+p9rc0vn3iXc-00410-00075884-00075967 Right.
+p9rc0vn3iXc-00411-00075967-00076051 You're a CEO. Exactly.
+p9rc0vn3iXc-00412-00076051-00076438 You're a CEO, you created your own business, right?
+p9rc0vn3iXc-00413-00076438-00076552 All right.
+p9rc0vn3iXc-00414-00076552-00076701 I mean that always- Okay.
+p9rc0vn3iXc-00415-00076701-00076901 I always try to get that across to our clients
+p9rc0vn3iXc-00416-00076901-00077034 when they first start working with us
+p9rc0vn3iXc-00417-00077034-00077300 is really think about it, there was nothing,
+p9rc0vn3iXc-00418-00077300-00077669 and then you created it, and so it wasn't there
+p9rc0vn3iXc-00419-00077669-00077774 before you did it, right?
+p9rc0vn3iXc-00420-00077774-00077861 And now you did it.
+p9rc0vn3iXc-00421-00077861-00078005 And most people don't do that.
+p9rc0vn3iXc-00422-00078005-00078102 Right.
+p9rc0vn3iXc-00423-00078102-00078210 That most people do not do that, so.
+p9rc0vn3iXc-00424-00078210-00078293 Right.
+p9rc0vn3iXc-00425-00078293-00078376 That's an eagle.
+p9rc0vn3iXc-00426-00078376-00078665 Yep, okay, you're up there soaring, okay?
+p9rc0vn3iXc-00427-00078665-00078805 So, you need to speak it.
+p9rc0vn3iXc-00428-00078805-00078987 Another thing that we do, you know,
+p9rc0vn3iXc-00429-00078987-00079194 we get into about the idea of,
+p9rc0vn3iXc-00430-00079194-00079427 I talk about your daily GPS, okay?
+p9rc0vn3iXc-00431-00079427-00079596 For those in the academy know what I'm talking about,
+p9rc0vn3iXc-00432-00079596-00079952 or whatever of writing down daily, you know,
+p9rc0vn3iXc-00433-00079952-00080215 you know, they're not necessarily goals,
+p9rc0vn3iXc-00434-00080215-00080368 but it's who you are. Right.
+p9rc0vn3iXc-00435-00080368-00080596 You know, or what you're going to achieve.
+p9rc0vn3iXc-00436-00080596-00080816 So you may not be a 10-truck company yet,
+p9rc0vn3iXc-00437-00080816-00080959 but you say, in a positive,
+p9rc0vn3iXc-00438-00080959-00081209 I have a 10-truck plumbing company.
+p9rc0vn3iXc-00439-00081209-00081292 Right.
+p9rc0vn3iXc-00440-00081292-00081375 That kind of a thing.
+p9rc0vn3iXc-00441-00081375-00081674 But key to that exercise, every morning,
+p9rc0vn3iXc-00442-00081674-00081839 I've done it every morning for years.
+p9rc0vn3iXc-00443-00081839-00082030 Did it this morning, okay?
+p9rc0vn3iXc-00444-00082030-00082243 Is that once you write it out, you speak it out.
+p9rc0vn3iXc-00445-00082243-00082390 Out loud. Okay, out loud.
+p9rc0vn3iXc-00446-00082390-00082664 There's power in speaking it out.
+p9rc0vn3iXc-00447-00082664-00082747 Right.
+p9rc0vn3iXc-00448-00082747-00082840 I mean, that's how the universe was created,
+p9rc0vn3iXc-00449-00082840-00083110 when God created the universe, he spoke it into being
+p9rc0vn3iXc-00450-00083110-00083400 when he created the, you know, the Earth, you know,
+p9rc0vn3iXc-00451-00083400-00083703 and everything with it, and in it, and on it, whatever,
+p9rc0vn3iXc-00452-00083703-00083903 you know, above it, he's spoke it into being.
+p9rc0vn3iXc-00453-00083903-00084051 Right. Okay?
+p9rc0vn3iXc-00454-00084051-00084254 That is just one of those things.
+p9rc0vn3iXc-00455-00084254-00084409 You don't have to understand that, all right?
+p9rc0vn3iXc-00456-00084409-00084561 It's just like you don't, you know,
+p9rc0vn3iXc-00457-00084561-00084755 exactly understand how gravity works,
+p9rc0vn3iXc-00458-00084755-00085001 but you fall off, you know, it's a law of the universe.
+p9rc0vn3iXc-00459-00085001-00085179 You step off a cliff, you're going to drop, okay?
+p9rc0vn3iXc-00460-00085179-00085291 You gonna fall.
+p9rc0vn3iXc-00461-00085291-00085401 Darrell says, get your mind right,
+p9rc0vn3iXc-00462-00085401-00085632 and in the right, can you see that?
+p9rc0vn3iXc-00463-00085632-00085759 Place takes practice- Right.
+p9rc0vn3iXc-00464-00085759-00085904 You have to be intentional about your thought.
+p9rc0vn3iXc-00465-00085904-00086026 Dub, that's really good, DW.
+p9rc0vn3iXc-00466-00086026-00086266 That's wise right there.
+p9rc0vn3iXc-00467-00086266-00086381 Yes, you gotta be intentional about it.
+p9rc0vn3iXc-00468-00086381-00086478 Keyword, intentional.
+p9rc0vn3iXc-00469-00086478-00086687 Gotta be intentional, which means you,
+p9rc0vn3iXc-00470-00086687-00086793 'cause it's easy to not do that,
+p9rc0vn3iXc-00471-00086793-00087006 because your day can kinda get away from you, or whatever.
+p9rc0vn3iXc-00472-00087006-00087147 DW, I hope you don't mind.
+p9rc0vn3iXc-00473-00087147-00087336 DW is a great example.
+p9rc0vn3iXc-00474-00087336-00087549 When we first met him, I mean, he was wearing,
+p9rc0vn3iXc-00475-00087549-00087732 well, we'll get into that, but the whole mindset,
+p9rc0vn3iXc-00476-00087732-00087849 he looked, it was a plumber mindset.
+p9rc0vn3iXc-00477-00087849-00087969 Yes.
+p9rc0vn3iXc-00478-00087969-00088052 He ain't plumber.
+p9rc0vn3iXc-00479-00088052-00088136 He ain't a plumber no more.
+p9rc0vn3iXc-00480-00088136-00088231 He made so many plumber, he ain't no plumber.
+p9rc0vn3iXc-00481-00088231-00088461 He's probably one of the best plumbers out there.
+p9rc0vn3iXc-00482-00088461-00088637 I mean, he's had all the dues,
+p9rc0vn3iXc-00483-00088637-00088829 but he's all, he's business. He's business.
+p9rc0vn3iXc-00484-00088829-00089030 That comes from changing, you know-
+p9rc0vn3iXc-00485-00089030-00089113 Changing his mind.
+p9rc0vn3iXc-00486-00089113-00089252 At how you look at things, okay?
+p9rc0vn3iXc-00487-00089252-00089407 And then speaking these things out.
+p9rc0vn3iXc-00488-00089407-00089490 Yep, definitely, so speak it out loud.
+p9rc0vn3iXc-00489-00089490-00089827 Okay, so speak it out loud.
+p9rc0vn3iXc-00490-00089827-00089947 I am a CEO.
+p9rc0vn3iXc-00491-00089947-00090191 I am a CEO, okay? Correct.
+p9rc0vn3iXc-00492-00090191-00090405 Start looking at yourself that way, all right?
+p9rc0vn3iXc-00493-00090405-00090527 Okay. All right?
+p9rc0vn3iXc-00494-00090527-00090670 Number two.
+p9rc0vn3iXc-00495-00090670-00090845 Number two. Two, two,
+p9rc0vn3iXc-00496-00090845-00090961 there it is. All right.
+p9rc0vn3iXc-00497-00090961-00091123 Clothes make the man.
+p9rc0vn3iXc-00498-00091123-00091206 This one is one of my favorites.
+p9rc0vn3iXc-00499-00091206-00091289 Have you ever heard that saying?
+p9rc0vn3iXc-00500-00091289-00091373 I know. Uh-huh.
+p9rc0vn3iXc-00501-00091373-00091472 Did you make that up?
+p9rc0vn3iXc-00502-00091472-00091762 Yeah, sure.
+p9rc0vn3iXc-00503-00091762-00092022 Clothes make the man, okay?
+p9rc0vn3iXc-00504-00092022-00092107 So, what does that mean?
+p9rc0vn3iXc-00505-00092107-00092395 Okay, starting to dress the part, okay?
+p9rc0vn3iXc-00506-00092395-00092665 No longer wearing the plumbing clothes.
+p9rc0vn3iXc-00507-00092665-00092975 Even if you're still in the truck, wear a different uniform.
+p9rc0vn3iXc-00508-00092975-00093102 Yeah, especially a different uniform
+p9rc0vn3iXc-00509-00093102-00093191 than from what your guys wear.
+p9rc0vn3iXc-00510-00093191-00093324 Than what the guys are wearing.
+p9rc0vn3iXc-00511-00093324-00093537 Yeah, get a Polo or something that
+p9rc0vn3iXc-00512-00093537-00093647 makes you stand out a little bit more.
+p9rc0vn3iXc-00513-00093647-00093783 Okay, you're out of the truck,
+p9rc0vn3iXc-00514-00093783-00093926 or doing out of the truck things,
+p9rc0vn3iXc-00515-00093926-00094244 get rid of wearing the work boots, and the jeans,
+p9rc0vn3iXc-00516-00094244-00094471 or you know, that kind of stuff, okay?
+p9rc0vn3iXc-00517-00094471-00094592 Or your when you're in the office, you know,
+p9rc0vn3iXc-00518-00094592-00094721 maybe you're in, still one foot in the truck,
+p9rc0vn3iXc-00519-00094721-00095061 and one foot in the office, when you're in the office,
+p9rc0vn3iXc-00520-00095061-00095224 don't have the plumbing uniform on.
+p9rc0vn3iXc-00521-00095224-00095378 Have something else that makes you
+p9rc0vn3iXc-00522-00095378-00095605 feel like a business owner.
+p9rc0vn3iXc-00523-00095605-00095735 Right, okay?
+p9rc0vn3iXc-00524-00095735-00095883 Now, I'm not saying you gotta get all suited up and tie,
+p9rc0vn3iXc-00525-00095883-00096052 I mean, I didn't wanna, you know-
+p9rc0vn3iXc-00526-00096052-00096135 Three-piece suit.
+p9rc0vn3iXc-00527-00096135-00096291 And all yeah, and all that, and all that kind of stuff,
+p9rc0vn3iXc-00528-00096291-00096618 you still can find, you know, your groove, okay?
+p9rc0vn3iXc-00529-00096618-00096916 You know, I love, you know, well, we just mentioned,
+p9rc0vn3iXc-00530-00096916-00097179 that Ted and Mickey were in town, okay?
+p9rc0vn3iXc-00531-00097179-00097262 Ted and Mickey Thompson.
+p9rc0vn3iXc-00532-00097262-00097504 And they are just salt of the earth, great people.
+p9rc0vn3iXc-00533-00097504-00097587 Top Hat Plumbing.
+p9rc0vn3iXc-00534-00097587-00097677 Top Hat Plumbing.
+p9rc0vn3iXc-00535-00097677-00097868 Right, when he shared with me when he got married,
+p9rc0vn3iXc-00536-00097868-00098082 he wore, you know, a real nice, you know,
+p9rc0vn3iXc-00537-00098082-00098405 cowboy, you know, shirt and jeans, and whatever.
+p9rc0vn3iXc-00538-00098405-00098584 That's great, you be who you are, okay?
+p9rc0vn3iXc-00539-00098584-00098688 Right.
+p9rc0vn3iXc-00540-00098688-00099032 But it's so important to be different,
+p9rc0vn3iXc-00541-00099032-00099399 to set yourself apart from the frontline guys.
+p9rc0vn3iXc-00542-00099399-00099550 Or even, again-
+p9rc0vn3iXc-00543-00099550-00099633 From me being a plumber.
+p9rc0vn3iXc-00544-00099633-00099864 Right, even if it's still you in the truck,
+p9rc0vn3iXc-00545-00099864-00100173 doing things, even outside, once work's over,
+p9rc0vn3iXc-00546-00100173-00100403 put something different on, other than the regular-
+p9rc0vn3iXc-00547-00100403-00100501 Your regular clothes, like they wear,
+p9rc0vn3iXc-00548-00100501-00100633 that he wears a top hat.
+p9rc0vn3iXc-00549-00100633-00100850 I mean, so it's their personality, and that kind of stuff.
+p9rc0vn3iXc-00550-00100850-00100975 And cool, top and, you know.
+p9rc0vn3iXc-00551-00100975-00101211 But if it's khakis and different kind of shoes,
+p9rc0vn3iXc-00552-00101211-00101368 or something, it can still be boots,
+p9rc0vn3iXc-00553-00101368-00101472 but just not work boots,
+p9rc0vn3iXc-00554-00101472-00101730 and not work pants, and a work shirt,
+p9rc0vn3iXc-00555-00101730-00101841 something different that makes you
+p9rc0vn3iXc-00556-00101841-00102024 feel like a business owner.
+p9rc0vn3iXc-00557-00102024-00102122 Right, exactly.
+p9rc0vn3iXc-00558-00102122-00102206 'Cause the clothes make the man.
+p9rc0vn3iXc-00559-00102206-00102289 You know, so we see, you know, like our guy-
+p9rc0vn3iXc-00560-00102289-00102450 'Cause it's mental.
+p9rc0vn3iXc-00561-00102450-00102755 We had one client, we won't say the name.
+p9rc0vn3iXc-00562-00102755-00102910 They may, I don't know if they're on,
+p9rc0vn3iXc-00563-00102910-00103241 but we had him burn his, 'cause it was really a big deal
+p9rc0vn3iXc-00564-00103241-00103408 to kinda give up that.
+p9rc0vn3iXc-00565-00103408-00103517 'Cause it's what he's always been.
+p9rc0vn3iXc-00566-00103517-00103835 It's an identity, it traps you in with that identity
+p9rc0vn3iXc-00567-00103835-00103939 of being a plumber.
+p9rc0vn3iXc-00568-00103939-00104034 Look, if you wanna be a plumber,
+p9rc0vn3iXc-00569-00104034-00104144 there's nothing wrong with being a plumber.
+p9rc0vn3iXc-00570-00104144-00104227 Of course.
+p9rc0vn3iXc-00571-00104227-00104374 If you wanna be a plumber-
+p9rc0vn3iXc-00572-00104374-00104520 But you made the choice.
+p9rc0vn3iXc-00573-00104520-00104740 Go be a plumber for another company.
+p9rc0vn3iXc-00574-00104740-00104823 Right.
+p9rc0vn3iXc-00575-00104823-00105049 But you've made a choice to start a plumbing business.
+p9rc0vn3iXc-00576-00105049-00105132 Right, so you need to act like it.
+p9rc0vn3iXc-00577-00105132-00105416 And the business, the word business trumps plumbing.
+p9rc0vn3iXc-00578-00105416-00105561 Right. Okay?
+p9rc0vn3iXc-00579-00105561-00105647 So you need to act like it.
+p9rc0vn3iXc-00580-00105647-00105742 Right, okay?
+p9rc0vn3iXc-00581-00105742-00105825 You're not a plumber anymore.
+p9rc0vn3iXc-00582-00105825-00105916 And can you say, well, they can just accept me
+p9rc0vn3iXc-00583-00105916-00106016 the way I am, well, no, it doesn't doesn't work that way.
+p9rc0vn3iXc-00584-00106016-00106234 No, that's not how it works.
+p9rc0vn3iXc-00585-00106234-00106399 Yep, I've been nicer clothes,
+p9rc0vn3iXc-00586-00106399-00106667 it actually helps keep me from going out to a job.
+p9rc0vn3iXc-00587-00106667-00106799 All right, Joe.
+p9rc0vn3iXc-00588-00106799-00106922 Don't wanna wreck those nice jeans,
+p9rc0vn3iXc-00589-00106922-00107015 and then I can't see the rest of it,
+p9rc0vn3iXc-00590-00107015-00107218 but good for you, Melissa's encouraging him, I'm sure,
+p9rc0vn3iXc-00591-00107218-00107321 to not do that.
+p9rc0vn3iXc-00592-00107321-00107426 And that's true, right?
+p9rc0vn3iXc-00593-00107426-00107680 I mean, so you feel, and you just feel better,
+p9rc0vn3iXc-00594-00107680-00107838 and if you have something a little bit nicer on,
+p9rc0vn3iXc-00595-00107838-00108015 and you're like, okay, well, I'm not gonna go out there,
+p9rc0vn3iXc-00596-00108015-00108253 and start digging a ditch. Right, okay.
+p9rc0vn3iXc-00597-00108253-00108423 Now, of course, if you need to,
+p9rc0vn3iXc-00598-00108423-00108610 then that's what we have to do for, you know, whatever,
+p9rc0vn3iXc-00599-00108610-00108858 but try not to.
+p9rc0vn3iXc-00600-00108858-00109156 Right, hey, even when I was back in, you know,
+p9rc0vn3iXc-00601-00109156-00109440 we went through '08, and I've shared with you before,
+p9rc0vn3iXc-00602-00109440-00109665 you know, not everything goes, you know, up and up and up.
+p9rc0vn3iXc-00603-00109665-00109789 And sometimes you gotta do what you gotta do.
+p9rc0vn3iXc-00604-00109789-00109881 And I'm back in the band-
+p9rc0vn3iXc-00605-00109881-00110007 Like that housing everything crash.
+p9rc0vn3iXc-00606-00110007-00110126 Right, when everything crashed,
+p9rc0vn3iXc-00607-00110126-00110316 and even though we had, you know, lost half our guys,
+p9rc0vn3iXc-00608-00110316-00110421 but I was still back, you know,
+p9rc0vn3iXc-00609-00110421-00110723 I went back in the van to get, you know, whatever,
+p9rc0vn3iXc-00610-00110723-00110808 to make my own money.
+p9rc0vn3iXc-00611-00110808-00110986 Oh, he did burn his Chartharts.
+p9rc0vn3iXc-00612-00110986-00111234 I wore the Attaboy uniform,
+p9rc0vn3iXc-00613-00111234-00111379 but mine was all embroidered.
+p9rc0vn3iXc-00614-00111379-00111555 Okay, yeah, a little nicer.
+p9rc0vn3iXc-00615-00111555-00111701 Mine was, you know, it was nicer
+p9rc0vn3iXc-00616-00111701-00111834 than what the other guys were wearing, okay?
+p9rc0vn3iXc-00617-00111834-00111918 Right, yeah, to stand out.
+p9rc0vn3iXc-00618-00111918-00112129 Just it's more of a mind game, to remind me that I,
+p9rc0vn3iXc-00619-00112129-00112324 this isn't where I belong. Right.
+p9rc0vn3iXc-00620-00112324-00112631 I can, I've done that, I've paid my dues or whatever,
+p9rc0vn3iXc-00621-00112631-00112764 but that's just not what I belong, okay?
+p9rc0vn3iXc-00622-00112764-00112927 Right, and you also wanna make sure
+p9rc0vn3iXc-00623-00112927-00113134 that you do stand out from the guys.
+p9rc0vn3iXc-00624-00113134-00113303 So if you're not in the truck anymore,
+p9rc0vn3iXc-00625-00113303-00113620 you don't wanna, here's a good point
+p9rc0vn3iXc-00626-00113620-00113753 that I just came up with.
+p9rc0vn3iXc-00627-00113753-00114107 If you want, if someone walked into your office,
+p9rc0vn3iXc-00628-00114107-00114327 and all your crew was there, and you were there,
+p9rc0vn3iXc-00629-00114327-00114733 would they be able to tell who is the business owner?
+p9rc0vn3iXc-00630-00114733-00114817 Right.
+p9rc0vn3iXc-00631-00114817-00115091 Or would it just look like everybody's the plumber?
+p9rc0vn3iXc-00632-00115091-00115311 If the answer to that is everybody looks like a plumber,
+p9rc0vn3iXc-00633-00115311-00115482 you're wearing the wrong thing.
+p9rc0vn3iXc-00634-00115482-00115608 Right. You're in the wrong thing.
+p9rc0vn3iXc-00635-00115608-00115721 Right. You have to stand out,
+p9rc0vn3iXc-00636-00115721-00115805 you have to be separate.
+p9rc0vn3iXc-00637-00115805-00116046 Listen, you're not buddies with your plumbers.
+p9rc0vn3iXc-00638-00116046-00116281 You're not buddies, you have to separate it.
+p9rc0vn3iXc-00639-00116281-00116514 And one way to separate is by what you wear.
+p9rc0vn3iXc-00640-00116514-00116727 Right, you know, clothes make the man.
+p9rc0vn3iXc-00641-00116727-00116918 I mean, it is that simple.
+p9rc0vn3iXc-00642-00116918-00117207 People judge a book by its cover, period.
+p9rc0vn3iXc-00643-00117207-00117481 I know we're working on a society where we're even,
+p9rc0vn3iXc-00644-00117481-00117713 you know, we're even getting rid of Dr. Seuss, you know-
+p9rc0vn3iXc-00645-00117713-00117796 I haven't even seen what that's all about.
+p9rc0vn3iXc-00646-00117796-00117994 And I mean, all this cancel culture bull-
+p9rc0vn3iXc-00647-00117994-00118126 Crap. Yeah.
+p9rc0vn3iXc-00648-00118126-00118309 You know, and we're trying to make it all safe.
+p9rc0vn3iXc-00649-00118309-00118581 The reality, it will never be that way.
+p9rc0vn3iXc-00650-00118581-00118664 Mm-mm.
+p9rc0vn3iXc-00651-00118664-00118833 All right, people judge a book by its cover,
+p9rc0vn3iXc-00652-00118833-00119025 and they have their opinions.
+p9rc0vn3iXc-00653-00119025-00119408 So if you go into, well, I know that I, you know,
+p9rc0vn3iXc-00654-00119408-00119753 I remember when I got one of my first plumbing jobs,
+p9rc0vn3iXc-00655-00119753-00120079 when I was a plumber, I actually wore their uniform,
+p9rc0vn3iXc-00656-00120079-00120249 and I wore it clean.
+p9rc0vn3iXc-00657-00120249-00120395 And pressed. And pressed it.
+p9rc0vn3iXc-00658-00120395-00120718 I mean, and pressed it means I ironed it every morning.
+p9rc0vn3iXc-00659-00120718-00120801 Okay?
+p9rc0vn3iXc-00660-00120801-00120992 And so I looked clean and sharp.
+p9rc0vn3iXc-00661-00121140-00121438 I wasn't even a licensed plumber yet, I was an apprentice,
+p9rc0vn3iXc-00662-00121438-00121571 and I got my own truck.
+p9rc0vn3iXc-00663-00121730-00121828 Before other guys-
+p9rc0vn3iXc-00664-00121828-00121911 'Cause you were all clean and shiny.
+p9rc0vn3iXc-00665-00121911-00122016 Because yes, I could do the work.
+p9rc0vn3iXc-00666-00122016-00122296 Maybe not, I wasn't as good as other guys,
+p9rc0vn3iXc-00667-00122296-00122438 but I looked the part.
+p9rc0vn3iXc-00668-00122438-00122680 I mean, it's just, there's something that happens that way.
+p9rc0vn3iXc-00669-00122680-00122763 All right?
+p9rc0vn3iXc-00670-00122763-00123092 When you go to the bank and you're asking for money,
+p9rc0vn3iXc-00671-00123092-00123379 don't go there looking like death warmed over.
+p9rc0vn3iXc-00672-00123379-00123601 Or yeah, with dirty, crappy boots on.
+p9rc0vn3iXc-00673-00123601-00123698 Right.
+p9rc0vn3iXc-00674-00123698-00123781 Who knows what's on it?
+p9rc0vn3iXc-00675-00123781-00123864 Right, well, they can take me as I am.
+p9rc0vn3iXc-00676-00123864-00123947 Well, they will take you as you are,
+p9rc0vn3iXc-00677-00123947-00124081 and say, I ain't gonna give you any money.
+p9rc0vn3iXc-00678-00124081-00124164 Denied.
+p9rc0vn3iXc-00679-00124164-00124492 You're right, you know, it's just how it is, okay?
+p9rc0vn3iXc-00680-00124492-00124663 Yeah, and also, plumbing business owners,
+p9rc0vn3iXc-00681-00124663-00124908 if you are involved in like any BNI group,
+p9rc0vn3iXc-00682-00124908-00125097 or local community, Chamber of Commerce,
+p9rc0vn3iXc-00683-00125097-00125328 or anything like that, and they have their meetings,
+p9rc0vn3iXc-00684-00125328-00125520 don't wear your plumbing uniform.
+p9rc0vn3iXc-00685-00125520-00125754 If you are in the truck all the time,
+p9rc0vn3iXc-00686-00125754-00125972 and you have to go to the meeting, you know,
+p9rc0vn3iXc-00687-00125972-00126320 in between jobs, have a clean set of clothes
+p9rc0vn3iXc-00688-00126320-00126563 that aren't plumbing uniforms in your truck,
+p9rc0vn3iXc-00689-00126563-00126688 so you can change into that, and walk in.
+p9rc0vn3iXc-00690-00126688-00126975 Right, that, you know, I remember when I was still
+p9rc0vn3iXc-00691-00126975-00127150 in the truck, and when went to those kinds of things,
+p9rc0vn3iXc-00692-00127150-00127313 I was trying to learn, you know, this kinda stuff,
+p9rc0vn3iXc-00693-00127313-00127659 where I met my mentor that actually taught me business.
+p9rc0vn3iXc-00694-00127659-00127861 I was still in the truck, okay?
+p9rc0vn3iXc-00695-00127861-00127965 Before I became a million dollar, you know, the business-
+p9rc0vn3iXc-00696-00127965-00128180 A hundred years ago, the plumber.
+p9rc0vn3iXc-00697-00128180-00128433 Or whatever, by being with these other CEOs,
+p9rc0vn3iXc-00698-00128433-00128681 or the business leaders, that's what compelled me,
+p9rc0vn3iXc-00699-00128681-00128847 you know, move up. Right, hook you up.
+p9rc0vn3iXc-00700-00128847-00129113 But I'd go to those meetings, and I'd wear khakis,
+p9rc0vn3iXc-00701-00129113-00129376 which you won't catch me in khakis today, that's a little.
+p9rc0vn3iXc-00702-00129376-00129595 But that was different than wearing the plumbing uni,
+p9rc0vn3iXc-00703-00129595-00130002 but khakis, and had a white shirt that had Attaboy all,
+p9rc0vn3iXc-00704-00130002-00130126 I mean- Yes, embroidered.
+p9rc0vn3iXc-00705-00130126-00130209 So it looked more executive.
+p9rc0vn3iXc-00706-00130209-00130319 Right. Right.
+p9rc0vn3iXc-00707-00130319-00130579 No glue and primer stains, exactly right.
+p9rc0vn3iXc-00708-00130579-00130732 Yeah, exactly right.
+p9rc0vn3iXc-00709-00130732-00130892 Clean and fresh, right, Joseph.
+p9rc0vn3iXc-00710-00130892-00130975 There we go, all right?
+p9rc0vn3iXc-00711-00130975-00131087 'Cause clothes do make the man.
+p9rc0vn3iXc-00712-00131087-00131313 Clothes make the man, all right?
+p9rc0vn3iXc-00713-00131313-00131481 So go, and it's kinda fun.
+p9rc0vn3iXc-00714-00131481-00131704 Even we have some of our clients here, even a couple on
+p9rc0vn3iXc-00715-00131704-00131912 are talking about, even the car they're driving,
+p9rc0vn3iXc-00716-00131912-00132186 the truck, they've gone out and gotten their CEO-
+p9rc0vn3iXc-00717-00132186-00132269 CEO truck.
+p9rc0vn3iXc-00718-00132269-00132506 The CEO truck or vehicle, okay?
+p9rc0vn3iXc-00719-00132506-00132668 There's just something about that, guys.
+p9rc0vn3iXc-00720-00132668-00132797 It's a mindset.
+p9rc0vn3iXc-00721-00132797-00133053 And so, when you start seeing yourself that way,
+p9rc0vn3iXc-00722-00133053-00133397 and in doing and speaking out that way, it all of a sudden,
+p9rc0vn3iXc-00723-00133397-00133581 it's lifting your business up.
+p9rc0vn3iXc-00724-00133581-00133664 Right, it elevates you.
+p9rc0vn3iXc-00725-00133664-00133747 You start headed that way.
+p9rc0vn3iXc-00726-00133747-00134012 Let's see, Jeff says, when I dress for success,
+p9rc0vn3iXc-00727-00134012-00134240 people tell me all the time, you don't look like a plumber.
+p9rc0vn3iXc-00728-00134240-00134369 I tell them, I'm a small business owner.
+p9rc0vn3iXc-00729-00134369-00134452 Good for you, Jeff.
+p9rc0vn3iXc-00730-00134452-00134637 Right on, Jeff, right on, right on, you know?
+p9rc0vn3iXc-00731-00134637-00134823 From now on, Jeff, you're not gonna say,
+p9rc0vn3iXc-00732-00134823-00134906 small business owner,
+p9rc0vn3iXc-00733-00134906-00134990 you're gonna say, I'm a CEO. Right.
+p9rc0vn3iXc-00734-00134990-00135155 Because that is what you are.
+p9rc0vn3iXc-00735-00135155-00135520 That is what you are, that is your responsibility,
+p9rc0vn3iXc-00736-00135520-00135603 all right?
+p9rc0vn3iXc-00737-00135603-00135736 All right, and that's what you're doing.
+p9rc0vn3iXc-00738-00135736-00136016 So, there we go, all right? Yes, love that.
+p9rc0vn3iXc-00739-00136016-00136138 Yeah, it's always good when someone says,
+p9rc0vn3iXc-00740-00136138-00136258 well, you don't look like a plumber.
+p9rc0vn3iXc-00741-00136258-00136341 Yeah.
+p9rc0vn3iXc-00742-00136341-00136424 That's usually a good sign.
+p9rc0vn3iXc-00743-00136424-00136655 Right, hey, I'm just having a rough, I mean, you can be.
+p9rc0vn3iXc-00744-00136655-00136738 You can get dirty, but-
+p9rc0vn3iXc-00745-00136738-00136890 Right, right, whatever, but yeah.
+p9rc0vn3iXc-00746-00136890-00137039 Clean up before your next job,
+p9rc0vn3iXc-00747-00137039-00137140 or certainly before the next day.
+p9rc0vn3iXc-00748-00137140-00137326 Well, I'm the CEO, all right?
+p9rc0vn3iXc-00749-00137326-00137530 Yeah, I've done plumbing, I'm proud to be a plumber.
+p9rc0vn3iXc-00750-00137530-00137638 Okay?
+p9rc0vn3iXc-00751-00137638-00137775 All right?
+p9rc0vn3iXc-00752-00137775-00138035 'Cause plumbers do protect the health of the nation.
+p9rc0vn3iXc-00753-00138035-00138145 Yeah, they do. It's true.
+p9rc0vn3iXc-00754-00138145-00138265 Yeah, we do.
+p9rc0vn3iXc-00755-00138265-00138471 All right, hey, for joining us and for listening,
+p9rc0vn3iXc-00756-00138471-00138651 I have something for you, all right?
+p9rc0vn3iXc-00757-00138651-00138934 A Million Dollar Plumber blueprint, all right?
+p9rc0vn3iXc-00758-00138934-00139157 All the specs you need to know
+p9rc0vn3iXc-00759-00139157-00139327 to have a successful, self-sustaining,
+p9rc0vn3iXc-00760-00139327-00139521 and profitable plumbing business.
+p9rc0vn3iXc-00761-00139521-00139854 I laid it out all in there very simply for you, okay?
+p9rc0vn3iXc-00762-00139854-00139977 And it's free.
+p9rc0vn3iXc-00763-00139977-00140240 All right, simply type free in the comments,
+p9rc0vn3iXc-00764-00140240-00140681 or go to themilliondollarplumber.com/free,
+p9rc0vn3iXc-00765-00140681-00141053 and grab your free copy today.
+p9rc0vn3iXc-00766-00141053-00141182 Okay. All right,
+p9rc0vn3iXc-00767-00141182-00141377 so there we have it, all right, number one.
+p9rc0vn3iXc-00768-00141377-00141492 MDP mindsets.
+p9rc0vn3iXc-00769-00141492-00141575 All right.
+p9rc0vn3iXc-00770-00141575-00141729 How to be a plumbing business CEO, right?
+p9rc0vn3iXc-00771-00141729-00141854 Speak it, all right?
+p9rc0vn3iXc-00772-00141854-00141964 Speak it, say it.
+p9rc0vn3iXc-00773-00141964-00142132 This is what I am- Out loud.
+p9rc0vn3iXc-00774-00142132-00142215 You know? All the time.
+p9rc0vn3iXc-00775-00142215-00142342 Put it on your card, you know?
+p9rc0vn3iXc-00776-00142342-00142453 That kind of, you know, your business card.
+p9rc0vn3iXc-00777-00142453-00142606 Ooh, that's another good one, yeah, get a business.
+p9rc0vn3iXc-00778-00142606-00142788 Yeah, if you've got business cards, which you should,
+p9rc0vn3iXc-00779-00142788-00143057 put on there, CEO. Okay.
+p9rc0vn3iXc-00780-00143057-00143175 Instead of plumber or whatever.
+p9rc0vn3iXc-00781-00143175-00143452 All right, speak it out, okay?
+p9rc0vn3iXc-00782-00143452-00143683 Clothes make the man, dress like it.
+p9rc0vn3iXc-00783-00143683-00143839 Dress for success, you ever heard that?
+p9rc0vn3iXc-00784-00143839-00143975 Dress for success.
+p9rc0vn3iXc-00785-00143975-00144058 All right?
+p9rc0vn3iXc-00786-00144058-00144286 Hey, some of these things sound, you know, kinda cliche.
+p9rc0vn3iXc-00787-00144286-00144548 There's the reason why save some sayings, right?
+p9rc0vn3iXc-00788-00144548-00144672 'Cause it is, you know? 'Cause they're true.
+p9rc0vn3iXc-00789-00144672-00144763 Right.
+p9rc0vn3iXc-00790-00144763-00144846 They're true statements.
+p9rc0vn3iXc-00791-00144846-00145050 Right, a bird in the hand's worth two in the bush, okay?
+p9rc0vn3iXc-00792-00145050-00145137 All right.
+p9rc0vn3iXc-00793-00145137-00145252 Six in one, half dozen of the other.
+p9rc0vn3iXc-00794-00145252-00145335 That's right.
+p9rc0vn3iXc-00795-00145504-00145786 Wait, six dozen of one, say, half of that divided by-
+p9rc0vn3iXc-00796-00145786-00145995 What if it's a baker dozen? Oh.
+p9rc0vn3iXc-00797-00145995-00146145 Oh, okay. Or a baker's dozen.
+p9rc0vn3iXc-00798-00146145-00146228 Yeah, it really doesn't count.
+p9rc0vn3iXc-00799-00146228-00146319 So, how does the math work on that?
+p9rc0vn3iXc-00800-00146319-00146402 I don't know.
+p9rc0vn3iXc-00801-00146402-00146527 All right, carry the one, cross the, Okay.
+p9rc0vn3iXc-00802-00146527-00146809 All right, all right, and finally, number three,
+p9rc0vn3iXc-00803-00146809-00146987 the piece the resistance.
+p9rc0vn3iXc-00804-00146987-00147163 Like you never heard this one before.
+p9rc0vn3iXc-00805-00147273-00147593 It ain't about plumbing, all right?
+p9rc0vn3iXc-00806-00147692-00147776 Well, what does that have to do
+p9rc0vn3iXc-00807-00147776-00148022 with the Million Dollar Plumber mindset?
+p9rc0vn3iXc-00808-00148022-00148122 And being a CEO?
+p9rc0vn3iXc-00809-00148273-00148464 It isn't, okay?
+p9rc0vn3iXc-00810-00148464-00148743 It ain't, like, where I come from, we say ain't,
+p9rc0vn3iXc-00811-00148743-00148838 we don't say isn't.
+p9rc0vn3iXc-00812-00148838-00148944 I know I always tried to,
+p9rc0vn3iXc-00813-00148944-00149137 when we first started speaking that,
+p9rc0vn3iXc-00814-00149137-00149344 and we write that out, remember, we'd have our editor,
+p9rc0vn3iXc-00815-00149344-00149427 you know, that kind of stuff say,
+p9rc0vn3iXc-00816-00149427-00149676 well, it isn't, no, where I'm from, we say ain't, okay?
+p9rc0vn3iXc-00817-00149676-00149784 It ain't about plumbing.
+p9rc0vn3iXc-00818-00149784-00149932 It ain't about plumbing, okay?
+p9rc0vn3iXc-00819-00149932-00150136 All right, well, what do you mean in this?
+p9rc0vn3iXc-00820-00150136-00150437 Okay, your focus is not
+p9rc0vn3iXc-00821-00150437-00150615 with the plumbing business CEO, okay?
+p9rc0vn3iXc-00822-00150615-00150796 It's not about the plumbing.
+p9rc0vn3iXc-00823-00150796-00150946 I mean, I still hear it, you know,
+p9rc0vn3iXc-00824-00150946-00151190 the guys getting wrapped up in all the code and all that,
+p9rc0vn3iXc-00825-00151190-00151428 you know, all that, but that's not, you're driven, you know,
+p9rc0vn3iXc-00826-00151428-00151632 that's not what you're about, okay?
+p9rc0vn3iXc-00827-00151632-00151844 If you wanna still be all up on the code,
+p9rc0vn3iXc-00828-00151844-00152147 and all in the, you know, the latest equipment,
+p9rc0vn3iXc-00829-00152147-00152486 and techniques, and all the stuff, and the gear, and that,
+p9rc0vn3iXc-00830-00152486-00152702 all that kinda crap, okay?
+p9rc0vn3iXc-00831-00152702-00152932 Then go be a plumber for someone, okay?
+p9rc0vn3iXc-00832-00152932-00153399 As a plumbing business CEO, your job is the business, okay?
+p9rc0vn3iXc-00833-00153399-00153682 It ain't about, it has nothing to do with plumbing.
+p9rc0vn3iXc-00834-00153682-00153874 Okay, well, I don't know about business,
+p9rc0vn3iXc-00835-00153874-00153984 or I'm not that kinda guy.
+p9rc0vn3iXc-00836-00153984-00154067 Well-
+p9rc0vn3iXc-00837-00154067-00154150 Learn.
+p9rc0vn3iXc-00838-00154150-00154301 Right, neither am I, okay?
+p9rc0vn3iXc-00839-00154301-00154483 It doesn't mean you've gotta go,
+p9rc0vn3iXc-00840-00154483-00154630 I ain't sitting behind a desk.
+p9rc0vn3iXc-00841-00154630-00154865 Well, I didn't wanna sit behind a desk either.
+p9rc0vn3iXc-00842-00154865-00154996 That's why I got in trouble all the time.
+p9rc0vn3iXc-00843-00154996-00155079 That's true.
+p9rc0vn3iXc-00844-00155079-00155162 Where's Richard?
+p9rc0vn3iXc-00845-00155162-00155245 Gone.
+p9rc0vn3iXc-00846-00155245-00155329 Lovely Laura be looking for me.
+p9rc0vn3iXc-00847-00155329-00155683 I'd be at the track, the race track, or at the airport.
+p9rc0vn3iXc-00848-00155807-00155925 You know, going golfing.
+p9rc0vn3iXc-00849-00155925-00156067 Yeah, and I'm, phones are ringing off the hook,
+p9rc0vn3iXc-00850-00156067-00156256 and my hair's on fire.
+p9rc0vn3iXc-00851-00156256-00156464 All right. Here's Ricky.
+p9rc0vn3iXc-00852-00156464-00156547 Okay.
+p9rc0vn3iXc-00853-00156547-00156678 You are a CEO! Yeah, you are, brother!
+p9rc0vn3iXc-00854-00156678-00156814 That's right, Ricky, good for you.
+p9rc0vn3iXc-00855-00156814-00157053 And so, I'm not saying you gotta sit behind a desk,
+p9rc0vn3iXc-00856-00157053-00157273 and have a bit, you know, and be that, that's not.
+p9rc0vn3iXc-00857-00157273-00157404 And look at numbers all day.
+p9rc0vn3iXc-00858-00157404-00157521 Right, that's not what it's about, okay?
+p9rc0vn3iXc-00859-00157521-00157878 You can do it in your style, all right?
+p9rc0vn3iXc-00860-00157878-00158097 But it's about the business, okay?
+p9rc0vn3iXc-00861-00158097-00158217 About the business.
+p9rc0vn3iXc-00862-00158217-00158499 You're focusing on the things that will move the business,
+p9rc0vn3iXc-00863-00158499-00158764 this asset that you're working on for, you know,
+p9rc0vn3iXc-00864-00158764-00159205 that's the whole, you know, the whole MDP mindset, okay?
+p9rc0vn3iXc-00865-00159205-00159396 The Million Dollar Plumber mindset is all
+p9rc0vn3iXc-00866-00159396-00159553 about the freedom lifestyle, okay?
+p9rc0vn3iXc-00867-00159553-00159775 So how do we create a business?
+p9rc0vn3iXc-00868-00159775-00160000 How do we create this asset that we have
+p9rc0vn3iXc-00869-00160000-00160291 that allows us to live the lives that we wanna live, okay?
+p9rc0vn3iXc-00870-00160291-00160515 And live our lives on our terms, all right.
+p9rc0vn3iXc-00871-00160515-00160837 And then create real wealth, wealth being financial wealth,
+p9rc0vn3iXc-00872-00160837-00161053 and time wealth, okay.
+p9rc0vn3iXc-00873-00161053-00161169 Learning how to use a printer,
+p9rc0vn3iXc-00874-00161169-00161320 and send an email is the hardest thing
+p9rc0vn3iXc-00875-00161320-00161499 about being a plumber business owner.
+p9rc0vn3iXc-00876-00161499-00161886 Yeah, brother, DW, thank you for saying it.
+p9rc0vn3iXc-00877-00161886-00162187 I know, I know.
+p9rc0vn3iXc-00878-00162187-00162379 And that's all right, look, that's, guys,
+p9rc0vn3iXc-00879-00162379-00162500 that's why I say it.
+p9rc0vn3iXc-00880-00162500-00162663 And I mean this, if you watch this,
+p9rc0vn3iXc-00881-00162663-00162769 you know, who is this guy?
+p9rc0vn3iXc-00882-00162769-00162855 He thinks he's all that.
+p9rc0vn3iXc-00883-00162855-00162938 No, I don't think I'm, first of all,
+p9rc0vn3iXc-00884-00162938-00163239 I don't think I'm all that, I've just made every mistake.
+p9rc0vn3iXc-00885-00163354-00163596 And I, still to this day, have trouble.
+p9rc0vn3iXc-00886-00163596-00163811 I can't get the printer to work.
+p9rc0vn3iXc-00887-00163811-00164107 Even with the earlier today with the computer,
+p9rc0vn3iXc-00888-00164107-00164282 I was like, sweetheart!
+p9rc0vn3iXc-00889-00164282-00164424 Sweetheart! You know?
+p9rc0vn3iXc-00890-00164424-00164607 Exit strategy, exactly right, Darrell.
+p9rc0vn3iXc-00891-00164607-00164690 Okay, yes, exactly.
+p9rc0vn3iXc-00892-00164690-00164896 Every business should have an exit strategy,
+p9rc0vn3iXc-00893-00164896-00165012 it would be a number one sell.
+p9rc0vn3iXc-00894-00165012-00165275 Exactly, well, business is set up to sell,
+p9rc0vn3iXc-00895-00165275-00165469 to exit, even if you don't want to.
+p9rc0vn3iXc-00896-00165469-00165719 But you know, I just wanted to say that again,
+p9rc0vn3iXc-00897-00165719-00166001 if I can do it, I get it, guys, I get it.
+p9rc0vn3iXc-00898-00166001-00166330 I get, you know, the idea of not just being, you know,
+p9rc0vn3iXc-00899-00166330-00166663 technically minded, you know, and that kinda thing.
+p9rc0vn3iXc-00900-00166663-00166746 Baby, you can't be perfect at everything.
+p9rc0vn3iXc-00901-00166746-00166902 Okay, but you can learn it.
+p9rc0vn3iXc-00902-00166902-00167006 You gotta make it part, you know,
+p9rc0vn3iXc-00903-00167006-00167244 and you don't have to be an expert at everything, you don't.
+p9rc0vn3iXc-00904-00167244-00167395 It's not about being an expert.
+p9rc0vn3iXc-00905-00167395-00167478 Did you hear what I said?
+p9rc0vn3iXc-00906-00167478-00167587 What?
+p9rc0vn3iXc-00907-00167587-00167760 I said, you can't be perfect at everything.
+p9rc0vn3iXc-00908-00167760-00167958 Yeah, I'm not.
+p9rc0vn3iXc-00909-00167958-00168050 With technology, you said,
+p9rc0vn3iXc-00910-00168050-00168154 you have trouble with technology,
+p9rc0vn3iXc-00911-00168154-00168238 so I said, you can't be perfect at everything.
+p9rc0vn3iXc-00912-00168238-00168321 What am I perfect at?
+p9rc0vn3iXc-00913-00168321-00168404 Everything else.
+p9rc0vn3iXc-00914-00168404-00168685 Yeah, you hear that?
+p9rc0vn3iXc-00915-00168685-00168902 And he is a words of affirmation guy.
+p9rc0vn3iXc-00916-00168902-00169027 I still got it.
+p9rc0vn3iXc-00917-00169027-00169265 Okay.
+p9rc0vn3iXc-00918-00169265-00169435 Sorry, what were you saying?
+p9rc0vn3iXc-00919-00169435-00169630 You were saying something really smart and I cut you off.
+p9rc0vn3iXc-00920-00169630-00169780 Hey baby, that's what I'm saying.
+p9rc0vn3iXc-00921-00170040-00170201 All right, no, you just,
+p9rc0vn3iXc-00922-00170201-00170501 you don't have to be an expert at everything, okay?
+p9rc0vn3iXc-00923-00170501-00170688 But you just need to guide and direct it,
+p9rc0vn3iXc-00924-00170688-00170953 and that is now your focus.
+p9rc0vn3iXc-00925-00170953-00171268 And I get it, you know, the plumbing, you know,
+p9rc0vn3iXc-00926-00171268-00171466 when I was in the truck, and go walk on a job site,
+p9rc0vn3iXc-00927-00171466-00171619 I'd own it.
+p9rc0vn3iXc-00928-00171619-00171853 You know, I'm, you know, I could figure things out,
+p9rc0vn3iXc-00929-00171853-00172073 it was confidence, you know?
+p9rc0vn3iXc-00930-00172073-00172222 Get out of my way, Sparkies.
+p9rc0vn3iXc-00931-00172222-00172384 Sparkies, not our Sparkies.
+p9rc0vn3iXc-00932-00172384-00172537 The HVAC got, you know, message talk with me,
+p9rc0vn3iXc-00933-00172537-00172865 I could talk with all the guys, I could do, you know,
+p9rc0vn3iXc-00934-00172865-00173031 you know, whatever, okay?
+p9rc0vn3iXc-00935-00173031-00173233 And own the job site.
+p9rc0vn3iXc-00936-00173233-00173434 And so, there's a comfort level there, all right?
+p9rc0vn3iXc-00937-00173434-00173539 Sure.
+p9rc0vn3iXc-00938-00173539-00173693 And so, when we got into the business, it's like-
+p9rc0vn3iXc-00939-00173693-00173777 That's not comfortable.
+p9rc0vn3iXc-00940-00173777-00174026 You know, you mentioned, I'm a words of affirmation guy.
+p9rc0vn3iXc-00941-00174026-00174109 You are.
+p9rc0vn3iXc-00942-00174109-00174272 Well, I don't wanna put myself in a situation
+p9rc0vn3iXc-00943-00174272-00174461 where I don't know what the heck's going on.
+p9rc0vn3iXc-00944-00174461-00174773 But you know, it is your responsibility
+p9rc0vn3iXc-00945-00174773-00175093 as the plumbing business CEO, all right?
+p9rc0vn3iXc-00946-00175093-00175229 To know what's going on. Yes.
+p9rc0vn3iXc-00947-00175229-00175409 And it's not about the plumbing, you know,
+p9rc0vn3iXc-00948-00175409-00175570 these guys, you need to focus on the job,
+p9rc0vn3iXc-00949-00175570-00175717 what Joe's doing on the job.
+p9rc0vn3iXc-00950-00175943-00176102 The only reason you'll be checking up on Joe
+p9rc0vn3iXc-00951-00176102-00176213 is why hasn't he moved yet?
+p9rc0vn3iXc-00952-00176213-00176306 He's doing a water heater.
+p9rc0vn3iXc-00953-00176306-00176413 He's been there three hours.
+p9rc0vn3iXc-00954-00176413-00176532 Why hasn't he moved?
+p9rc0vn3iXc-00955-00176532-00176633 Time is money.
+p9rc0vn3iXc-00956-00176633-00176821 Not, well, lemme get out there and help him.
+p9rc0vn3iXc-00957-00176821-00177000 Lemme maybe give him another set of hands,
+p9rc0vn3iXc-00958-00177000-00177222 and I'll show him how to do it right, you know?
+p9rc0vn3iXc-00959-00177222-00177502 No, it's not, if you're thinking that way,
+p9rc0vn3iXc-00960-00177502-00177597 you're gonna fail.
+p9rc0vn3iXc-00961-00177597-00177811 Right, can't scale that, can't scale it.
+p9rc0vn3iXc-00962-00177811-00177903 Can't scale that. Mm-mm, not possible.
+p9rc0vn3iXc-00963-00177903-00178151 Certainly not with the freedom lifestyle.
+p9rc0vn3iXc-00964-00178151-00178255 Right.
+p9rc0vn3iXc-00965-00178255-00178338 Okay?
+p9rc0vn3iXc-00966-00178338-00178447 Okay!
+p9rc0vn3iXc-00967-00178447-00178531 So, there we go.
+p9rc0vn3iXc-00968-00178531-00178661 All right, so, the Million Dollar Plumber mindset.
+p9rc0vn3iXc-00969-00178661-00178937 We've got some other series coming up on how to get to that.
+p9rc0vn3iXc-00970-00178937-00179142 Right, to the million dollar-
+p9rc0vn3iXc-00971-00179142-00179225 Freedom lifestyle, right?
+p9rc0vn3iXc-00972-00179225-00179344 Right, to the freedom lifestyle.
+p9rc0vn3iXc-00973-00179344-00179533 Hey, and if you're serious about,
+p9rc0vn3iXc-00974-00179533-00179738 if that's what you're looking for, the freedom lifestyle,
+p9rc0vn3iXc-00975-00179738-00179838 okay, if that's something you want, all right?
+p9rc0vn3iXc-00976-00179838-00179926 And why wouldn't you?
+p9rc0vn3iXc-00977-00179926-00180220 And you're wanting, you're wanting to create, right,
+p9rc0vn3iXc-00978-00180220-00180380 with your plumbing business,
+p9rc0vn3iXc-00979-00180380-00180643 and looking to create real wealth, okay?
+p9rc0vn3iXc-00980-00180643-00180861 Create financial security, all right?
+p9rc0vn3iXc-00981-00180861-00181146 And creating the time, the time that you can do
+p9rc0vn3iXc-00982-00181146-00181273 what you wanna do, okay?
+p9rc0vn3iXc-00983-00181273-00181401 Time is wealth as well.
+p9rc0vn3iXc-00984-00181401-00181623 Oh, and it, right, even more so today.
+p9rc0vn3iXc-00985-00181623-00181760 Yes, absolutely.
+p9rc0vn3iXc-00986-00181760-00181878 Then you need to check out
+p9rc0vn3iXc-00987-00181878-00182268 my Million Dollar Plumber business success academy,
+p9rc0vn3iXc-00988-00182268-00182351 all right?
+p9rc0vn3iXc-00989-00182351-00182748 Everything in there that you need to know, all right?
+p9rc0vn3iXc-00990-00182748-00182855 To run through everything- Everything.
+p9rc0vn3iXc-00991-00182855-00183156 Then to have a successful plumbing business
+p9rc0vn3iXc-00992-00183156-00183355 that will create that freedom lifestyle for you.
+p9rc0vn3iXc-00993-00183355-00183484 Look, you have two choices.
+p9rc0vn3iXc-00994-00183484-00183567 What?
+p9rc0vn3iXc-00995-00183567-00183667 You can either continue to do
+p9rc0vn3iXc-00996-00183667-00183800 the stupid stuff you're doing.
+p9rc0vn3iXc-00997-00183925-00184059 I'm just gonna lay it out there for you, okay?
+p9rc0vn3iXc-00998-00184059-00184175 Lovingly said.
+p9rc0vn3iXc-00999-00184175-00184451 Continue to do, what do they say about insanity, you know?
+p9rc0vn3iXc-01000-00184451-00184638 Doing the same thing over and over,
+p9rc0vn3iXc-01001-00184638-00184776 and expecting different results.
+p9rc0vn3iXc-01002-00184776-00185102 Right, you continue to do that, okay, all right?
+p9rc0vn3iXc-01003-00185102-00185518 Or you can learn what you need to learn, all right?
+p9rc0vn3iXc-01004-00185518-00185842 To create the life and the business
+p9rc0vn3iXc-01005-00185842-00185975 that you truly desire, okay?
+p9rc0vn3iXc-01006-00185975-00186169 It's all there in the success academy.
+p9rc0vn3iXc-01007-00186169-00186252 Type success.
+p9rc0vn3iXc-01008-00186252-00186355 And type success. In the comments.
+p9rc0vn3iXc-01009-00186355-00186438 In the comments.
+p9rc0vn3iXc-01010-00186438-00186562 Or go to-
+p9rc0vn3iXc-01011-00186562-00186817 Themilliondollarplumber.com/success
+p9rc0vn3iXc-01012-00186817-00187149 for your free 14-day trial. It's true.
+p9rc0vn3iXc-01013-00187149-00187232 All right?
+p9rc0vn3iXc-01014-00187232-00187316 So check it out. It's true!
+p9rc0vn3iXc-01015-00187316-00187477 All right, there we go, all right.
+p9rc0vn3iXc-01016-00187477-00187709 And I wanna remind you also, before I let you go,
+p9rc0vn3iXc-01017-00187709-00187866 that you were purposefully and wonderfully created,
+p9rc0vn3iXc-01018-00187866-00188035 and you were created to do great things.
+p9rc0vn3iXc-01019-00188035-00188147 So plumb like a champion.
+p9rc0vn3iXc-01020-00188147-00188279 [Announcer] Thank you for tuning in
+p9rc0vn3iXc-01021-00188279-00188538 to "Potty Talk Live" with Richard Behney,
+p9rc0vn3iXc-01022-00188538-00188698 the Million Dollar Plumber.
+p9rc0vn3iXc-01023-00188698-00188941 If you enjoyed this show, share it now.
+p9rc0vn3iXc-01024-00188941-00189044 Just do it.
+p9rc0vn3iXc-01025-00189044-00189173 [Announcer] To catch past episodes,
+p9rc0vn3iXc-01026-00189173-00189525 and get alerts for new episodes, turn notifications on now,
+p9rc0vn3iXc-01027-00189525-00189762 and join us again for more tips, techniques,
+p9rc0vn3iXc-01028-00189762-00190092 and strategies to help you grow your plumbing business.
+pqBNAWwX2cU-00000-00000003-00000163 If you knew that there was a stock
+pqBNAWwX2cU-00001-00000163-00000553 that hedge funds were going to buy in the coming months, would you buy it?
+pqBNAWwX2cU-00002-00000606-00000743 Of course you should buy it.
+pqBNAWwX2cU-00003-00000743-00000856 You buy it now
+pqBNAWwX2cU-00004-00000856-00001243 and then later when billions of dollars are going to flow into the stock.
+pqBNAWwX2cU-00005-00001256-00001486 You can take your profits, a huge profit.
+pqBNAWwX2cU-00006-00001506-00001846 This is how most of the time you make money on ten baggers,
+pqBNAWwX2cU-00007-00001846-00002146 five baggers, because you buy when nobody is buying.
+pqBNAWwX2cU-00008-00002146-00002466 And eventually, you know, all these big hedge funds. Because you look at their
+pqBNAWwX2cU-00009-00002466-00002773 portfolio, most of the time they move together.
+pqBNAWwX2cU-00010-00002803-00002880 They make their moves.
+pqBNAWwX2cU-00011-00002880-00003103 And the same stocks they for you look alike.
+pqBNAWwX2cU-00012-00003110-00003450 So when they are buying, then you are the one going to take profits,
+pqBNAWwX2cU-00013-00003490-00003583 huge profits.
+pqBNAWwX2cU-00014-00003583-00004000 So you always buy before the hedge funds and they are several.
+pqBNAWwX2cU-00015-00004026-00004356 That also just said to me that maybe hedge funds are going to buy this stock.
+pqBNAWwX2cU-00016-00004393-00004693 This one recently has been suggested to me and it seems that
+pqBNAWwX2cU-00017-00004693-00004990 maybe hedge funds can buy this stock in the coming month.
+pqBNAWwX2cU-00018-00005026-00005083 I'm sure
+pqBNAWwX2cU-00019-00005083-00005386 if you have been following this channel, if you have been following my portfolio,
+pqBNAWwX2cU-00020-00005386-00005526 my work, You will know
+pqBNAWwX2cU-00021-00005526-00005833 which stock I'm talking about and I will ask you to smash the like button
+pqBNAWwX2cU-00022-00005833-00006146 so that you can spread this video throughout the YouTube universe.
+pqBNAWwX2cU-00023-00006183-00006533 So why do I say that all hedge funds invest in the same way?
+pqBNAWwX2cU-00024-00006563-00006827 It's because of the way that they are structured.
+pqBNAWwX2cU-00025-00006827-00007197 So these hedge funds, they make money on people investing in them.
+pqBNAWwX2cU-00026-00007210-00007650 So they have their clients. The more assets under management they have,
+pqBNAWwX2cU-00027-00007730-00007870 The more money they are going to make.
+pqBNAWwX2cU-00028-00007870-00008140 You may say that this is the same thing that I'm doing on eToro.
+pqBNAWwX2cU-00029-00008140-00008333 Yes, in some way it is the same thing
+pqBNAWwX2cU-00030-00008333-00008687 because the more assets under management, the more money I make on eToro.
+pqBNAWwX2cU-00031-00008713-00008920 But those hedge funds unlike me on eToro
+pqBNAWwX2cU-00032-00008920-00009070 They don't have skin in the game.
+pqBNAWwX2cU-00033-00009070-00009387 So they can manage billions of dollars for other people.
+pqBNAWwX2cU-00034-00009387-00009700 And they can take huge risk on these other people's money,
+pqBNAWwX2cU-00035-00009730-00009877 but not on their own money.
+pqBNAWwX2cU-00036-00009877-00010077 So most hedge funds, they act like this.
+pqBNAWwX2cU-00037-00010077-00010297 Warren Buffett, he never started a hedge fund
+pqBNAWwX2cU-00038-00010310-00010547 when he was managing the Buffett Partnership.
+pqBNAWwX2cU-00039-00010570-00010737 He had skin in the game.
+pqBNAWwX2cU-00040-00010737-00010970 So let's go back to the way that these hedge funds work.
+pqBNAWwX2cU-00041-00010993-00011237 So these hedge funds, if they don't have skin in the game,
+pqBNAWwX2cU-00042-00011260-00011517 what they are targeting is to maximize profits.
+pqBNAWwX2cU-00043-00011530-00011800 It doesn't really matter how much risk they are taking.
+pqBNAWwX2cU-00044-00011800-00012010 And that's why it is so easy for them
+pqBNAWwX2cU-00045-00012010-00012390 to give unreasonable expectations to their clients.
+pqBNAWwX2cU-00046-00012410-00012803 So they are going to tell clients, I'm going to make 20% a year for the next
+pqBNAWwX2cU-00047-00012803-00012900 ten years.
+pqBNAWwX2cU-00048-00012910-00013153 They say to another client, I'll make 50% this year.
+pqBNAWwX2cU-00049-00013153-00013504 I will make 50% every year for the next five years.
+pqBNAWwX2cU-00050-00013530-00013930 So these clients, they don't watch my videos, they are not subscribed.
+pqBNAWwX2cU-00051-00013944-00014250 And if you are one of hedge funds so please subscribe,
+pqBNAWwX2cU-00052-00014274-00014477 Those clients, they want to make that much money
+pqBNAWwX2cU-00053-00014477-00014830 and maybe the hedge fund, they have some track records
+pqBNAWwX2cU-00054-00014830-00015054 maybe for one year or two years they had great returns
+pqBNAWwX2cU-00055-00015074-00015514 and now they sell this idea to their clients that they can keep making that much money
+pqBNAWwX2cU-00056-00015517-00015610 so the hedge funds,
+pqBNAWwX2cU-00057-00015610-00016004 then they are always going to go and target stocks that are doing very well
+pqBNAWwX2cU-00058-00016017-00016337 so that you can actually make the amount of money they have promised.
+pqBNAWwX2cU-00059-00016337-00016557 Because if you don't meet the expectations,
+pqBNAWwX2cU-00060-00016557-00016724 the clients are going to go away.
+pqBNAWwX2cU-00061-00016724-00017217 So they focus on returns mostly rather than making good investments.
+pqBNAWwX2cU-00062-00017230-00017550 And when they see a stock that has gained 100% in three month,
+pqBNAWwX2cU-00063-00017550-00017837 they will see, okay, maybe it is going to gain more.
+pqBNAWwX2cU-00064-00017857-00018097 It's going to gain 200% in the next three months.
+pqBNAWwX2cU-00065-00018097-00018370 I will invest in this and they see their friends investing in
+pqBNAWwX2cU-00066-00018370-00018534 the same stocks. Other hedge funds.
+pqBNAWwX2cU-00067-00018534-00018717 They are going to invest in the same stocks
+pqBNAWwX2cU-00068-00018717-00018987 because if they see other hedge funds investing, the clients also
+pqBNAWwX2cU-00069-00019010-00019217 they start asking questions.
+pqBNAWwX2cU-00070-00019217-00019427 Why is everybody investing in the stock?
+pqBNAWwX2cU-00071-00019440-00019747 You are not? why everybody has Tesla in their portfolio?
+pqBNAWwX2cU-00072-00019794-00019994 Why don't you have Tesla in your portfolio?
+pqBNAWwX2cU-00073-00019994-00020114 So those hedge funds,
+pqBNAWwX2cU-00074-00020114-00020454 because they are scared that their clients are going to take their money away.
+pqBNAWwX2cU-00075-00020464-00020647 They are going to invest in this company.
+pqBNAWwX2cU-00076-00020647-00020991 I'm not saying that I as a popular investor, an Elite Popular
+pqBNAWwX2cU-00077-00021001-00021321 Investor on eToro, I don't have such requests.
+pqBNAWwX2cU-00078-00021351-00021691 People ask me, why don't I invest in such and such company?
+pqBNAWwX2cU-00079-00021707-00021851 Because everybody is investing in it.
+pqBNAWwX2cU-00080-00021851-00022321 It will go up, but it doesn't meet my requirements as a value investor.
+pqBNAWwX2cU-00081-00022321-00022511 I don't invest really for the returns.
+pqBNAWwX2cU-00082-00022511-00022831 I invest because this is a good company and I expect over the long
+pqBNAWwX2cU-00083-00022831-00023121 term it's going to do better than the market.
+pqBNAWwX2cU-00084-00023137-00023544 So I do not really invest in a company because I expected to make 100%
+pqBNAWwX2cU-00085-00023544-00023777 in three months. This is not what I do.
+pqBNAWwX2cU-00086-00023834-00024151 I invest in companies and eventually, because value investing,
+pqBNAWwX2cU-00087-00024151-00024534 this has been something proven for nearly a century.
+pqBNAWwX2cU-00088-00024537-00024704 It has been proven that it works.
+pqBNAWwX2cU-00089-00024704-00024897 So I as a value investor.
+pqBNAWwX2cU-00090-00024897-00025061 I invest in companies like this.
+pqBNAWwX2cU-00091-00025061-00025251 So which this company that I'm talking about
+pqBNAWwX2cU-00092-00025251-00025534 that all hedge funds will probably buy in the coming month?
+pqBNAWwX2cU-00093-00025557-00025707 It is GEO Group.
+pqBNAWwX2cU-00094-00025707-00025917 And why do I see that they will buy this?
+pqBNAWwX2cU-00095-00025937-00026257 First, we need to understand how hedge funds buy stocks.
+pqBNAWwX2cU-00096-00026304-00026551 Why do they go and target specific stocks?
+pqBNAWwX2cU-00097-00026641-00026904 Those hedge funds, as I told you, they move together
+pqBNAWwX2cU-00098-00026918-00027194 and they are mostly influenced by the market.
+pqBNAWwX2cU-00099-00027221-00027627 So they are going to look at what the Federal Reserve System is doing,
+pqBNAWwX2cU-00100-00027627-00027764 how the market is doing,
+pqBNAWwX2cU-00101-00027764-00028021 because they know that they are all correlated with each other.
+pqBNAWwX2cU-00102-00028048-00028154 They are not stupid people.
+pqBNAWwX2cU-00103-00028154-00028291 those hedge fund managers,
+pqBNAWwX2cU-00104-00028291-00028678 they are intelligent. Maybe more intelligent than you and me, more intelligent
+pqBNAWwX2cU-00105-00028678-00028861 than the value investors, but it doesn't matter.
+pqBNAWwX2cU-00106-00028861-00028971 They don't make much money.
+pqBNAWwX2cU-00107-00028971-00029191 What matters as an investor is not how smart you are,
+pqBNAWwX2cU-00108-00029191-00029331 but how much money you are making.
+pqBNAWwX2cU-00109-00029331-00029551 So those hedge fund, they are going to write reports
+pqBNAWwX2cU-00110-00029571-00029924 which they will give to their clients and they will say, okay,
+pqBNAWwX2cU-00111-00029941-00030488 the market, you can do this next year, the market is going to go 5% up next year.
+pqBNAWwX2cU-00112-00030501-00030821 So I will invest in this sector because this sector is going 10% up
+pqBNAWwX2cU-00113-00030854-00031201 and then they are going to put their thesis  why they think this sector
+pqBNAWwX2cU-00114-00031201-00031311 is going to do well.
+pqBNAWwX2cU-00115-00031311-00031641 Of course, this thesis, other hedge funds are working on the same thing
+pqBNAWwX2cU-00116-00031641-00031771 because they think alike.
+pqBNAWwX2cU-00117-00031771-00031891 And when they are going to look
+pqBNAWwX2cU-00118-00031891-00032334 for investments, it depends on where we are in the market cycle, where we are
+pqBNAWwX2cU-00119-00032334-00032701 in the business cycle, in the long term, medium term business cycle,
+pqBNAWwX2cU-00120-00032711-00032878 if we are just
+pqBNAWwX2cU-00121-00032878-00033278 recovering from a recession, the business... new business cycle has just started.
+pqBNAWwX2cU-00122-00033314-00033505 So companies are profitable again.
+pqBNAWwX2cU-00123-00033505-00033755 Most companies are valued on earnings.
+pqBNAWwX2cU-00124-00033755-00034228 So earnings per share, PE ratio, this is what most hedge funds are going to use.
+pqBNAWwX2cU-00125-00034228-00034575 In order for them to look for stocks, they are going to look for stocks.
+pqBNAWwX2cU-00126-00034575-00034831 We look low Earnings multiples, low PE ratios.
+pqBNAWwX2cU-00127-00034835-00035035 So if the average is ten
+pqBNAWwX2cU-00128-00035035-00035355 because the market is recovering, everything is a little cheap.
+pqBNAWwX2cU-00129-00035358-00035451 So let's average.
+pqBNAWwX2cU-00130-00035451-00035661 is ten they find a stock which is eight.
+pqBNAWwX2cU-00131-00035681-00036058 They are going to invest in the stock, but gradually things get expensive, more
+pqBNAWwX2cU-00132-00036058-00036365 and more expensive. PE ratio now the average is already
+pqBNAWwX2cU-00133-00036365-00036765 15, is already 20 and there are less and less.
+pqBNAWwX2cU-00134-00036778-00037041 Companies available with low PE ratio.
+pqBNAWwX2cU-00135-00037058-00037295 They are going to invest now on Price to Sales ratio
+pqBNAWwX2cU-00136-00037318-00037478 they they don't consider earnings at all.
+pqBNAWwX2cU-00137-00037478-00037768 What's important is revenues before going to revenues.
+pqBNAWwX2cU-00138-00037768-00038118 Of course, they go for operating income through EBITDA and then the revenues
+pqBNAWwX2cU-00139-00038125-00038331 But after revenues, now it's other metrics.
+pqBNAWwX2cU-00140-00038355-00038678 Because there are some companies when we are very late in the cycle,
+pqBNAWwX2cU-00141-00038711-00039001 there are some companies revenues are not even here,
+pqBNAWwX2cU-00142-00039018-00039258 but they are other metrics, for example, number of users.
+pqBNAWwX2cU-00143-00039278-00039568 But once the recession hits or there is something bad,
+pqBNAWwX2cU-00144-00039568-00039938 like stagflation, like inflation, like we are seeing today, these hedge funds,
+pqBNAWwX2cU-00145-00039938-00040352 they are going to invest back into what we typically call the value stocks.
+pqBNAWwX2cU-00146-00040385-00040722 So stocks that can still make money during the recession.
+pqBNAWwX2cU-00147-00040752-00041278 So if let's say, we are probably going to see more stagflation.
+pqBNAWwX2cU-00148-00041302-00041492 We are already in stagflationary phase
+pqBNAWwX2cU-00149-00041492-00041852 even though the Federal Reserve System, the White House doesn't  admit it. 
+pqBNAWwX2cU-00150-00041852-00042195 that we have a recession, the economy is not growing that much.
+pqBNAWwX2cU-00151-00042255-00042395 We have high inflation.
+pqBNAWwX2cU-00152-00042395-00042742 So if this persists  and it seems that it will persist,
+pqBNAWwX2cU-00153-00042768-00043005 those hedge funds will invest in those companies
+pqBNAWwX2cU-00154-00043018-00043395 that are still capable of making money in this environment.
+pqBNAWwX2cU-00155-00043395-00043665 And one such company is GEO Group.
+pqBNAWwX2cU-00156-00043708-00044035 GEO Group is a private prison company
+pqBNAWwX2cU-00157-00044055-00044325 and those private prisons, they are always going to make money.
+pqBNAWwX2cU-00158-00044325-00044605 Whatever the market is doing, whatever the economy is doing,
+pqBNAWwX2cU-00159-00044635-00044788 they are always going to make money
+pqBNAWwX2cU-00160-00044788-00045132 because bad people will always be bad people and will be in jail.
+pqBNAWwX2cU-00161-00045175-00045522 Of course, there's political pressures on private prisons now.
+pqBNAWwX2cU-00162-00045522-00045925 President Joe Biden signed an executive order banning the DOJ,
+pqBNAWwX2cU-00163-00045925-00046242 the Department of Justice, from renewing their contracts
+pqBNAWwX2cU-00164-00046242-00046392 with the private prisoners.
+pqBNAWwX2cU-00165-00046392-00046839 So they are going to lose sales from the DOJ, from the US marshals, that is.
+pqBNAWwX2cU-00166-00046839-00047179 And also the Federal Bureau of Prisons.
+pqBNAWwX2cU-00167-00047212-00047452 But what about other sources? the states?
+pqBNAWwX2cU-00168-00047499-00047732 they are still going to make money from state governments.
+pqBNAWwX2cU-00169-00047755-00047992 They are still going to make money from ICE, especially.
+pqBNAWwX2cU-00170-00047992-00048209 We have a crisis on the southern border.
+pqBNAWwX2cU-00171-00048209-00048645 And now looking at how it is, expected that people will vote in November.
+pqBNAWwX2cU-00172-00048645-00048742 It seems that
+pqBNAWwX2cU-00173-00048742-00049172 the Republicans will take back the House, the Senate, it's seems the Democrats 
+pqBNAWwX2cU-00174-00049172-00049439 keep it. This is risky for the Democrats.
+pqBNAWwX2cU-00175-00049439-00049642 The Republicans take back the Senate.
+pqBNAWwX2cU-00176-00049672-00049872 Then it would be good for this business.
+pqBNAWwX2cU-00177-00049872-00050085 The hedge funds, they focus on all of this.
+pqBNAWwX2cU-00178-00050085-00050455 And for them right now, if you think like a hedge fund manager,
+pqBNAWwX2cU-00179-00050502-00050752 GEO Group is a stock that you should be buying.
+pqBNAWwX2cU-00180-00050779-00050919 You will tell me that
+pqBNAWwX2cU-00181-00050919-00051189 because of the controversies, they're not going to invest in it.
+pqBNAWwX2cU-00182-00051215-00051439 Believe me, the hedge funds, they are here to make money.
+pqBNAWwX2cU-00183-00051479-00051622 When the controversy is working
+pqBNAWwX2cU-00184-00051622-00051832 in their favor, they are going to invest in the company.
+pqBNAWwX2cU-00185-00051832-00051892 When the
+pqBNAWwX2cU-00186-00051892-00052255 controversies don't work in their favor, but they are going to short the company.
+pqBNAWwX2cU-00187-00052269-00052535 So it really depends where we are in the cycle.
+pqBNAWwX2cU-00188-00052565-00052895 And right now we are in a place in the cycle where it makes sense
+pqBNAWwX2cU-00189-00052895-00053255 for these hedge funds to invest in a company such as GEO Group.
+pqBNAWwX2cU-00190-00053279-00053562 And one hedge fund manager who is a contrarian
+pqBNAWwX2cU-00191-00053589-00053746 is already investing in this company.
+pqBNAWwX2cU-00192-00053746-00054002 That is Michael Burry, Scion Capital.
+pqBNAWwX2cU-00193-00054042-00054212 But we know he's a value investor.
+pqBNAWwX2cU-00194-00054212-00054376 Even though he's a hedge fund manager.
+pqBNAWwX2cU-00195-00054376-00054536 He's a value investor.
+pqBNAWwX2cU-00196-00054536-00054809 I'm not saying that all hedge funds are the same.
+pqBNAWwX2cU-00197-00054809-00054966 Some of them can be value investors.
+pqBNAWwX2cU-00198-00054966-00055129 Some of them are great investors.
+pqBNAWwX2cU-00199-00055129-00055332 But when I'm talking about hedge funds
+pqBNAWwX2cU-00200-00055352-00055679 investing in the company, I'm talking about hedge funds in general.
+pqBNAWwX2cU-00201-00055726-00055879 So would I invest in the company?
+pqBNAWwX2cU-00202-00055879-00056202 I'm already investing in GEO Group and you can find
+pqBNAWwX2cU-00203-00056202-00056539 the whole analysis on my research platform if you're interested.
+pqBNAWwX2cU-00204-00056539-00056959 You can have all my analyzes, all my research, and even my investing course
+pqBNAWwX2cU-00205-00056959-00057182 For beginners. It's completely for free there.
+pqBNAWwX2cU-00206-00057182-00057442 If you are interested with the research platform,
+pqBNAWwX2cU-00207-00057456-00057642 you can have a one month free trial.
+pqBNAWwX2cU-00208-00057642-00057836 So on the research platform I calculated
+pqBNAWwX2cU-00209-00057836-00058176 the intrinsic value of GEO Group and according to me it is undervalued.
+pqBNAWwX2cU-00210-00058212-00058566 This is what I'm focusing on when investing in GEO Group.
+pqBNAWwX2cU-00211-00058566-00058776 I do not think who will buy the stock in the future.
+pqBNAWwX2cU-00212-00058789-00058946 It doesn't really matter to me.
+pqBNAWwX2cU-00213-00058946-00059359 But I know because this is market psychology, this is what investors do.
+pqBNAWwX2cU-00214-00059399-00059689 If you as a value investor, you are buying a stock today
+pqBNAWwX2cU-00215-00059719-00059819 It is because you know
+pqBNAWwX2cU-00216-00059819-00059959 that at some point in the future
+pqBNAWwX2cU-00217-00059959-00060189 other people will be greedy and will be buying the stock.
+pqBNAWwX2cU-00218-00060226-00060723 So be greedy when others are fearful. People have been fearful about GEO Group for years.
+pqBNAWwX2cU-00219-00060723-00060973 I have been buying, but now it seems there is a change.
+pqBNAWwX2cU-00220-00060986-00061146 Soon people will be greedy.
+pqBNAWwX2cU-00221-00061146-00061363 Then you take your profits on GEO Group.
+pqBNAWwX2cU-00222-00061363-00061642 So this is something very common with most stocks you will see.
+pqBNAWwX2cU-00223-00061653-00061823 This is something that happens.
+pqBNAWwX2cU-00224-00061823-00061996 It's not just GEO Group.
+pqBNAWwX2cU-00225-00061996-00062149 It is something very common.
+pqBNAWwX2cU-00226-00062149-00062296 If there is a stock
+pqBNAWwX2cU-00227-00062296-00062663 All hedge funds will short it together and they are going to buy it together.
+pqBNAWwX2cU-00228-00062673-00062969 So you as a value investor, as a contrarian in the market.
+pqBNAWwX2cU-00229-00063036-00063179 Don't focus on all of this.
+pqBNAWwX2cU-00230-00063179-00063453 Focus on the numbers. Look at GEO Group today.
+pqBNAWwX2cU-00231-00063466-00063806 The company is trading at four times owner's earnings.
+pqBNAWwX2cU-00232-00063809-00064039 So this is an undervalued stock, according to me.
+pqBNAWwX2cU-00233-00064039-00064296 You calculate the intrinsic value, discount these earnings.
+pqBNAWwX2cU-00234-00064319-00064443 It is undervalued.
+pqBNAWwX2cU-00235-00064443-00064826 So this is what I'm focusing on, on the business, investing in the business.
+pqBNAWwX2cU-00236-00064826-00065063 And then later, those hedge fund, those investors.
+pqBNAWwX2cU-00237-00065063-00065366 They are going to invest in the company like they always do.
+pqBNAWwX2cU-00238-00065406-00065549 I'm not saying that it will happen
+pqBNAWwX2cU-00239-00065549-00065856 with GEO Group, but in general, this is how hedge funds work.
+pqBNAWwX2cU-00240-00065873-00066023 And now looking at GEO Group.
+pqBNAWwX2cU-00241-00066023-00066436 It is one of their probable targets, if we think like a hedge fund
+pqBNAWwX2cU-00242-00066446-00066589 in what they may invest in.
+pqBNAWwX2cU-00243-00066589-00066783 Another interesting thing we need to mention about
+pqBNAWwX2cU-00244-00066783-00067156 GEO Group, about the business is that now the credit agencies,
+pqBNAWwX2cU-00245-00067200-00067580 they have rerated, they have upgraded the credit rating of the company.
+pqBNAWwX2cU-00246-00067630-00068020 So another thing that hedge funds focus on. What the credit agencies
+pqBNAWwX2cU-00247-00068040-00068446 are saying. So when the credit agencies upgrade the debt of a company.
+pqBNAWwX2cU-00248-00068446-00068713 Hedge funds are always going to buy this company
+pqBNAWwX2cU-00249-00068726-00069000 this company, GEO Group, they have done debt exchange.
+pqBNAWwX2cU-00250-00069010-00069396 They are able in the future eventually to pay more debt and return cash to
+pqBNAWwX2cU-00251-00069396-00069713 Shareholders may be through buybacks or dividends. When they do that,
+pqBNAWwX2cU-00252-00069730-00070050 Once again, hedge funds are going to invest in such a company.
+pqBNAWwX2cU-00253-00070053-00070293 So let me know in the comments. What do you think about GEO Group?
+pqBNAWwX2cU-00254-00070293-00070413 Are you investing in the company?
+pqBNAWwX2cU-00255-00070413-00070676 Do you think hedge funds are going to invest in the company?
+pqBNAWwX2cU-00256-00070700-00070830 Thank you for watching this video.
+pqBNAWwX2cU-00257-00070830-00071010 Please like subscribe and share.
+pqBNAWwX2cU-00258-00071010-00071143 Please watch these two videos here
+pqBNAWwX2cU-00259-00071143-00071383 If you have missed them. Have a nice day and goodbye.
+psrBGIHDh30-00000-00000100-00000500 Hey, it's Yahor from Anywaywell. I hope you enjoy the beat!
+psrBGIHDh30-00001-00000500-00000900 Subscribe for more beats and freestyle videos!
+psrBGIHDh30-00002-00000900-00001200 Thank you. Let's go!
+pwlUKXNZwS8-00000-00000256-00000783 saya melamar ke harvard untuk 10 kali ditolak ketika kfc datang
+pwlUKXNZwS8-00001-00000783-00001471 cina datang ke negara saya 20 24 orang pergi untuk pekerjaan itu
+pwlUKXNZwS8-00002-00001471-00001927 23 orang menerima bahwa saya adalah satu-satunya pria
+pwlUKXNZwS8-00003-00002143-00002448 salah satu sumber rahasia kesuksesan alibaba
+pwlUKXNZwS8-00004-00002448-00003032 adalah bahwa kami memiliki banyak wanita 47 wanita perusahaan kami jika Anda ingin menang
+pwlUKXNZwS8-00005-00003032-00003376 di abad 21st memberdayakan orang lain untuk memastikan
+pwlUKXNZwS8-00006-00003376-00003679 orang lain lebih baik dari Anda maka Anda akan sukses
+pwlUKXNZwS8-00007-00003679-00004032 jadi saya menemukan wanita yang mereka pikirkan tentang orang lain lebih dari yang mereka pikirkan
+pwlUKXNZwS8-00008-00004032-00004287 diri
+pwlUKXNZwS8-00009-00004360-00004936 Cina hari ini adalah ekonomi terbesar kedua di dunia yang harus dibayar Cina
+pwlUKXNZwS8-00010-00004936-00005407 memperhatikan kualitas ekonomi seperti pertumbuhan manusia
+pwlUKXNZwS8-00011-00005407-00005687 waktu tertentu lambatnya pertumbuhan tubuh Anda akan
+pwlUKXNZwS8-00012-00005687-00006135 lambat tetapi Anda harus menumbuhkan pikiran Anda menumbuhkan budaya Anda menumbuhkan lembah Anda
+pwlUKXNZwS8-00013-00006135-00006415 kunjungan Anda saya pikir Cina sedang bergerak ke sana
+pwlUKXNZwS8-00014-00006415-00006664 arah
+pwlUKXNZwS8-00015-00006743-00007240 ketika Anda memiliki satu miliar dolar, itu bukan uang Anda
+pwlUKXNZwS8-00016-00007240-00007847 itulah kepercayaan yang diberikan masyarakat kepada Anda [Musik]
+pwlUKXNZwS8-00017-00007847-00008159 begitu banyak orang yang saya ajak bicara saat itu untuk alipay, kata mereka
+pwlUKXNZwS8-00018-00008159-00008560 ini adalah ide terbodoh yang pernah Anda miliki
+pwlUKXNZwS8-00019-00008560-00009015 tetapi saya mengatakan saya tidak memakai kolaborasi bodoh selama orang menggunakannya
+pwlUKXNZwS8-00020-00009015-00009768 sekarang ada 800 juta orang yang menggunakan alipay ini
+pwlUKXNZwS8-00021-00009872-00010504 Tai chi adalah tentang bagaimana Anda menyeimbangkan cara Anda bekerja seperti orang yang bersaing
+pwlUKXNZwS8-00022-00010504-00010960 Katakanlah ketika saya bersaing dengan ebay melihat Anda membenci ebay tidak, saya tidak membenci ebay
+pwlUKXNZwS8-00023-00010960-00011311 itu perusahaan yang hebat kau tahu mereka datang aku pergi kau tahu taiji
+pwlUKXNZwS8-00024-00011311-00011711 adalah seperti Anda berkelahi di sini saya pergi ke sana Anda meletakkan di atas saya pergi
+pwlUKXNZwS8-00025-00011711-00012200 hak sialan jadi keseimbangan saya
+pwlUKXNZwS8-00026-00012320-00012688 Bertahun-tahun yang lalu saya ingin mengubah dunia sekarang, saya pikir jika kita ingin berubah
+pwlUKXNZwS8-00027-00012688-00013095 dunia kita mengubah diri kita sendiri mengubah diri kita sendiri lebih penting
+pwlUKXNZwS8-00028-00013095-00013495 lebih mudah daripada mengubah dunia dan kedua adalah itu
+pwlUKXNZwS8-00029-00013495-00013831 Saya ingin meningkatkan dunia karena mungkin mengubah dunia
+pwlUKXNZwS8-00030-00013831-00014431 pekerjaan obama karena pekerjaanku adalah memastikan
+pwlUKXNZwS8-00031-00014431-00014791 bahwa tim saya bahagia karena tim saya bahagia
+pwlUKXNZwS8-00032-00014791-00015144 mereka bisa membuat kebiasaan saya bahagia
+pwlUKXNZwS8-00033-00015200-00015680 saya ingat salah satu manajemen senior dari orang-orang yang lebih hangat datang ke hamzo
+pwlUKXNZwS8-00034-00015680-00016095 lima tahun yang lalu saya berkata mungkin dalam 10 tahun akan lebih hangat dari yang Anda katakan muda
+pwlUKXNZwS8-00035-00016095-00016551 Pria, Anda memiliki harapan yang baik jika Anda ingin memiliki 10
+pwlUKXNZwS8-00036-00016551-00017008 000 pelanggan baru Anda harus membangun gudang baru dan ini itu
+pwlUKXNZwS8-00037-00017008-00018176 untuk saya layani
+p-c8Iniua4u-00000-00000500-00000997 It is very frightening to hear when the moderate party leader, Ulf Kristersson, 
+p-c8Iniua4u-00001-00000997-00001561 talks about crime, and measures against crime,
+p-c8Iniua4u-00002-00001561-00001978 when he has no idea --- why crime occurs!
+p-c8Iniua4u-00003-00002062-00002379 The moderates suggest that we should have harder prison sentences, 
+p-c8Iniua4u-00004-00002405-00002762 which we know will not help at all.
+p-c8Iniua4u-00005-00002862-00003106 We can take a look at the United States,
+p-c8Iniua4u-00006-00003106-00003503 which has one of the longest prison sentences in the world,
+p-c8Iniua4u-00007-00003503-00003880 and yet the United States is one of the countries
+p-c8Iniua4u-00008-00003880-00004177 with the highest crime rates in the world!
+p-c8Iniua4u-00009-00004180-00004581 So harsher prison sentences do not help against crime,
+p-c8Iniua4u-00010-00004581-00005378 instead we must influence the root causes of why a person becomes a criminal.
+p-c8Iniua4u-00011-00005378-00005702 To do the same thing, over and over again
+p-c8Iniua4u-00012-00005702-00005979 and expect a different result,
+p-c8Iniua4u-00013-00005979-00006379 that is the definition of madness!
+p-c8Iniua4u-00014-00006379-00006529 So harsher punishment does not help,
+p-c8Iniua4u-00015-00006529-00007227 when people are in urgent need of housing, money, and a financial security, 
+p-c8Iniua4u-00016-00007227-00007403 which is enough for the rest of their lives.
+p-c8Iniua4u-00017-00007500-00007971 We can look at the Jaguars in the Amazon rainforests, 
+p-c8Iniua4u-00018-00007971-00008268 who cuddle, socialize and thrive in each other's company
+p-c8Iniua4u-00019-00008268-00008582 when there is an abundance of food, 
+p-c8Iniua4u-00020-00008582-00008982 then become enemies and divide the rainforest into their own territories, 
+p-c8Iniua4u-00021-00008982-00009182 when there is a shortage of food.
+p-c8Iniua4u-00022-00009182-00009783 So crime arises when people have to compete for resources,
+p-c8Iniua4u-00023-00009783-00010560 ie when people are financially difficult, with a lack of money, food, or housing,
+p-c8Iniua4u-00024-00010560-00010820 Therefore people become desperate 
+p-c8Iniua4u-00025-00010820-00011301 And some commit crimes to secure their own, and their family's future.
+p-c8Iniua4u-00026-00011401-00011775 If the state were to allow the Swedish people to sell food,
+p-c8Iniua4u-00027-00011775-00012222 accessories and other small goods in public places
+p-c8Iniua4u-00028-00012222-00013019 Up to SEK 100,000 per year tax-free, without having to start complicated business forms,
+p-c8Iniua4u-00029-00013019-00013396 Then our young people would be able to make money in a legal way,
+p-c8Iniua4u-00030-00013396-00013693 instead of have to commit crimes.
+p-c8Iniua4u-00031-00013693-00014064 The state has made it popular to be a criminal,
+p-c8Iniua4u-00032-00014064-00014734 through the American culture of violence that was introduced on Swedish television in 1987,
+p-c8Iniua4u-00033-00014734-00015135 through the moderate company kinnevik.
+p-c8Iniua4u-00034-00015135-00015435 The fact that the state allows serious violence,
+p-c8Iniua4u-00035-00015435-00015986 murder and other destructive acts to be shown on television, every day,
+p-c8Iniua4u-00036-00015986-00016389 gives our children and young people a distorted picture of reality,
+p-c8Iniua4u-00037-00016389-00016800 and that violence is normalized in their minds.
+p-c8Iniua4u-00038-00016800-00017350 Violence on television works just like advertising for our children and young people,
+p-c8Iniua4u-00039-00017350-00017921 who often act just as violently as their role models on television.
+p-c8Iniua4u-00040-00018021-00018491 It is not news, that people admire what is shown on TV,
+p-c8Iniua4u-00041-00018491-00018822 states are very well aware of this,
+p-c8Iniua4u-00042-00018822-00019185 therefore they wont film, when football supporters 
+p-c8Iniua4u-00043-00019185-00019636 run onto football fields to hug their idols.
+p-c8Iniua4u-00044-00019736-00020190 If the state bans all violence on television, just as it did in the 80's,
+p-c8Iniua4u-00045-00020190-00020537 and instead shows comedies and constructive programs,
+p-c8Iniua4u-00046-00020537-00020794 and nature conservation programs,
+p-c8Iniua4u-00047-00020794-00021147 then the violence in society would probably disappear by itself,
+p-c8Iniua4u-00048-00021147-00021348 within a 10-year period.
+p-c8Iniua4u-00049-00021448-00022005 So in order to get rid of all the crimes in society, we must first create a just society,
+p-c8Iniua4u-00050-00022005-00022369 and give people what all people need,
+p-c8Iniua4u-00051-00022369-00023059 ie free food, free housing, and free energy, as well as our other welfare.
+p-c8Iniua4u-00052-00023059-00023610 Thereafter, the majority of all crimes will disappear automatically.
+p-c8Iniua4u-00053-00023610-00023810 How should this go, one might wonder?
+p-c8Iniua4u-00054-00023810-00024214 We have already built up our hydropower plants in Norrland 
+p-c8Iniua4u-00055-00024214-00024564 that produce electricity completely free of charge for the entire population of Sweden,
+p-c8Iniua4u-00056-00024564-00024818 we have already built up the main grid for electricity,
+p-c8Iniua4u-00057-00024818-00025218 and when we shut down ALL non-vital production,
+p-c8Iniua4u-00058-00025218-00025608 the electricity is good enough for all of Sweden's households,
+p-c8Iniua4u-00059-00025608-00026209 our infrastructure and the factories that will produce our vital needs,
+p-c8Iniua4u-00060-00026209-00026706 and we also get some electricity left over, which we can export to our neighboring countries,
+p-c8Iniua4u-00061-00026706-00027103 and thus get some extra money into the treasury.
+p-c8Iniua4u-00062-00027103-00027470 We have machines that produce all the food we need,
+p-c8Iniua4u-00063-00027470-00027771 almost completely without manpower, 
+p-c8Iniua4u-00064-00027771-00028321 we can imagine that a tractor can plow several thousand hectares of arable land,
+p-c8Iniua4u-00065-00028321-00028775 with only one person driving the tractor.
+p-c8Iniua4u-00066-00028775-00029235 We can also imagine what it was like about 200 years ago,
+p-c8Iniua4u-00067-00029235-00029883 when several hundred people were needed to sow, plow and harvest a single field,
+p-c8Iniua4u-00068-00029883-00030727 so a jobless society, with almost no labor and still get all the comfort we have today,
+p-c8Iniua4u-00069-00030727-00031087 we can achieve already today if the state wants it.
+p-c8Iniua4u-00070-00031087-00031411 We have already built up our homes through the "million housing program"
+p-c8Iniua4u-00071-00031411-00031808 and our parents have already built up our villas and other residential buildings, 
+p-c8Iniua4u-00072-00031808-00032295 so this costs nothing anymore once we write off our housing loans,
+p-c8Iniua4u-00073-00032295-00032782 which the banks have received back, through repayments and interest over the years.
+p-c8Iniua4u-00074-00032782-00033263 Therefore, we have the right to write off all mortgages,
+p-c8Iniua4u-00075-00033263-00034250 and thus get a fair world with free housing, and free food and energy,
+p-c8Iniua4u-00076-00034250-00034908 just as we have had in all times before bankers, business leaders and states
+p-c8Iniua4u-00077-00034908-00035358 started to take charge for things that are actually free,
+p-c8Iniua4u-00078-00035358-00035655 and can be produced with very little labor.
+p-c8Iniua4u-00079-00035655-00036132 Vote green, and feel free to vote for Nature's Party in the upcoming elections!
+p-c8Iniua4u-00080-00036132-00036299 Thank you for listening!!
+pAAO5UqiCPM-00000-00000000-00000200 Thanks for watching!! :D
+pBZOYULXJEg-00000-00001016-00001384 I'm Jay Berkley, I'm a Professor of Paediatric infectious diseases;
+pBZOYULXJEg-00001-00001440-00001968 I've been based at the overseas Programme in Kilifi for about 21 years.
+pBZOYULXJEg-00002-00002048-00002696 My research is really in causes of death and things that we can do about
+pBZOYULXJEg-00003-00002696-00003247 preventing deaths in the most vulnerable groups of children in developing countries.
+pBZOYULXJEg-00004-00003304-00004144 In a broad sense it's addressing groups such as children with malnutrition or new-born infants,
+pBZOYULXJEg-00005-00004200-00004784 but more specifically, my main focus is on infections and antibiotic trials
+pBZOYULXJEg-00006-00004784-00005224 to try and improve treatment of severe infections in children.
+pBZOYULXJEg-00007-00005224-00005944 So although there's been progress in reducing child mortality, we still see a very high
+pBZOYULXJEg-00008-00006032-00006720 mortality rate, particularly in new-borns, but there are also other high-risk groups in whom we
+pBZOYULXJEg-00009-00006720-00007320 don't seem to see any reduction in the fatality rates when children get admitted to hospital.
+pBZOYULXJEg-00010-00007384-00007744 The other thing that's emerged from recent research from our group and from
+pBZOYULXJEg-00011-00007744-00008152 other groups is that after children are discharged from hospital,
+pBZOYULXJEg-00012-00008240-00008616 there's still considerable mortality so some of the studies that we've done
+pBZOYULXJEg-00013-00008672-00009288 have shown that for every one child who dies in hospital another one dies post discharge, and so
+pBZOYULXJEg-00014-00009288-00010016 a big focus of our research is also understanding the post-discharge period - is it a problem that
+pBZOYULXJEg-00015-00010016-00010544 hasn't been treated sufficiently in hospital, or is it something about the child's environment or
+pBZOYULXJEg-00016-00010608-00011392 another infection that they're exposed to, so understanding those helps us understand
+pBZOYULXJEg-00017-00011392-00012088 how to make the treatments more effective - so that's our aim, to fight tropical illnesses.
+pBZOYULXJEg-00018-00012312-00012863 The largest study we do is the CHAIN Network which is the Childhood Acute Illness and Nutrition
+pBZOYULXJEg-00019-00012863-00013624 Network, it's a study that's at typical hospitals across nine sites in Africa and in south Asia,
+pBZOYULXJEg-00020-00013696-00014352 and in those sites we're looking at what are the potentially preventable causes of mortality
+pBZOYULXJEg-00021-00014352-00014896 in children, after all of the recommended treatments have been applied, so at the sites we
+pBZOYULXJEg-00022-00014991-00015480 make sure that they're following guidelines and treatments are available, we then follow a
+pBZOYULXJEg-00023-00015480-00016512 large cohort of about 4,000 children who are at increased risk of mortality to see what factors
+pBZOYULXJEg-00024-00016512-00017224 are leading to an increased risk of mortality in hospital and after discharge, with the purpose of
+pBZOYULXJEg-00025-00017224-00017936 then taking some of the findings into clinical trials in order to directly improve outcomes.
+pBZOYULXJEg-00026-00017992-00018312 We've just finished a small but intensive trial
+pBZOYULXJEg-00027-00018312-00019080 of a potentially new antibiotic treatment for neonatal sepsis, and we're doing a very
+pBZOYULXJEg-00028-00019080-00019640 large trial of first-line antibiotics for severely ill malnourished children.
+pBZOYULXJEg-00029-00019848-00020432 We're doing those trials because there's an emerging, increasing problem with antimicrobial
+pBZOYULXJEg-00030-00020432-00020936 resistance - in the settings where we work typically people don't have any information on
+pBZOYULXJEg-00031-00020936-00021592 what exact bacteria are causing illness and what the sensitivity or resistance to antibiotics is,
+pBZOYULXJEg-00032-00021592-00022416 so we need policies which are going to be able to be used in those typical settings,
+pBZOYULXJEg-00033-00022416-00022816 and these are settings where the vast majority of children are treated.
+pBZOYULXJEg-00034-00022920-00023360 There are some information around, but they typically come from university hospitals or
+pBZOYULXJEg-00035-00023360-00024152 private hospitals where only a tiny proportion of children get treated, so by investing in research,
+pBZOYULXJEg-00036-00024152-00024728 in that more generalisable population, with we're providing information and
+pBZOYULXJEg-00037-00024728-00025376 improvements in guidelines and knowledge which will improve outcomes in a much broader sense.
+pBZOYULXJEg-00038-00025376-00025936 With regards to translation, we do a lot of work in the laboratory: firstly, we do a lot of
+pBZOYULXJEg-00039-00025936-00026512 work on characterising what kinds of infections children have - sometimes that uses the normal
+pBZOYULXJEg-00040-00026512-00027136 microbiology techniques and sometimes that uses much more advanced, modern molecular techniques
+pBZOYULXJEg-00041-00027136-00027695 for detecting bacteria, for detecting resistance and detecting other kinds of infections.
+pBZOYULXJEg-00042-00027768-00028200 We also do a lot of work with partners overseas and other labs,
+pBZOYULXJEg-00043-00028200-00028920 looking at the effect on the nutrition and infections have on children's metabolism.
+pBZOYULXJEg-00044-00028920-00029408 By getting this very detailed data and putting it together with the detailed information that
+pBZOYULXJEg-00045-00029408-00030064 we collect from the children themselves, we're then able to look at what pathways
+pBZOYULXJEg-00046-00030064-00030712 might be intervenable, so what pathways could be interrupted or changed, which would then improve
+pBZOYULXJEg-00047-00030712-00031360 outcomes and so we can translate those findings from the laboratory into initially small trials,
+pBZOYULXJEg-00048-00031360-00031936 where we look at safety or look at whether or not the pathways are being affected,
+pBZOYULXJEg-00049-00031936-00032383 and then into larger trials to go into policy to improve outcomes.
+pH8Z7znu2CM-00000-00001096-00002985 Joseph and Kendra Duggar have announced they are expecting their first child. The newlyweds shared the big news in a video message posted to TLC's website, along with a photo of themselves kissing each other as they held up his and her mugs that read Mom and Dad.
+pH8Z7znu2CM-00001-00003135-00004969 'We're expecting,' father-to-be Joseph, 22, proclaimed happily as he sat beside his wife, 19. 'I've always dreamed of having my own family,' the star explained in the video. 'It's really surreal that it's actually here.'
+pH8Z7znu2CM-00002-00005119-00006578 Joseph speculated they may be having a boy.  'Either way, I'm happy just to have a baby,' he said.  'I think it could be a girl!' Kendra added. 'You never know!' 
+pH8Z7znu2CM-00003-00006728-00007947 'I hope the baby has Kendra's laugh,' Joseph mused, to the delight of his wife.  'I hope the baby will be patient like Joe,' Kendra said. 
+pH8Z7znu2CM-00004-00008097-00009167 Joseph, who is the son of Jim Bob and Michelle Duggar, married Kendra in Siloam Springs, Arkansas in September. 
+pI2qMPIkDO8-00000-00000000-00000630 The Asian Arowana, also known as the Dragon Fish or Asian Bonytongue, is a captivating and highly
+pI2qMPIkDO8-00001-00000630-00001194 sought-after species of freshwater fish. With its vibrant colors, unique scales, and graceful
+pI2qMPIkDO8-00002-00001194-00001680 movements, it has become a popular choice for fish enthusiasts and collectors worldwide.
+pI2qMPIkDO8-00003-00001680-00002244 In this video, we will explore the fascinating world of Asian Arowana, including its appearance,
+pI2qMPIkDO8-00004-00002244-00002838 behavior, habitat, and care requirements. The Asian Arowana is renowned for its striking
+pI2qMPIkDO8-00005-00002838-00003354 appearance. It has a long, slender body with large scales that shimmer and reflect light,
+pI2qMPIkDO8-00006-00003354-00003929 giving it a majestic and dragon-like appearance. The fish comes in various colors, including red,
+pI2qMPIkDO8-00007-00003929-00004596 golden, green, and silver. Each color variation is highly prized and often associated with different
+pI2qMPIkDO8-00008-00004596-00005184 qualities, such as wealth, luck, and prosperity. Asian Arowanas are known for their intelligent
+pI2qMPIkDO8-00009-00005184-00005652 and curious nature. They are active swimmers and are capable of leaping out of the water to
+pI2qMPIkDO8-00010-00005652-00006084 catch prey. They have a powerful jaw and sharp teeth, making them skilled
+pI2qMPIkDO8-00011-00006084-00006594 predators. These fish are also known for their territorial behavior and may exhibit aggression
+pI2qMPIkDO8-00012-00006594-00007026 towards other fish in their environment. Asian Arowanas are native to the freshwater
+pI2qMPIkDO8-00013-00007026-00007518 habitats of Southeast Asia, particularly in countries like Malaysia, Indonesia,
+pI2qMPIkDO8-00014-00007518-00008106 and Thailand. They prefer slow-moving rivers, flooded forests, and swamps. In the wild,
+pI2qMPIkDO8-00015-00008106-00008604 they are often found in densely vegetated areas where they can hide and ambush their prey.
+pI2qMPIkDO8-00016-00008604-00009168 Keeping Asian Arowanas requires specific care to ensure their health and well-being. They need
+pI2qMPIkDO8-00017-00009168-00009678 a spacious aquarium with ample swimming space and a secure lid to prevent jumping. The water
+pI2qMPIkDO8-00018-00009678-00010554 temperature should be maintained between 76-82°F (24-28°C), and the pH level should be slightly
+pI2qMPIkDO8-00019-00010554-00011118 acidic to neutral. Arowanas are carnivorous and should be fed a diet consisting of live or frozen
+pI2qMPIkDO8-00020-00011118-00011682 foods like insects, small fish, and shrimp. It is important to note that Asian Arowanas
+pI2qMPIkDO8-00021-00011682-00012240 are protected species in many countries due to overfishing and habitat destruction. Therefore,
+pI2qMPIkDO8-00022-00012240-00012762 owning or trading these fish may be subject to legal regulations and restrictions. It
+pI2qMPIkDO8-00023-00012762-00013254 is essential to research and adhere to local laws and obtain fish from reputable sources.
+pI2qMPIkDO8-00024-00013254-00013686 Asian Arowanas hold great cultural significance in many Asian countries.
+pI2qMPIkDO8-00025-00013686-00014226 They are often considered symbols of luck, prosperity, and power. In some cultures,
+pI2qMPIkDO8-00026-00014226-00014682 they are believed to bring good fortune to their owners and are kept as prized possessions.
+pI2qMPIkDO8-00027-00014682-00015336 Due to habitat loss and overfishing, many Asian Arowana species are classified as endangered or
+pI2qMPIkDO8-00028-00015336-00015828 critically endangered. Conservation organizations and governments are working together to protect
+pI2qMPIkDO8-00029-00015828-00016278 their natural habitats and regulate their trade to ensure their survival in the wild.
+pI2qMPIkDO8-00030-00016278-00016890 In conclusion, the Asian Arowana is a captivating and revered fish species with its stunning colors,
+pI2qMPIkDO8-00031-00016890-00017460 graceful movements, and cultural significance. While they require specific care and are subject
+pI2qMPIkDO8-00032-00017460-00017988 to legal considerations, they can bring joy and fascination to dedicated fish enthusiasts.
+pI2qMPIkDO8-00033-00017988-00018420 By understanding their unique characteristics and supporting conservation efforts,
+pI2qMPIkDO8-00034-00018420-00018936 we can appreciate and help preserve the beauty of the Asian Arowana for future generations to come.
+pIZMu3I0xuy-00000-00001216-00001372 Human Population
+pIZMu3I0xuy-00001-00001372-00001752 Modern humans evolved in Africa about 200,000 years ago.
+pIZMu3I0xuy-00002-00001752-00002318 About 100,000 years ago, we began migrating across the globe.
+pIZMu3I0xuy-00003-00003490-00004460 Our population remained low—probably less than 1 million people.
+pIZMu3I0xuy-00004-00004460-00004804 With the advent of farming, growth picked up.
+pIZMu3I0xuy-00005-00004804-00005462 By AD 1, world population reached approximately 170 million people.
+pIZMu3I0xuy-00006-00006200-00006618 Year: AD 1 Human Population: 170 m
+pIZMu3I0xuy-00007-00006620-00007026 Dot = 1 million people
+pIZMu3I0xuy-00008-00007252-00007445 Han Dynasty
+pIZMu3I0xuy-00009-00007445-00007622 Roman Empire
+pIZMu3I0xuy-00010-00008480-00008656 Human population: 177 m
+pIZMu3I0xuy-00011-00008660-00008820 Silk Road
+pIZMu3I0xuy-00012-00010308-00010358 Year: 300 AD
+pIZMu3I0xuy-00013-00010360-00010518 Human population: 180 m
+pIZMu3I0xuy-00014-00010520-00010780 Golden Age of India
+pIZMu3I0xuy-00015-00012292-00012452 Year: 500 AD
+pIZMu3I0xuy-00016-00012452-00012560 Human population: 177 m
+pIZMu3I0xuy-00017-00012560-00012760 Peak Mayan Civilization
+pIZMu3I0xuy-00018-00013463-00013641 Year: 600 AD
+pIZMu3I0xuy-00019-00013641-00013778 Human population: 181 m
+pIZMu3I0xuy-00020-00013778-00013978 Birth of Islam
+pIZMu3I0xuy-00021-00014719-00014838 Year: 750 AD Human Population: 195 m
+pIZMu3I0xuy-00022-00014838-00014986 Smallpox in Japan
+pIZMu3I0xuy-00023-00015812-00015934 Year: 850 AD Human population: 215m
+pIZMu3I0xuy-00024-00015934-00016084 Gunpowder invented
+pIZMu3I0xuy-00025-00017700-00017900 Year: 1050 AD Human population: 283 m
+pIZMu3I0xuy-00026-00017900-00018100 Navigational compass
+pIZMu3I0xuy-00027-00019512-00019712 Year: 1200 AD Human population: 362 m
+pIZMu3I0xuy-00028-00019712-00019912 Mongol Empire
+pIZMu3I0xuy-00029-00020790-00020916 Year: 1350 Human population: 364 m
+pIZMu3I0xuy-00030-00020916-00021398 Bubonic plague Rare decline in world population
+pIZMu3I0xuy-00031-00022108-00022252 Year: 1490 AD Human population: 405 m
+pIZMu3I0xuy-00032-00022252-00022452 Europeans arrive
+pIZMu3I0xuy-00033-00022570-00022668 Year: 1520 AD Human population: 456 m
+pIZMu3I0xuy-00034-00022668-00022840 Transatlantic slave trade
+pIZMu3I0xuy-00035-00024868-00025058 Year: 1740 Human population: 714 m
+pIZMu3I0xuy-00036-00025058-00025258 Industrial Revolution Modern technology and medicine bring faster growth
+pIZMu3I0xuy-00037-00026404-00026592 Year: 1910 Human population: 1.7 b
+pIZMu3I0xuy-00038-00026592-00026792 World Wars
+pIZMu3I0xuy-00039-00027577-00027810 Year: 2015 Human population: 7.4 b
+pIZMu3I0xuy-00040-00027810-00028226 Year: 2050 Human population (projected): 9.5 b
+pIZMu3I0xuy-00041-00028808-00029058 LEVELING OFF
+pIZMu3I0xuy-00042-00029058-00029498 If current trends continue, global populations will peak at 11 billion around 2100.
+pIZMu3I0xuy-00043-00029874-00030170 Growth is slowing because average fertility rates are falling in nearly every country.
+pIZMu3I0xuy-00044-00030170-00030370 1950: 5 babies/woman
+pIZMu3I0xuy-00045-00030370-00030570 2100: 2 babies/woman
+pIZMu3I0xuy-00046-00031027-00031494 If fertility forecasts are slightly off, our peak population will peak lower...
+pIZMu3I0xuy-00047-00031610-00031924 ...or higher.
+pIZMu3I0xuy-00048-00032536-00032896 It took 200,000 years for our population to reach 1 billion.
+pIZMu3I0xuy-00049-00032896-00033276 And only 200 years to reach 7 billion.
+pIZMu3I0xuy-00050-00033898-00034518 As our population has grown, so has our use of Earth's resources.
+pIZMu3I0xuy-00051-00034728-00034890 Choices we make today
+pIZMu3I0xuy-00052-00034890-00035090 family planning
+pIZMu3I0xuy-00053-00035090-00035290 reduced consumption
+pIZMu3I0xuy-00054-00035290-00035522 pollution controls
+pIZMu3I0xuy-00055-00035522-00035722 habitat protection
+pIZMu3I0xuy-00056-00035928-00036318 Choices we make today affect the future of our species—and all life on Earth.
+pJiMEWkbLRk-00000-00000411-00000532 WELCOME TO ALL HANDS UPDATE.
+pJiMEWkbLRk-00001-00000532-00000900 I'M PETTY OFFICER IAN COTTER WITH YOUR HEADLINES FROM AROUND THE FLEET.
+pJiMEWkbLRk-00002-00000900-00001301 THE HARRY S. TRUMAN CARRIER STRIKE GROUP RELIEVED THE NIMITZ CARRIER STRIKE GROUP IN THE ARABIAN
+pJiMEWkbLRk-00003-00001301-00001695 SEA AUGUST 26TH, ASSUMING THE DUTIES AS COMMANDER, TASK FORCE 50.
+pJiMEWkbLRk-00004-00001695-00002221 CTF 50 IS RESPONSIBLE FOR CONDUCTING MARITIME SECURITY OPERATIONS, BUILDING TRUST AND CONFIDENCE
+pJiMEWkbLRk-00005-00002221-00002491 IN THE REGION AND SUPPORTING OPERATION ENDURING FREEDOM.
+pJiMEWkbLRk-00006-00002491-00003013 OVER IN THE CORAL SEA, U.S. NAVY AND AUSTRALIAN DEFENCE FORCE RECOVERED AND DISPOSED OF LIVE
+pJiMEWkbLRk-00007-00003013-00003514 ORDINANCE THAT WAS EMERGENCY JETTISONED BY TWO AV-8 BRAVO HARRIER AIRCRAFT JULY 16TH
+pJiMEWkbLRk-00008-00003514-00003731 DURING OPERATION TALISMAN SABER.
+pJiMEWkbLRk-00009-00003731-00004194 EXPLOSIVE ORDINANCE DISPOSAL MOBILE UNIT FIVE DIVERS RETRIEVED TWO UNARMED LIVE ROUNDS OFF
+pJiMEWkbLRk-00010-00004194-00004607 THE COAST OF QUEENSLAND AUSTRALIA AUGUST 29TH AND 30TH IN AN EFFORT TO PROTECT THE GREAT
+pJiMEWkbLRk-00011-00004607-00004766 BARRIER REEF.
+pJiMEWkbLRk-00012-00004766-00005028 YOU CAN READ MORE ABOUT THESE STORIES AT NAVY.MIL.
+pJiMEWkbLRk-00013-00005028-00005197 FROM ALL HANDS UPDATE, THANKS FOR WATCHING.
+pL7hSe4ua-c-00000-00000000-00000200 Es campur
+pL7hSe4ua-c-00001-00000200-00000400 Mix ice
+pL7hSe4ua-c-00002-00000400-00000600 Indonesian street food 
+pQfOChL-tSg-00000-00000062-00000252 - Once you find your dream home
+pQfOChL-tSg-00001-00000252-00000492 we need to make sure you get to move in.
+pQfOChL-tSg-00002-00000492-00000772 Here are 10 things not to do
+pQfOChL-tSg-00003-00000772-00001091 once you're under contract on your dream home.
+pQfOChL-tSg-00004-00001091-00001398 Hi I'm Danielle Scott with Keller Williams Realty.
+pQfOChL-tSg-00005-00001398-00001887 Once you're under contract don't do any of the following.
+pQfOChL-tSg-00006-00001887-00002087 One, don't change jobs,
+pQfOChL-tSg-00007-00002087-00002377 become self-employed or quit your job.
+pQfOChL-tSg-00008-00002377-00002729 Two, don't buy a new car, van or truck
+pQfOChL-tSg-00009-00002729-00002965 or you may end up living in it.
+pQfOChL-tSg-00010-00002965-00003261 Three, don't use any charge cards
+pQfOChL-tSg-00011-00003261-00003588 or let your accounts fall behind.
+pQfOChL-tSg-00012-00003588-00004033 Four, don't spend any money you have set aside for closing.
+pQfOChL-tSg-00013-00004033-00004379 Five, don't omit any debt or liabilities
+pQfOChL-tSg-00014-00004379-00004555 from your loan application.
+pQfOChL-tSg-00015-00004555-00004904 Six, don't buy any furniture even if the payment
+pQfOChL-tSg-00016-00004904-00005164 is deferred until after closing.
+pQfOChL-tSg-00017-00005164-00005464 Seven, don't allow any additional inquiries
+pQfOChL-tSg-00018-00005464-00005756 on your credit except for insurance purposes.
+pQfOChL-tSg-00019-00005756-00006018 Eight, don't make any large deposits
+pQfOChL-tSg-00020-00006018-00006365 or take out any large deposits without first
+pQfOChL-tSg-00021-00006365-00006620 checking with your lender or your agent.
+pQfOChL-tSg-00022-00006620-00006892 Nine, don't change bank accounts.
+pQfOChL-tSg-00023-00006892-00007294 10, don't co-sign on a loan for anyone no matter what
+pQfOChL-tSg-00024-00007294-00007476 and be sure to tell your lender
+pQfOChL-tSg-00025-00007476-00007745 if you already are co-signed for anyone.
+pQfOChL-tSg-00026-00007745-00008004 That debt goes against you too.
+pQfOChL-tSg-00027-00008004-00008442 I hope this helps I want to make sure that you understand
+pQfOChL-tSg-00028-00008442-00008778 how to keep your home and actually close
+pQfOChL-tSg-00029-00008778-00008946 once we help you find it.
+pQfOChL-tSg-00030-00008946-00009254 Feel free to reach out to me if you'd like to go
+pQfOChL-tSg-00031-00009254-00009603 into more detail on any of these points.
+pQfOChL-tSg-00032-00009603-00009712 I'm Danielle Scott,
+pQfOChL-tSg-00033-00009712-00010037 let's make it great home buying and selling day.
+pR87HZRPhGU-00000-00000072-00000872 Hey uh.. so today I'm going to be talking about the media and the way the media
+pR87HZRPhGU-00001-00001080-00001592 interprets or depicts countries
+pR87HZRPhGU-00002-00001768-00002240 outside, internationally and nationally within the country.
+pR87HZRPhGU-00003-00002392-00003240 When you see a media web page or you read a newspaper, if.. if anyone's still doing it um..,
+pR87HZRPhGU-00004-00003296-00003879 but let's just say you're looking at a media web page and the web page's title is,
+pR87HZRPhGU-00005-00003960-00004872 "China Decides la la la.." um.. "America.. America Plans To na na na", um.. "Russia.
+pR87HZRPhGU-00006-00004984-00005592 Russia a b c d whatever", "Russia Wants To Do This", "Russia Is Doing This", "Ukraine
+pR87HZRPhGU-00007-00005592-00006464 Is Doing That", "Burma Wants To la la la.." Every time, every time you see a whole country's name..
+pR87HZRPhGU-00008-00006552-00007231 a whole country's name in the beginning of a headline or in a headline at all,
+pR87HZRPhGU-00009-00007231-00007984 know that they're lying through their teeth. It's an outright, blatant lie. Think of it this way.
+pR87HZRPhGU-00010-00008064-00008744 Imagine if it was the name of your country, the country that you're in and people abroad were
+pR87HZRPhGU-00011-00008744-00009440 reading a very incendiary or very hot topic kind of headline about your country, and all.. it's
+pR87HZRPhGU-00012-00009440-00010040 like, for example my country is South Africa. If all of a sudden um.. people in China were reading
+pR87HZRPhGU-00013-00010040-00010896 an article, and it was titled, "South Africa Sets Massive Sanctions Against China". South Africa
+pR87HZRPhGU-00014-00011080-00011792 sets massive sanctions against China. Is it true? No, it's really not. No, it's really not. You know
+pR87HZRPhGU-00015-00011792-00012504 why? Because it's not South Africa - it is a very small tiny little party of South Africa
+pR87HZRPhGU-00016-00012504-00013232 that has decided. It decided this on behalf of the whole of the rest of South Africa, all the plebs
+pR87HZRPhGU-00017-00013232-00013944 ,all of the public, the tax paying public, all the little people - even though those people are
+pR87HZRPhGU-00018-00013944-00014680 probably completely unaware of what's going on, and this is true for all countries at the moment.
+pR87HZRPhGU-00019-00014791-00015528 The media is not representing us accurately, and this is true in all countries. So whenever you
+pR87HZRPhGU-00020-00015528-00016056 see that, whenever you see one particular country saying they're doing whatever,
+pR87HZRPhGU-00021-00016240-00016888 it's not true. Maybe it is one small part of.. of that country. Maybe it's the.. the government or
+pR87HZRPhGU-00022-00016888-00017504 whatever, but it's not the whole country and very, very frequently, it's definitely not the majority.
+pR87HZRPhGU-00023-00017632-00018264 It's that.. that ruling part - it's that.. that.. whether it's dictatorial, whether it's democratic,
+pR87HZRPhGU-00024-00018264-00018760 whether it's.. I don't know first world or third world or whatever.
+pR87HZRPhGU-00025-00018760-00019240 There's a very small handful of people that are deciding for that whole country what they're doing
+pR87HZRPhGU-00026-00019240-00019696 and very frequently there are no referendums, there are no questions going out to the public
+pR87HZRPhGU-00027-00019696-00020064 like, should we do this or not, should we go ahead with this or.. No, no, no,
+pR87HZRPhGU-00028-00020064-00020584 that doesn't happen. They just go ahead and decide it, and now.. now everybody's in the boat with us;
+pR87HZRPhGU-00029-00020584-00021120 everyone's going to get judged for the decisions that we make because we're all
+pR87HZRPhGU-00030-00021120-00021664 lumped into this.. this word that is the country name. Know that this is happening everywhere
+pR87HZRPhGU-00031-00022024-00022128 and the way that
+pR87HZRPhGU-00032-00022352-00023144 you will be judged for being lumped in with your country is the same that is.. is exactly
+pR87HZRPhGU-00033-00023144-00023576 the same that.. that's happening to other people, for example in Russia at the moment.
+pR87HZRPhGU-00034-00023576-00024056 There are so many Russians that are against this war. There are so many Russians that have been
+pR87HZRPhGU-00035-00024056-00024576 protest.. protesting in the streets uh.. how much, what.. what can they do, what more can
+pR87HZRPhGU-00036-00024576-00024968 they do besides screaming and shouting and maybe being online and maybe endangering..
+pR87HZRPhGU-00037-00025024-00025560 I don't know, their social profile, or you know in Russia it's.. it's not um.. hang on.
+pR87HZRPhGU-00038-00025856-00026304 I was saying um.. about Russia. I mean it's not easy being in Russia. I mean
+pR87HZRPhGU-00039-00026304-00026976 those people have been trained for the last two generations at least that, you know,
+pR87HZRPhGU-00040-00026976-00027408 what the government says, the government does and you don't question it, or you know hey, maybe
+pR87HZRPhGU-00041-00027408-00027992 there's a knock at the door at night so 12.. 12.. 12.. o'clock at night, and somebody disappears.
+pR87HZRPhGU-00042-00027992-00028624 Really? You know anyone's saying anything.. like these people are actually really, really brave
+pR87HZRPhGU-00043-00028624-00028992 being out there in the street actually protesting like that. They are really brave.
+pR87HZRPhGU-00044-00029208-00029608 They're risking quite a lot just doing that. They're probably on a social profile. They're
+pR87HZRPhGU-00045-00029608-00029912 probably on a list, a list somewhere,
+pR87HZRPhGU-00046-00029912-00030264 a red list. You know what I mean? Yeah, you do know what I mean.
+pR87HZRPhGU-00047-00030544-00030983 So they're not part of that 'Russia Decides', 'Russia Does This, Russia Does That'. Really?
+pR87HZRPhGU-00048-00031064-00031360 Really? Yeah, whatever.
+pR87HZRPhGU-00049-00031639-00032088 I think this kind of thing happens to China a lot. Now I live in China at the moment,
+pR87HZRPhGU-00050-00032088-00032776 but like the way that the media portrays China in certain countries is very much 'the bad guy',
+pR87HZRPhGU-00051-00032832-00033480 very much the big threat, very much like, "Oh the crouching tiger gonna beat us up one day!"
+pR87HZRPhGU-00052-00033672-00034312 Really? Whenever you see "China", know that whatever they're assuming
+pR87HZRPhGU-00053-00034376-00035112 or whatever action is happening, it's probably from the government. It's not from the masses,
+pR87HZRPhGU-00054-00035176-00035744 it's not from the masses and very frequently the masses don't even know what's going on.
+pR87HZRPhGU-00055-00035952-00036624 The masses are being given.. the masses get a very very very
+pR87HZRPhGU-00056-00036904-00037560 portioned view of what's going on. The government decides everything. There's a lot going on that we
+pR87HZRPhGU-00057-00037560-00038168 don't know about. I'm talking about 'we', I mean me and us.. us foreigners with the Chinese people.
+pR87HZRPhGU-00058-00038168-00038568 There's a lot that we don't.. we see stuff that's going on in our media here,
+pR87HZRPhGU-00059-00038672-00039384 but we very frequently don't know what's really going on. We will see numbers and we'll see like,
+pR87HZRPhGU-00060-00039480-00040072 headlines and stuff, but like, we just go, "Okay well, you know uhh..." So we have a very,
+pR87HZRPhGU-00061-00040072-00040904 very limited access to the entire body of what.. knowledge of what's really going on out there,
+pR87HZRPhGU-00062-00040904-00041456 and so when they say, "China la la la", "China Is Doing This" in another foreign um..
+pR87HZRPhGU-00063-00041592-00042344 media website, etc. to people of other nations, you have got to know, you have got to know that
+pR87HZRPhGU-00064-00042344-00042912 that's a lie because when they say 'China', they don't mean all of China. They mean the government
+pR87HZRPhGU-00065-00042912-00043592 of China. They mean the.. the deciding parties. The rest of china probably doesn't have a clue
+pR87HZRPhGU-00066-00043592-00044208 what's going on and that's the reality, and I would go ahead and say the same case is the
+pR87HZRPhGU-00067-00044208-00044760 same for many other countries too. When they say 'Ukraine', when they say 'Russia', when they say
+pR87HZRPhGU-00068-00044760-00045256 'America', when they say 'Indonesia', whatever - they put the whole country name - they are
+pR87HZRPhGU-00069-00045256-00045864 lying through their teeth. It's the government; it's a certain small party. It has nothing to do
+pR87HZRPhGU-00070-00045864-00046432 with more than 50 percent of that population, but the way it's worded, you would think it is. It's
+pR87HZRPhGU-00071-00046432-00046928 not. Very frequently most of those people have no clue about what's going on, and what is being
+pR87HZRPhGU-00072-00046928-00047728 decided for them and other countries. This is.. this is just a fact and we have to be aware of
+pR87HZRPhGU-00073-00047728-00048312 this. You have to be aware of this. The reason why is because a lot there's a lot of manipulation,
+pR87HZRPhGU-00074-00048312-00049112 there's a lot of manipulation going on to have us against each other, to build us up with
+pR87HZRPhGU-00075-00049112-00049672 these stories about what's going on in other countries so that we butt heads, and this can
+pR87HZRPhGU-00076-00049736-00050328 validate fighting against each other or validate a war. This.. this happens all the time and you
+pR87HZRPhGU-00077-00050328-00050896 must know that it's going on. You can only take these headlines with a.. a pinch of salt. What am
+pR87HZRPhGU-00078-00050896-00051455 I saying? A pinch of salt? Or you can't take it seriously. You just can't take it seriously. If
+pR87HZRPhGU-00079-00051455-00052216 the headline says, "Chinese Government Decides la la la" okay, fair. That's probably getting closer
+pR87HZRPhGU-00080-00052216-00052776 to the truth. If the headline says, "Russian Government la la la" or "Putin la la la",
+pR87HZRPhGU-00081-00052776-00053416 okay cool, cool, fine. Well, let's work with that. If it says uh.. "The American Government.." or,
+pR87HZRPhGU-00082-00053416-00053920 I don't know, "Obama.." well, choose.. choose a president that happens to be there these
+pR87HZRPhGU-00083-00053920-00054440 particular years. Fine, whatever - if they use that name and if they use the government kind of
+pR87HZRPhGU-00084-00054440-00055120 area, fine; that's working, but if they just use the whole country name, oh my god, alarm bells,
+pR87HZRPhGU-00085-00055120-00055664 red herrings, na ah ah. Something is really off here because know that most of that population,
+pR87HZRPhGU-00086-00055728-00056248 I would say I'm going to go ahead and say most populations of nations worldwide at the moment
+pR87HZRPhGU-00087-00056248-00056800 are not in a good place especially post-COVID. Economically even, people don't know where they
+pR87HZRPhGU-00088-00056800-00057336 stand, they don't know what they're going to be doing - this is a high area of instability. People
+pR87HZRPhGU-00089-00057336-00057824 are just trying to hold their lives together, they are just trying to hold their lives together.
+pR87HZRPhGU-00090-00058016-00058455 So basically people at their.. at the moment are kind of fighting for their lives.
+pR87HZRPhGU-00091-00058704-00059688 They just.. I think the dream, a high dream for most populations at the moment
+pR87HZRPhGU-00092-00059896-00060055 is a sense of normality
+pR87HZRPhGU-00093-00060288-00061104 and stability. That's a high hope. A stable job for most, I mean is that a thing
+pR87HZRPhGU-00094-00061104-00061455 that the average person can even wish or ask for.. for these days?
+pR87HZRPhGU-00095-00061696-00062304 Um.. to.. to not be in any kind of particular conflict, to have a roof over your head, to..
+pR87HZRPhGU-00096-00062304-00062648 to have a stable roof over your head that you can count on from one year to the next.
+pR87HZRPhGU-00097-00062920-00063640 These are the things that are not just bothering one nation, but nations worldwide. will this is
+pR87HZRPhGU-00098-00063640-00064384 not a time that humanity is doing well, so in any one particular country, they're just worried
+pR87HZRPhGU-00099-00064384-00064767 about the clothes on their back. They're worried about getting through the week, getting through
+pR87HZRPhGU-00100-00064767-00065520 the months, getting through the year. They have no energy, time or wherewithal to be thinking,
+pR87HZRPhGU-00101-00065624-00066064 "Let's.. let's.. let's get together and do something about that country over
+pR87HZRPhGU-00102-00066064-00066728 there because we have a thing out for you! We don't like you." Nobody cares. Nobody
+pR87HZRPhGU-00103-00066856-00067384 cares. They're just all trying to survive. They're all trying to just make ends meet,
+pR87HZRPhGU-00104-00067456-00067688 and get to the end of the month, get to the end of the year,
+pR87HZRPhGU-00105-00067688-00068096 and sense of stability for the kids, and so on and so forth, get.. get their kids to
+pR87HZRPhGU-00106-00068168-00069032 to school on time, have a.. have a job that keeps rolling through, save up for whatever - that's..
+pR87HZRPhGU-00107-00069032-00069632 that's a high dream for many people at the moment. Say so.. though.. they.. you must know,
+pR87HZRPhGU-00108-00069736-00070520 most people on the planet have war or any kind of conflict, or any kind of like, poking motion,
+pR87HZRPhGU-00109-00070520-00071184 any kind of incendiary poking motion towards another country or any decisions made to.. to
+pR87HZRPhGU-00110-00071240-00071848 raise hackles or you know, to.. to have people on the on their guard or on their defensive,
+pR87HZRPhGU-00111-00071848-00072472 in any other country, know that this is what most people do not want to even start. They don't want
+pR87HZRPhGU-00112-00072472-00072928 to go there; they don't even want to entertain the thought because they're just trying as hard
+pR87HZRPhGU-00113-00072928-00073480 as they can to hold their own lives together. Why would they want to start to dig at another
+pR87HZRPhGU-00114-00073480-00074088 country? Why would they want to start to even cause problems with any other country? They really
+pR87HZRPhGU-00115-00074088-00074696 really don't - they're just trying to survive. So if you are reading something that's getting your..
+pR87HZRPhGU-00116-00074696-00075216 your heat up, that's getting your emotions up about the people in another country,
+pR87HZRPhGU-00117-00075216-00075904 read very very carefully, and if they say the name of the whole country in the headline * immediate
+pR87HZRPhGU-00118-00075904-00076608 red flag *, it's a lie. Yes, it's true for one small group, but that is probably an under 10%
+pR87HZRPhGU-00119-00076608-00077184 of the country that's deciding that for the whole country, and this is true for almost all of them.
+pR87HZRPhGU-00120-00077504-00078368 So uh.. yeah. Look before you leap. Don't hate too much. Take a second to think. Take a few seconds,
+pR87HZRPhGU-00121-00078368-00078984 take a few minutes, take an hour to think, "Okay, who is this good for? What am I meant
+pR87HZRPhGU-00122-00078984-00079440 to be feeling here?" Let the.. take note of your emotions when you're reading a news article.
+pR87HZRPhGU-00123-00079504-00079912 What's it stirring up? So I'm stirring some intense sense of fear or..
+pR87HZRPhGU-00124-00079912-00080448 What's it gearing you towards? What should.. what is it pushing you towards? What kind of actions do
+pR87HZRPhGU-00125-00080448-00081136 they want you to take or acquiesce to if you're in line with that statement they're making?
+pR87HZRPhGU-00126-00081312-00081920 Now watch your back with this because the last thing any of us want is to be in a war or to be in
+pR87HZRPhGU-00127-00081920-00082472 a conflict or be on one side where we're attacking a whole bunch of people very similar to us
+pR87HZRPhGU-00128-00082472-00083248 with very very normal dreams for just a regular life - that would be.. to me, that would be the
+pR87HZRPhGU-00129-00083248-00083712 worst thing. For me to realise that I'm in a battle or a conflict with other people that
+pR87HZRPhGU-00130-00083712-00084232 turn out to be just like me, and they are as clueless as I am about what's really going on.
+pR87HZRPhGU-00131-00084232-00084928 We live in an era of ignorance, of perpetuated ignorance. It is ignorance on purpose.
+pR87HZRPhGU-00132-00085080-00085632 It's a kind of dark age - we live in a dark age. This is a dark age. Do you think.. I mean yes,
+pR87HZRPhGU-00133-00085632-00086192 we have smartphones. Yes, we live in towers, yes, la la la whatever, uh.. freeways,
+pR87HZRPhGU-00134-00086192-00086896 oh yeah good. "This is sane modern life" - no, we live in a dark age. It's another dark age so..
+pR87HZRPhGU-00135-00086896-00088016 so just know this before you.. you judge a whole group of people for the decisions of a very few.
+pSAQ9fbqn2E-00000-00000495-00000835 So let's now talk a little bit about classification of dolomite!
+pSAQ9fbqn2E-00001-00000835-00001331 Just like we have classification of limestone,  we also have a classification system for dolomite
+pSAQ9fbqn2E-00002-00001331-00002063 which is based on two things: first on looking at the plane of the crystal of dolomite
+pSAQ9fbqn2E-00003-00002063-00002966 either they can be straight: then this is called a planar dolomite, so these examples here on the top,
+pSAQ9fbqn2E-00004-00002966-00003468 or they can be nonplanar and that's the example at the bottom.
+pSAQ9fbqn2E-00005-00003468-00004078 Then we have several cases that can happen. Let's review first the planar dolomites.
+pSAQ9fbqn2E-00006-00004078-00005185 So a planar dolomite can be euhedral, or known as planar-e if the minerals are clearly expressed, they're euhedral minerals.
+pSAQ9fbqn2E-00007-00005185-00005848 So you can clearly see those minerals planes,  you can clearly see the minerals shape and their habits.
+pSAQ9fbqn2E-00008-00005890-00006688 You can have a subhedral dolomite where you know,  it's still a planar dolomite, it's the 'planar-s' for subhedral.
+pSAQ9fbqn2E-00009-00006688-00007636 Because you can still recognize those minerals very well but in some cases you have more rounding and less well expressed mineralogic form.
+pSAQ9fbqn2E-00010-00007636-00008488 Then you can have cement of dolomite. So you can have evidence that the dolomite has been precipitated
+pSAQ9fbqn2E-00011-00008488-00009090 as a planar dolomite as a cement  in a cavity or vug or something like this,
+pSAQ9fbqn2E-00012-00009090-00009455 and this is known as a planar-c dolomite.
+pSAQ9fbqn2E-00013-00009501-00010488 You can also have a massive limestone with only a few  well expressed minerals of dolomites popping in within the structure.
+pSAQ9fbqn2E-00014-00010488-00011151 So that's like a porphyr and so it's known  as a porphyrotopic dolomite or planar-p dolomite.
+pSAQ9fbqn2E-00015-00011238-00011448 Now let's look at the non-planar dolomites!
+pSAQ9fbqn2E-00016-00011448-00011995 So you can have a case where you essentially have a subhedral to nonplanar dolomite.
+pSAQ9fbqn2E-00017-00011995-00012635 So that would be known as a planar-s to nonplanar-a  or a transitional dolomite.
+pSAQ9fbqn2E-00018-00012700-00013391 Then you can have a true anhedral dolomites. Where the minerals are there, you can see the grain of dolomites,
+pSAQ9fbqn2E-00019-00013391-00014213 but the are all rounded and you don't really see the mineral planes. So that's a nonplanar-a for anhedral dolomite.
+pSAQ9fbqn2E-00020-00014213-00014700 There's a special case of nonplanar dolomite,  known as saddle dolomite.
+pSAQ9fbqn2E-00021-00014700-00015190 Saddle dolomite is relatively large,  you know a few centimetres, millimetre at least.
+pSAQ9fbqn2E-00022-00015190-00015856 So this is a large crystal of dolomite, and it's actually a very interesting dolomite to recognize
+pSAQ9fbqn2E-00023-00015856-00016288 because it only forms at high temperature, so above 60˚C.
+pSAQ9fbqn2E-00024-00016288-00016950 So it's always associated with some form of hot fluid: maybe hydrothermal fluid or subsurface condition,
+pSAQ9fbqn2E-00025-00016950-00017365 it is not an early dolomite. So a very interesting one to recognize.
+pSAQ9fbqn2E-00026-00017365-00018083 And you can also have a nonplanar-p dolomite, so a porphyrotopic dolomite that is very similar to planar-p,
+pSAQ9fbqn2E-00027-00018083-00018793 the difference of course, that the minerals themselves  don't have those nice planar surfaces but they're subrounded.
+pSAQ9fbqn2E-00028-00018793-00019000 So let's look at a few examples!
+pSAQ9fbqn2E-00029-00019000-00019536 We'll start with the transitional dolomite, this is a nice example of the transitional dolomite
+pSAQ9fbqn2E-00030-00019536-00020553 you can recognize some mineral planes, and some area that are sort of subhedral but there's also a lot of areas that are anhedral.
+pSAQ9fbqn2E-00031-00020553-00021268 So where you really cannot see any plane of minerals. So that's a transitional dolomite.
+pSAQ9fbqn2E-00032-00021268-00021953 That's a good example here of a porphyrotopic dolomite.  Notice the rhombic shape of the dolomite
+pSAQ9fbqn2E-00033-00021953-00022526 this is really typical for the dolomite, this rhom shapes. You don't always find it so nicely expressed.
+pSAQ9fbqn2E-00034-00022526-00023301 But this is an example where it is nicely expressed  and it's a porphyrotopic dolomite, planar-p dolomite
+pSAQ9fbqn2E-00035-00023301-00024191 because those minerals are basically expressed in the matrix of limestone. So they just popped up as a porphyr in that matrix.
+pSAQ9fbqn2E-00036-00024191-00024835 And now let's look at some nice planar dolomite. And you can see the difference between this picture
+pSAQ9fbqn2E-00037-00024835-00025353 of planar dolomite, all the planes of the dolomites are well expressed.
+pSAQ9fbqn2E-00038-00025353-00025988 And compared this to the previous picture I've showed you,  where the planes of the dolomites and the edges of the minerals
+pSAQ9fbqn2E-00039-00025988-00026228 Are not so easy to spot.
+pSAQ9fbqn2E-00040-00026301-00027208 And finally, let's look at the nonplanar saddle dolomite! And you can recognize here the saddle dolomite because of its saddle shape
+pSAQ9fbqn2E-00041-00027208-00028306 this slightly bend shape, and look at the scale this is 400μ here, so this is a relatively large dolomite mineral.
+pSAQ9fbqn2E-00042-00028306-00028506 And it is again typical for warm diagenetic condition.
+pTBPXZ-Vgv0-00000-00000211-00000614 - --how to view quizzes in the Canvas student app on an Apple
+pTBPXZ-Vgv0-00001-00000614-00000791 device.
+pTBPXZ-Vgv0-00002-00000791-00001022 First step is from your dashboard.
+pTBPXZ-Vgv0-00003-00001022-00001319 You'll tap and select on the course that you wish to view.
+pTBPXZ-Vgv0-00004-00001586-00002179 You'll then tap on the Quizzes option on the left-hand side.
+pTBPXZ-Vgv0-00005-00002179-00002421 You'll then tap on the name of the specific quiz
+pTBPXZ-Vgv0-00006-00002421-00002564 that you want to view and take.
+pTBPXZ-Vgv0-00007-00002850-00003124 To take the quiz, you'll tap on the Take Quiz option
+pTBPXZ-Vgv0-00008-00003124-00003304 at the bottom.
+pTBPXZ-Vgv0-00009-00003304-00003582 And that's how you view quizzes in the Canvas student
+pTBPXZ-Vgv0-00010-00003582-00003829 app on an Apple device.
+pXtjSBgZ6lc-00000-00000000-00000454 Harry asks, "Who are the maintainers of Bitcoin Core?"
+pXtjSBgZ6lc-00001-00000454-00001016 "Is it possible to know what privileges they each have at any given time?"
+pXtjSBgZ6lc-00002-00001016-00001566 First of all, let's disambiguate the term 'maintainer' a bit.
+pXtjSBgZ6lc-00003-00001566-00002140 There is a difference between contributors, committers, and maintainers in an open-source project,
+pXtjSBgZ6lc-00004-00002140-00002610 [partly] hosted on GitHub in the style that the Bitcoin Core project is.
+pXtjSBgZ6lc-00005-00002610-00004216 Let's de-obfuscate those [roles]. Contributors are people who make pull-requests and actively contribute code.
+pXtjSBgZ6lc-00006-00004216-00004570 All changes to Bitcoin Core happen through pull-requests.
+pXtjSBgZ6lc-00007-00004570-00005154 Nobody commits [new changes] directly to a [release] branch without a full review of their pull request.
+pXtjSBgZ6lc-00008-00005154-00005608 All changes go through a pull-request process.
+pXtjSBgZ6lc-00009-00005608-00006470 Maintainers are people responsible for gathering all of the change request that have been approved...
+pXtjSBgZ6lc-00010-00006470-00007018 by the broad social consensus of the development community.
+pXtjSBgZ6lc-00011-00007018-00007552 Changes must have broad consensus and  acceptability to be included into the next release.
+pXtjSBgZ6lc-00012-00007552-00008302 [after discussion about] how to sequence them, when to package them, and what each release should include.
+pXtjSBgZ6lc-00013-00008302-00009234 Even [those] decisions are heavily influenced by the collaborative communication between contributors,
+pXtjSBgZ6lc-00014-00009234-00009994 who are commenting on the pull-requests and [fostering] open discussions about [development timelines].
+pXtjSBgZ6lc-00015-00009994-00010590 A maintainer's job is to schedule the releases and project manage them in a timetable,
+pXtjSBgZ6lc-00016-00010590-00011472 ensuring that the right changes are included, putting them through quality assurance testing,
+pXtjSBgZ6lc-00017-00011472-00012148 and then finally producing a release as a bundle of changes with a specific version number.
+pXtjSBgZ6lc-00018-00012148-00012800 That [comes with] release notes about what changed, who changed it, who contributed, what effects it has,
+pXtjSBgZ6lc-00019-00012800-00013426 why the changes were [made], and any upgrade or migration steps that are necessary.
+pXtjSBgZ6lc-00020-00013426-00013654 Maintainers collect all of that documentation.
+pXtjSBgZ6lc-00021-00013654-00014126 Today, the [lead] maintainer for Bitcoin Core is Wladimir van der Laan.
+pXtjSBgZ6lc-00022-00014126-00015108 They have another maintainer who is now working to assist Wladimir, mostly as the wallet maintainer.
+pXtjSBgZ6lc-00023-00015108-00015680 [His name is Samuel Dobson, @meshcollider]. He was recently given that responsibility.
+pXtjSBgZ6lc-00024-00015680-00016332 But the [lead] maintainer is [still] Wladimir van der Laan, @orionwl.
+pXtjSBgZ6lc-00025-00016332-00017150 The number of contributors is anywhere from thirty to forty people who contribute regularly...
+pXtjSBgZ6lc-00026-00017150-00018044 to almost every release, to more than five hundred [people] who have contributed to different extents.
+pXtjSBgZ6lc-00027-00018044-00018434 People come and go; this is a vast, sprawling project.
+pXtjSBgZ6lc-00028-00018434-00019026 There is another question, which is: who has commit access to the repository?
+pXtjSBgZ6lc-00029-00019026-00020098 That [question] comes up all the time, but it is not really a relevant question.
+pXtjSBgZ6lc-00030-00020098-00020850 The people who have commit access to the repository don't use that commit access.
+pXtjSBgZ6lc-00031-00020850-00021160 If they do have it, they don't use it.
+pXtjSBgZ6lc-00032-00021160-00021848 All commits [are done] by the project  maintainers, [who manage releases].
+pXtjSBgZ6lc-00033-00021848-00022328 It is a very different perspective [from most projects].
+pXtjSBgZ6lc-00034-00022328-00023002 Who can sign with a trusted PGP key? Who can commit to the repository?
+pXtjSBgZ6lc-00035-00023002-00024166 Who can register those commits in the repository? Who decides what changes happen for each release?
+pXtjSBgZ6lc-00036-00024166-00024752 [Essentially], who decides versus who contributes the changes are all different roles.
+pXtjSBgZ6lc-00037-00024752-00025376 At the moment, there are specifically five keys. I am looking at the documentation here.
+pXtjSBgZ6lc-00038-00025376-00025956 [The keys] belong to van der Laan, Pieter Wuille, Jonas Schnelli, Marco Falke, and Dobson.
+pXtjSBgZ6lc-00039-00025956-00026582 These are the five people who have PGP keys that can sign commits into the repository,
+pXtjSBgZ6lc-00040-00026582-00027420 but they are not necessarily the people who are making the commits.
+pXtjSBgZ6lc-00041-00027420-00028062 The release scheduling and committing of  changes [is done] by Wladimir van der Laan,
+pXtjSBgZ6lc-00042-00028062-00028418 who doesn't make decisions about what changes go in.
+pXtjSBgZ6lc-00043-00028418-00029170 Wladimir simply reflects, and does administrative work for, decisions which have already been made...
+pXtjSBgZ6lc-00044-00029170-00029344 by the community.
+pXtjSBgZ6lc-00045-00029344-00029939 In fact, he exercises very little authority over what goes into a [Bitcoin Core] release.
+pXtjSBgZ6lc-00046-00029939-00030439 He reflects decisions that have already been made in a very collaborative and open manner,
+pXtjSBgZ6lc-00047-00030439-00031392 through discussions on the pull-requests and project notes [around] the code.
+pXtjSBgZ6lc-00048-00031392-00031868 So, it is very different than what you see at the surface.
+pXtjSBgZ6lc-00049-00031868-00033094 By the way, there is a really nice article by Jameson Lopp about the maintenance of Bitcoin Core: how it works,
+pXtjSBgZ6lc-00050-00033094-00033628 who has commit access, who has privileges, etc.
+pXtjSBgZ6lc-00051-00033628-00034288 It is a Medium article called "Who Controls Bitcoin Core?" by Jameson Lopp, if you want to read more.
+pXtjSBgZ6lc-00052-00034288-00034510 That is just one of the sources you might use.
+pXtjSBgZ6lc-00053-00034510-00035114 [AUDIENCE] You told us that Bitcoin is [becoming] stronger when it is attacked. [ANDREAS] Yes.
+pXtjSBgZ6lc-00054-00035116-00036194 [AUDIENCE] Bitcoin has antibodies that make it stronger. Even the hardest threats seem to make it grow better.
+pXtjSBgZ6lc-00055-00036194-00036696 Do you think there are threats that are maybe not sophisticated, not so scary,
+pXtjSBgZ6lc-00056-00036696-00037130 but for which Bitcoin is less prepared to react?
+pXtjSBgZ6lc-00057-00037130-00038230 Something that can be technical or social, something subtle or unexpected, toxic and unhealthy for Bitcoin?
+pXtjSBgZ6lc-00058-00038230-00038696 [ANDREAS] Yeah, I do think those [threats] exist.
+pXtjSBgZ6lc-00059-00038696-00039542 For the most part, we have seen a number of social attacks against Bitcoin, against the community.
+pXtjSBgZ6lc-00060-00039542-00040330 Attempts to create unnecessary drama  that distracts people from the real focus.
+pXtjSBgZ6lc-00061-00040330-00041120 What has become clear is, one of the ways to slow down [the development of] a cryptocurrency...
+pXtjSBgZ6lc-00062-00041120-00041388 is to create drama inside the communities.
+pXtjSBgZ6lc-00063-00041388-00041678 That has been attempted.
+pXtjSBgZ6lc-00064-00041678-00042320 I would say, in some ways, it was successful. It did not stop us, but it has slowed us down.
+pXtjSBgZ6lc-00065-00042320-00042648 I try not to feed the drama.
+pXtjSBgZ6lc-00066-00042648-00043246 I think it is important to remain focused on the technology and stop paying attention to the people.
+pXtjSBgZ6lc-00067-00043246-00044106 The more time you spend talking about people... That is gossip. It is not [technical work]. So don't.
+pXtjSBgZ6lc-00068-00044106-00044870 Ignore the personalities. Focus on the technology and what is actually happening in the space.
+pXtjSBgZ6lc-00069-00044870-00045362 [The gossip] is not real, it is fake drama.
+pXtjSBgZ6lc-00070-00045362-00046252 One of the ways in which Bitcoin has changed is... It is not just Bitcoin anymore.
+pXtjSBgZ6lc-00071-00046252-00046834 Even if an attack was successful against  one cryptocurrency, that will [lead to...
+pXtjSBgZ6lc-00072-00046834-00047090 the creation of] other cryptocurrencies.
+pXtjSBgZ6lc-00073-00047090-00047612 All of these cryptocurrencies are trying different approaches.
+pXtjSBgZ6lc-00074-00047612-00048440 Communities focus on different applications, monetary policies, and security principles for [their] codebase.
+pXtjSBgZ6lc-00075-00048440-00049470 That diversity, like biological diversity, means we are much more resistant to any single type of attack.
+pXtjSBgZ6lc-00076-00049470-00050200 Even if [some malicious actors] manage to stop one, they [can't] stop everything.
+pXtjSBgZ6lc-00077-00050200-00051050 The ones that [can't be] stopped expand to fill the [hole] in the ecosystem and [become] stronger.
+pXtjSBgZ6lc-00078-00051050-00051750 Now we have this adaptation to threats. It is not just Bitcoin that is anti-fragile.
+pXtjSBgZ6lc-00079-00051750-00052686 It is the entire crypto ecosystem, with all of its competing interests, developers, and models.
+pXtjSBgZ6lc-00080-00052686-00053417 It is advancing as a whole and even more unstoppable than Bitcoin itself.
+pXtjSBgZ6lc-00081-00053417-00054141 I find it harder and harder to come up with realistic scenarios of how [Bitcoin could be] stopped. I really do.
+pXtjSBgZ6lc-00082-00054141-00054546 [AUDIENCE] Hello, Andreas. Thank you for being such a great teacher. [ANDREAS] Thank you.
+pXtjSBgZ6lc-00083-00054546-00055379 [AUDIENCE] I have a question about governments, governance, and decisions in this system.
+pXtjSBgZ6lc-00084-00055379-00056144 How do we make decisions in a decentralized protocol?
+pXtjSBgZ6lc-00085-00056144-00056364 [ANDREAS] That is a great question.
+pXtjSBgZ6lc-00086-00056364-00056846 Government, governance, and how do we make decisions in these decentralized systems?
+pXtjSBgZ6lc-00087-00056846-00057274 We don't know yet. This is the beauty of it.
+pXtjSBgZ6lc-00088-00057274-00057946 We do know that we have created a system of rules without rulers, a system that is extremely resistant...
+pXtjSBgZ6lc-00089-00057946-00058391 to external change and very difficult to hijack.
+pXtjSBgZ6lc-00090-00058391-00058728 That is the starting point. Where can that system take us?
+pXtjSBgZ6lc-00091-00058728-00059008 How do we change it when we want to change it?
+pXtjSBgZ6lc-00092-00059008-00059514 How do we achieve consensus? How do we arrive at common decisions?
+pXtjSBgZ6lc-00093-00059514-00060382 We don't know yet, but we are trying a dozen different ways; some of them are working, some of them not.
+pXtjSBgZ6lc-00094-00060382-00061086 That is the beauty of this. We have a new field of computer science [around] consensus algorithms...
+pXtjSBgZ6lc-00095-00061086-00061394 that did not exist ten years ago.
+pXtjSBgZ6lc-00096-00061394-00061986 That field is inventing incredible things every single day.
+pXtjSBgZ6lc-00097-00061990-00063444 The best part of it, what most people haven't quite grasped, is that we are working on a deployed system,
+pXtjSBgZ6lc-00098-00063444-00064472 a $150 billion live payment system, that is [one of the] largest civilian deployments of applied cryptography...
+pXtjSBgZ6lc-00099-00064472-00064800 on this planet.
+pXtjSBgZ6lc-00100-00064800-00065329 When I was a young cypherpunk in 1991, I was enamored with Pretty Good Privacy (PGP)...
+pXtjSBgZ6lc-00101-00065329-00065908 and wanted to teach everybody how to  use PGP to encrypt their emails (I failed).
+pXtjSBgZ6lc-00102-00065908-00066244 [Laughter] Nobody was paying any attention.
+pXtjSBgZ6lc-00103-00066244-00067894 If you told me that in two or three decades (god, I'm old), we would have a practical, usable, deployed network...
+pXtjSBgZ6lc-00104-00067894-00068386 of digital signatures and using some of the most amazing applied cryptography,
+pXtjSBgZ6lc-00105-00068386-00068656 with new technology that was never seen before...
+pXtjSBgZ6lc-00106-00068656-00069390 A real production network that is growing  and has involved [tens of millions] of people?
+pXtjSBgZ6lc-00107-00069390-00069984 We don't even know. I wouldn't have believed you then. That is the platform on which we're trying this stuff.
+pXtjSBgZ6lc-00108-00069984-00070206 There is no [established] theory about this.
+pXtjSBgZ6lc-00109-00070206-00070892 You can develop your lovely pristine idea in the lab and gently put it out [into the wild internet],
+pXtjSBgZ6lc-00110-00070892-00071796 and it [could be] ravaged by hackers in milliseconds, because it [can't] survive under adversarial conditions.
+pXtjSBgZ6lc-00111-00071796-00072402 The honeypot is $150 billion dollars, the prize that awaits anyone [who can break it].
+pXtjSBgZ6lc-00112-00072402-00073126 Bitcoin isn't here today because no one tried to hack it. Bitcoin is here today because everyone tried.
+pXtjSBgZ6lc-00113-00073126-00074202 Again and again, each time they somewhat succeed, it adapts, learns, evolves, and becomes stronger.
+pXtjSBgZ6lc-00114-00074202-00075380 Like an immune system. Governance played out on an evolutionary-like playing field with real money at stake.
+pXtjSBgZ6lc-00115-00075380-00075872 We have a chance, for the first time, to do it differently.
+pXtjSBgZ6lc-00116-00075872-00076284 People ask me, "Will Bitcoin be regulated?"
+pXtjSBgZ6lc-00117-00076284-00076956 Bitcoin is regulated; it is regulated by  mathematics, instead of human committees.
+pXtjSBgZ6lc-00118-00076956-00077516 We already have systems regulated by committee. We are trying to do [something different] now.
+pXtjSBgZ6lc-00119-00077516-00078232 That is, regulating by algorithm. We think it will be better, but we don't know yet.
+pYu4YrSWQFE-00000-00000052-00000152 Hi, everyone!
+pYu4YrSWQFE-00001-00000152-00000258 I'm Amanda Howell.
+pYu4YrSWQFE-00002-00000258-00000720 I'm the Business Librarian here at UW-Whitewater, and today I'm going to talk to you about citing
+pYu4YrSWQFE-00003-00000720-00000884 using APA style.
+pYu4YrSWQFE-00004-00000884-00001519 We're going to focus on citing books, citing articles, and citing websites, 3 sources you're
+pYu4YrSWQFE-00005-00001519-00002154 most likely to use for your research during your time at UW-Whitewater.
+pYu4YrSWQFE-00006-00002154-00002511 It's important that you give credit to other for the work that they've done so that you
+pYu4YrSWQFE-00007-00002511-00002695 can avoid plagiarism.
+pYu4YrSWQFE-00008-00002695-00003315 Not doing so, intentionally or not, is misrepresenting your own work and may result in serious academic
+pYu4YrSWQFE-00009-00003315-00003669 and legal ramifications.
+pYu4YrSWQFE-00010-00003669-00004218 When citing a book you need to find the following information: the name of the author, the date
+pYu4YrSWQFE-00011-00004218-00004789 of publication, the title of the book, the publisher, and the publishing city.
+pYu4YrSWQFE-00012-00004789-00005115 One thing to note is the capitalization on the title of the book.
+pYu4YrSWQFE-00013-00005115-00005602 You're only going to capitalize the first word of the title and the first word of the
+pYu4YrSWQFE-00014-00005602-00005893 subtitle, as well as any proper nouns.
+pYu4YrSWQFE-00015-00005893-00006296 Most of this information can be found on the title page.
+pYu4YrSWQFE-00016-00006296-00006640 Let's take a look at the title page of a book by Scott Goodson.
+pYu4YrSWQFE-00017-00006640-00007026 The book is called Uprising: How to Build a Brand - and Change the World - by Sparking
+pYu4YrSWQFE-00018-00007026-00007200 Cultural Movements.
+pYu4YrSWQFE-00019-00007200-00007573 Just under that is the name of the author; sometimes you'll have more than one.
+pYu4YrSWQFE-00020-00007573-00007946 And then at the bottom of the page you'll usually find the publisher, in this case McGraw
+pYu4YrSWQFE-00021-00007946-00008046 Hill.
+pYu4YrSWQFE-00022-00008046-00008310 We then see a lot of different publishing cities.
+pYu4YrSWQFE-00023-00008310-00008679 Generally if you see more than one, you're just going to take the first one.
+pYu4YrSWQFE-00024-00008679-00009065 The only element we're not finding on this page is the date of publication.
+pYu4YrSWQFE-00025-00009065-00009699 So what we can do is go to the Copyright page and see that it was published in 2012.
+pYu4YrSWQFE-00026-00009699-00010113 Now that we found all the information we need we can create our citation.
+pYu4YrSWQFE-00027-00010113-00010567 Note that when using APA style you only include the author's initials after you've typed out
+pYu4YrSWQFE-00028-00010567-00010667 their last name.
+pYu4YrSWQFE-00029-00010667-00011195 Also, make note of the fact that I only capitalized the first word of the title and the first
+pYu4YrSWQFE-00030-00011195-00011454 word of the subtitle.
+pYu4YrSWQFE-00031-00011454-00011888 Citing articles is similar to citing books except you need a little bit more information.
+pYu4YrSWQFE-00032-00011888-00012360 In addition to the author and the date of publication you have to get the article title,
+pYu4YrSWQFE-00033-00012360-00012988 the journal title, the volume and issue number, if applicable, the page range of the article,
+pYu4YrSWQFE-00034-00012988-00013575 and either the digital object identifier, known as the DOI, or the journal's homepage.
+pYu4YrSWQFE-00035-00013575-00013902 This is not the same as the URL where you found the article.
+pYu4YrSWQFE-00036-00013902-00014297 It is actually the website for the journal that published the article.
+pYu4YrSWQFE-00037-00014297-00014797 Note that the article title is exactly like the book title where we only capitalized the
+pYu4YrSWQFE-00038-00014797-00015072 first word of the title and the first word of the subtitle.
+pYu4YrSWQFE-00039-00015072-00015459 However the entire journal title is capitalized.
+pYu4YrSWQFE-00040-00015459-00015756 Let's take a look at an example of an article.
+pYu4YrSWQFE-00041-00015756-00016144 We can see that for this particular article the journal title is actually listed in two
+pYu4YrSWQFE-00042-00016144-00016693 places: once at the very top of the page as well as once just below that.
+pYu4YrSWQFE-00043-00016693-00017198 At the very top of the page we also see the volume, the date of publication, as well as
+pYu4YrSWQFE-00044-00017198-00017479 the page range of the article.
+pYu4YrSWQFE-00045-00017479-00017985 Underneath the second instance of the journal title we have the article title, and underneath
+pYu4YrSWQFE-00046-00017985-00018455 that we have the names of both the authors.
+pYu4YrSWQFE-00047-00018455-00018935 The last thing you'll want to look for is the DOI, which will be at either the top or
+pYu4YrSWQFE-00048-00018935-00019191 the bottom of the title page.
+pYu4YrSWQFE-00049-00019191-00019690 This is preferred to the journal's homepage so you'll want to make note of this if it
+pYu4YrSWQFE-00050-00019690-00019790 exists.
+pYu4YrSWQFE-00051-00019790-00020264 I was able to find mine at the bottom of the page.
+pYu4YrSWQFE-00052-00020264-00020691 Now that we've gathered all of this information, we can create our citation.
+pYu4YrSWQFE-00053-00020691-00021075 You'll notice that there is an ampersand between my two authors.
+pYu4YrSWQFE-00054-00021075-00021495 You'll want to use that rather than typing out the word "and" if you have multiple authors
+pYu4YrSWQFE-00055-00021495-00021765 for a specific article or book.
+pYu4YrSWQFE-00056-00021765-00022178 You'll also notice that I didn't have an issue number for this particular article so I just
+pYu4YrSWQFE-00057-00022178-00022635 skipped over that particular portion.
+pYu4YrSWQFE-00058-00022635-00023021 The last citation example we're going to look at is for a website.
+pYu4YrSWQFE-00059-00023021-00023465 This is probably the hardest type of citation to create because a lot of the information
+pYu4YrSWQFE-00060-00023465-00023781 can be hard to find, if it exists at all.
+pYu4YrSWQFE-00061-00023781-00024294 It's important to keep in mind that if you can't find this information it's hard to validate
+pYu4YrSWQFE-00062-00024294-00024643 that you are in fact using a reliable source.
+pYu4YrSWQFE-00063-00024643-00025163 What you want to look for is the name of the author or authors, the date that the information
+pYu4YrSWQFE-00064-00025163-00025710 on the page was published, the title of the page you were actually on (unless you're citing
+pYu4YrSWQFE-00065-00025710-00026224 the entire website), and finally the link for the page.
+pYu4YrSWQFE-00066-00026224-00026700 If we take a look at this example, we can see that the title of the page is near the
+pYu4YrSWQFE-00067-00026700-00026800 top.
+pYu4YrSWQFE-00068-00026800-00026985 It is Who We Are.
+pYu4YrSWQFE-00069-00026985-00027358 Usually the author and the date of publication would be found just below the name of the
+pYu4YrSWQFE-00070-00027358-00027469 page.
+pYu4YrSWQFE-00071-00027469-00027831 We don't see either of those under the title of this page.
+pYu4YrSWQFE-00072-00027831-00028069 In this case, however, we do have a corporate author.
+pYu4YrSWQFE-00073-00028069-00028430 It's going to be the Direct Marketing Association.
+pYu4YrSWQFE-00074-00028430-00028906 When you're looking at a website for an association or a government entity, like the United States
+pYu4YrSWQFE-00075-00028906-00029427 Census Bureau, that's going to be your author unless you can find the name of an individual.
+pYu4YrSWQFE-00076-00029427-00029987 As I mentioned earlier, there is no date of publication for this particular page so we'll
+pYu4YrSWQFE-00077-00029987-00030410 have to make a modification when we actually create our citation.
+pYu4YrSWQFE-00078-00030410-00030644 Now we can create our citation.
+pYu4YrSWQFE-00079-00030644-00030980 If you have a corporate author, you'll want to capitalize every word of their name and
+pYu4YrSWQFE-00080-00030980-00031355 avoid using the short form, in this case DMA.
+pYu4YrSWQFE-00081-00031355-00031805 If you don't have an author, you're going to put the name of the webpage first, then
+pYu4YrSWQFE-00082-00031805-00031997 the date, and then the link.
+pYu4YrSWQFE-00083-00031997-00032553 You'll notice that because I didn't have a date for this particular page I just put n.d.
+pYu4YrSWQFE-00084-00032553-00032834 for "no date."
+pYu4YrSWQFE-00085-00032834-00033183 Once you have all of your citations you can create your Reference list, which will go
+pYu4YrSWQFE-00086-00033183-00033291 at the end of your paper.
+pYu4YrSWQFE-00087-00033291-00033713 You'll want to make sure that your references are in alphabetical order and that they each
+pYu4YrSWQFE-00088-00033713-00034166 have a hanging indent, so that the first line of each citation is longer than the subsequent
+pYu4YrSWQFE-00089-00034166-00034266 lines.
+pYu4YrSWQFE-00090-00034266-00034537 Your reference list should also be double spaced.
+pYu4YrSWQFE-00091-00034537-00034958 The top of your page should say References.
+pYu4YrSWQFE-00092-00034958-00035315 In addition to citing your sources at the end of your paper, you'll also need to cite
+pYu4YrSWQFE-00093-00035315-00035499 them throughout your paper.
+pYu4YrSWQFE-00094-00035499-00035934 In order to do this you need the author name, the date, and the page number for the source
+pYu4YrSWQFE-00095-00035934-00036083 you're citing.
+pYu4YrSWQFE-00096-00036083-00036391 Notice that before the page number I just have the letter p.
+pYu4YrSWQFE-00097-00036391-00036941 For the Direct Marketing Association I added DMA in brackets so that each subsequent time
+pYu4YrSWQFE-00098-00036941-00037453 I cite this source I can just use DMA in my in-text citation.
+pYu4YrSWQFE-00099-00037453-00038255 For documents without pages, you can indicate paragraph number by using p-a-r-a period.
+pYu4YrSWQFE-00100-00038255-00038643 This page contains some other useful sources you might want to check out if you get stuck
+pYu4YrSWQFE-00101-00038643-00039043 as you're creating your citations.
+pYu4YrSWQFE-00102-00039043-00039472 Please feel free to contact me if you have any questions about citing sources using APA
+pYu4YrSWQFE-00103-00039472-00039522 style.
+parLlrmqHfg-00000-00000786-00001176 Improvising alone on a standard jazz tune
+parLlrmqHfg-00001-00001242-00002090 Welcome ! We'll speak about an important topic for jazz musicians :
+parLlrmqHfg-00002-00002146-00002734 improvising alone, thinking about the chord changes of a tune without any backing track
+parLlrmqHfg-00003-00002778-00003215 but with a metronome on 2 and 4 (or with our foot)
+parLlrmqHfg-00004-00003272-00003928 when we improvise we have to be aware of what's going on
+parLlrmqHfg-00005-00003964-00004794 we must hear the tempo, the chords, without being too much "the head in the guitar"
+parLlrmqHfg-00006-00004870-00005710 But ! That being said, it's important to anticipate what is going to happen next
+parLlrmqHfg-00007-00005764-00006744 and this is achievable with a good visualization and memorization of the chord changes
+parLlrmqHfg-00008-00006795-00007466 it will allow us to anticipate the next chords and adapt our improvisation
+parLlrmqHfg-00009-00007598-00008510 this is the topic of this video : I'll improvise on a tune without any backing track
+parLlrmqHfg-00010-00008650-00009228 I'll play on Autumn Leaves (in Gm)
+parLlrmqHfg-00011-00009244-00009706 here are the first notes and the first chord
+parLlrmqHfg-00012-00009726-00010590 and the first thing to do is to simply play the chords of the song, 1 for each bar or on every beat
+parLlrmqHfg-00013-00010590-00011300 with the thumb in order to have a sweet sound on an archtop guitar
+parLlrmqHfg-00014-00011332-00011818 and with a metronome on 2 and 4
+parLlrmqHfg-00015-00018326-00019174 you can play just thinking about the chords, but you can also sing or hum the tune
+parLlrmqHfg-00016-00019200-00020426 because singing a tune is THE BEST way to memorise how the chords fit the melody
+parLlrmqHfg-00017-00020448-00021204 and the most precise point of reference you can have is the head of the tune
+parLlrmqHfg-00018-00021244-00022598 ok, this was the first step. Now we know how many measures each chord last and where are the cadences are
+parLlrmqHfg-00019-00022678-00023514 now I'll improvise, with only a few notes : the chord tones
+parLlrmqHfg-00020-00023514-00024222 2 or 3 notes for each chord : the simplest, the better
+parLlrmqHfg-00021-00024222-00025092 It is to make sure I here the chord changes
+parLlrmqHfg-00022-00025134-00025968 I'll do it on the whole song, so it implies learning all the chord tones
+parLlrmqHfg-00023-00025968-00026482 and it's a big amount of work : this video is for advanced guitarists
+parLlrmqHfg-00024-00026514-00027374 (By the way there's a 1h video course on Autumn Leaves on my french website guitare-improvisation.com)
+parLlrmqHfg-00025-00027374-00027718 Let's go !
+parLlrmqHfg-00026-00027742-00028124 1, 2, 1 2 3 4
+parLlrmqHfg-00027-00034090-00034372 this was the 1st beat of the 1st measure
+parLlrmqHfg-00028-00034372-00034954 now, a few tips to help you do this exercise :
+parLlrmqHfg-00029-00034986-00035546 if the metronome on 2 and 4 is too difficult for you, put it on the 1 and 3
+parLlrmqHfg-00030-00035586-00036084 also, I sometimes quote the head
+parLlrmqHfg-00031-00036084-00036670 you have to learn to play the head of the tune !
+parLlrmqHfg-00032-00036670-00037438 as I said it helps a lot if I whant to know where I am in the changes
+parLlrmqHfg-00033-00037498-00038350 (and it's really nice to sometimes quote the head in a chorus)
+parLlrmqHfg-00034-00038640-00039652 now, third exercise : doing something a bit more subtle, playing more personnal licks
+parLlrmqHfg-00035-00039652-00040140 (even if that's kind of what I did in the last impro, but let's do it more clearly)
+parLlrmqHfg-00036-00040356-00041106 (without the chords on the screen !)
+parLlrmqHfg-00037-00046640-00047350 you may have noticed that I made a little mistake : the E7 lasted a little too long
+parLlrmqHfg-00038-00047376-00048720 so if you play the chords while I'm playing (I advise you to do so) you'll here it
+parLlrmqHfg-00039-00049180-00051000 now, let's do a fun excercise ! I improvise and you have to tell when I'm making a mistake in the changes
+parLlrmqHfg-00040-00051092-00052014 I may make a chord last too long, or not long enough for example
+parLlrmqHfg-00041-00052176-00053216 there will be 3 improvisations (the answers are in the free documents in the description of the video)
+parLlrmqHfg-00042-00053317-00053636 first "mistake exercise"
+parLlrmqHfg-00043-00055150-00056096 try to here my mistake without playing the chords with me. If it's too hard for you, play the chords
+parLlrmqHfg-00044-00056148-00056646 second "mistake exercise"
+parLlrmqHfg-00045-00058336-00058782 last "mistake exercise"
+parLlrmqHfg-00046-00060594-00061612 try to find where I made a mistake, IF I made a mistake : it's also possible that I didn't make any
+parLlrmqHfg-00047-00061612-00062038 answers are freely available (see the description of the video)
+parLlrmqHfg-00048-00062088-00063388 last exercise, for advanced musicians : try to anticipate or delay the chords
+parLlrmqHfg-00049-00063408-00063888 try rythmical surprises, insist on the 2 and 4 or on upbeats
+parLlrmqHfg-00050-00063888-00064429 to make it more subtle and less academic
+parLlrmqHfg-00051-00064460-00064932 let's go !
+parLlrmqHfg-00052-00071160-00072146 in this impro, play the chords with me : you'll see where I move the chords and where I alter the harmony
+parLlrmqHfg-00053-00072164-00072914 I hope this video will help you to make this exercise (maybe you discovered it)
+parLlrmqHfg-00054-00072940-00073894 and to understand will this exercise is so important for jazz musicians
+parLlrmqHfg-00055-00074190-00075292 on my youtube channel you'll find thousands of jazz backing tracks and tutorials
+pcWD_sfZRju-00000-00000106-00000236 WELCOME TO ALL HANDS UPDATE.
+pcWD_sfZRju-00001-00000236-00000437 HERE'S WHAT'S HAPPENING AROUND YOUR NAVY.
+pcWD_sfZRju-00002-00000477-00000760 THE 75TH COMMEMORATION OF THE JAPANESE ATTACK ON
+pcWD_sfZRju-00003-00000760-00001061 PEARL HARBOR AND OTHER US MILITARY FACILITIES ON
+pcWD_sfZRju-00004-00001061-00001378 O-WA-WHO IS AN OPPORTUNITY TO HONOR THE COURAGE,
+pcWD_sfZRju-00005-00001378-00001638 SERVICE AND SACRIFICE OF THE MILITARY PERSONNEL
+pcWD_sfZRju-00006-00001638-00001801 PRESENT DURING THE ATTACKS.
+pcWD_sfZRju-00007-00001871-00002072 THE THEME OF THIS YEAR'S COMMEMORATION,
+pcWD_sfZRju-00008-00002072-00002312 "HONORING THE PAST, INSPIRING THE FUTURE,
+pcWD_sfZRju-00009-00002312-00002525 " IS HIGHLIGHTED THROUGH EVENTS THIS WEEK THAT WILL
+pcWD_sfZRju-00010-00002525-00002679 CONTINUE THROUGH DECEMBER 11TH.
+pcWD_sfZRju-00011-00002786-00003016 TAKE A MOMENT OF SILENCE TODAY TO REMEMBER OUR
+pcWD_sfZRju-00012-00003016-00003340 SHIPMATES LOST ON "A DAY THAT WILL LIVE IN INFAMY."
+pcWD_sfZRju-00013-00003486-00003596 DWIGHT D.
+pcWD_sfZRju-00014-00003596-00003793 EISENHOWER CARRIER STRIKE GROUP ENTERED 6TH FLEET,
+pcWD_sfZRju-00015-00003793-00004007 CONTINUING SUPPORT OF OPERATION INHERENT
+pcWD_sfZRju-00016-00004007-00004104 RESOLVE.
+pcWD_sfZRju-00017-00004214-00004411 ELEMENTS OF THE TWENTY SECOND MARINE
+pcWD_sfZRju-00018-00004411-00004724 EXPEDITIONARY UNIT ABOARD USS WASP CONTINUED
+pcWD_sfZRju-00019-00004724-00004944 PRECISION AIR STRIKES AGAINST ISIL TARGETS IN
+pcWD_sfZRju-00020-00004944-00005291 LIBYA AS PART OF OPERATION "ODYSSEY LIGHTNING" .
+pcWD_sfZRju-00021-00005291-00005492 YOU CAN READ MORE ABOUT WHAT'S HAPPENING AROUND
+pcWD_sfZRju-00022-00005492-00005709 YOUR NAVY AT NAVY.MIL.
+pduRD2JW094-00000-00000712-00001344 Discover Das Es Salaam, 10 tourist attractions. Tanzania is filled with beautiful and amazing
+pduRD2JW094-00001-00001344-00001928 cities that are good tourist destinations. Dar-Es-salaam is a major city and commercial port
+pduRD2JW094-00002-00001928-00002376 on the coast of the Indian Ocean in Tanzania that developed from a fishing village.
+pduRD2JW094-00003-00002432-00003016 The word Das Es Salaam is developed from an Arabic word which means the abode of peace and it’s the
+pduRD2JW094-00004-00003016-00003576 busiest and largest city in this country. This city is a burgeoning port city and commercial
+pduRD2JW094-00005-00003576-00004152 centerpiece for much of Eastern and Central Africa. Dar es Salaam is a rich melting pot
+pduRD2JW094-00006-00004152-00004720 of ancient cultures and modern lifestyles. This city is home to Tanzania’s major port.
+pduRD2JW094-00007-00004776-00005216 Das es salam is filled with desirable location and natural harbor,
+pduRD2JW094-00008-00005216-00005824 coast line, natural reserves, botanic gardens stunning natural parks and much more. Here
+pduRD2JW094-00009-00005824-00006392 some interesting place to visit and things to know about Das es salaam. Please take a moment to like,
+pduRD2JW094-00010-00006392-00006840 share, comment and subscribe to Africa reload for more interesting updates.
+pduRD2JW094-00011-00007104-00007616 10. St Joseph’s Cathedral. This cathedral was built by German missionaries
+pduRD2JW094-00012-00007616-00008352 from 1897 through 1902. This Gothic-style Roman Catholic Church looks a little incongruous amid
+pduRD2JW094-00013-00008352-00008904 the steel and glass skyscrapers bristling along the harbor front. It is a cool and calming place
+pduRD2JW094-00014-00008904-00009544 to sit and reflect on a hot day. The Dar es Salaam archdiocese's seat, St. Joseph Cathedral,
+pduRD2JW094-00015-00009544-00010048 has several notable features that make it worthwhile to visit. Its shingled spire,
+pduRD2JW094-00016-00010048-00010552 vaulted interior, and exquisite stained-glass windows are the most remarkable features.
+pduRD2JW094-00017-00010608-00011088 Many of the original German inscriptions and artwork may be found throughout the cathedral,
+pduRD2JW094-00018-00011088-00011672 including a carved relief over the main altar. St. Joseph's Metropolitan Cathedral is located on
+pduRD2JW094-00019-00011672-00012312 Sokoine Drive, directly across from the Zanzibar ferry terminal and within the Dar es Salaam CBD.
+pduRD2JW094-00020-00012368-00012752 It's also close to the white father's home, Magogoni Creek,
+pduRD2JW094-00021-00012752-00013296 and BRT routes. Visit this church to marvel at the beauty of this gold-plated structure.
+pduRD2JW094-00022-00013552-00014184 9. National Museum & House of Culture. Tanzania's Museum and House of Culture is the
+pduRD2JW094-00023-00014184-00014784 first and largest of these institutions. It can be found on Dar Es Salaam's Shaban Robert Street.
+pduRD2JW094-00024-00014784-00015640 It was built between 1938 and 1939 and opened to the public in 1940. Since 1940, the National
+pduRD2JW094-00025-00015640-00016232 Museum of Tanzania, formally known as the King George V Museum of England, has inherited the
+pduRD2JW094-00026-00016232-00016768 collections and other resources of the National Museum of Tanzania, which is the country's largest
+pduRD2JW094-00027-00016768-00017376 and oldest museum. The National Museum and House of Culture hosts both indoor (permanent exhibition
+pduRD2JW094-00028-00017376-00018136 galleries) and outdoor (aquarium, trees, butterfly catch gardens, Hope out of Sorrow, Historical
+pduRD2JW094-00029-00018136-00018832 State automobiles, and so on) exhibitions. It now has four permanent exhibitions, comprising
+pduRD2JW094-00030-00018832-00019440 galleries dedicated to art, history, human evolution, and rock art. The permanent exhibition
+pduRD2JW094-00031-00019440-00020048 halls for Biology and Ethnography are now being built. In the same vein, the National Museum and
+pduRD2JW094-00032-00020048-00020672 House of Culture oversees six storage rooms for collections of art, archaeology, biology,
+pduRD2JW094-00033-00020672-00021296 ethnography, history, and paleontology, as well as a Strong Room for vital national treasures.
+pduRD2JW094-00034-00021560-00022088 8. Mwenge Woodcarvers Market. The Mwenge Woodcarvers Market,
+pduRD2JW094-00035-00022088-00022624 located in the heart of Dar es Salaam, is bustling with artists hard at work crafting
+pduRD2JW094-00036-00022624-00023256 beautiful and distinctive goods. Wooden masks are particularly popular, but artisan shops selling
+pduRD2JW094-00037-00023256-00023880 a wide range of items from textiles to sandals, beaded necklaces, and more will also have gems.
+pduRD2JW094-00038-00023936-00024456 Unlike numerous tourist marketplaces across the world, merchants in this market do not bother
+pduRD2JW094-00039-00024456-00024952 customers. They will encourage you to visit their store, but you will not be pressured to make a
+pduRD2JW094-00040-00024952-00025512 purchase. Remember to haggle and browse around in the market because prices are adjustable and
+pduRD2JW094-00041-00025512-00026144 you may be able to locate a similar item for half the price at another vendor. Mwenge Woodcarvers
+pduRD2JW094-00042-00026144-00026648 Market appears unassuming at first glance, but walk through the central opening and you'll
+pduRD2JW094-00043-00026648-00027183 find yourself in the midst of over 100 shops selling a diverse range of beautiful woodwork.
+pduRD2JW094-00044-00027424-00028048 7. Askari Monument. Dar es Salaam, Tanzania's largest commercial city,
+pduRD2JW094-00045-00028048-00028720 has a variety of German colonial memory sites, which include buildings, statues, and open spaces.
+pduRD2JW094-00046-00028720-00029320 The origins of the Askari monument, a life-size bronze construction of the King's African Rifleman
+pduRD2JW094-00047-00029320-00029848 near the intersection of Samora Avenue and Maktaba Street, can be traced back to German
+pduRD2JW094-00048-00029848-00030439 times. The monument was built in the 1920s to replace the Wissmann Monument, which was built
+pduRD2JW094-00049-00030439-00031024 in 1906 in honor of Major Hauptmann Hermann von Wissmann (1853-1905), a German commander
+pduRD2JW094-00050-00031024-00031664 (Reichskommissar) who suppressed African rebellion in East Africa. Coastal resistance, particularly
+pduRD2JW094-00051-00031664-00032280 that led by Swahili-Arab trader Abushiri ibn Salim al-Harthi (1889), provided a significant threat to
+pduRD2JW094-00052-00032280-00032928 the Deutsch-Ostafrikanische Gesellschaft (OAG). The Abushiri soldiers, for example, set fire
+pduRD2JW094-00053-00032928-00033784 to the Benedictine Mission Station at Pugu on January 13, 1889, killing three missionaries.
+pduRD2JW094-00054-00033904-00034376 6. Kunduchi Water Park. Wet 'n' Wild is a 6-acre
+pduRD2JW094-00055-00034376-00034968 entertainment and recreation facility (the largest in East and Central Africa) that offers a wide
+pduRD2JW094-00056-00034968-00035608 range of activities. Water slides (some as high as six stories), children's water play systems,
+pduRD2JW094-00057-00035608-00036176 sports grounds, restaurants, Go-karts and a quad bike track (the first and only in
+pduRD2JW094-00058-00036176-00036368 Tanzania), water sports, conference facilities, and more are among the attractions. Kunduchi
+pduRD2JW094-00059-00036368-00036752 Water Parks is in the middle of Dar es Salaam, Tanzania.
+pduRD2JW094-00060-00036752-00037272 It's an amusement park that provides room for you to arrange activities that you'd like to do
+pduRD2JW094-00061-00037272-00037704 or be a part of. It has the manpower and capacity to do so for you.
+pduRD2JW094-00062-00037928-00038408 5. Kivukoni Fish Market. Dar es Salaam is primarily a
+pduRD2JW094-00063-00038408-00038984 corporate and government hub, with manufacturing also contributing significantly to the economy.
+pduRD2JW094-00064-00038984-00039648 We wanted to see the markets since they are, in my opinion, the heart and soul of a city. Fishing has
+pduRD2JW094-00065-00039648-00040216 always been a significant part of Dar's culture. The city was founded as a fishing community,
+pduRD2JW094-00066-00040216-00040848 therefore the first stop was the Kivukoni fish markets. The markets, unlike any other I'd seen,
+pduRD2JW094-00067-00040848-00041432 are a big jumble of tables where the fish has been strewn. On one table, there were similar species
+pduRD2JW094-00068-00041432-00042032 of fish, but on another, there were fish of all sizes. Some were being auctioned, while others
+pduRD2JW094-00069-00042032-00042624 were being sold. Fish that had been purchased were being cleaned and fried in another building.
+pduRD2JW094-00070-00042888-00043208 4. Mbudya Island. Mbudya,
+pduRD2JW094-00071-00043208-00043560 an uninhabited island north of Tanzania's major city,
+pduRD2JW094-00072-00043560-00044072 Dar es Salaam, is one of the four islands that make up the Dar es Salaam Marine Reserve.
+pduRD2JW094-00073-00044128-00044664 This beautiful island in the middle of the Indian Ocean resembles Dar Es Salaam's Zanzibar.
+pduRD2JW094-00074-00044664-00045264 Mbudya Island, with its white sand beaches and crystal clear turquoise waters, is the ideal
+pduRD2JW094-00075-00045264-00045944 getaway for anyone looking to escape the rush and bustle of Dar es Salaam. Sunbathing, snorkeling,
+pduRD2JW094-00076-00045944-00046528 swimming, and simply hanging out on the beach with friends are all popular activities on this island.
+pduRD2JW094-00077-00046528-00047208 Mbudya Island is only a 20-minute boat ride from Dar es Salaam's mainland. To catch a local speed
+pduRD2JW094-00078-00047208-00047840 boat from Dar es Salaam to Mbudya Island, go to Kibo beach or Jangwani sea breeze resort
+pduRD2JW094-00079-00047840-00048464 in the Mbezi region and pay the Mbudya island government fees and boat fees to a local boatman.
+pduRD2JW094-00080-00048464-00049040 Every half-hour, a boat is available. Depending on how many people you're bringing, there are a
+pduRD2JW094-00081-00049040-00049672 variety of sizes available. The biggest boats can accommodate up to 40 people. The boat ride can be
+pduRD2JW094-00082-00049672-00050232 a little bumpy. If you get seasick easily, take some anti-seasickness medication before you go.
+pduRD2JW094-00083-00050464-00050904 3. Coco Beach. Coco Beach is a length of beachfront in
+pduRD2JW094-00084-00050904-00051512 Dar es Salaam, Tanzania's Oyster Bay neighborhood. It is one of the city's most well-known beaches,
+pduRD2JW094-00085-00051512-00051952 with citizens flocking to it for beach parties and festivals during the holidays.
+pduRD2JW094-00086-00051952-00052304 As a result, there are a variety of businesses on the beach,
+pduRD2JW094-00087-00052304-00052800 including restaurants, food stands, and shops that rent beach equipment like floaters.
+pduRD2JW094-00088-00052855-00053296 Joggers frequent Coco Beach, many of whom come from adjacent neighborhoods
+pduRD2JW094-00089-00053296-00053896 and can be seen in greater numbers in the evening. At the northern end, there is only one large,
+pduRD2JW094-00090-00053896-00054488 well-established restaurant; the rest are smaller, tarp-covered eateries that provide decent food.
+pduRD2JW094-00091-00054728-00055344 2. Udzungwa Mountains National Park. Udzugwa Highlands National Park is a wildlife
+pduRD2JW094-00092-00055344-00055936 refuge in central Tanzania, between the Iringa and Morogoro regions of the eastern arc mountains,
+pduRD2JW094-00093-00055936-00056544 60 kilometers south of Mikumi National Park on the Mikumi Ifakara road. The national park was
+pduRD2JW094-00094-00056544-00057191 established in 1992 by Prince Bernhard of the Netherlands, the founder of the World Wide Fund,
+pduRD2JW094-00095-00057191-00057896 and covers a total area of 1,990 square kilometers, with vegetation types such as Miombo
+pduRD2JW094-00096-00057896-00058496 woodland, grassland, tropical rainforest, and mountain forest serving as habitats for abundant
+pduRD2JW094-00097-00058496-00059312 wildlife such as elephants, bushbucks, palm civets, leopards, duikers, Miombo genets, hyenas,
+pduRD2JW094-00098-00059312-00060104 over 400 bird species, 2500 ploughed animals, and Wasungwa, a Kihehe word that means "those
+pduRD2JW094-00099-00060104-00060632 who live on the edges of the mountains," inspired the creation of Udzugwa Mountains National Park.
+pduRD2JW094-00100-00060840-00060991 1. Oyster Bay.
+pduRD2JW094-00101-00061048-00061624 In the bustling metropolis of Dar es Salaam, the majestic Oyster Bay Hotel is located.
+pduRD2JW094-00102-00061624-00062191 Only eight tastefully designed suites are available, each with fantastic ocean views,
+pduRD2JW094-00103-00062191-00062808 plush king-sized mattresses, stunning ensuite bathrooms, air conditioning, a ceiling fan,
+pduRD2JW094-00104-00062808-00063400 television, and exquisite French doors leading to a private terrace. Delicious meals are served in
+pduRD2JW094-00105-00063400-00064024 the elegant dining room, as well as al fresco on the veranda and pool deck. A welcoming study room
+pduRD2JW094-00106-00064024-00064632 allows guests to catch up on emails or read an unfinished book, while a dazzling swimming pool
+pduRD2JW094-00107-00064632-00065120 offers a calm respite to enjoy a refreshing cocktail while soaking up the Tanzanian sun.
+pduRD2JW094-00108-00065376-00065968 You'll probably fall in love with Dar es Salaam if you give it some time. Dar is a vibrant seaside
+pduRD2JW094-00109-00065968-00066608 city that blends African, Indian, and Arab influences to create a unique urban culture.
+pduRD2JW094-00110-00066608-00067216 Dar has grown chaotically in all directions in recent decades, yet it retains much of its charm.
+pduRD2JW094-00111-00067216-00067816 With a city center full of charming colonial buildings, bougainvillea-lined streets, and a slew
+pduRD2JW094-00112-00067816-00068448 of palm-fringed beaches. Take trip to this nation and enjoy the beauty of Africa in one nation.
+pduRD2JW094-00113-00068504-00068992 Thanks for taking out the time to watch this video till the end. Please make sure to like,
+pduRD2JW094-00114-00068992-00069608 share, subscribe, comment and turn on your notifications for new Africa reloaded videos.
+pfLP92UVzsy-00000-00000012-00000608 SAM: Hello everyone, and welcome to October's edition of Talking Maths in (not quite) Lockdown;
+pfLP92UVzsy-00001-00000608-00001336 um, for those - I think all of you have been with us before at some point, but as a reminder, uh,
+pfLP92UVzsy-00002-00001336-00001928 myself, Katie, Kevin and Ben make up Talking Maths in Public, which is not so much just a
+pfLP92UVzsy-00003-00001928-00002400 conference for maths communicators but also a bit of a support group, and we've been doing
+pfLP92UVzsy-00004-00002400-00003104 these Talking Maths in Lockdown sessions, um, since March-April time when lockdown started.,
+pfLP92UVzsy-00005-00003104-00003856 um, today's session is all about Geogebra and it will not be the same format as the sessions we've
+pfLP92UVzsy-00006-00003856-00004248 had before, where we've had more of a panel discussion and things like that - this will
+pfLP92UVzsy-00007-00004248-00004696 be more of a workshop, so you will actually have to do stuff; although we're not actually
+pfLP92UVzsy-00008-00004696-00005112 looking over your shoulders on your computers and we won't know if you've done it or not,
+pfLP92UVzsy-00009-00005176-00005896 um, so you can join in as little or as much as you like. As a reminder, we are recording these
+pfLP92UVzsy-00010-00005896-00006392 sessions, so if you don't want to be recorded, turn off your mic, turn off your camera; um,
+pfLP92UVzsy-00011-00006392-00006936 if you have any questions at any point that you want to ask but you don't want to be the one to
+pfLP92UVzsy-00012-00006936-00007584 ask them, stick them in the chat or send them to us directly and we will get them asked for you.
+pfLP92UVzsy-00013-00007672-00008248 Please make sure that you are following our code of conduct at all times - so, our code of conduct
+pfLP92UVzsy-00014-00008248-00008752 is here [pastes into chat] - essentially, be kind to people; use inclusive language;
+pfLP92UVzsy-00015-00008856-00009280 make sure you're not sort of inadvertently saying things that might be offensive or upsetting to
+pfLP92UVzsy-00016-00009280-00009984 anyone, and just just have a good time really. Um, is there anything else I need to tell you?
+pfLP92UVzsy-00017-00010280-00010816 Oh yes - uh, some people might be sharing some things within the call that they don't want to be
+pfLP92UVzsy-00018-00010816-00011384 shared more widely, so if you see anything either in it or that someone says that they're sharing
+pfLP92UVzsy-00019-00011440-00012120 and you want to share it beyond today, do ask them first please, just in case it's not possible for
+pfLP92UVzsy-00020-00012120-00012904 that to be shared more widely. And with that I think we'll introduce ourselves, before we get
+pfLP92UVzsy-00021-00012904-00013472 cracking with what's happening today, so: my name is Sam Durbin; I work at the Royal Institution;
+pfLP92UVzsy-00022-00013472-00013976 I am currently on furlough for just the last couple of days, and then back in November
+pfLP92UVzsy-00023-00013976-00014376 at which point I'll go back to my normal day job of looking after the secondary
+pfLP92UVzsy-00024-00014376-00014904 maths masterclasses all around the country., um, Katie, do you want to introduce yourself?
+pfLP92UVzsy-00025-00015040-00015368 KATIE: Can do, yeah! I'm Katie Steckles; I'm a mathematician who talks about maths
+pfLP92UVzsy-00026-00015368-00015856 for a living - so I do talks and workshops - increasingly online - uh, do bits of writing
+pfLP92UVzsy-00027-00015856-00016288 about maths and bits of lecturing at universities and whatever else people will ask me to do,
+pfLP92UVzsy-00028-00016392-00017000 and I will pass on to... Kevin! KEVIN: Right, hello, yes, I'm Kevin Houston; I'm the, um,
+pfLP92UVzsy-00029-00017064-00017616 um, well, I'm a lecturer at the University of Leeds, and I'm the education secretary of
+pfLP92UVzsy-00030-00017616-00018352 the London Mathematical Society, um, and I do bits and pieces in - for - for that I do bits
+pfLP92UVzsy-00031-00018352-00019016 and pieces for outreach and public engagement. So uh, I guess I hand over to Ben now, yes? BEN: Hi,
+pfLP92UVzsy-00032-00019016-00019632 I'm Ben; I'm also one of the TMiP team; as you will have already known; I work for the AMSP - the
+pfLP92UVzsy-00033-00019632-00019992 Advanced Maths Support Programme - for half of my time; I'm based at the University of Bath,
+pfLP92UVzsy-00034-00019992-00020512 and I teach on a maths communication course there and I'm freelance doing similar stuff
+pfLP92UVzsy-00035-00020512-00020984 to Katie - for example in the, sort of, maths communication world, and, similar to Kevin,
+pfLP92UVzsy-00036-00020984-00021440 best described as 'bits and pieces' I think - but in this case bits and pieces to do with maths
+pfLP92UVzsy-00037-00021440-00021984 education and outreach particularly, occasionally including things like YouTube with Numberphile
+pfLP92UVzsy-00038-00022064-00022352 but I also enjoy quite a lot of my time tinkering on a piece of software called
+pfLP92UVzsy-00039-00022352-00022648 Geogebra - coincidentally we're going to look at that later on.
+pfLP92UVzsy-00040-00022772-00023376 KATIE: Good!, um, so I can introduce our speakers for today; so, uh to give you a
+pfLP92UVzsy-00041-00023376-00023944 sense of what this session is about: so, if you do not know what Geogebra is: brave; uh,
+pfLP92UVzsy-00042-00023944-00024504 coming to a session with absolutely no knowledge of what it is about! but this, um, it's a dynamic
+pfLP92UVzsy-00043-00024504-00024984 geometry tool - so it's a piece of software, uh, which allows you to make constructions and
+pfLP92UVzsy-00044-00024984-00025480 kind of manipulate things geometrically and do lots of interesting maths stuff, and and it's
+pfLP92UVzsy-00045-00025480-00026104 increasingly helpful in learning and teaching in the world of doing a lot of things online,
+pfLP92UVzsy-00046-00026104-00026536 erm, because it's got lots of nice features that you can use to do that and it's really great for
+pfLP92UVzsy-00047-00026536-00026880 demonstrating things. So we thought it would be useful for people to put together a session
+pfLP92UVzsy-00048-00026880-00027320 in which you can just have a bit of a play with it - and I'm hoping that people either have a little
+pfLP92UVzsy-00049-00027320-00027792 bit of experience of Geogebra - I guess to be essentially in the same situation as me, which is
+pfLP92UVzsy-00050-00027792-00028216 that I am aware that Geogebra is a thing and I've seen some really cool things done on Geogebra,
+pfLP92UVzsy-00051-00028216-00028672 but I wouldn't have a clue how to create those things and I've done bits of clicking around and
+pfLP92UVzsy-00052-00028672-00029168 sort of modifying other people's Geogebras but not really got a handle on it myself. So if you're in
+pfLP92UVzsy-00053-00029168-00029560 that situation - don't worry, that is the perfect level; erm, if you've got a bit more experience
+pfLP92UVzsy-00054-00029560-00029944 with Geogebra, then that's fantastic and you might even be able to chip in and contribute as well.
+pfLP92UVzsy-00055-00030024-00030680 So, um, I'm gonna hand over, um, to Ben who's got a plan I believe for what's gonna happen
+pfLP92UVzsy-00056-00030680-00031032 in the session - as I say, it's more of a workshop than a panel thing, so you might be
+pfLP92UVzsy-00057-00031032-00031680 asked to do stuff yourselves and get involved. BEN: Let me uh quickly say hi to Becky because
+pfLP92UVzsy-00058-00031680-00032144 it's very much going to be the two of us leading this session - so I've mentioned my background
+pfLP92UVzsy-00059-00032144-00032672 already; Becky, is it fair to ask you to say 30 seconds about your background? BECKY: oh sure, um,
+pfLP92UVzsy-00060-00032792-00033320 yeah so uh, I used to be a maths teacher and I've just started a new job as a data analyst,
+pfLP92UVzsy-00061-00033320-00033844 but in - in the gap in between;, um, actually my Geogebra stuff -
+pfLP92UVzsy-00062-00033896-00034480 we'll go into this more later. Anyway, um, in the gap between I've done, um, a lot of different
+pfLP92UVzsy-00063-00034480-00035112 freelancing maths education stuff; written some, um, resources for MEI, made videos - but
+pfLP92UVzsy-00064-00035112-00035568 the thing that I've been doing most recently is an online Geogebra for Artists course, um, so
+pfLP92UVzsy-00065-00035568-00035944 yeah I think that's - I think that's why I've been invited to talk today. BEN: It is indeed! I'm glad
+pfLP92UVzsy-00066-00035944-00036376 we agree on that - but also Becky knows what's going on with Geogebra in general and with maths
+pfLP92UVzsy-00067-00036376-00036992 in general so we're really glad to have her; more from her in detail in a moment., um, I am going to
+pfLP92UVzsy-00068-00036992-00037448 reiterate what Katie said, which is that this is going to be at least in part a hands-on workshop:
+pfLP92UVzsy-00069-00037448-00037856 we'd like you to get involved and I'll show you how you're going to do that in a moment; although
+pfLP92UVzsy-00070-00037856-00038424 there also will be some demonstrations. We are trying to find the balance between showcasing
+pfLP92UVzsy-00071-00038424-00039040 what Geogebra can do uh and giving you time to tinker with it which is almost always the only
+pfLP92UVzsy-00072-00039040-00039472 way to learn a new piece of software is tinkering time. We cannot give you enough tinkering time in
+pfLP92UVzsy-00073-00039472-00039832 this meeting uh, in that you will never have enough tinkering time but we'll give you a
+pfLP92UVzsy-00074-00039832-00040248 chance to tinker with some guided tasks and we recommend, if you think it's going to be useful,
+pfLP92UVzsy-00075-00040248-00040792 that you go and find some other tinkering time. Uh, by way of introduction let me show you what
+pfLP92UVzsy-00076-00040792-00041280 the Geogebra sort of starting point looks like and then i'll show you a quick demo and then give you
+pfLP92UVzsy-00077-00041280-00041896 a task to do. So I'm going to share a random web browser screen - just confirm that you're seeing a
+pfLP92UVzsy-00078-00041896-00042320 google page? Anyone who's on video - if you're nodding at that - so if you type in Geogebra
+pfLP92UVzsy-00079-00042384-00042776 to any google thing - geogebra.org - you come to a website that looks like this.
+pfLP92UVzsy-00080-00042776-00043240 This is where Geogebra lives and you can in fact start a calculator by clicking on that button
+pfLP92UVzsy-00081-00043240-00043664 there and you get a diagram that looks like this. This is running in a browser; there is no download
+pfLP92UVzsy-00082-00043664-00044192 happening. That said, the browser version - this happens to be the calculator suite, or the
+pfLP92UVzsy-00083-00044192-00044680 graphing calculator - has got some limitations and there are lots of versions of Geogebra. It
+pfLP92UVzsy-00084-00044680-00045064 is a bit of a sore point for people starting off Gepgebra is there's lots of different versions.
+pfLP92UVzsy-00085-00045064-00045512 I will happily talk about that but I don't want to complicate things, so for the sake
+pfLP92UVzsy-00086-00045512-00046008 of today I'm not going to be using that particular version and if you're interested I'm going to be
+pfLP92UVzsy-00087-00046008-00046488 using one that would be called Geogebra classic and I would probably want to be using - if you
+pfLP92UVzsy-00088-00046488-00046960 click on the 'download', there's loads of options here; Geogebra Classic 5 would be the version I
+pfLP92UVzsy-00089-00047056-00047472 would prefer to use. There are all sorts of problems with there being lots of versions;
+pfLP92UVzsy-00090-00047528-00047880 i'll get into that discussion later on if anyone wants me to, but I'm going to try and avoid that.
+pfLP92UVzsy-00091-00047880-00048200 What I'd like to do first though is uh stop sharing this screen
+pfLP92UVzsy-00092-00048200-00048640 and just show you one example use of Geogebra for something extremely basic - and remember,
+pfLP92UVzsy-00093-00048640-00049048 we're trying to find the balance between teaching you how to work this software and also showing you
+pfLP92UVzsy-00094-00049104-00049472 inspiring examples, and Becky's going to show you some inspiring examples; this one I'm not
+pfLP92UVzsy-00095-00049472-00049880 going to claim is inspiring but I do think it's important so bear with me here. I'm going to share
+pfLP92UVzsy-00096-00049880-00050360 a Geogebra Classic 5 screen; it should look fairly boring - it looks a little bit different from the
+pfLP92UVzsy-00097-00050360-00050816 web interface but it's essentially the same. Give me a nod on video if you are seeing that?
+pfLP92UVzsy-00098-00050880-00051408 Good - thank you. I'm going to type in a classic piece of maths education -I'm aware that I'm not
+pfLP92UVzsy-00099-00051408-00051824 just talking to teachers here, but inevitably when we're demonstrating mathematics it's sort
+pfLP92UVzsy-00100-00051824-00052104 of - there's an overlap with the skills needed for a teacher. So let me show you
+pfLP92UVzsy-00101-00052216-00052616 something which is classic piece of school maths. I'm typing into the input bar at the bottom: I'm
+pfLP92UVzsy-00102-00052616-00053344 typing y equals, which means Geogebra recognizes y as a piece of coordinate axis sort-of notation;
+pfLP92UVzsy-00103-00053344-00053967 I'm going to type in m times x plus c; I hope that is a familiar setup - that's the sort of British,
+pfLP92UVzsy-00104-00054064-00054408 sort of standard for a straight line graph. So when I press return
+pfLP92UVzsy-00105-00054472-00054840 I don't think it should be too much of a surprise if Geogebra plots me a straight line. However,
+pfLP92UVzsy-00106-00054920-00055479 first, two comments is that it's colored y and x in different colors; it recognizes those letters
+pfLP92UVzsy-00107-00055479-00056000 as the standard notation for the y-axis and the x-axis; and m and c is colored gray - that's
+pfLP92UVzsy-00108-00056000-00056344 because it doesn't know what those things; are as far as it's concerned they're just letters.
+pfLP92UVzsy-00109-00056344-00056904 In particular they're not numbers. So when I ask Geogebra to plot this, like any computer software
+pfLP92UVzsy-00110-00056904-00057279 it should complain - it should say 'I don't know what m and c are'. In this case though Geogebra
+pfLP92UVzsy-00111-00057279-00057736 actually reacts slightly differently: when I press return it says 'do you want to create sliders for
+pfLP92UVzsy-00112-00057736-00058272 c and m?' and I wanted to show you this first of all because having variables is a crucial part
+pfLP92UVzsy-00113-00058272-00058752 of mathematics, and it's also a crucial part of Geogebra. In particular these variables,
+pfLP92UVzsy-00114-00058752-00059367 or parameters, or algebraic representations, c and m; uh if I create them as sliders we'll have
+pfLP92UVzsy-00115-00059367-00059767 a visual representation. So first of all it's plotted a line - in fact it's plotted the line
+pfLP92UVzsy-00116-00059767-00060328 y equals x plus one; that's because at the moment c and m are one but I can change these
+pfLP92UVzsy-00117-00060328-00060855 values. This is the heart of Geogebra: not only can I plot things, I can make them move. So if I
+pfLP92UVzsy-00118-00060855-00061208 try again - before I drag it I would like you to consider what's going to happen when I drag this
+pfLP92UVzsy-00119-00061208-00061655 m - it's going to change the value of m and this line will change. I'm not going to give you time
+pfLP92UVzsy-00120-00061655-00062032 to type in the chat box because I think a lot of you are very familiar with this piece of maths,
+pfLP92UVzsy-00121-00062032-00062384 but I'd like you to predict nevertheless: if I drag m to the right what will happen? Hold it in
+pfLP92UVzsy-00122-00062384-00062904 your head ;and, lo and behold something probably like what you were expecting is happening. If I
+pfLP92UVzsy-00123-00062904-00063312 drag it to the left it goes down that way - in fact to save my thumb clicking I'm just going to
+pfLP92UVzsy-00124-00063312-00063679 right click on it and click on animation on, and then I'm just going to walk away. And genuinely I
+pfLP92UVzsy-00125-00063679-00064216 wanted to show you this example because I once had to walk into a year 7 lesson that I had to cover
+pfLP92UVzsy-00126-00064216-00064591 in maths; they were doing straight line equations and I had no chance to plan a lesson. I walked
+pfLP92UVzsy-00127-00064591-00065055 into the room, typed in y equals mx plus c to Geogebra and then animated m and just left it on
+pfLP92UVzsy-00128-00065055-00065528 the board as they were coming in. And the bulk of the lesson was us discussing what happened next.
+pfLP92UVzsy-00129-00065584-00065968 In particular I want to make the point that using Geogebra to draw this straight line
+pfLP92UVzsy-00130-00065968-00066392 lets you make it move, and suddenly the movement makes people discuss mathematics.
+pfLP92UVzsy-00131-00066496-00066912 For example one of the comments I had immediately from one of the year sevens in the room was,
+pfLP92UVzsy-00132-00066968-00067608 'sir, why does it go quicker when it's flatter?' Just want you to decide if you agree.
+pfLP92UVzsy-00133-00067984-00068512 Is that line moving quicker in some way when it's flatter? Does it seem to slow down when it gets
+pfLP92UVzsy-00134-00068512-00069024 steep? and the answer is yes and no, and we had a whole lot of discussion about it; it's all we
+pfLP92UVzsy-00135-00069024-00069464 could - we didn't get into the uh the gradient of line being correlated to the tangent of the
+pfLP92UVzsy-00136-00069464-00069840 angle uh although it is, but there's all sorts of depth going on and my first point is that if you
+pfLP92UVzsy-00137-00069840-00070320 make something move mathematically you provoke discussion uh whether this is for educational
+pfLP92UVzsy-00138-00070320-00070712 purposes or inspirational purposes or whether - sort of - just sharing interesting ideas,
+pfLP92UVzsy-00139-00070712-00071080 it is pretty powerful. So the reason why I think this session on Geogebra is important is not
+pfLP92UVzsy-00140-00071080-00071520 because Geogebra is important; it's that if we can make maths dynamic we have a very powerful
+pfLP92UVzsy-00141-00071520-00072008 tool and Geogebra is a freely available way to do it. There are other tools for example Desmos which
+pfLP92UVzsy-00142-00072008-00072448 we should probably do another Talking Maths in Lockdown session on. Uh, obviously you're crying
+pfLP92UVzsy-00143-00072448-00072720 out for me to animate the other one as well so let's just leave this going and that just
+pfLP92UVzsy-00144-00072776-00073136 basically caused everyone's head to explode in the year 7 lesson we were talking about.
+pfLP92UVzsy-00145-00073224-00073584 Making maths move is what this session is about ,and it happens to be we're going to work on
+pfLP92UVzsy-00146-00073584-00074032 Geogebra to do that and I would like to give you a couple of exercises to work on in Geogebra
+pfLP92UVzsy-00147-00074032-00074496 in order to get to know the tools; once you know the tools the creativity is the only
+pfLP92UVzsy-00148-00074496-00074912 thing that limits you, so you just need to come up with something - like any old tool;
+pfLP92UVzsy-00149-00074912-00075288 it's a little bit like programming with geometry and Becky's going to show you some
+pfLP92UVzsy-00150-00075288-00075680 more artistic examples of that. Before I leave this particular example are there any questions
+pfLP92UVzsy-00151-00075680-00076064 about what I've just built in Geogebra now? If they're questions about 'how did you do that?',
+pfLP92UVzsy-00152-00076064-00076424 save them but I will give time in the chat if anyone wants to put anything in there now.
+pfLP92UVzsy-00153-00077168-00077632 Okay I'm going to stop sharing and I'm going to crack on with a different task. So I am going
+pfLP92UVzsy-00154-00077632-00078112 to give you a maths question and - you are going to have a chance to build this file in a moment
+pfLP92UVzsy-00155-00078112-00078608 yourself, but I would like you to just watch this for now. I - I'm going to build something;
+pfLP92UVzsy-00156-00078608-00079160 please don't try and copy this while I'm doing it - you will inevitably be looking at your screen
+pfLP92UVzsy-00157-00079216-00079608 while - if you're trying to copy this, and you won't notice what I'm doing;
+pfLP92UVzsy-00158-00079608-00079880 I'm going to give you a chance to copy this, but I'm going to share a screen again
+pfLP92UVzsy-00159-00079944-00080352 and this is going to be a - another Geogebra Classic 5 screen that's sort of just blank
+pfLP92UVzsy-00160-00080352-00080656 and I'm going to build a triangle. I'm going to do that by clicking on
+pfLP92UVzsy-00161-00080656-00081136 the polygon tool and I hope you realise if you hover over a tool it tells you how to use it.
+pfLP92UVzsy-00162-00081200-00081632 First tip for learning how to do something is make sure you read documentation; in Geogebra's case,
+pfLP92UVzsy-00163-00081632-00082040 whether you lose - use the online version or the classic one there are tool tips available; they
+pfLP92UVzsy-00164-00082040-00082504 tell you how to use things. So I'm going to draw a triangle and it says to draw a polygon I select
+pfLP92UVzsy-00165-00082504-00082976 all vertices and then the first vertex again. If I was feeling in a patronising mood I would
+pfLP92UVzsy-00166-00082976-00083448 ask someone to tell me what to do, but I think you'd probably tell me to click three corners
+pfLP92UVzsy-00167-00083448-00083848 and then finish it by clicking the first vertex again. And, lo and behold, there is a triangle.
+pfLP92UVzsy-00168-00083848-00084344 First comment though: this is not just *that* triangle, because this is a triangle on a dynamic
+pfLP92UVzsy-00169-00084344-00084784 geometry package - so it means it's actually any triangle and I can move this around and I have
+pfLP92UVzsy-00170-00084784-00085224 genuinely a general triangle available to me; that's the difference between what the
+pfLP92UVzsy-00171-00085224-00085648 Greeks were doing scratching into stone these triangles ,and what we can do on a dynamic piece
+pfLP92UVzsy-00172-00085648-00086016 of geometry. Okay, I would like all of you now - this is the most exciting interaction you'll
+pfLP92UVzsy-00173-00086016-00086464 have all day - to point to the screen; I will see you - you're on camera - I'm not going to be able
+pfLP92UVzsy-00174-00086464-00086984 to see where you're pointing accurately, but I'd like you to point to the centre of that triangle.
+pfLP92UVzsy-00175-00087264-00087656 I am being deliberately vague - you might know where this is going - but the reason I'm asking
+pfLP92UVzsy-00176-00087656-00088072 you to point is that you had a commitment to make about where the centre was; you had some intuition
+pfLP92UVzsy-00177-00088136-00088520 and I would ask someone: I'm gonna pick on Sam - I'm sorry to do this
+pfLP92UVzsy-00178-00088520-00088960 to you Sam - would you be happy to tell me in words, if you turn your microphone on,
+pfLP92UVzsy-00179-00088960-00089568 where you thought the centre was roughly, and why? SAM: I thought it was in the middle... sorry,
+pfLP92UVzsy-00180-00089568-00090064 I couldn't resist saying that!, um, BEN: I agree - I think a lot of people are thinking that - why
+pfLP92UVzsy-00181-00090064-00090744 would it be anywhere else Ben? Whereabouts in the middle... about there? SAM: About there,
+pfLP92UVzsy-00182-00090744-00091144 yeah. BEN: About there? SAM: No, about the first one. BEN: It was about there. Did you have
+pfLP92UVzsy-00183-00091144-00091848 any reason for picking that one? SAM: Uh, it's about halfway along each of the lines. BEN: Ah,
+pfLP92UVzsy-00184-00091848-00092240 so you're thinking about midpoints of these segments. SAM: I was! BEN: In that case let's
+pfLP92UVzsy-00185-00092240-00092600 ask Geogebra to build those midpoints - I'm going to this menu; going to the midpoint tool - oh look
+pfLP92UVzsy-00186-00092600-00092984 there's a tool - if you want to know how to use a tool, hover over it: 'select two points'. okay,
+pfLP92UVzsy-00187-00092984-00093368 I'll select a and b, and it's made the midpoint which Sam is just referred to - I'm going to
+pfLP92UVzsy-00188-00093368-00093736 do that for another one here; but that's not defining a centre Sam, so say a little bit more
+pfLP92UVzsy-00189-00093968-00094216 (by the way I do realise there are two Sams in the room, and I didn't make it clear but
+pfLP92UVzsy-00190-00094216-00094568 luckily one of them answered so that's okay) SAM: I should have just stayed quiet and let the other
+pfLP92UVzsy-00191-00094568-00095440 Sam answer!, um, if you then join up the midpoint of a line with its opposite corner, where those
+pfLP92UVzsy-00192-00095536-00096208 further three lines intersect was about where I was thinking. BEN: So that diagram. SAM: Yeah;
+pfLP92UVzsy-00193-00096208-00096536 but with the other - with the other line as well. BEN: Wait, hold on a minute: let me just
+pfLP92UVzsy-00194-00096536-00096920 make something profoundly obvious to everyone - if you draw two lines on a screen the fact that they
+pfLP92UVzsy-00195-00096920-00097352 intersect is not a surprise (at least I hope it's not to anyone who's done a little bit of playing
+pfLP92UVzsy-00196-00097352-00097712 with lines) However- SAM: As long as they're not parallel! BEN: I agree, there is a possibility
+pfLP92UVzsy-00197-00097712-00098040 that they don't intersect but if they're not parallel they will intersect somewhere - in
+pfLP92UVzsy-00198-00098040-00098440 fact if I draw line segments, maybe we're dodging the issue; maybe they won't intersect, but if I
+pfLP92UVzsy-00199-00098440-00098952 draw three lines there is no reason why they would intersect in the same place - and yet if I click
+pfLP92UVzsy-00200-00099096-00099600 the third line in here it does look like they intersect in the same place and that is a shock.
+pfLP92UVzsy-00201-00099712-00100184 This also means this centre is an important point, uh, and by the way remember this is not just any
+pfLP92UVzsy-00202-00100184-00100488 particular triangle, this is all possible triangles, because this moves with it;
+pfLP92UVzsy-00203-00100488-00100888 and again I'm going to keep doing this movement to remind you that the reason to use Geogebra or
+pfLP92UVzsy-00204-00100888-00101320 Desmos is that it's dynamic; it's general - this is not just true for the one drawing I've done,
+pfLP92UVzsy-00205-00101320-00101880 it's for all of them. So this point that we've just created is a centre of this triangle,
+pfLP92UVzsy-00206-00101880-00102376 but it is not the only one. In fact, Sam, I agree - is the centre and it has a particular sense:
+pfLP92UVzsy-00207-00102376-00102856 this is called the centroid it is the centre of mass of this triangle. Other options
+pfLP92UVzsy-00208-00102856-00103392 are available. So to cut a long story short, I'm gonna "here's one I made earlier" moment
+pfLP92UVzsy-00209-00103392-00103896 with my Blue Peter sort of best hat on; I'm gonna go to my second screen
+pfLP92UVzsy-00210-00103896-00104359 over here and show you a file which I am going to give you a chance to play with in a moment;
+pfLP92UVzsy-00211-00104504-00104840 so this is the demo file I made in Geogebra; I'm not going to go into great detail about
+pfLP92UVzsy-00212-00104840-00105120 how I made it, but I'm going to give you a chance to try and make it in a moment; can you confirm
+pfLP92UVzsy-00213-00105120-00105464 you're seeing a blank looking triangle on the screen at the moment - nod if you can;
+pfLP92UVzsy-00214-00105464-00105872 so what we just built earlier was the centroid and I've got a tick box to show that, and it's
+pfLP92UVzsy-00215-00105872-00106335 made by joining the medians, if you like - medians being the midpoints to the opposite corners which
+pfLP92UVzsy-00216-00106335-00106792 Sam described. There are other options for the centre of a triangle - for example I could maybe
+pfLP92UVzsy-00217-00106792-00107400 do the perpendicular bisectors of the sides; and if you do that they all meet at a point - in fact,
+pfLP92UVzsy-00218-00107400-00107872 that one's called the circumcentre and it is in fact the centre of the circle around a triangle;
+pfLP92UVzsy-00219-00107872-00108264 and every triangle has a circle it's called the circumcircle, and it has a centre of that circle;
+pfLP92UVzsy-00220-00108264-00108759 but this centre of the triangle can be outside the triangle, which is slightly alarming for
+pfLP92UVzsy-00221-00108759-00109111 people who were pointing to the middle of the triangle. That is also not the only option;
+pfLP92UVzsy-00222-00109111-00109496 maybe another option could be - and if I had longer I'd love to get you to suggest things - but
+pfLP92UVzsy-00223-00109496-00109992 instead of bisecting the sides I could bisect the angles: so I could bisect the angles like that,
+pfLP92UVzsy-00224-00109992-00110328 and then I get an incentre which is the centre of a different circle - in fact it's the centre of
+pfLP92UVzsy-00225-00110328-00110696 the incircle which stays inside the triangle at all times; and again I want to move this around
+pfLP92UVzsy-00226-00110696-00111096 to show that this is not just one triangle: this is all triangles have these properties.
+pfLP92UVzsy-00227-00111176-00111559 So to cut a long story short I could wax lyrical on centres of triangles - lots of them are
+pfLP92UVzsy-00228-00111559-00112072 available; in fact the Wikipedia page on centres of triangles has a frightening number of centres;
+pfLP92UVzsy-00229-00112072-00112568 but there are four in particular which are quite well known: the incentre; the orthocentre,
+pfLP92UVzsy-00230-00112568-00113016 which is the centre of the altitudes, where you drop perpendiculars from coin- corners to the
+pfLP92UVzsy-00231-00113016-00113511 opposite side; and the centroid. So those four are famous centres and if I just uh hide some of the
+pfLP92UVzsy-00232-00113511-00113952 rest of these, they're all different places. So basically I'm saying wherever you pointed
+pfLP92UVzsy-00233-00113952-00114400 at the triangle you're probably close to one of these centres. But building this in Geogebra
+pfLP92UVzsy-00234-00114400-00114935 and making it dynamic is quite a good exercise to get used to the geometric aspect of Geogebra.
+pfLP92UVzsy-00235-00114935-00115335 So here's what I'd like you to do: I'm going to share a link - in fact, if someone in the,
+pfLP92UVzsy-00236-00115335-00115983 um, team could paste the link to a Geogebra classroom page for me into the chat, what I've
+pfLP92UVzsy-00237-00115983-00116400 done is given you access to a - what Geogebra will also do is provide a classroom environment
+pfLP92UVzsy-00238-00116400-00116896 where you can ask people to go and have a play, and I can see what you're doing - which made
+pfLP92UVzsy-00239-00116896-00117208 me wince earlier when Sam said "we're not going to be watching over your shoulder";
+pfLP92UVzsy-00240-00117280-00117832 I actually am in this case. So that link is now in the classroom: I'd like you to click on that;
+pfLP92UVzsy-00241-00117832-00118304 you're going to have to type in your name and once you're in there I would like you to click
+pfLP92UVzsy-00242-00118304-00118759 through and follow - I'm going to stop sharing my screen now - follow the instructions. The first
+pfLP92UVzsy-00243-00118759-00119200 task you get to see will be the file I was just playing with ,so you can have a play with that,
+pfLP92UVzsy-00244-00119200-00119808 and the second task is a blank triangle - and I would like you to try and build your own centres
+pfLP92UVzsy-00245-00119808-00120304 of triangles. You can do the centroid, like Sam just did; you can try and do the incentre
+pfLP92UVzsy-00246-00120304-00120664 which is the centre of the incircle, and that's coming from angle bisectors
+pfLP92UVzsy-00247-00120664-00120992 and you're going to have to experiment with the tools available to you to do that;
+pfLP92UVzsy-00248-00121048-00121335 I'm not going to say any more - I want to give you a full five minutes where I shut up,
+pfLP92UVzsy-00249-00121392-00121744 but I will come back in five minutes and see how you're getting on. At any point
+pfLP92UVzsy-00250-00121744-00122183 if you want to jump on a mic and ask a question, you will have to do that publicly - it's not
+pfLP92UVzsy-00251-00122183-00122568 like I can just sidle over to you and, uh, whisper to you but please feel free
+pfLP92UVzsy-00252-00122568-00123052 to do that. I'm going to stop talking for five minutes: build your own centres of triangles.
+pfLP92UVzsy-00253-00123052-00123552 SAM: Do we get background music? BEN: Do you want background music?
+pfLP92UVzsy-00254-00123708-00124696 SAM: There's some nodding going on. BEN: I- I fear I should spare you that, so no,
+pfLP92UVzsy-00255-00124696-00125416 but hey - thanks for the request. I'm going to stop talking for a bit.
+pfLP92UVzsy-00256-00125416-00125856 Everyone's at different levels of expertise here - there are two questions about this task:
+pfLP92UVzsy-00257-00125935-00126383 task 3 and task 4; they should be on the same page as the the task you're working on - feel
+pfLP92UVzsy-00258-00126383-00126856 free to put responses to that if you wish but that is not compulsory. I would suggest
+pfLP92UVzsy-00259-00126856-00127583 not going beyond that; there's a- stay on the triangle blank activity for now.
+pfLP92UVzsy-00260-00127952-00128159 I'd just like to say, Fran: I really enjoyed what
+pfLP92UVzsy-00261-00128159-00128352 I saw from your construction just there - that was looking good!
+pfLP92UVzsy-00262-00128592-00128983 ...or maybe I'm looking at the wrong page.
+pfLP92UVzsy-00263-00129316-00129856 FRAN: Can I ask a question Ben? BEN: You can indeed. FRAN: I got to the point, um, so I
+pfLP92UVzsy-00264-00129856-00130208 constructed the perpendicular bisectors BEN: I can see that, well done! FRAN: I found the point where
+pfLP92UVzsy-00265-00130208-00130896 they intersected yeah and then I tried to get hold of like a point a b or c and move it and it didn't
+pfLP92UVzsy-00266-00130896-00131240 do what I thought it was going to do so I was like oh. BEN: I'm glad someone asked that and everyone
+pfLP92UVzsy-00267-00131240-00131616 else can either listen to what I'm saying or carry on tinkering; you need to make sure, to move a
+pfLP92UVzsy-00268-00131616-00132112 point, that you are on the select tool, which is the arrow tool on the far left of the toolbar;
+pfLP92UVzsy-00269-00132112-00132432 and there is a shortcut available to get you to that, because you want to go back to that tool
+pfLP92UVzsy-00270-00132432-00132904 repeatedly all the time, so: pressing escape, if you're in a Geogebra window, will jump you back to
+pfLP92UVzsy-00271-00132904-00133424 that tool and that is super helpful - I basically hover over the escape button all the time when I'm
+pfLP92UVzsy-00272-00133424-00133816 working. FRAN: Thank you very much indeed. BEN: Tell me if that works. BECKY: That's a really
+pfLP92UVzsy-00273-00133816-00134408 good tip that I didn't know, thank you! BEN: Oh I've taught Becky something! Yes! SAM: How do you
+pfLP92UVzsy-00274-00134408-00134752 find the centre of a circle you've just made? BEN: Indeed: how do you?
+pfLP92UVzsy-00275-00134848-00135184 I'm doing full teacher mode here! if a circle is on the screen, uh,
+pfLP92UVzsy-00276-00135184-00135600 where is the centre? Now this is actually a classic A-level algebra question,
+pfLP92UVzsy-00277-00135672-00136328 but let me give you one option: you know about circle theorems; you know that chords of a circle
+pfLP92UVzsy-00278-00136328-00136904 and the perpendicular bisectors of a chord always go through the centre, so here's one option:
+pfLP92UVzsy-00279-00136904-00137400 draw two chords on that circle and bisect them; they will intersect at the centre.
+pfLP92UVzsy-00280-00137400-00138028 In fact what you'll be doing is very very close to finding the circumcircle of a triangle. SAM:
+pfLP92UVzsy-00281-00138080-00138432 So there's not- not a special button to click on the circle and go "show me the
+pfLP92UVzsy-00282-00138432-00138848 centre"? BEN: There- you could- you could make one in Geogebra, but what's interesting about working
+pfLP92UVzsy-00283-00138848-00139280 in Geogebra is that you have to be doing maths in order to construct stuff. This is not the same as
+pfLP92UVzsy-00284-00139280-00139672 drawing stuff - this is construction ,and a lot of you are realising that from what I'm seeing now.
+pfLP92UVzsy-00285-00139832-00140160 I am going to stop you, and I'm sorry this is going to frustrate you - Geogebra classroom
+pfLP92UVzsy-00286-00140160-00140552 does give me a pause button - I'm just watching your faces and all of you are looking slightly
+pfLP92UVzsy-00287-00140552-00140896 frustrated I've stopped you interacting with Geogebra - this is something they borrowed
+pfLP92UVzsy-00288-00140896-00141320 from Desmos classroom. I will come back to another chance to play with Geogebra classroom in a bit;
+pfLP92UVzsy-00289-00141320-00141864 first of all though, let me show you something slightly meta - uh, what I'm seeing is this
+pfLP92UVzsy-00290-00141864-00142296 screen here: let me just share my screen - so if you could come back to a Zoom window,
+pfLP92UVzsy-00291-00142440-00142928 um, give me a nod on the video if you are seeing a screen that says Task 2. So I've anonymized all
+pfLP92UVzsy-00292-00142928-00143248 your names -it's a bit like Desmos, if you play with that, and I can see that everyone's been
+pfLP92UVzsy-00293-00143248-00143632 working; I see student number 4 has, uh, played with the buttons making some polygons, thank
+pfLP92UVzsy-00294-00143632-00144152 you student number 4. Let me zoom in on student number 10 because I like what happened here; I-
+pfLP92UVzsy-00295-00144216-00144608 I think I know who this is but, uh, you can own up later on - I can have a look at what
+pfLP92UVzsy-00296-00144608-00145088 everyone's working on, which is really lovely from an online teaching point of view or enrichment
+pfLP92UVzsy-00297-00145088-00145536 point of view, but I can also interact with the file here that student 10 has played with,
+pfLP92UVzsy-00298-00145536-00145944 and can you see - something I wanted pointed out here: first of all they've made j which is what
+pfLP92UVzsy-00299-00145944-00146392 they - I think they've made the circumcentre, that's from the perpendicular bisectors;
+pfLP92UVzsy-00300-00146392-00147152 they've also made h which is the incentre, but the incircle is not quite working; which is annoying,
+pfLP92UVzsy-00301-00147152-00147424 and I hope the person who actually made this is feeling the annoyance
+pfLP92UVzsy-00302-00147528-00147816 because I want to make a final point before I hand over to Becky here is that when you're
+pfLP92UVzsy-00303-00147816-00148448 working in Geogbra with geometry you are needing to construct rather than draw. And that difference
+pfLP92UVzsy-00304-00148448-00148984 is subtle and inherently mathematical and difficult to explain, but once you've
+pfLP92UVzsy-00305-00148984-00149368 experienced this problem you really know what we're talking about. So what this person has done
+pfLP92UVzsy-00306-00149368-00149816 is drawn the centre - they've constructed the incentre and there is the centre of an incircle,
+pfLP92UVzsy-00307-00149816-00150368 but when they try and draw the circle, they have just drawn it, and this i dot is not in the right
+pfLP92UVzsy-00308-00150368-00151032 place. So allow me to do one further demo of this: let me just bring up a new Geogebra file.
+pfLP92UVzsy-00309-00151176-00151568 Now, give it a moment, you are seeing the file I started demonstrating on earlier.
+pfLP92UVzsy-00310-00151680-00152112 Lovely. So I'm actually gonna - in Geogebra you can hide and show things - I'm just gonna turn
+pfLP92UVzsy-00311-00152112-00152464 off all the things I built earlier which are all these buttons - these little buttons on the left,
+pfLP92UVzsy-00312-00152520-00152960 if they don't crash on me... that's unexpected! Let's just turn these buttons off and they
+pfLP92UVzsy-00313-00152960-00153392 disappear; so th- those little blue buttons are hide/show buttons. Actually I wanted the
+pfLP92UVzsy-00314-00153392-00153760 triangle to remain. So I'm going to build the angle bisectors very quickly using the
+pfLP92UVzsy-00315-00153760-00154184 angle bisector tool; we could talk lots about all these tools but let's just do a couple of them;
+pfLP92UVzsy-00316-00154344-00154752 angle bisector again; and this will build the incentre where they
+pfLP92UVzsy-00317-00154752-00155472 meet. But if I just draw - oh I missed there, I need to delete that; if I construct the
+pfLP92UVzsy-00318-00155472-00155976 intersection of those two then the circle I want does seem to look like this.
+pfLP92UVzsy-00319-00156264-00156760 Where I click now will define the edge of the circle and I need to know where it has to touch
+pfLP92UVzsy-00320-00156760-00157104 and what the person we did before clicked in - somewhat arbitrary - they just drew it so
+pfLP92UVzsy-00321-00157104-00157568 it looked right. So the question is how do you mathematically know where this circle touches?
+pfLP92UVzsy-00322-00157568-00157992 The answer, to cut a long story short, is that it must be the closest distance from h
+pfLP92UVzsy-00323-00157992-00158400 to one of these sides, which is actually going to be tied up with a perpendicular.
+pfLP92UVzsy-00324-00158400-00158752 So if I construct the point it must touch, by clicking that point and this line, it's
+pfLP92UVzsy-00325-00158752-00159252 giving me the perpendicular; I want the circle to touch at that point - which I missed again -
+pfLP92UVzsy-00326-00159480-00160000 but then I can build the circle which is centred there and touches that point, and now when I
+pfLP92UVzsy-00327-00160000-00160512 move it around the incircle moves with it. So I wanted you to experience some of this frustration,
+pfLP92UVzsy-00328-00160512-00160848 and the fact that you're different- you're not drawing you're constructing, is actually
+pfLP92UVzsy-00329-00160848-00161280 a helpful thing Geogebra teaches you but it is quite frustrating until you get the hang of it.
+pfLP92UVzsy-00330-00161344-00161744 Okay: quick summary of what I've shown you: uh I've shown you that Geogebra does
+pfLP92UVzsy-00331-00161744-00162136 geometry, basically; I've shown you how Geogebra classroom works, and you can use that in an online
+pfLP92UVzsy-00332-00162136-00162472 environment to see what everyone else is doing; and I have given you a chance to tinker - not long
+pfLP92UVzsy-00333-00162472-00162904 enough. There will be another chance later on, but that was our first introduction; I'm going to stop
+pfLP92UVzsy-00334-00162904-00163288 talking now and hand over to Becky, if that's alright, because what I'd like her to show you,
+pfLP92UVzsy-00335-00163288-00163784 uh, is sort of fast forwarding -when you've done enough working with the tools, what you can end
+pfLP92UVzsy-00336-00163784-00164200 up doing just for the hell of it in Geogebra to make some lovely pictures. Is that okay Becky?
+pfLP92UVzsy-00337-00164480-00164848 Yeah all good yep, um, so I mean to be honest I'm
+pfLP92UVzsy-00338-00164848-00165048 just going to carry on from what Ben has said because
+pfLP92UVzsy-00339-00165320-00165872 the reason I have been using Geogebra to create art, um, is because of it's- the-
+pfLP92UVzsy-00340-00165872-00166248 the construction that you can do; so on a computer, on other, you know, um,
+pfLP92UVzsy-00341-00166352-00166880 image editing programs and things like that, um, there are some simple, you know, geometric
+pfLP92UVzsy-00342-00166880-00167312 tools that you can use - you can often rotate or reflect things, um, but it doesn't have the full,
+pfLP92UVzsy-00343-00167368-00167912 you know, suite of construction things so, um, I've been using Geogebra since,
+pfLP92UVzsy-00344-00167912-00168616 um, I was a teacher but, um, over lockdown there was a lot of kind of math arts / maths art, um,
+pfLP92UVzsy-00345-00168616-00169144 challenges and things like that and and I've moved to America and I didn't bring a lot of my
+pfLP92UVzsy-00346-00169144-00169640 artistic stuff with me so I've kind of used what I had, and one of the things I had was Geogebra.
+pfLP92UVzsy-00347-00169640-00170344 Um, so I'm going to do a couple of demonstrations, um, one slightly more simple and a slightly more
+pfLP92UVzsy-00348-00170344-00170808 complicated one just to kind of show you some things; um, like as Ben said,
+pfLP92UVzsy-00349-00170808-00171480 um, I wouldn't recommend following along, um, because I will go reasonably fast;
+pfLP92UVzsy-00350-00171480-00171928 however these are- these are being recorded so if you want to kind of go back, um, and with
+pfLP92UVzsy-00351-00171928-00172352 the pause button you can kind of go through and see how I've constructed some of these designs.
+pfLP92UVzsy-00352-00172472-00172952 So I'm going to start sharing my screen. I'm going to do a demonstration and I'm going to-
+pfLP92UVzsy-00353-00172952-00173604 I'm going to carry on talking over the top, um, so: can you all see my Geogebra window? BEN:
+pfLP92UVzsy-00354-00173656-00174726 I can see it! BECKY: Good, okay. So with the tools here we've got, um, you know, the point tool,
+pfLP92UVzsy-00355-00174726-00175448 the circle tool, the line tool - so anything that you can construct with, um, a ruler and
+pfLP92UVzsy-00356-00175448-00175984 [compass] using those tools you can also do in Geogebra - but the nice thing about Geogebra
+pfLP92UVzsy-00357-00176112-00176776 is that you can do some shortcuts. So for example I do quite like the regular polygon tool,
+pfLP92UVzsy-00358-00176776-00177160 um, so rather than - so for example a polygon I could construct
+pfLP92UVzsy-00359-00177160-00177760 using a ruler and compass is a 12-sided polygon, um, but I'm just gonna you know do a quick, um,
+pfLP92UVzsy-00360-00177816-00178288 oh I think I've shared the the Geogebra window so there is a pop-up where I type in the number of
+pfLP92UVzsy-00361-00178288-00178816 vertices because I haven't shared my screen as such, um, so I just typed in the box,
+pfLP92UVzsy-00362-00178816-00179528 um, the number of vertices I want is 12. So here we go - I've got a nice dodecagon, um,
+pfLP92UVzsy-00363-00179768-00180448 and, um, I actually- I'm just going to use this to kind of be the basis of my design,
+pfLP92UVzsy-00364-00180520-00181448 um, I'm going to use the polygon tool and I'm just gonna do a quick, um, star, so I just - you know,
+pfLP92UVzsy-00365-00181448-00181880 I'm building my polygons. This is kind of - kind of like, you know, the undo button is
+pfLP92UVzsy-00366-00181880-00182496 my friend here, so: one two three four five, one two three four five ,one two three four five...
+pfLP92UVzsy-00367-00182720-00183440 So you know not only can we do, um, geometric designs but yeah the ability to kind of,
+pfLP92UVzsy-00368-00183440-00183888 um, have I done it wrong again? I think I have. I think I can - if I click I can
+pfLP92UVzsy-00369-00184024-00184656 undo some of these, um, oh dear. okay I'll choose a different star - you know, that
+pfLP92UVzsy-00370-00184656-00185312 you can do - a star that I can see a bit better. I'm going to do the star that's made up of, um,
+pfLP92UVzsy-00371-00185520-00185992 equilateral triangles. So, uh, in Geogebra but there's always multiple ways of doing
+pfLP92UVzsy-00372-00185992-00186304 things - so I could have started with an equilateral triangle and rotated it.
+pfLP92UVzsy-00373-00186304-00186832 I've got this nice 12-pointed star and then what I like to do is I'm going to do some,
+pfLP92UVzsy-00374-00186832-00188056 um, transformations in a minute - I seem to have seemed to have coloured my polygon,
+pfLP92UVzsy-00375-00188200-00188856 uh, so yeah so, um, I - I started using Geogebra to create some artistic designs
+pfLP92UVzsy-00376-00188856-00189360 and it kind of got a little bit of interest and so I - I set up a an online course
+pfLP92UVzsy-00377-00189440-00190136 to teach people how to do the things that I do. Well no, not really - it was kind of using
+pfLP92UVzsy-00378-00190240-00191008 art, artistic designs, geometric designs to, um, learn about Geogebra.
+pfLP92UVzsy-00379-00191328-00192168 I can see so, um, as Ben was saying about constructions, I've managed to miss a point here,
+pfLP92UVzsy-00380-00192168-00192912 so this is a kind of a drawing point rather than a construction point; but I can just do a quick fix
+pfLP92UVzsy-00381-00192912-00193328 so this is the other thing about Geogebra is that because we can fix it as we go along,
+pfLP92UVzsy-00382-00193328-00193988 it gives us kind of a bit more flexibility. So I'm just going to quickly redefine -
+pfLP92UVzsy-00383-00194128-00194456 so, um, yeah I'm just very giving a brief
+pfLP92UVzsy-00384-00194576-00195208 overview of, um, the kinds of things that I've used Geogebra for recently, um,
+pfLP92UVzsy-00385-00195408-00196648 and I've - I've spent some time exploring some of the tools, um, but,
+pfLP92UVzsy-00386-00196824-00197408 um, it's very powerful - I think that - and we can use we can use some of these,
+pfLP92UVzsy-00387-00197464-00198304 um, so here I'm using the translate by vector tool, um, I've got my basic
+pfLP92UVzsy-00388-00198304-00198744 shape; I can, you know, I can - you know, in this case tile it in a particular
+pfLP92UVzsy-00389-00198960-00199304 pattern, um, so yeah this is just a kind of, um,
+pfLP92UVzsy-00390-00199400-00199824 the kinds of things that you can do in Geogebra. I wrote down some things I said I was going to say,
+pfLP92UVzsy-00391-00199880-00200432 um, so yeah so I i started using Geogebra because, um, in knockdown because I didn't have a lot of,
+pfLP92UVzsy-00392-00200496-00201032 um, tools with me and so this was kind of like, you know, Geogebra is free and, you know,
+pfLP92UVzsy-00393-00201032-00201496 you can do all the constructions and you can play around as well so it doesn't - you know if you
+pfLP92UVzsy-00394-00201496-00201904 make a mistake it's very easy to undo whereas if you've, you know, drawn something on a page
+pfLP92UVzsy-00395-00201904-00202360 and, you know, if - trying to fix it can be quite messy, um, so here I, you know, once I've
+pfLP92UVzsy-00396-00202360-00202912 constructed one part doing the transformations to create all the other parts can be quite quick, um,
+pfLP92UVzsy-00397-00203048-00203448 but yeah that's just kind of using the basic geometry tools and in some ways that's not
+pfLP92UVzsy-00398-00203568-00204160 that's not the most powerful aspect of Geogebra but it does allow you to be incredibly precise
+pfLP92UVzsy-00399-00204160-00204664 with your kind of mathematical construction - so a lot of for example Islamic designs are based on
+pfLP92UVzsy-00400-00204664-00205264 you know ruler and compass constructions, um, and you know if you do it, um, if you do it
+pfLP92UVzsy-00401-00205671-00205992 you know like this then you get the very precise things rather than, you know, if you're using
+pfLP92UVzsy-00402-00205992-00206448 other image editing software, um, you know you can kind of click things together and do some
+pfLP92UVzsy-00403-00206448-00206928 of it but it's kind of - the power in this is the mathematical nature of it so you can create these
+pfLP92UVzsy-00404-00206928-00207976 beautiful designs very precisely, um, and often much quicker than you can, um, by hand. So that
+pfLP92UVzsy-00405-00207976-00208704 was just a quick, um, demo of how - how quick it is to build something - you know, a simple
+pfLP92UVzsy-00406-00208704-00209176 repeating pattern I wanted to kind of, um, show you some inspiration so I'm going to, um, stop
+pfLP92UVzsy-00407-00209176-00209664 sharing that screen and, um, I've got a little powerpoint presente- presentation to share so, um,
+pfLP92UVzsy-00408-00209816-00210471 I'm going to just quickly start with that. So here, um,
+pfLP92UVzsy-00409-00210640-00211352 I've started with - these are two pentagons that I've rotated, um, round so I've got a
+pfLP92UVzsy-00410-00211352-00211848 kind of ten-pointed shape and I'm going to kind of go through this quite quickly and kind of talk
+pfLP92UVzsy-00411-00211848-00212319 over it, um, but obviously again there's a lot FRAN: Becky, I can't see what you can see I don't
+pfLP92UVzsy-00412-00212319-00212832 think BECKY: oh hang on a second - thank you let me just see BEN: Just the share screen has gone,
+pfLP92UVzsy-00413-00212832-00213512 that's all BECKY: Yeah i'll just, uh SAM: It might be worth turning off your video as well -you're,
+pfLP92UVzsy-00414-00213512-00213871 um, breaking up slightly again BECKY: Yeah, let me see if I can do that,
+pfLP92UVzsy-00415-00214032-00215160 um, there we go. Okay, so here we go, here are my pentagons I was just mentioning, um, and I'm
+pfLP92UVzsy-00416-00215160-00215567 just going to kind of quickly go through this - so this is an Islamic - a traditional Islamic design
+pfLP92UVzsy-00417-00215688-00216432 and, um, uh there are some points that can be built; well they have to be constructed
+pfLP92UVzsy-00418-00216432-00217216 from from the pentagons, um, and then some segments being built from - from those points,
+pfLP92UVzsy-00419-00217319-00218144 um, and then using the polygon tool this was, um, filling in the the gaps to create, um, I've hidden
+pfLP92UVzsy-00420-00218144-00218728 some of the the points in that case so Ben showed how to hide the points using the points on the
+pfLP92UVzsy-00421-00218728-00219216 left-hand side, um, and hiding the construction lines so I come up with this design at the end,
+pfLP92UVzsy-00422-00219271-00219744 um, so this is a traditional Islamic decorative pattern, (BEN: So nice)
+pfLP92UVzsy-00423-00219744-00220208 um, I also using the same construction this is kind of like a standalone design
+pfLP92UVzsy-00424-00220208-00220623 with 10-fold symmetry but using the same construction I created a different polygon
+pfLP92UVzsy-00425-00220680-00221400 from the same points so it's this kind of like half - half a design and then doubled it using
+pfLP92UVzsy-00426-00221400-00221936 the rotation tool and then extended it so this - because it's in the rhombus shape it can be
+pfLP92UVzsy-00427-00221936-00222384 tessellated, um, and then you know this is my kind of final - final piece for that one was a
+pfLP92UVzsy-00428-00222384-00223052 kind of black and white tiling pattern which can be kind of, um, decorated however you like. Um,
+pfLP92UVzsy-00429-00223152-00223592 and then so that's one of the pieces; I'm going to show you a couple more, kind of for inspiration,
+pfLP92UVzsy-00430-00223592-00224367 of things that I've done; so these are - some of them are based on Islamic traditional designs; the
+pfLP92UVzsy-00431-00224367-00224952 one on the top rights is a decoration of a dome but it's got really nice mathematical properties,
+pfLP92UVzsy-00432-00224952-00225696 um, and I spent quite a lot of time working on how to do - rather than - so a lot of the instructions
+pfLP92UVzsy-00433-00225696-00226112 I found were very much "use the compass and ruler and do this, do that" and then I was kind of
+pfLP92UVzsy-00434-00226112-00226519 delving into the, you know, the underlying geometry of it in order to be able to do it
+pfLP92UVzsy-00435-00226519-00227119 quicker in Geogebra, um, the bottom left one there is a random design of my own making, um, as you
+pfLP92UVzsy-00436-00227119-00227664 know we were looking at the stars earlier so I believe that's a a star in - a 12-pointed star,
+pfLP92UVzsy-00437-00227736-00228400 um, and then tessellated - well translated I suppose yeah, tiling, um, and coloured
+pfLP92UVzsy-00438-00228400-00228767 in a particular way; and then on the bottom right I think that was another traditional
+pfLP92UVzsy-00439-00228767-00229336 Islamic design. But also, you know ,you can use some kind of mathematical constructions
+pfLP92UVzsy-00440-00229336-00229952 as well - so here starting with something that I wouldn't be able to construct by hand,
+pfLP92UVzsy-00441-00230040-00230704 um, this is a you know a nice spiral and then so, um, some equations of lines which I could have
+pfLP92UVzsy-00442-00230704-00231048 drawn geometrically but in this case I drew them out directly essentially;
+pfLP92UVzsy-00443-00231136-00231744 and then I kind of decorated it so I filled in some intersection points and built up - so here
+pfLP92UVzsy-00444-00231744-00232232 I'm building circles; the centre of the circle is on the spiral and the edge of the circle is
+pfLP92UVzsy-00445-00232232-00232944 at the centre of my my screen it's actually the origin of the coordinate points, um,
+pfLP92UVzsy-00446-00232944-00233416 and you know, then I decorated it in various ways - and again I use the same construction:
+pfLP92UVzsy-00447-00233416-00233952 so here's one decorated version, um, but I could also go back to the construction points
+pfLP92UVzsy-00448-00233952-00234456 and redecorate them in a different way, so using the same kind of underlying structure in more than
+pfLP92UVzsy-00449-00234456-00234904 one way and I quite like that about Geogebra it kind of gives you that flexibility. Um,
+pfLP92UVzsy-00450-00235167-00235608 and yeah there's some formatting that I can do in Geogebra - some of the formatting and decorative
+pfLP92UVzsy-00451-00235608-00236280 work I've done outside, but the the being able to capture these kind of very mathematical shapes
+pfLP92UVzsy-00452-00236280-00236784 and patterns, um, is what I've been using Geogebra for and, um, just to kind of, um,
+pfLP92UVzsy-00453-00236856-00237392 give you some idea I thought we'd, um, this here are some animations that I've done, um, this was
+pfLP92UVzsy-00454-00237392-00237840 actually kind of more for a demonstration but you can imagine how we could use this artistically,
+pfLP92UVzsy-00455-00237928-00238848 um, and then here's one that I was exploring so some of the the points in this are seen in - in
+pfLP92UVzsy-00456-00238848-00239352 geometric arts, particularly Islamic art and I wanted to kind of see how the family of
+pfLP92UVzsy-00457-00239528-00239928 of shape [sound cuts off]
+pfLP92UVzsy-00458-00239928-00240648 you know some other tiling shapes, um, so it's quite, you know, being able to animate stuff - you
+pfLP92UVzsy-00459-00240648-00241256 can do kind of moving art, um, and here's one: this is kind of based on a kind of retro
+pfLP92UVzsy-00460-00241312-00241848 poster, um, and I quite like that being able to do randomized things - so that's the other thing
+pfLP92UVzsy-00461-00241848-00242760 in Geogebra, you can do generative art using some kind of randomized, um, I think, um, oh no this
+pfLP92UVzsy-00462-00242760-00243256 is another one - oh this was this was something we did in my class. So I've been running this class,
+pfLP92UVzsy-00463-00243256-00243623 um, for the last 10 weeks and I've got another class that's just started; so we're doing online
+pfLP92UVzsy-00464-00243680-00244176 and, um, we're yeah we're going through various designs and learning some tools that go along
+pfLP92UVzsy-00465-00244176-00244664 with it, um, so it's quite it's quite a good quite good introduction I think to Geogebra and gives
+pfLP92UVzsy-00466-00244664-00245104 some kind of purpose to the learning so you know a bit more structured if if you kind of prefer
+pfLP92UVzsy-00467-00245104-00245584 that kind of learning. So i'll leave my contact details I'm intending to start another course,
+pfLP92UVzsy-00468-00245584-00246160 um, in January at a time that would be good for UK people, um, so if you drop me a message
+pfLP92UVzsy-00469-00246264-00246656 I will, um, put you on a waiting list and let you know when bookings are open for that.
+pfLP92UVzsy-00470-00246728-00247616 Um, I found teaching Geogebra online mostly good, um, but as as Ben said you know not being able to
+pfLP92UVzsy-00471-00247616-00248128 stand over and watch what someone's doing has been challenging and I haven't used Geogebra classroom
+pfLP92UVzsy-00472-00248128-00248696 because it does have limitations and I mentioned in the chat but I much prefer, um, the Classic 5,
+pfLP92UVzsy-00473-00248760-00249088 um, particularly for art because it's got good
+pfLP92UVzsy-00474-00249088-00249784 exporting, um, facilities basically that's my main, um, the main thing that I - I really
+pfLP92UVzsy-00475-00249784-00250344 tried out in 6 and couldn't couldn't get on board with. Everyone has their preferences I think! Um,
+pfLP92UVzsy-00476-00250471-00250992 I - if you if you're interested in the kind of the artistic stuff we've kind of
+pfLP92UVzsy-00477-00250992-00251480 started a #GeogebraArt hashtag so feel free to kind of peruse that to see what you know
+pfLP92UVzsy-00478-00251480-00251904 we've been doing in class, or people who have been in the class have then gone on to do, um,
+pfLP92UVzsy-00479-00252040-00252488 it's it's been it's been really good I really enjoyed teaching it and I think that you know
+pfLP92UVzsy-00480-00252488-00252976 it's covered quite a lot of, um, the basic tools so if you you know if you want to have
+pfLP92UVzsy-00481-00252976-00253776 a look at Geogebra, um, some of the art stuff is a place where you could start. Um, yeah I
+pfLP92UVzsy-00482-00253776-00254256 think - I - Ithink I'm nearly done with all my - oh I did have one last thing I wanted to show,
+pfLP92UVzsy-00483-00254312-00255160 um, I'm gonna have to do something else in a minute but I wanted to quickly show, um,
+pfLP92UVzsy-00484-00255304-00256152 I'm gonna - I think I've got to do a new share; let's try this; so this was something that I,
+pfLP92UVzsy-00485-00256152-00257119 um, invented recently? Um, so one of the things that Geogebra has is a spreadsheet, so
+pfLP92UVzsy-00486-00257208-00258104 I thought perhaps I could use the spreadsheet to create some interesting thing, um, and let's see
+pfLP92UVzsy-00487-00258104-00258671 if I can get this right so I've - I'm creating a slider right now, so in - in the same way that Ben
+pfLP92UVzsy-00488-00258671-00259376 used the slider for his, um, y = mx + c, um, I'm going to create a little slider with some options;
+pfLP92UVzsy-00489-00259376-00259936 so my slider goes between nought and one. Okay and then in my spreadsheet, um,
+pfLP92UVzsy-00490-00260119-00260704 so here - these are just descriptions of the points that I've already created, um,
+pfLP92UVzsy-00491-00260704-00261623 I'm then going to - I can draw a line; I can draw a polygon using the kind of commands.
+pfLP92UVzsy-00492-00261828-00262280 BEN: What's amazing to me is that the spreadsheet in Geogebra is fully functional and includes all
+pfLP92UVzsy-00493-00262280-00262640 of Geogebra's commands in it, and once you realise that the sort of -the potential is
+pfLP92UVzsy-00494-00262640-00263192 just amazing. BECKY: Yeah I seem to have, uh, typed something wrong... what have I done?
+pfLP92UVzsy-00495-00263296-00264240 Oh, I've got a polygon A2 B2 C2... it tells me it's not a polygon BEN: C2 is a point?
+pfLP92UVzsy-00496-00264320-00264744 um BECKY: There we go, um, so the nice thing about this obviously is if you've clearly made
+pfLP92UVzsy-00497-00264744-00265128 a mistake you can usually go back and fix it! So I've typed in the points a b and c here,
+pfLP92UVzsy-00498-00265192-00265648 and I've created the polygon; so now what I'm going to do, um, I'm going to show you where
+pfLP92UVzsy-00499-00265648-00266232 the function is so that I can type it in and you can all see it: so here, I want to create a point
+pfLP92UVzsy-00500-00266312-00266976 and I want it to be, um, between a and b and so, um,
+pfLP92UVzsy-00501-00267272-00267992 I've just got to - so I want it to be, um, based on the
+pfLP92UVzsy-00502-00268080-00268984 cells in the in the, um, spreadsheet so A2, um, plus a lots of
+pfLP92UVzsy-00503-00269232-00270280 um B2 minus A2. Now I'll just check - and hopefully it slides along that side,
+pfLP92UVzsy-00504-00270280-00270800 which it does, good. Okay, nice, and so I'm going to do the same over here, so I've got a point that
+pfLP92UVzsy-00505-00270800-00271280 slides up and down that side and then I kind of want to do the same here except there's going
+pfLP92UVzsy-00506-00271280-00271896 to be an error because I don't want it at point D2 I want to go back around my triangle. So as
+pfLP92UVzsy-00507-00271896-00272352 we find we've got this nice these three points which move proportionally along
+pfLP92UVzsy-00508-00272512-00273000 along those edges. sS now what I can do is I can say, "okay I'm going to, you know,
+pfLP92UVzsy-00509-00273000-00273480 like in a spreadsheet drag down my polygon" and some of you might have some idea where
+pfLP92UVzsy-00510-00273480-00274136 this is going - so with these quick commands I can just, you know, drag down a few times,
+pfLP92UVzsy-00511-00274240-00274792 and then I get something beautiful like this! And my favourite thing about this really
+pfLP92UVzsy-00512-00274880-00275496 is, you know, that you can play with the slider, but also, you know, if I want to
+pfLP92UVzsy-00513-00275552-00276224 rearrange how my original triangle was, um, I can move those points around. So I think that's
+pfLP92UVzsy-00514-00276224-00276840 I think that's done - done with my demos, um, (BEN: Thank you Becky!) there's a little, um,
+pfLP92UVzsy-00515-00277048-00277448 a little - if you want to - if you want to join my maths art - uh my Geogebra art class, do drop us a
+pfLP92UVzsy-00516-00277448-00277792 line - I'll put you on the waiting list but there are plenty of other people who are doing great
+pfLP92UVzsy-00517-00277792-00278480 things out there with Geogebra, um, and Alison is online and she told me she was going to be running
+pfLP92UVzsy-00518-00278480-00279112 a Geogebra workshop upcoming, um, sometime in December - didn't you say Alison, um, so
+pfLP92UVzsy-00519-00279168-00279808 I'm sure she will keep everyone posted on that as well, check the hashtag, follow me on, um, Twitter
+pfLP92UVzsy-00520-00279912-00280720 oh sorry Alison I've - I've just dumped you in it sorry, um, so at Twitter I'm @beckykwarren,
+pfLP92UVzsy-00521-00280720-00281296 um, and my email address is, um, i'll put my email address in there [types in chat]. BEN:
+pfLP92UVzsy-00522-00281296-00281792 While Becky's typing that I - I know that Becky has to go very shortly; if there are questions
+pfLP92UVzsy-00523-00281864-00282280 particularly for her we're going to have a little bit longer to tinker and I'll do some other demos
+pfLP92UVzsy-00524-00282280-00282680 for you, if anyone's still around, but before Becky goes now is the time to get your questions
+pfLP92UVzsy-00525-00282680-00283008 in the chat for Becky - or jump on a microphone to be honest, because there's not loads of us
+pfLP92UVzsy-00526-00283008-00283344 here - I hope that's okay Becky, if you've got a couple of minutes to see if there's anyone
+pfLP92UVzsy-00527-00283344-00283640 wanting to ask you anything based on what you've shown us? BECKY: Yes, five more minutes I've got!
+pfLP92UVzsy-00528-00283756-00284144 BEN: It turns out Becky has to do real work in California! JOHN: Can I just ask - John Bibby
+pfLP92UVzsy-00529-00284144-00284496 here - there's one specific question but the context is, um,
+pfLP92UVzsy-00530-00284560-00284992 this is a project I've rather parked because I've got myself in several dead ends,
+pfLP92UVzsy-00531-00284992-00285384 some of them technological and this has been very useful - it's helped me brush off some sort of 10
+pfLP92UVzsy-00532-00285384-00285952 year old cobwebs from my knowledge of Geogebra but, um, I'm trying to build a bridge between on
+pfLP92UVzsy-00533-00285952-00286448 the one hand statistical visualisations where you've got data, and on the other hand things
+pfLP92UVzsy-00534-00286448-00287056 like a generalised version of Ulam's spiral, which I'm sure you know about, and the technology I want
+pfLP92UVzsy-00535-00287056-00287512 is in Excel and I'm just wondering whether it's here - it's uh to have a spreadsheet in which you
+pfLP92UVzsy-00536-00287512-00288000 can get the cells basically square and then you can colour the cells according to the value that
+pfLP92UVzsy-00537-00288000-00288584 is in the cell. Is that something you can do in Geogebra? BEN: Do you want to tackle that Becky?
+pfLP92UVzsy-00538-00288816-00289168 BECKY: Ooh, I'm not sure - I haven't really played with that - the spreadsheet functionality
+pfLP92UVzsy-00539-00289168-00289696 in that way, but potentially? BEN: I think what John is referring to is conditional formatting
+pfLP92UVzsy-00540-00289696-00290424 which certainly exists in Excel I don't think it works in the Geogebra spreadsheet; but what
+pfLP92UVzsy-00541-00290488-00290872 the Geogebra spreadsheet has an advantage for is that anything you put in a spreadsheet could
+pfLP92UVzsy-00542-00290872-00291368 be represented in the graphics view visually - so there'd be a way to make those cells refer
+pfLP92UVzsy-00543-00291368-00291936 to something in a coordinate axis but not quite as simple as colouring the cells according to
+pfLP92UVzsy-00544-00291936-00292344 what's in there. That -that, you're probably better off staying in Excel - Excel is still
+pfLP92UVzsy-00545-00292344-00292728 a valid spreadsheet. JOHN: Yeah there's other things I want to do that I can't do in Excel,
+pfLP92UVzsy-00546-00292728-00293128 but anyway I might come back to you with an Excel question, thank you very much. BEN:
+pfLP92UVzsy-00547-00293128-00293456 There's - I don't know if Necky's found this but there's almost certainly - if someone says
+pfLP92UVzsy-00548-00293456-00293896 "can Geogebra do such and such?" the answer is "yes", but, uh, you've got to find the right
+pfLP92UVzsy-00549-00293896-00294216 button - or if you can't find the right button you write to Michael Borcherds,
+pfLP92UVzsy-00550-00294216-00294488 who's the lead developer of Geogebra and say "I wish Geogebra would do this", and usually within
+pfLP92UVzsy-00551-00294488-00294824 20 minutes he responds on Twitter and says "yeah, just update your version"
+pfLP92UVzsy-00552-00294824-00295256 Yeah. I'm kind of exaggerating slightly. JOHN: The answers I've had so far are just "oh, John,
+pfLP92UVzsy-00553-00295256-00295856 it's easy you can write it in Python" but for that I have to brush away my Python cobwebs BEN: I feel
+pfLP92UVzsy-00554-00295856-00296400 those cobwebs as well! Geogebra is like learning a programming language - their similarity is quite
+pfLP92UVzsy-00555-00296400-00296792 strong but it's kind of a geometrical programming language and an algebraic programming language,
+pfLP92UVzsy-00556-00296792-00297128 more of which i'll show in a minute; before Becky has to go, uh, thanks for your question
+pfLP92UVzsy-00557-00297128-00297544 John - I'm sorry we can't answer it satisfactorily yet. Anything else for Becky before she goes?
+pfLP92UVzsy-00558-00297704-00298064 Katie has just put a comment - you can make points change colour based on their properties;
+pfLP92UVzsy-00559-00298064-00298480 it's called dynamic colouring, and I can show an example of that in a moment, uh, if needed.
+pfLP92UVzsy-00560-00298760-00299488 Questions for Becky are, um, now - now is the time. BECKY: You can always email me questions
+pfLP92UVzsy-00561-00299488-00300244 if you've got anything that comes up later, um, always happy to hear from Geogebra fans. BEN:
+pfLP92UVzsy-00562-00300312-00300808 I thoroughly recommend following Becky on Twitter, even if it's just to see nice artwork regularly
+pfLP92UVzsy-00563-00300808-00301312 animating all over your Twitter feed - it's wonderful, and I haven't done her course yet on
+pfLP92UVzsy-00564-00301312-00301768 using art to make Geogebra but even if you're happy with Geogebra, being sort of given the
+pfLP92UVzsy-00565-00301768-00302200 discipline of making art with it is a really good way to get better at anything and that that's true
+pfLP92UVzsy-00566-00302200-00302768 whether it's Geogebra or using paints - or so I'm told but, um, Becky is one of the leading lights
+pfLP92UVzsy-00567-00302768-00303320 of making nice art on Geogebra with - in Twitter and, so go and follow her I think. SAM: It's
+pfLP92UVzsy-00568-00303320-00303792 probably worth saying as well how pleasing that was just to watch, so thank you very very much
+pfLP92UVzsy-00569-00303792-00304256 Becky - that totally made my day. BEN: Did I - did I say it the wrong way round? Kevin's just
+pfLP92UVzsy-00570-00304256-00304640 asking - "using art to make Geogebra"! That's clearly why Geogebra exists - it's - it's an
+pfLP92UVzsy-00571-00304640-00305220 artistic creation. Becky: anything you want to say before we we leave you to your real job? BECKY:
+pfLP92UVzsy-00572-00305272-00305688 No, this has been - it's been really lovely though and yeah, maybe i'll just you know start a
+pfLP92UVzsy-00573-00305688-00306080 YouTube streaming where I just do things on Geogebra and you can just like have it on in
+pfLP92UVzsy-00574-00306080-00306496 the background as a kind of relaxation - you know? BEN: I genuinely think there's a market
+pfLP92UVzsy-00575-00306496-00306952 for that! Let's keep moving because I want to give people another chance to vote on what
+pfLP92UVzsy-00576-00306952-00307168 else they want to ask about - BECKY: Well, it's been lovely to see you all BEN: Becky thank you
+pfLP92UVzsy-00577-00307592-00307944 and uh yeah Becky over in California (BECKY: Thanks everyone!) has to start a real day
+pfLP92UVzsy-00578-00307944-00308472 of work now. We, officially, uh, these previous sessions have been running for an hour;
+pfLP92UVzsy-00579-00308472-00308816 we did warn, and I hope you got the message that this one might run a bit longer. I have
+pfLP92UVzsy-00580-00308816-00309440 got other demos I can show but I don't want to do things that are irrelevant for people, so I think
+pfLP92UVzsy-00581-00309440-00309912 I would like to give you a vote on what I show you next; you are of course free to leave at any point
+pfLP92UVzsy-00582-00309912-00310416 if you are, uh, Geogebra-ed out, although I don't know how that happens - it's never happened to me
+pfLP92UVzsy-00583-00310416-00310784 SAM: Or actually have real world things to do? BEN: Well, there's always that;
+pfLP92UVzsy-00584-00310784-00311072 so I'm just gonna - there are several things I'd like to show; um,
+pfLP92UVzsy-00585-00311128-00311472 Katie mentioned to me in a private message and I think it is worth demonstrating that you - you can
+pfLP92UVzsy-00586-00311472-00311936 do colouring things, and the other thing I haven't mentioned a lot of is that the the name Geogebra
+pfLP92UVzsy-00587-00311936-00312496 is a portmanteau of geometry and algebra, and a lot of what we've talked about so far
+pfLP92UVzsy-00588-00312496-00313016 is more geometrical. And I want to point out that the algebraic functions in geogebra are strong
+pfLP92UVzsy-00589-00313016-00313408 and fundamentally integrated with the geometry, and that's what makes it strong and I would like
+pfLP92UVzsy-00590-00313408-00313896 to demo something of that, but: I would - if I put this poll up -I'm actually going to share
+pfLP92UVzsy-00591-00313896-00314584 four options that I've just written in Word, so let's just do this, um, okay are you seeing,
+pfLP92UVzsy-00592-00314584-00315224 uh, four options with with four words and I wonder if we could start a poll? Would you like me to
+pfLP92UVzsy-00593-00315312-00315664 do a demo that involves something to do with the Mandelbrot set? Uh, do a demo that
+pfLP92UVzsy-00594-00315664-00316048 involves something with sunflower spirals? And those two are demos - I won't necessarily talk
+pfLP92UVzsy-00595-00316048-00316576 about how I built those files ,although I can do if there's long enough; uh I would
+pfLP92UVzsy-00596-00316576-00317088 also like to show you things that are much more sort of algebraic, so the gradient function,
+pfLP92UVzsy-00597-00317088-00317520 which is code for differentiation maybe from first principles; and maybe a Maclaurin or Taylor
+pfLP92UVzsy-00598-00317520-00318136 series -these are not necessarily, um, enrichment ideas but they are nice ways of visualising bits
+pfLP92UVzsy-00599-00318136-00318472 of maths that you might already know about but seeing them move makes a difference; and those
+pfLP92UVzsy-00600-00318472-00318960 two I can build from scratch, uh, live in front of you at quite some speed - but remember this is
+pfLP92UVzsy-00601-00318960-00319472 being recorded. So if you would like to vote, um, why is the poll not working? That's interesting.
+pfLP92UVzsy-00602-00319784-00320168 I'm gonna - I'm gonna launch it again; hopefully you're now able to vote for one of those things.
+pfLP92UVzsy-00603-00320168-00320624 I'm not gonna take this as gospel - I won't commit myself to doing all of these but if - if
+pfLP92UVzsy-00604-00320624-00321096 two of them are sort of sharing the vote I'll try and do those two before we finish,
+pfLP92UVzsy-00605-00321096-00321600 and I'll make sure that I leave a little bit of time before 5.30 for some further questions.
+pfLP92UVzsy-00606-00321600-00322112 There are lots of things to learn about Geogebra, but as long as you don't treat this as your only
+pfLP92UVzsy-00607-00322112-00322432 chance to learn it then you'll be fine. This is an introduction to Geogebra in the end;
+pfLP92UVzsy-00608-00322488-00323032 the, uh, I'm going to give a couple more seconds to the voting -several people - at the moment,
+pfLP92UVzsy-00609-00323032-00323408 you can't see the results, but at the moment the gradient function construction is winning, but the
+pfLP92UVzsy-00610-00323408-00323904 votes are spread around all of those options; so I think I'm going to start by showing you a gradient
+pfLP92UVzsy-00611-00323904-00324432 function construction and if we have time I'll do another demo possibly of possibly the Maclaurin
+pfLP92UVzsy-00612-00324432-00324848 series - that's also doing well on the votes, um, and there'll be time for further questions
+pfLP92UVzsy-00613-00324848-00325408 as well, so let me stop sharing a screen; Sam, are you happy for me to carry on with a demo? I'm
+pfLP92UVzsy-00614-00325408-00325816 just checking that there's nothing else burning in the chat that needs to be addressed. Okay,
+pfLP92UVzsy-00615-00325992-00326440 uh Alison is commenting about singing Smash Mouth: that's an entirely different
+pfLP92UVzsy-00616-00326440-00326952 workshop. Okay, uh, I'm trying to SAM:Wait, you do a workshop on that?
+pfLP92UVzsy-00617-00327084-00327544 BEN: Not yet but you know, say the word and send me, uh, send me the money and
+pfLP92UVzsy-00618-00327656-00328256 I will pollute your ears. I am hopefully sharing a blank Geogebra screen once again - give me a nod
+pfLP92UVzsy-00619-00328256-00328792 if you're seeing that. Okay, quick context, uh, the vote was strongest for the gradient function
+pfLP92UVzsy-00620-00328792-00329280 construction, so I'm going to construct a file which allows me to talk about what the gradient of
+pfLP92UVzsy-00621-00329280-00329744 a function might mean, particularly for students - or just generally interested people - who are
+pfLP92UVzsy-00622-00329744-00330248 used to gradients being of straight lines and not of curved functions. So the algebraic things of
+pfLP92UVzsy-00623-00330248-00330824 Geogebra are strong; I can do functions I can type f of x equals; I'm going to pick x cubed subtract
+pfLP92UVzsy-00624-00330824-00331240 x which is a nice curvy looking function. A couple of things about Geogebra to realise:
+pfLP92UVzsy-00625-00331240-00331664 using a mouse wheel - and again, this is why I like the downloaded version; the- the keyboard
+pfLP92UVzsy-00626-00331664-00332064 shortcuts are better - but I could just zoom in, so I'm scrolling my mouse wheel to zoom in. There
+pfLP92UVzsy-00627-00332064-00332472 are zoom tools but you don't need to click on them - just use a mouse wheel; um, you can change the
+pfLP92UVzsy-00628-00332472-00332920 scale - cubic functions tend to go quickly; I want to spread this out - if I just drag around
+pfLP92UVzsy-00629-00332920-00333384 the screen I'm moving it, but if I hold down shift (you can't see me holding down shift)
+pfLP92UVzsy-00630-00333448-00333888 and I drag one of the axes, you can change the scale of one axis. And that's quite a crucial
+pfLP92UVzsy-00631-00333888-00334376 keyboard shortcut to be able to scale what you see to fit nicely; uh, and actually I want to go up
+pfLP92UVzsy-00632-00334376-00334920 again. Maybe there: so there's the nice curvy bit of a cubic function. I'm going to create a point
+pfLP92UVzsy-00633-00334920-00335280 on this function; I'm going to use the point tool - if you use the point tool on a background you
+pfLP92UVzsy-00634-00335280-00335808 just create a point, but if I put a point on the function it makes a point on the function and it
+pfLP92UVzsy-00635-00335808-00336280 only moves on the function. That's already quite useful; what I'd really like to know is what's the
+pfLP92UVzsy-00636-00336280-00336856 gradient of this function and to do that we have to realise that it's going to change because it's
+pfLP92UVzsy-00637-00336856-00337440 curved - it's not a straight line. There is a tangent tool in Geogebra; so if I click on the
+pfLP92UVzsy-00638-00337440-00337864 third option here the tangent tool is here; if I don't know how to use it I can hover over it;
+pfLP92UVzsy-00639-00337864-00338280 "select point or line then circle, conic or function"; it's a little bit cryptic to be honest,
+pfLP92UVzsy-00640-00338280-00338648 but in this case it's really saying pick a point and then a function and it'll draw the
+pfLP92UVzsy-00641-00338648-00339144 tangent to that function at that point in this case, that point, this function. Boom!
+pfLP92UVzsy-00642-00339408-00339792 And it moves. And to be honest if I didn't like dragging it around,
+pfLP92UVzsy-00643-00339792-00340200 I right click on it and click animate on, and I just - again I walk away and let people watch
+pfLP92UVzsy-00644-00340200-00340704 this. Already I think we have a fertile discussion of what the gradient of a curve function is;
+pfLP92UVzsy-00645-00340704-00341024 because it's moving, it's capturing all the generality at once and I could go on about
+pfLP92UVzsy-00646-00341024-00341488 this - I really think we have responsibility as educators and demonstrators of mathematics
+pfLP92UVzsy-00647-00341488-00342000 to make it move, because we haven't been able to do that in previous generations of maths outreach
+pfLP92UVzsy-00648-00342088-00342472 but now it's free on Geogebra and Desmos and all those things. I'm aware this is
+pfLP92UVzsy-00649-00342472-00342936 not yet the gradient function; what I'd like to do is track the gradient of this function
+pfLP92UVzsy-00650-00342936-00343248 at a particular point, and there's a couple of subtleties, so I'm going to stop the animation
+pfLP92UVzsy-00651-00343248-00343664 just for now - I'm going to maybe drag it back up here. There's a command buried in Geogebra which
+pfLP92UVzsy-00652-00343664-00344072 it took me ages to find - and it's brilliant, but it's not in the help files because it's so basic;
+pfLP92UVzsy-00653-00344128-00344640 and that command is how to grab the x-coordinate or in fact the y-coordinate of a point on your
+pfLP92UVzsy-00654-00344640-00345112 screen. I would like to get the x coordinate of a, because that's kind of where I am in the function
+pfLP92UVzsy-00655-00345112-00345656 and the command for that is x brackets a. Uh it's such a short command that it's really not in there
+pfLP92UVzsy-00656-00345656-00346032 but if I press return on that you can see it's made a little thing in the algebra window here:
+pfLP92UVzsy-00657-00346032-00346408 negative 0.8, which is roughly where the x coordinate is. I didn't need to make that - I
+pfLP92UVzsy-00658-00346408-00346864 could have just used the command x of a whenever I want to grab the x coordinate of a; what I would
+pfLP92UVzsy-00659-00346864-00347488 like to do is plot the point which has the same x coordinate as a, but in fact has the same y
+pfLP92UVzsy-00660-00347488-00348152 coordinate as the slope of that line. So to do it the long way around there's a slope tool which is
+pfLP92UVzsy-00661-00348152-00348504 nice - remember what I said about Geogebra: if you're wondering "can it do that?" the answer
+pfLP92UVzsy-00662-00348504-00348952 is "yes, probably it's in one of these buttons", and if it's not it might be in a command - there's
+pfLP92UVzsy-00663-00348952-00349392 literally thousands of commands buried in there; I'm not going to expand all these - there's loads
+pfLP92UVzsy-00664-00349392-00349744 of commands in here and you can start typing them. For example if you start typing that you
+pfLP92UVzsy-00665-00349744-00350296 get a list of command options coming up. There's a lot to explore here; we can't do it all today;
+pfLP92UVzsy-00666-00350296-00350624 what I really want is the slope command, and there is a button, there's also a command for
+pfLP92UVzsy-00667-00350624-00351008 it and I want the slope of this line. And you can see it's drawn me the slope of that line,
+pfLP92UVzsy-00668-00351008-00351432 which is nice - and it changes as we move around. It's drawn the sort of gradient triangle
+pfLP92UVzsy-00669-00351536-00351968 and it's given a value m - that's presumably because of the old mx + c convention;
+pfLP92UVzsy-00670-00352032-00352408 in that case though what I would like to do is plot a coordinate which has the x coordinate
+pfLP92UVzsy-00671-00352408-00352952 the same as a but its y coordinate is in fact the gradient, i.e. m - and this is the sort
+pfLP92UVzsy-00672-00352952-00353408 of subtle bit: just type it in I would like a coordinate, that's normally got two brackets;
+pfLP92UVzsy-00673-00353408-00353800 I would like the x bit to be x of a, and I'd like the y coordinate to be the number m.
+pfLP92UVzsy-00674-00354064-00354472 There's no command for this: I just type it in. Once you realise that Geogebra will just let
+pfLP92UVzsy-00675-00354472-00354808 you build what you want, then you start seeing the potential power here,
+pfLP92UVzsy-00676-00354864-00355120 and it's going to move - so as I drag a you can see b moves,
+pfLP92UVzsy-00677-00355192-00355504 and I hope you recognise what sort of shape it's moving in - just to save a bit of time
+pfLP92UVzsy-00678-00355504-00356024 I'm going to animate a and you can see that b is moving in what I might call the gradient function
+pfLP92UVzsy-00679-00356024-00356464 of the cubic curve. I hope you recognise what shape b is: it would be lovely if we could
+pfLP92UVzsy-00680-00356464-00357040 leave b sort of making a trace wouldn't it? Good - luckily Geogebra can do that. Right click on b,
+pfLP92UVzsy-00681-00357040-00357472 turn on the trace and as it comes back on the screen you'll see b leaving a track wherever it
+pfLP92UVzsy-00682-00357472-00357992 goes. I hope - here it comes. Whee! Don't know why I said whee - it feels like a nice moving thing.
+pfLP92UVzsy-00683-00358136-00358576 Okay: cut long story short I've basically differentiated the cubic function, by tracking the
+pfLP92UVzsy-00684-00358576-00358936 gradient of the cubic function, and lo and behold it looks like a quadratic function. There's a lot
+pfLP92UVzsy-00685-00358936-00359352 to talk about - you can imagine teaching this is quite powerful; the movement is good - uh,
+pfLP92UVzsy-00686-00359352-00359784 one other command though you might want in Geogebra is: first of all, the trace function is
+pfLP92UVzsy-00687-00359784-00360216 very ephemeral - it disappears as soon as you move around; it's just a really - sort of - it's there
+pfLP92UVzsy-00688-00360216-00360616 until you drag and then it vanishes. What I'd like is sort of more concrete version and to do that
+pfLP92UVzsy-00689-00360616-00361048 there's a locus command. This is surprisingly useful. Anything that moves as a result of
+pfLP92UVzsy-00690-00361048-00361504 something else moving ,you can leave a permanent trace if you like. So this locus command asks
+pfLP92UVzsy-00691-00361504-00362080 for 'Select a locus point and then a point on an object or slider'. In this case the point that you
+pfLP92UVzsy-00692-00362080-00362656 want to trace out, and the point that controls it. So in this case I want to trace out b as a moves;
+pfLP92UVzsy-00693-00362744-00363440 there it is, and that is the gradient function as a locus. To cut a long story short: I could change
+pfLP92UVzsy-00694-00363440-00364184 the original function - double click this; let's make it sin x. What will the gradient function be?
+pfLP92UVzsy-00695-00364328-00364880 It will be cosine x, and because I've constructed it, everything updates as I change the original
+pfLP92UVzsy-00696-00364880-00365224 inputs. Uh, obviously we have to deal with the fact that we're working in radians here - I'm
+pfLP92UVzsy-00697-00365224-00365640 not going to tackle that unless anyone wants me to talk about degrees versus radians later on;
+pfLP92UVzsy-00698-00365640-00366064 nice way to explore it. Final thing though: Geogebra is actually an algebraic program as
+pfLP92UVzsy-00699-00366064-00366472 well. This - I've kind of constructed this geometrically by talking about coordinates,
+pfLP92UVzsy-00700-00366472-00366984 but it will actually differentiate a function - f is sin x; I can type in f dashed of x
+pfLP92UVzsy-00701-00367080-00367496 and it will plot it anyway. So if I hide the locus and everything else you can see that it knows the
+pfLP92UVzsy-00702-00367496-00367896 differential of sine is cosine and everything else. So Geogebra can do differentiation
+pfLP92UVzsy-00703-00367896-00368368 actually analytically as well - in fact it has a computer algebra system built into it under the
+pfLP92UVzsy-00704-00368368-00368800 CAS thing here, but that's another can of worms which I don't want to open right now because we
+pfLP92UVzsy-00705-00368800-00369488 will need all night and I don't have all night. All I've done here is build a gradient function
+pfLP92UVzsy-00706-00369488-00369984 construction. And actually I feel like doing this live - actually building the construction
+pfLP92UVzsy-00707-00369984-00370472 in front of an audience - has a little bit more value than presenting a finished file like this.
+pfLP92UVzsy-00708-00370552-00371000 But the classroom link I shared earlier does have a finished version of this gradient
+pfLP92UVzsy-00709-00371000-00371408 function thing, and if you want to follow that classroom link I'll make sure it's unpaused after
+pfLP92UVzsy-00710-00371408-00371784 this session; you can go and have a play with my demonstration file and maybe try and build your
+pfLP92UVzsy-00711-00371784-00372216 own version. Obviously you could just do that on your standalone copy of Geogebra as well,
+pfLP92UVzsy-00712-00372216-00372736 but I'll make sure the demo file for this is available. So I promised I'd do that one,
+pfLP92UVzsy-00713-00372736-00373216 um, are there any questions about this gradient function construction or anything you wish that
+pfLP92UVzsy-00714-00373216-00373688 it could do now we've built this - anything else you want to ask? I've got more demos in me but
+pfLP92UVzsy-00715-00373688-00374248 I don't want to overload you. B- by the way, if you want to ask a question it's probably
+pfLP92UVzsy-00716-00374248-00374648 quicker to jump on the microphone than the chat and you're welcome just to jump on the mic and
+pfLP92UVzsy-00717-00374648-00375096 ask me anything you want. KEVIN: Yeah, I was I was going to be wrangling the questions but we
+pfLP92UVzsy-00718-00375096-00375712 haven't got any questions in the cha,t so yeah just just feel free to uh ask Ben a question.
+pfLP92UVzsy-00719-00375836-00376376 BEN: I'm interested in, um, what people didn't know already who - I know that Alison and Sam
+pfLP92UVzsy-00720-00376376-00376704 have used Geogebra before but there's a few things like that shift drag is- is a useful
+pfLP92UVzsy-00721-00376704-00377144 tip I'm glad that was new; uh Sam has written down the summary of the - that x coordinate
+pfLP92UVzsy-00722-00377144-00377576 piece of advice is really really helpful to get a piece of algebra out of a point...
+pfLP92UVzsy-00723-00377736-00378296 Okay, the the second demonstration that won on the vote was the Maclaurin Taylor series
+pfLP92UVzsy-00724-00378296-00378640 and since that is a construction I can do from scratch I will I will try and do that in the
+pfLP92UVzsy-00725-00378640-00379088 next five minutes, and then we've officially got a quarter of an hour for any other questions and
+pfLP92UVzsy-00726-00379088-00379536 if you want me just to do more demos - you know, I did warn everyone that, uh, we could
+pfLP92UVzsy-00727-00379536-00380016 be here all night but I will finish at half past unless someone wants to pay me to stay over...
+pfLP92UVzsy-00728-00380132-00380672 FRAN: Ben, are you able to build into whatever you're going to show us the bit that Katie
+pfLP92UVzsy-00729-00380672-00381072 mentioned about being able to... dynamic colouring for points? BEN: That's a very good point - so let
+pfLP92UVzsy-00730-00381072-00381400 me show you the demo file for this one I built, which I spent a little bit time- more time
+pfLP92UVzsy-00731-00381400-00381832 prettying up and including some dynamic colouring; I'm going to stop sharing this and show you my
+pfLP92UVzsy-00732-00381888-00382472 demo file. I mean there's a lot to be said for the difference between one you build on the fly
+pfLP92UVzsy-00733-00382472-00383056 and the one you spend time prettying up, which is my code for paying attention to the aesthetics.
+pfLP92UVzsy-00734-00383296-00383816 Okay let me just share the right thing here: sharing my second screen, this is the, uh,
+pfLP92UVzsy-00735-00383816-00384104 the gradient function demo - give me a nod if you're seeing a more complicated looking screen.
+pfLP92UVzsy-00736-00384264-00384632 Okay in the online - you can always upload online resources to Geogebra and then they're
+pfLP92UVzsy-00737-00384632-00385000 free to share; you can see this thing is animating away, I hope, in a similar
+pfLP92UVzsy-00738-00385000-00385384 way but I've done a little bit more care and attention - you can see some tick boxes exist,
+pfLP92UVzsy-00739-00385384-00385840 and I've got an input box to change the function I've got, sort of, I can click the trace on,
+pfLP92UVzsy-00740-00385840-00386152 I can show b and that so it turns on b. Now something else is happening with the
+pfLP92UVzsy-00741-00386152-00386696 colouring of that b point, and that's what I want to demo - although one other comment is that when
+pfLP92UVzsy-00742-00386696-00387048 you upload a file to Geogebra, there's always this little button on the bottom right which
+pfLP92UVzsy-00743-00387048-00387512 maximises it onto a screen and that's really useful for demonstrating in a classroom
+pfLP92UVzsy-00744-00387512-00387984 or when you're sharing over Zoom because it sort of fills the available space.
+pfLP92UVzsy-00745-00388088-00388552 So can you just uh confirm to me by nodding on the video again if you are seeing this animating file
+pfLP92UVzsy-00746-00388552-00389064 now - and the trace seems to change colour. Uh was it Fran who asked about the changing colour?
+pfLP92UVzsy-00747-00389144-00389624 So Fran, what you'll notice is that I've made the point b, which is my gradient trace,
+pfLP92UVzsy-00748-00389688-00390216 change colour at a certain point and I wonder if it's obvious to everyone when it changes
+pfLP92UVzsy-00749-00390216-00390684 colour - jump on the mic and tell me: just, I need some interaction at this point. ALISON:
+pfLP92UVzsy-00750-00390760-00391352 Whether it's positive or negative. BEN: Yeah, and when you said it I know you mean the gradient of
+pfLP92UVzsy-00751-00391352-00391744 the original function. And the reason I've done that is if I was trying to get a class introduced
+pfLP92UVzsy-00752-00391744-00392120 to the concept of gradient, I might, before I end up tracing b (and I'm going to just turn off
+pfLP92UVzsy-00753-00392120-00392592 that trace now) I might ask them to sort of put their hands up whenever they think the gradient
+pfLP92UVzsy-00754-00392592-00393008 is positive and then put them down when it's negative, and just get the whole class doing that,
+pfLP92UVzsy-00755-00393008-00393424 and even an audience if I was going to do that - and it's kind of a fun way to force them to
+pfLP92UVzsy-00756-00393424-00393992 engage with the dynamic nature. So, hand up and down, and it's still down, it's still down...
+pfLP92UVzsy-00757-00393992-00394376 and now it's up again - and it forces them to realise when it's positive and negative, and
+pfLP92UVzsy-00758-00394376-00394904 then I can capture that by showing b that point I've made and I've made it change colour when
+pfLP92UVzsy-00759-00394904-00395424 that changes. So how did I do that? Let me quickly show you that to do that. I need to get hold of b.
+pfLP92UVzsy-00760-00395656-00395968 If you right click on anything in Geogebra, as long as you're in a sort of decent version of
+pfLP92UVzsy-00761-00395968-00396304 Geogebra and it's not sort of locked down web version, then you can get to the properties of
+pfLP92UVzsy-00762-00396304-00396696 the thing - and the web version is a bit clunky; it kind of takes up loads of screen space which
+pfLP92UVzsy-00763-00396696-00397176 is another reason why I like the downloaded version, but in the advanced tab of the properties
+pfLP92UVzsy-00764-00397232-00397744 there are some dynamic colours options, and the crucial thing is here if you understand about RGB
+pfLP92UVzsy-00765-00397744-00398200 values - you can change it to HSV values - that's just a way of specifying a colour. The way you
+pfLP92UVzsy-00766-00398200-00398728 can specify the colour can have variables in it. And inn this case the variable a is involved and
+pfLP92UVzsy-00767-00398728-00399352 I've got a situation here where a is - I think - a hundred: I've chosen that, um, and let's have a
+pfLP92UVzsy-00768-00399352-00400056 look at what a is; a is an if statement: I've made it if the gradient is zero - bigger than zero,
+pfLP92UVzsy-00769-00400056-00400536 then a is a hundred and if it is not bigger than zero a becomes zero. So this is a sort
+pfLP92UVzsy-00770-00400536-00401000 of almost a command in a programming language - it's actually a Geogebra command: an if statement;
+pfLP92UVzsy-00771-00401080-00401440 you have to get used to that, but then I'm using that value a to determine the properties
+pfLP92UVzsy-00772-00401440-00402040 of the colour of b. Jumping back to that, you can see that when a is 100 that becomes one
+pfLP92UVzsy-00773-00402112-00402552 and the others becomes zero, I think, and then when a is zero that becomes zero and these become
+pfLP92UVzsy-00774-00402552-00402960 100 so it changes from fully red to a mixture of green and blue
+pfLP92UVzsy-00775-00402960-00403552 which apparently is - whatever colour it is (being slightly colourblind I'm not sure)
+pfLP92UVzsy-00776-00403728-00404352 What color is that? Teal? Should we go with teal? I'm just guessing words. FRAN: Aqua! BEN:
+pfLP92UVzsy-00777-00404352-00404872 Aqua - that's a new word for me. Good. So, um, I realise that was not a comprehensive tutorial but
+pfLP92UVzsy-00778-00404872-00405320 the fact that there is a dynamic colouring option buried in the settings is kind of what you need to
+pfLP92UVzsy-00779-00405320-00405696 know, and then go and play with it and make it change based on another variable. Everything on
+pfLP92UVzsy-00780-00405696-00406128 Geogebra is based on variables -almost everything you can type into you can set - you can type in a
+pfLP92UVzsy-00781-00406128-00406488 number, or you could type in a variable that changes based on something else changing.
+pfLP92UVzsy-00782-00406552-00406936 The other thing to mention is that I've done some tick boxes - uh, someone, I think, uh,
+pfLP92UVzsy-00783-00407056-00407408 who was it in the, uh, was it Claire in the chat was saying tick boxes look
+pfLP92UVzsy-00784-00407408-00407944 complicated - there is a tool for tick boxes and it's relatively easy to use. There is a slight
+pfLP92UVzsy-00785-00407944-00408320 catch in how it does it behind the scenes but to cut a long story short
+pfLP92UVzsy-00786-00408472-00409024 every func- every object has a property buried behind it; so I'm going to get the properties of
+pfLP92UVzsy-00787-00409024-00409616 this function x... f, sorry - and in the advanced tab there is a condition to show the object.
+pfLP92UVzsy-00788-00409688-00410064 And what you actually build with the tick box is a boolean - a true or false, whether it's
+pfLP92UVzsy-00789-00410064-00410568 ticked or not - and you make that tick or cross change this thing here and that's how to change
+pfLP92UVzsy-00790-00410672-00411392 a hide or show thing. But the tech - the tick box button in here actually does that for you, um,
+pfLP92UVzsy-00791-00411392-00411864 which Katie learned the hard way, by doing it really the proper way ,behind the scenes,
+pfLP92UVzsy-00792-00411864-00412352 and then realising there is eventually a shortcut to it. Happy to explain that more later on, um,
+pfLP92UVzsy-00793-00412480-00412856 I said I'd do one more demo but since that question from Fran was a helpful way to make sure
+pfLP92UVzsy-00794-00412856-00413296 I said I'd do what I said about the colouring, any other questions in that vein while I'm demoing
+pfLP92UVzsy-00795-00413296-00414039 this? SAM H: Ben, in the number there you used an if statement; (BEN: Yes.) Can you do that in point
+pfLP92UVzsy-00796-00414039-00414568 definitions and in other places as well? Can you do it anywhere? BEN: Yes. Remember: the questions
+pfLP92UVzsy-00797-00414568-00415216 that start with "can Geogebra do...?" - yes. Yes, they can. So I could create a point which is -
+pfLP92UVzsy-00798-00415216-00415839 I'm gonna say - if m is greater than zero, I could create a point at (1,0) - I don't-
+pfLP92UVzsy-00799-00416080-00416472 this is- I don't know why you'd want to - and if it's not, make a point at (2,0);
+pfLP92UVzsy-00800-00416600-00417160 what I'm doing is writing an if statement there that depends on m and will create a point at (1,0)
+pfLP92UVzsy-00801-00417239-00417743 or at (2,0) based on that thing - so I'm kind of making points come out of the if statement;
+pfLP92UVzsy-00802-00417743-00418184 so if I - if I do that, you can see that point c is turned up and it will jump over there.
+pfLP92UVzsy-00803-00418320-00418824 So the - the- that point c depends on an if statement I built. Once you realise that
+pfLP92UVzsy-00804-00418824-00419160 sort of power is available you realise this is a proper programming language - in fact everything
+pfLP92UVzsy-00805-00419160-00419752 in Geogebra is a command in this input bar; what they've actually done is made the tools at the top
+pfLP92UVzsy-00806-00419752-00420168 shortcuts for some commands but there's a lot of tools available that are not picture tools
+pfLP92UVzsy-00807-00420168-00420568 but they are in the command bar available; uh, some really complicated stuff, one of which
+pfLP92UVzsy-00808-00420568-00421047 i'll show you in a moment to do with a Taylor series expansion - there is no button for that,
+pfLP92UVzsy-00809-00421112-00421424 but that's a really nice question from Sam - 'if' statements are powerful and
+pfLP92UVzsy-00810-00421424-00421760 you can use them for anything. Any other questions on this one?
+pfLP92UVzsy-00811-00421992-00422184 I am rushing you through Geogebra 101 here.
+pfLP92UVzsy-00812-00422424-00422704 Ooh, Katie's called the colour 'cyan' just to update you on that. I'm just,
+pfLP92UVzsy-00813-00422704-00423088 uh, catching up. Okay I'd like to do one more demo and then I'm going to stop
+pfLP92UVzsy-00814-00423088-00423400 demoing at you even though I love it so much.
+pfLP92UVzsy-00815-00423728-00424128 Keep the questions coming - you're of course free just to interrupt me at any point just to uh
+pfLP92UVzsy-00816-00424208-00424504 ask me something interesting. I'm going to start with a brand new
+pfLP92UVzsy-00817-00424504-00424743 Geogebra file and I'm going to share that with you now.
+pfLP92UVzsy-00818-00425168-00425784 Can you see how much I'm sweating on the camera? I shouldn't shouldn't have pointed that out.
+pfLP92UVzsy-00819-00425880-00426176 Okay give me a nod if you're seeing a blank looking Geogebra file again?
+pfLP92UVzsy-00820-00426304-00427072 Lovely. Okay this time I'm going to create a - let's - let's do - let's do sin x again; I think
+pfLP92UVzsy-00821-00427072-00427696 that's a good function to play with. And I want to point out that you can create a series - this
+pfLP92UVzsy-00822-00427696-00428192 is a function which is clearly not polynomial but you can pretend that it is a polynomial:
+pfLP92UVzsy-00823-00428192-00428768 for example the polynomial y equals x is quite close to it. Can you see that? I - I
+pfLP92UVzsy-00824-00428768-00429128 realise it's not very obviously the same - it doesn't wiggle, but if I zoom in close enough
+pfLP92UVzsy-00825-00429192-00429496 they're actually ridiculously close and if you didn't know already that sine x
+pfLP92UVzsy-00826-00429496-00429984 is approximated quite well by the function x at small values then now you can see it visually.
+pfLP92UVzsy-00827-00430039-00430647 It's better approximated by the function x subtract x cubed over three,
+pfLP92UVzsy-00828-00430943-00431247 because it's got kind of the wiggles in the same place. Now, it's not perfect
+pfLP92UVzsy-00829-00431247-00431624 but actually does a little bit better than straight line - although not a lot better,
+pfLP92UVzsy-00830-00431624-00432032 um, so why is that better and how could I come up with other polynomials that make it? Well
+pfLP92UVzsy-00831-00432032-00432480 that's the whole point of a Taylor series or a Maclaurin series, and if you teach A-level maths
+pfLP92UVzsy-00832-00432480-00432735 that's when they first meet it; but it's a really lovely thing that even
+pfLP92UVzsy-00833-00432735-00433176 functions which are not polynomials can indeed be polynomials and Geogebra can do this for you,
+pfLP92UVzsy-00834-00433176-00433616 and the fact that it's just a command, I love. So the first thing I'm going to do is create a slider
+pfLP92UVzsy-00835-00433728-00434160 which I'm going to call n, and this is going to be the number of terms I'm going to create for the
+pfLP92UVzsy-00836-00434160-00434696 polynomial; I'm going to let it go from 0 up to - I don't know - 30, and I'm gonna - I'd like it to
+pfLP92UVzsy-00837-00434696-00435272 go up in ones. So you can see this button - this option box for sliders lets you define a slider
+pfLP92UVzsy-00838-00435272-00435680 to do whatever you want. You can see that slides from 1 - 0 to 30. It doesn't do anything - it's
+pfLP92UVzsy-00839-00435680-00436056 just a slider at the moment, but now I'm going to type in the magic - I'm going to start typing the
+pfLP92UVzsy-00840-00436056-00436608 word taylor and you can see that - ah, 'would you like a taylor polynomial?' Yes, Geogebra,
+pfLP92UVzsy-00841-00436608-00437064 yes I would. You can press return to select the functions or you can use arrow keys if there's
+pfLP92UVzsy-00842-00437064-00437535 more than one available from the suggestions; in this case it's asking for a function, an x value,
+pfLP92UVzsy-00843-00437535-00437976 and an order number. The function I want to use is my f which I defined earlier so I'm just typing f;
+pfLP92UVzsy-00844-00437976-00438416 I would like to expand it around zero - which is incidentally what they call a
+pfLP92UVzsy-00845-00438488-00438816 Maclaurin series, when a Taylor series expands around zero - let's not worry about that;
+pfLP92UVzsy-00846-00438920-00439216 and then I'm going to say how many terms I'd like the polynomial - that's the n I
+pfLP92UVzsy-00847-00439216-00439600 made earlier. I've left it parked on 11 for no good reason, but let's just type in n
+pfLP92UVzsy-00848-00439720-00440088 and there we have it. One other tip :any algebraic representation on the left if
+pfLP92UVzsy-00849-00440088-00440328 I would like it visible on the screen I can just drag that onto the screen
+pfLP92UVzsy-00850-00440488-00441008 and there I've got the algebraic version of this function. Here's the original one: so sine x and
+pfLP92UVzsy-00851-00441008-00441392 that polynomial are remarkably similar -if I zoom out you can see that they're pretty good.
+pfLP92UVzsy-00852-00441392-00441704 I'm just going to stop these moving around by clicking 'absolute position on the screen'
+pfLP92UVzsy-00853-00441776-00442096 and you're all dying to see what happens when I change the slider: if I reduce the number
+pfLP92UVzsy-00854-00442096-00442520 of terms you can see the polynomial gets worse and worse - starts off with zero, there's the
+pfLP92UVzsy-00855-00442520-00443112 one we talked about earlier there's the cubic one, um, oh it's three factorial that's why it
+pfLP92UVzsy-00856-00443112-00443720 was rubbish earlier no one pointed out my error! And if I keep dragging - I love that oscillation.
+pfLP92UVzsy-00857-00443904-00444272 I mean I knew about Maclaurin and Taylor series, but the first time I saw this little animation,
+pfLP92UVzsy-00858-00444272-00444816 which you can build in about 30 seconds flat, I was like, "that's why the more terms you've got
+pfLP92UVzsy-00859-00444816-00445352 it's better!" And actually if you look at other polynomials - for example, uh, someone give me
+pfLP92UVzsy-00860-00445352-00446008 another polynomia,l uh... e to the x? Let's change the original function to be e to the x
+pfLP92UVzsy-00861-00446256-00446576 and you can see that this also gets better and better - but the right hand side is always pretty
+pfLP92UVzsy-00862-00446576-00446912 good; it's just the left hand side that's oscillating in a slightly kind of weird and
+pfLP92UVzsy-00863-00446912-00447400 comical way; I'm not sure why I find that funny but I kind of do. Um, the reason I want to demo
+pfLP92UVzsy-00864-00447400-00447880 this is this is a fundamentally algebraic and sort of calculus based topic, but Geogebra really helps
+pfLP92UVzsy-00865-00447880-00448232 you demonstrate it and it's doing that because it's good at algebra and it has that Taylor
+pfLP92UVzsy-00866-00448232-00448800 command - and, you can make things move; you can generalise things, and that visual approximation
+pfLP92UVzsy-00867-00448800-00449192 of things with movement involved helps all of us understand what's going on as things change.
+pfLP92UVzsy-00868-00449280-00449808 Okay, end of demo, time for questions if you want any - I'm just going to leave this animating.
+pfLP92UVzsy-00869-00450124-00450632 JOHN: Ben, I'm sorry to monopolise - before you go I've got a bit of a sort of meta question
+pfLP92UVzsy-00870-00450632-00450992 I'd like to ask - not specific to this. BEN: Ask it now and I'll tell
+pfLP92UVzsy-00871-00450992-00451247 you if I think it's worth asking answering in front of everyone,
+pfLP92UVzsy-00872-00451247-00451528 and if it's not then I'll loiter for five minutes at the end and we'll tackle it.
+pfLP92UVzsy-00873-00451600-00452088 JOHN: Yesterday I went to an Autograph seminar BEN: Good! Autograph is also getting better and
+pfLP92UVzsy-00874-00452088-00452952 better. JOHN: Right and, um, having been, that was wiping away some Autograph cobwebs, um, and after
+pfLP92UVzsy-00875-00452952-00453392 them both I've come to the conclusion more or less that I don't need both of them. Now firstly,
+pfLP92UVzsy-00876-00453392-00453784 is that correct? I mean is there anything that one does that the other doesn't and vice versa?
+pfLP92UVzsy-00877-00453856-00454335 and secondly, sort of - personal preferences aside - can you put your finger on anything
+pfLP92UVzsy-00878-00454335-00455064 that Autograph does that Geogebra doesn't or vice versa? BEN: That is, I think, a worthwhile
+pfLP92UVzsy-00879-00455064-00455576 question to answer now to everybody unless there are other burning questions, um, I will comment
+pfLP92UVzsy-00880-00455576-00456184 on this because it is a common concern, like - basically other packages are available. That's
+pfLP92UVzsy-00881-00456184-00456664 the important headline here; we chose to focus on one today because actually if I try to demonstrate
+pfLP92UVzsy-00882-00456664-00456935 three different packages - and I'm thinking of three particularly different packages in
+pfLP92UVzsy-00883-00456935-00457424 my head which I'll say in a moment -there's no way we could show you the depth of all of them
+pfLP92UVzsy-00884-00457424-00457776 so it may be that we should run another session on Autograph, and we should run another session
+pfLP92UVzsy-00885-00457776-00458208 on Desmos, having just done one on Geogebra. Those three, I think, are worth knowing about.
+pfLP92UVzsy-00886-00458304-00458752 Geogebra is free - it now works in a browser and online and downloaded and a phone;
+pfLP92UVzsy-00887-00458752-00459208 Desmos is free it works in a browser - not as a downloaded version, but it does work on a phone;
+pfLP92UVzsy-00888-00459208-00459712 Autograph is now free thanks to, I think, Complete Maths, um, and they're doing lots of work on
+pfLP92UVzsy-00889-00459712-00460216 Autograph; and so here's a quick potted summary of which ones are useful for what: Autograph was
+pfLP92UVzsy-00890-00460216-00460704 the first one to get a 3D mode that functioned. Changed our worlds, if you played with lots of
+pfLP92UVzsy-00891-00460704-00461439 3D things. In my opinion Geogebra 3D mode is now superior. Autograph still has a very good 3D mode,
+pfLP92UVzsy-00892-00461439-00461752 and Desmos doesn't, for example - so there's one comparison between the two.
+pfLP92UVzsy-00893-00461888-00462272 If you were going to use Desmos to plot complex numbers, it's difficult; Geogebra has them built
+pfLP92UVzsy-00894-00462272-00462696 in - for a long time Autograph didn't, but now it does, so Autograph has caught up on that;
+pfLP92UVzsy-00895-00462776-00463272 and Geogebra as well. If you want to do simple graphing, Desmos is a very clean interface - it
+pfLP92UVzsy-00896-00463272-00463720 works very naturally and instinctively, but if you want to do complicated functions like I've been
+pfLP92UVzsy-00897-00463720-00464247 modeling, say the Taylor polynomial here - Desmos won't do that; and I don't know if Autograph will
+pfLP92UVzsy-00898-00464247-00464888 do that. I suspect there would be a way to make it happen. (JOHN: Yeah) Now, getting away from facts:
+pfLP92UVzsy-00899-00464888-00465416 I find Autograph less intuitive. And I find the fact that all the commands are available
+pfLP92UVzsy-00900-00465416-00465760 in Geogebra - I can go and look at the help file - a bit more intuitive; but that could just be a
+pfLP92UVzsy-00901-00465760-00466104 product of the fact that I've spent more time with it. So I think all those three are worth
+pfLP92UVzsy-00902-00466104-00466568 knowing about - they're all getting better all the time and now they're all free so: choose
+pfLP92UVzsy-00903-00466568-00466928 one and dust the cobwebs off and don't stress too much about it is my summary. JOHN: Yeah,
+pfLP92UVzsy-00904-00466928-00467288 thank you. KEVIN: Right, right - okay, okay - Ben I'm gonna have to stop you there because we're
+pfLP92UVzsy-00905-00467288-00467816 gonna run out of time quite soon because there was a couple of things you know we were gonna say, um,
+pfLP92UVzsy-00906-00467816-00468232 and we're also going to, uh, Sam was going to say something - Sam Hartburn was going to say
+pfLP92UVzsy-00907-00468232-00468792 something - I mean we kind of, uh... BEN: Yeah, good point. SAM H: Yeah, um, it's a shame Becky
+pfLP92UVzsy-00908-00468792-00469296 didn't hang around really because I did Becky's, um, online Geogebra course and I just wanted to
+pfLP92UVzsy-00909-00469296-00469928 tell you how fantastic it was. Um, I'd done a fair bit of Geogebra stuff before, but I learned lots
+pfLP92UVzsy-00910-00469928-00470639 of new tricks to make things more efficient and to do things quicker, but the main thing was it's
+pfLP92UVzsy-00911-00470639-00471056 just really inspiring -she showed us lots more designs than she showed us today - obviously,
+pfLP92UVzsy-00912-00471056-00471696 it was a 10 week course - and, um, just lots of really inspiring designs, um, enabled me to go and
+pfLP92UVzsy-00913-00471696-00472272 do lots of new and different things and, yeah, really highly recommend - highly recommend it;
+pfLP92UVzsy-00914-00472272-00472743 it was really a social thing as well, um, just to get together with other people who liked
+pfLP92UVzsy-00915-00472743-00473176 playing with Geogebra and like playing with art and we share our designs on Twitter and
+pfLP92UVzsy-00916-00473272-00473743 work on things together - it's just been a really nice thing to do. So, yeah definitely sign up
+pfLP92UVzsy-00917-00473743-00474232 if - uh, when - if/when she runs another one. BEN: Let me back that up by saying I didn't do Becky's
+pfLP92UVzsy-00918-00474232-00474728 course but watching the output from Sam Hartburn for example on Twitter shows me what an amazing
+pfLP92UVzsy-00919-00474728-00475080 course it is, because I can see lots of people having a really good time making amazing stuff
+pfLP92UVzsy-00920-00475080-00475535 and loving doing so and if you want to go - I think second following suggestion :follow Sam
+pfLP92UVzsy-00921-00475535-00475943 Hartburn, because some some of the art she's produced off the back of Becky's course is
+pfLP92UVzsy-00922-00475943-00476372 lovely and it's just worth looking at. KEVIN: Indeed, indeed SAM H: Aww, thanks Ben. KEVIN:
+pfLP92UVzsy-00923-00476464-00476968 Yes thanks Sam, um, so I think we really should wrap it up, so, uh, thanks to Becky who's uh
+pfLP92UVzsy-00924-00476968-00477600 unfortunately, uh, had had to, uh - has got herself a job now so, um, well sorry a uh - a
+pfLP92UVzsy-00925-00477600-00477952 job BEN: Another job! KEVIN: Another job - it's not freelancing yes I mean that's uh.
+pfLP92UVzsy-00926-00478208-00478800 Yeah, um, so yes thanks to Ben and Becky, um, there are a couple of things I should say;
+pfLP92UVzsy-00927-00478800-00479376 the last video that we did or rather last session that we did, we've got it on video it is up on
+pfLP92UVzsy-00928-00479376-00480104 the YouTube channel, um, but unfortunately it doesn't have the proper, um, uh captions
+pfLP92UVzsy-00929-00480104-00480472 on it yet because we haven't had time, so if anybody wants to volunteer and do the captions,
+pfLP92UVzsy-00930-00480535-00481288 um, then if they could please that'd be fantastic, um, so volunteers for that. Next time we've got
+pfLP92UVzsy-00931-00481288-00481839 Inclusivity in Maths Communication so that'll probably be towards the end of November, um,
+pfLP92UVzsy-00932-00481839-00482439 so probably a month from now more or less, and we'll look - well look out on the the email that
+pfLP92UVzsy-00933-00482439-00482920 we send around for that. I think I should just end by saying well thank you very much everybody for
+pfLP92UVzsy-00934-00482920-00483639 coming and asking your questions and, um, I'm sure Ben, um, will will be available for questions,
+pfLP92UVzsy-00935-00483704-00484135 uh, by by email, um BEN: I'll loiter for a couple of minutes now, although I won't stay
+pfLP92UVzsy-00936-00484135-00484447 too long, if anyone needs KEVIN: ...for a few minutes because - because, as - we've got an
+pfLP92UVzsy-00937-00484447-00484943 important insight into Ben's mind: that, uh, art is there for doing Geogebra! Well that's - that's
+pfLP92UVzsy-00938-00484943-00485392 uh important insight I think for.. to Ben BEN: I don't see the contradiction. KEVIN: So, so
+pfLP92UVzsy-00939-00485392-00485935 thanks very much for everybody and I think as Ben says he'll be he'll be hanging around uh for five
+pfLP92UVzsy-00940-00485935-00486368 minutes or so if you want any more. But anyway thanks all for coming and we'll see you next time.
+pj07t8ckaaU-00000-00000096-00000500 Here's another important note about naming alkenes-- sometimes they can
+pj07t8ckaaU-00001-00000500-00000629 simultaneously be
+pj07t8ckaaU-00002-00000629-00000948 alcohols in which case we refer to them as
+pj07t8ckaaU-00003-00000948-00001498 "enols", and the ranking as you see in this molecule favors the location of
+pj07t8ckaaU-00004-00001498-00001523 the
+pj07t8ckaaU-00005-00001522-00002016 -OH group in the alcohol so the alcohol group outranks the double bond
+pj07t8ckaaU-00006-00002016-00002338 so this chain is number backwards, and notice
+pj07t8ckaaU-00007-00002338-00002718 the name reflects that the 2 is where the alcohol
+pj07t8ckaaU-00008-00002718-00003181 group is and that puts the carbon-carbon double bond starting at carbon number
+pj07t8ckaaU-00009-00003181-00003234 4
+pj07t8ckaaU-00010-00003234-00003538 so that becomes 4-penten-2-ol
+pj07t8ckaaU-00011-00003538-00004238 numbers to tell us where both of those important features are located. this next
+pj07t8ckaaU-00012-00004371-00004464 slide shows
+pj07t8ckaaU-00013-00004464-00004952 a couple more important alkyll groups that have double bonds
+pj07t8ckaaU-00014-00004952-00005305 we've seen things like methyl, ethyl, tertiary butyl
+pj07t8ckaaU-00015-00005305-00005787 back in Chapters 2 and 3 and 4
+pj07t8ckaaU-00016-00005787-00006218 and now we see this ethenyl group that has
+pj07t8ckaaU-00017-00006218-00006565 a carbon-carbon double bond.  If that were hanging off with a
+pj07t8ckaaU-00018-00006565-00007029 ring for example that might be called ethenyl, but it's much more commonly
+pj07t8ckaaU-00019-00007029-00007172 referred to as vinyl
+pj07t8ckaaU-00020-00007172-00007645 that's a very old and traditional name for this.  a lot of vinyl plastics you might be
+pj07t8ckaaU-00021-00007645-00007745 familiar with
+pj07t8ckaaU-00022-00007745-00008199 are made starting with compounds that have this structural feature
+pj07t8ckaaU-00023-00008199-00008649 if you add another CH2 group that vinyl becomes allyl
+pj07t8ckaaU-00024-00008649-00008999 also more formally known as propenyl
+pj07t8ckaaU-00025-00008999-00009299 so these molecules at the bottom can be
+pj07t8ckaaU-00026-00009299-00009749 named based on using these substituent names
+pj07t8ckaaU-00027-00009749-00010109 vinyl chloride is more I guess officially named as
+pj07t8ckaaU-00028-00010109-00010420 chloroethene, and this allyl alcohol is
+pj07t8ckaaU-00029-00010420-00010884 a common name for a what would be 2-propen-1-ol
+pj07t8ckaaU-00030-00010884-00011289 and remember we start numbering where the -OH group is
+pj07t8ckaaU-00031-00011289-00011989 in preference to where the double bond is.  If you build a molecular model of an
+pj07t8ckaaU-00032-00012029-00012203 alkene you do find that
+pj07t8ckaaU-00033-00012203-00012561 you get these 120 degree bond angles around the carbons
+pj07t8ckaaU-00034-00012561-00012904 all these atoms will be in the same plane and
+pj07t8ckaaU-00035-00012904-00013393 as it mentions there a double bond is shorter than a single bond, it's stronger
+pj07t8ckaaU-00036-00013393-00013681 than a single bond,although it's not twice as strong
+pj07t8ckaaU-00037-00013681-00014119 but because of that extra bonding that means there's not any conformational
+pj07t8ckaaU-00038-00014119-00014369 changes we can talk about here these atoms
+pj07t8ckaaU-00039-00014369-00014763 surrounding this double bond are pretty much stuck in place
+pj07t8ckaaU-00040-00014763-00015216 These next slides talk about the hybridization
+pj07t8ckaaU-00041-00015216-00015776 in alkenes, similar to what was done for alkanes back in Chapter two
+pj07t8ckaaU-00042-00015776-00016113 we're not going to get much into this-- I
+pj07t8ckaaU-00043-00016113-00016525 included these slides just to point you in the right direction in case you're
+pj07t8ckaaU-00044-00016525-00016579 taking
+pj07t8ckaaU-00045-00016579-00017153 some standardized tests that require you to know something about hybridization
+pj07t8ckaaU-00046-00017153-00017490 with alkanes with all the single bonds it was
+pj07t8ckaaU-00047-00017490-00017798 sp3, it was mixing an s-orbital with
+pj07t8ckaaU-00048-00017798-00018287 all 3 p-orbitals for carbon to come up with those four tetrahedral
+pj07t8ckaaU-00049-00018287-00018756 orbitals you just mix one of the s's with one of the p-orbitals
+pj07t8ckaaU-00050-00018756-00019151 when you're talking about alkenes and these sp2 orbitals that have already
+pj07t8ckaaU-00051-00019151-00019304 been mixed here at the bottom
+pj07t8ckaaU-00052-00019304-00019779 that gives us three equivalent orbitals and they would naturally be
+pj07t8ckaaU-00053-00019779-00020268 coplanar and 120 degrees apart. it's
+pj07t8ckaaU-00054-00020268-00020656 the overlap of carbons with the single-electron in this
+pj07t8ckaaU-00055-00020656-00021228 oun-hybridized p-orbital that's where that second bond the double bond comes from
+pj07t8ckaaU-00056-00021228-00021592 and here's all of that kinda superimposed on one structure
+pj07t8ckaaU-00057-00021592-00021962 with the double bond we have
+pj07t8ckaaU-00058-00021962-00022468 as it shows here what's a sideways overlap up these un-hybridize p-orbitals
+pj07t8ckaaU-00059-00022468-00022781 Orbitals usually overlap end-to-end
+pj07t8ckaaU-00060-00022781-00023107 but this sideways overlap means these orbitals are
+pj07t8ckaaU-00061-00023107-00023476 overlapping in 2 areas--both lobes down here on the bottom left
+pj07t8ckaaU-00062-00023476-00023824 that's not as strong as an interaction
+pj07t8ckaaU-00063-00023824-00024204 as the head to head bonding of the sp2 orbitals
+pj07t8ckaaU-00064-00024204-00024657 but it is the source of that extra bond that we draw, that extra line to
+pj07t8ckaaU-00065-00024657-00024788 indicate a double bond
+pj07t8ckaaU-00066-00024788-00025443 and these head to head orbital overlaps are refered to as sigma bonds that's a small case
+pj07t8ckaaU-00067-00025443-00025666 sigma down here in the purple region
+pj07t8ckaaU-00068-00025666-00026013 and the pi bonds here that's the term used to
+pj07t8ckaaU-00069-00026013-00026497 refer to overlap when it is sideway, not end-to-end.
+pj07t8ckaaU-00070-00026497-00026872 and here's a few more related
+pj07t8ckaaU-00071-00026872-00027187 comments about that kind of
+pj07t8ckaaU-00072-00027187-00027597 molecular explanation of bonding with molecular
+pj07t8ckaaU-00073-00027597-00027643 orbitals
+pjQWvXejRug-00000-00000003-00000537 Hey, you got a minute? -I gotta go. -That's okay. That's okay. This'll only take a
+pjQWvXejRug-00001-00000537-00001080 minute. It's a Crappy One Minute Wednesday. -What? One cache. -One minute.
+pjQWvXejRug-00002-00001080-00001398 Cue the title. -Start the clock. -Go!
+pjQWvXejRug-00003-00001416-00001476 [ Loon calls ]
+pjQWvXejRug-00004-00001598-00001660 [ Loon calls ]
+pjQWvXejRug-00005-00002150-00002568 [ laughter ] Wow! Look at this. This is great!
+pjQWvXejRug-00006-00002710-00002986 Oh there's a couple critters in here. -That's awesome.
+pjQWvXejRug-00007-00003002-00003247 Is that a trackable? -Yeah. -That's a big travel bug.
+pjQWvXejRug-00008-00003247-00003450 We have the cache owner with us.
+pjQWvXejRug-00009-00003460-00003804 How you doing? -Hey, how are ya? -This is an awesome cache.
+pjQWvXejRug-00010-00003804-00003878 -Thank you very much.
+pjQWvXejRug-00011-00003878-00004222 -It's really, really, cool. We have to take a close look at it. -Yeah. -How long has it
+pjQWvXejRug-00012-00004223-00004708 been here? -Uh. I think, I don't know. -You don't know. -Four or five years.
+pjQWvXejRug-00013-00004708-00005021 -It's gotta have a lot of Favourite Points. -I believe we have over 300
+pjQWvXejRug-00014-00005021-00005405 Favourite Points. -Yeah, oh it's gonna have more after today. That's awesome.
+pjQWvXejRug-00015-00005406-00005776 -Every time you come out here the cache keeps growing. People add. -Yeah.
+pjQWvXejRug-00016-00005794-00006282 -They sign the walls, add all these buttons. This dog was left by another friend.
+pjQWvXejRug-00017-00006282-00006688 -For a good time call.  -Yeah. Don't call me.
+pjQWvXejRug-00018-00006752-00007150 [ laughter ] Oh, oh, there's Tsmola, right there. -Yeah. -Yeah,  he did
+pjQWvXejRug-00019-00007154-00007540 a video on this one. That's cool. That's awesome.
+pjQWvXejRug-00020-00007702-00007802 [ Loon calls ]
+pjQWvXejRug-00021-00007888-00007976 [ Loon calls ]
+pjQWvXejRug-00022-00007990-00008494 See what I mean. It was one Crappy Cache! -That's true. -And that's what it's called.
+pjQWvXejRug-00023-00008494-00008940 It's called a Crappy Cache. It's awesome to meet the CO here too. -Yeah.
+pjQWvXejRug-00024-00008940-00009363 -And get some trackables. -Lots of TB's here. -Yeah it was really,
+pjQWvXejRug-00025-00009363-00009810 really, awesome. We signed our name up beside Darkflair we know from Canada,
+pjQWvXejRug-00026-00009810-00010452 and Tsmola. -And Favourite Points. -Oh absolutely. Definitely two more here.
+pjQWvXejRug-00027-00010452-00010808 Well over 300 [ favourite points ] gonna go for four pretty soon. -Yeah.
+pjQWvXejRug-00028-00010808-00011056 Worth coming for a visit. -Absolutely.
+pjQWvXejRug-00029-00011070-00011448 Thanks for watching, and don't forget to Like, Share, and -Subscribe.
+pjQWvXejRug-00030-00011450-00011886 And ring the bell. -Right. -Keep watching there's more caches to be found.
+pjQWvXejRug-00031-00011886-00012162 -Where will Geocaching take you?
+pjQWvXejRug-00032-00012162-00012452 All the way to Michigan. And we've got our smiley.
+pjQWvXejRug-00033-00012452-00013033 This is a one. This is a one. This is a, this is a Crappy
+pjQWvXejRug-00034-00013033-00013319 One a Minute Wednesday.
+pkffes_ISAA-00000-00000000-00000910 This video will explain how to use concept maps to break down complex topics into their individual related parts. 
+pkffes_ISAA-00001-00000910-00001726 This is a good way for a visual learner to understand the correlation between parts of a concept or idea.
+pkffes_ISAA-00002-00001726-00002283 Begin by writing a main topic at the top of your page, and draw a circle around it.
+pkffes_ISAA-00003-00002283-00002843 Break your main topic into components, and arrange them in a hierarchy. 
+pkffes_ISAA-00004-00002843-00003643 As you get farther down the page, away from your topic, the concepts become more and more specific.
+pkffes_ISAA-00005-00003643-00004529 Add linking lines between each component. Add in words that explain the relationship between two connected bubbles. 
+pkffes_ISAA-00006-00004530-00005386 When you read the bubbles in descending order including the linking words you will usually get sentences relating the entire topic.
+pkffes_ISAA-00007-00005386-00006009 Try concept mapping when you are trying to understand complex topics. Happy studying!
+pnD_XefPkMg-00000-00000373-00000540 Damn it!
+pnD_XefPkMg-00001-00002723-00002886 It’s so frustrating...
+pnD_XefPkMg-00002-00003656-00003836 You've got more, don't you?
+pnD_XefPkMg-00003-00003913-00004010 Yes!
+pnD_XefPkMg-00004-00004176-00004750 I'll go against you again.
+pnD_XefPkMg-00005-00004873-00005096 I’ll get you ready!
+pnD_XefPkMg-00006-00005136-00005293 Make the most of it.
+pnD_XefPkMg-00007-00005433-00005546 Thank you!
+pnD_XefPkMg-00008-00005616-00005783 Nakanishi, thank you!
+pnD_XefPkMg-00009-00006850-00006996 Nagata...
+pnD_XefPkMg-00010-00007080-00007359 This pro-wrestling world...
+pnD_XefPkMg-00011-00007733-00008080 He was the one who introduced me to this world.
+pnD_XefPkMg-00012-00008216-00008913 After I met him, my life was changed.
+pnD_XefPkMg-00013-00008990-00009136 I won't give up.
+pnD_XefPkMg-00014-00009286-00010050 I won't waste this second chance from Nakanishi.
+pnD_XefPkMg-00015-00010213-00011450 He's not human, a monster, but I'll give him hell.
+pnD_XefPkMg-00016-00012386-00013466 I could see that he's grown, little by little,  mastering each hold and strike...
+pnD_XefPkMg-00017-00013743-00014390 He understands how to  take advantage of his physique.
+pnD_XefPkMg-00018-00014433-00014540 But…
+pnD_XefPkMg-00019-00014553-00016030 Pro-wrestling is a real fight. He needs to  maintain his composure when he's on the defensive.
+pnD_XefPkMg-00020-00016066-00016386 That’s what makes a pro-wrestler.
+pnD_XefPkMg-00021-00016633-00017953 It’s all about what you do in that ring.  He has to channel his emotions effectively.
+pnD_XefPkMg-00022-00018026-00018203 That’s what I ask of him.
+pnD_XefPkMg-00023-00018210-00018446 I know there's greatness in there.
+pnD_XefPkMg-00024-00018603-00019850 I taught him the same way my mentors taught me.
+pnD_XefPkMg-00025-00019970-00021406 I believe I showed him  everything in that ring today.
+pnD_XefPkMg-00026-00021513-00022830 That’s how I was trained by Hase and Sasaki. They  knocked me down and taught me how to stand up.
+pnD_XefPkMg-00027-00023126-00023556 Kitamura needs more of that.
+pnD_XefPkMg-00028-00023563-00024270 I don't think anyone gave him that, before.
+pnD_XefPkMg-00029-00024280-00024483 So, I did it.
+pnD_XefPkMg-00030-00024516-00025560 It would have been great if he could blow me off,  but I’m tougher than I look. 26 years going...
+pnD_XefPkMg-00031-00025796-00026093 Kitamura needs more of this.
+pnD_XefPkMg-00032-00026130-00026470 And he’ll become a real pro-wrestler.
+pnD_XefPkMg-00033-00026506-00026733 That’s why I was hard on him today.
+pnD_XefPkMg-00034-00026836-00027023 That’s all.
+pnXJANjDacM-00000-00000126-00000290 ♫
+pnXJANjDacM-00001-00000512-00000872 This video will provide you with a few ways to share your Zoom cloud recordings
+pnXJANjDacM-00002-00000872-00001216 with your students from within the Zoom tool in your Blackboard course site.
+pnXJANjDacM-00003-00001216-00001544 It is important to note that these steps will only work if you have chosen 'record
+pnXJANjDacM-00004-00001544-00001735 to cloud' when recording your Zoom meeting.
+pnXJANjDacM-00005-00001735-00002039 If you have a recording that was saved directly to your computer hard drive,
+pnXJANjDacM-00006-00002039-00002311 there are alternate steps required to share with your students and we will
+pnXJANjDacM-00007-00002311-00002624 cover that in a separate video. The best way to share cloud recordings
+pnXJANjDacM-00008-00002624-00002840 is to publish them from within the Zoom tool
+pnXJANjDacM-00009-00002840-00003184 in Blackboard. To be able to publish a recording to a class you will have to
+pnXJANjDacM-00010-00003184-00003568 have created the Zoom meeting within that specific Blackboard course site.
+pnXJANjDacM-00011-00003568-00003879 Once you have completed your recording and ended the zoom meeting, you will
+pnXJANjDacM-00012-00003879-00004232 receive an email when the recording has completed processing and is available to
+pnXJANjDacM-00013-00004232-00004464 share. When you navigate to the Zoom tool in
+pnXJANjDacM-00014-00004464-00004832 Blackboard, either through a tool link or through the tools section of your site,
+pnXJANjDacM-00015-00004832-00005096 you will see several tabs of content available to you.
+pnXJANjDacM-00016-00005096-00005464 Select the rightmost tab, 'cloud recordings'. From here you will see a full
+pnXJANjDacM-00017-00005464-00005800 list of all cloud recordings you have created using meetings in this specific
+pnXJANjDacM-00018-00005800-00006135 course. On the right hand side of the interface you will see a publish column
+pnXJANjDacM-00019-00006135-00006447 with a toggle button that you can move to the on position to share that
+pnXJANjDacM-00020-00006447-00006815 specific recording with students. All recordings are not shared by default
+pnXJANjDacM-00021-00006815-00007040 and you must manually activate the publish button
+pnXJANjDacM-00022-00007040-00007376 to share a video with students. Once you have published a recording,
+pnXJANjDacM-00023-00007376-00007728 students will be able to access the Zoom tool, again, either through the tool link
+pnXJANjDacM-00024-00007728-00008072 or through the course tools. They will also find a cloud recordings tab.
+pnXJANjDacM-00025-00008072-00008440 Your published videos will appear here in order of date and time recorded. When
+pnXJANjDacM-00026-00008440-00008672 a student clicks on the title of a certain recording
+pnXJANjDacM-00027-00008672-00009000 they will be brought to the video and audio files for that session.
+pnXJANjDacM-00028-00009000-00009336 When the student clicks on a certain file the passcode to view the file will
+pnXJANjDacM-00029-00009336-00009568 automatically be copied to their computer's clipboard.
+pnXJANjDacM-00030-00009568-00009872 They will need to paste that code into the following screen to access the
+pnXJANjDacM-00031-00009872-00010104 recording. Additionally, they can click on the 'eye'
+pnXJANjDacM-00032-00010104-00010464 icon underneath the recording to view the passcode and copy it down for later
+pnXJANjDacM-00033-00010464-00010704 use. If you have a recording that you do not
+pnXJANjDacM-00034-00010704-00011024 wish to publish to an entire class but you do want to share with specific
+pnXJANjDacM-00035-00011024-00011352 individuals, you can share a video link directly in a few ways.
+pnXJANjDacM-00036-00011352-00011679 First, when your cloud recording has finished processing on the Zoom server,
+pnXJANjDacM-00037-00011679-00011968 you will receive an email notification with two hyperlinks.
+pnXJANjDacM-00038-00011968-00012384 The first link will navigate to your bridgew.zoom.us profile and provide
+pnXJANjDacM-00039-00012384-00012632 access to that specific recording session.
+pnXJANjDacM-00040-00012632-00013016 The second link is the shareable invite link and the default password required
+pnXJANjDacM-00041-00013016-00013304 to view the recording. To share a video with a student you can
+pnXJANjDacM-00042-00013304-00013519 simply copy and paste both the second link
+pnXJANjDacM-00043-00013519-00013887 and its password into a separate email or private message to allow a student to
+pnXJANjDacM-00044-00013887-00014152 view the recording. You can also access the share link
+pnXJANjDacM-00045-00014152-00014344 directly through the Blackboard Zoom tool,
+pnXJANjDacM-00046-00014344-00014640 navigate to the cloud recordings tab and click on the title of the recording
+pnXJANjDacM-00047-00014640-00014872 you'd like to share. Here you will have access to download
+pnXJANjDacM-00048-00014872-00015216 the file directly or share the file. When you select share, you will be provided
+pnXJANjDacM-00049-00015216-00015495 with link details that you can copy and share with students directly.
+pnXJANjDacM-00050-00015495-00015759 These have been a few ways you can share your Zoom meeting recordings with
+pnXJANjDacM-00051-00015759-00015976 students, either through the publish feature or
+pnXJANjDacM-00052-00015976-00016208 through directly sharing the hyperlink for the recording.
+pnXJANjDacM-00053-00016208-00016520 For more information on this and many other topics, there are additional help
+pnXJANjDacM-00054-00016520-00016884 videos throughout the Teaching and Technology Center website and YouTube
+pnXJANjDacM-00055-00016884-00017184 page.
+pnXJANjDacM-00056-00017184-00017234 ♫
+qoAOPGGzaZ4-00000-00004940-00005301 Imagine for a moment you’re just playing CS:GO and getting paid for it
+qoAOPGGzaZ4-00001-00005301-00005457 Sounds like something unreal?
+qoAOPGGzaZ4-00002-00005468-00005685 But it’s more than possible with Exeedme
+qoAOPGGzaZ4-00003-00005704-00006306 This is a unique blockchain-powered tournament platform allowing gamers at all skill-levels to monetise their skills
+qoAOPGGzaZ4-00004-00006330-00006462 What do you need to begin?
+qoAOPGGzaZ4-00005-00006479-00006600 sign up on the Website
+qoAOPGGzaZ4-00006-00006618-00006741 pass the verification
+qoAOPGGzaZ4-00007-00006755-00007031 connect Steam and Faceit accounts to your profile
+qoAOPGGzaZ4-00008-00007059-00007224 It’s time to make some money!
+qoAOPGGzaZ4-00009-00007237-00007643 You can do it by participating in free tournaments with varying prize pools
+qoAOPGGzaZ4-00010-00007660-00008194 In Daily Ladders for each win you receive 5 points, for each loss you lose 3 points
+qoAOPGGzaZ4-00011-00008218-00008729 In this way, the top 10 players by points will get a monetary prize at the end of the tournament
+qoAOPGGzaZ4-00012-00008741-00009210 In addition, you can participate in single elimination tournaments by yourself or with friends
+qoAOPGGzaZ4-00013-00009225-00009800 Be sure to enter a unique HS Top CS:GO Tournament, which will be specially held on Exeedme!
+qoAOPGGzaZ4-00014-00009800-00010000 The battle promises to be hot!
+qoAOPGGzaZ4-00015-00010000-00010461 After winning, you just need to withdraw the money in any available cryptocurrency to your wallet
+qoAOPGGzaZ4-00016-00010473-00010800 There are also tournaments for Dota2 and Wazone lovers.
+qoAOPGGzaZ4-00017-00010812-00010941 Don’t waste your time!
+qoAOPGGzaZ4-00018-00010948-00011502 Play your favorite games and earn money with Exeedme! The link is in the description.
+q4Csg78XzWQ-00000-00001008-00001592 oh
+q4Csg78XzWQ-00001-00002056-00002984 oh my god
+q4Csg78XzWQ-00002-00003568-00004208 hi guys you're here again welcome to yb plays music today it's time again for dimash the
+q4Csg78XzWQ-00003-00004208-00004808 monthly reaction because i said in the last one from unforgettable day that didn't count
+q4Csg78XzWQ-00004-00004808-00005520 as the monthly reaction to dimash but this one is it's december and it's almost christmas time so
+q4Csg78XzWQ-00005-00005520-00006184 before we start i want to ask if there is a specific performance from dimash that i
+q4Csg78XzWQ-00006-00006184-00006823 perhaps can react to specifically for christmas so if you guys have any suggestions for that
+q4Csg78XzWQ-00007-00006823-00007472 leave them in the comments below that said the next one that i'm going to do is daybreak and
+q4Csg78XzWQ-00008-00007472-00008023 i have done a poll yesterday on uh on my channel because a lot of people suggested
+q4Csg78XzWQ-00009-00008023-00008768 that i do both the um the singer version and the busto version and i believe in the singer version
+q4Csg78XzWQ-00010-00008768-00009264 there are some behind the scenes so that's why i asked you guys which ones should i do first
+q4Csg78XzWQ-00011-00009328-00009888 and most of you guys suggested that i first do the one on the singer competition with some behind the
+q4Csg78XzWQ-00012-00009888-00010472 scenes and such and then afterwards we're going to react to the bastow version so if some of you
+q4Csg78XzWQ-00013-00010472-00010904 guys only want to see the bastello version or the singer version there will be timestamps in
+q4Csg78XzWQ-00014-00010904-00011440 the description below if you want to check that out so first let's check out the one on gloria
+q4Csg78XzWQ-00015-00011440-00011976 wu's channel that's the one with the behind the scenes and there are subtitles in both of the
+q4Csg78XzWQ-00016-00011976-00012696 videos before we start off i'm gonna read the description in this video first so it says here
+q4Csg78XzWQ-00017-00012696-00013288 dimash fell ill before the eighth round in the singer competition after receiving treatment from
+q4Csg78XzWQ-00018-00013288-00013912 a local doctor he defied the odds and delivered one of his most stunning performances find out
+q4Csg78XzWQ-00019-00013912-00014504 the gripping true story behind daybreak and experience the song with translated lyrics
+q4Csg78XzWQ-00020-00014616-00015352 sounds very interesting so he had problems with his health at that point in time i'm very curious
+q4Csg78XzWQ-00021-00015352-00016063 to see what he was able to do in that situation and i assume in the bastard version he was not ill
+q4Csg78XzWQ-00022-00016063-00016512 um so we're gonna see him in two different situations okay let's start off with
+q4Csg78XzWQ-00023-00016512-00016984 this one with behind the scenes on the singer competition here we go
+q4Csg78XzWQ-00024-00017336-00017448 2017
+q4Csg78XzWQ-00025-00017688-00018384 story behind daybreak
+q4Csg78XzWQ-00026-00019008-00019784 the night before the competition he lost his voice that's not good
+q4Csg78XzWQ-00027-00020024-00021184 the music also gives a lot more emotion
+q4Csg78XzWQ-00028-00021328-00021760 so they have this situation that they had this situation beforehand with other singers as well
+q4Csg78XzWQ-00029-00022240-00023288 he says oh yeah this is in china right so he doesn't understand so dimash doesn't
+q4Csg78XzWQ-00030-00023288-00024008 understand what he's saying i assume uh or does he speak chinese as well i don't know
+q4Csg78XzWQ-00031-00024200-00024488 he says i'm a bit worried now let's return a little here we go
+q4Csg78XzWQ-00032-00025104-00025384 his high note
+q4Csg78XzWQ-00033-00025839-00026520 so they say don't force yourself arriving at the competition after treatment he looks very
+q4Csg78XzWQ-00034-00026712-00028183 uh star like if you know what i mean like a pop star let's say
+q4Csg78XzWQ-00035-00031856-00032464 i don't know what the last couple of rounds of the competition were in his case and why
+q4Csg78XzWQ-00036-00032464-00033824 it was so tough for him but let me know in the comment section below he says good luck in english
+q4Csg78XzWQ-00037-00033824-00033920 is this another round
+q4Csg78XzWQ-00038-00034416-00035184 oh this is the other guy probably or not
+q4Csg78XzWQ-00039-00035928-00036472 so the song he performs is the first chinese song he has ever heard in his life
+q4Csg78XzWQ-00040-00036472-00036976 and that on a bad treatment let's say for his voice i mean not bad treatment but
+q4Csg78XzWQ-00041-00036976-00037200 a treatment that didn't work that well
+q4Csg78XzWQ-00042-00037424-00037984 okay
+q4Csg78XzWQ-00043-00038696-00039336 yeah exactly like he says the high note that the high notes that dimash sings are not
+q4Csg78XzWQ-00044-00039336-00039984 to show off but to express his feelings and that it that that's exactly where people
+q4Csg78XzWQ-00045-00039984-00040704 can can be mistaken about him like he's not showing off let's say i mean he's able to
+q4Csg78XzWQ-00046-00040704-00041384 show off of course but it's a way for him to express his feelings because like if you uh
+q4Csg78XzWQ-00047-00041384-00041952 noticed in in a lot of videos of him after the performance he's so uh
+q4Csg78XzWQ-00048-00042048-00042952 he's so humble and and um and gentle that it's probably hard for him to really express
+q4Csg78XzWQ-00049-00043056-00043832 all these emotions to the fullest whilst in these performances he can really lay it all on the floor
+q4Csg78XzWQ-00050-00043888-00044800 and go full out and that's probably a good way for him to to release all those emotions i think
+q4Csg78XzWQ-00051-00046216-00047784 okay let's see this
+q4Csg78XzWQ-00052-00048840-00049184 okay
+q4Csg78XzWQ-00053-00049992-00050584 i'm just trying to read everything here guys loved ones are alive and well
+q4Csg78XzWQ-00054-00051255-00051984 who is that is that family of history
+q4Csg78XzWQ-00055-00052496-00052696 oh man i'm getting goosebumps already just
+q4Csg78XzWQ-00056-00053191-00053720 is he calling to the original uh songwriter oh wow
+q4Csg78XzWQ-00057-00054488-00054784 dude
+q4Csg78XzWQ-00058-00056112-00056888 so in 1991 a cable car accident took the lives of 14 tourists in china one couple lifted their son
+q4Csg78XzWQ-00059-00056888-00057632 out of the falling cable car to safety they did not survive the accident oh man
+q4Csg78XzWQ-00060-00058096-00058504 now what does this has to do with this song is that why the song is written
+q4Csg78XzWQ-00061-00058504-00059264 or why he performs this song let's see hanh hong wrote daybreak to commemorate
+q4Csg78XzWQ-00062-00059264-00059655 to commemorate the couple's ultimate sacrifice for their child
+q4Csg78XzWQ-00063-00059760-00060528 she later adopted the orphan so the child did survive oh man she later adopted the door
+q4Csg78XzWQ-00064-00060608-00061784 jesus so han hong is the woman that wrote daybreak and then adopted the orphan oh my god okay
+q4Csg78XzWQ-00065-00062160-00062200 let's see
+q4Csg78XzWQ-00066-00062879-00063184 he's already feeling it
+q4Csg78XzWQ-00067-00064079-00064200 he must be thinking
+q4Csg78XzWQ-00068-00064496-00065984 about what happened before
+q4Csg78XzWQ-00069-00066432-00067288 guys i'm sorry to pause already but i gotta point out the subtle um string plucking i don't know of
+q4Csg78XzWQ-00070-00067288-00067936 what instrument it is is it from the violence or from from some kind of shallows i don't know
+q4Csg78XzWQ-00071-00068032-00068984 but you hear the subtle staccato uh plucking uh on the strings which gives such a fragile feeling an
+q4Csg78XzWQ-00072-00068984-00069752 intimate feeling to this intro to the song and he's starting off so how should you say it like
+q4Csg78XzWQ-00073-00069976-00070800 intimate really intimate and i'm feeling it already and i think he feels totally what happened
+q4Csg78XzWQ-00074-00070800-00071448 before was it han hong who wrote this song which gave her the inspiration to write this song
+q4Csg78XzWQ-00075-00071448-00072104 and then in combination with the struggle he has gone through just before this show i think this
+q4Csg78XzWQ-00076-00072104-00072984 must have been a very emotional um performance for him i want to hear that from the beginning again
+q4Csg78XzWQ-00077-00073552-00074384 he's breathing there you see it
+q4Csg78XzWQ-00078-00076048-00077184 foreign
+q4Csg78XzWQ-00079-00077712-00078584 foreign
+q4Csg78XzWQ-00080-00082728-00084184 foreign
+q4Csg78XzWQ-00081-00086632-00086984 my god these lyrics
+q4Csg78XzWQ-00082-00088552-00091184 oh gosh
+q4Csg78XzWQ-00083-00092096-00092584 oh god
+q4Csg78XzWQ-00084-00093520-00094232 okay oh guys i i really didn't want to pause it i have to though but oh my god i don't
+q4Csg78XzWQ-00085-00094232-00094736 know if you guys saw that but that was very emotional at a certain point for me as well
+q4Csg78XzWQ-00086-00094736-00095784 uh it's just if you place yourself in that situation the lyrics are kind of written in the
+q4Csg78XzWQ-00087-00095784-00096784 eyes of that or from the eyes from the orphan from that child that lost his or her parents
+q4Csg78XzWQ-00088-00096992-00097768 wow this must be incredibly hard for him to to perform even with his voice
+q4Csg78XzWQ-00089-00097768-00098424 completely intact but now with this treatment it's even harder for him to bring this and control this
+q4Csg78XzWQ-00090-00098424-00099032 it's such an emotional story it's really good i think that i watch this one first
+q4Csg78XzWQ-00091-00099032-00099656 so i have a good understanding of the backstory behind it it gives so much more value to this
+q4Csg78XzWQ-00092-00099656-00100984 performance you guys were right i want to return a little so we can hear the end in its entirety
+q4Csg78XzWQ-00093-00101232-00102384 is
+q4Csg78XzWQ-00094-00104776-00105183 this sounds so helpless
+q4Csg78XzWQ-00095-00106744-00107983 the breath control
+q4Csg78XzWQ-00096-00109272-00109576 let's return a little there was a technical difficulty over there
+q4Csg78XzWQ-00097-00109576-00110783 in that point in time that's too bad
+q4Csg78XzWQ-00098-00110904-00112183 oh it's in the video itself
+q4Csg78XzWQ-00099-00112240-00112383 all the breath in there
+q4Csg78XzWQ-00100-00113416-00113583 everybody is feeling that
+q4Csg78XzWQ-00101-00115280-00115911 this is oh no is that is that the
+q4Csg78XzWQ-00102-00116032-00116688 original let's return is that the original songwriter with the orphan
+q4Csg78XzWQ-00103-00117032-00117816 oh my god it's the first time that i get like this emotional uh with one of the master songs
+q4Csg78XzWQ-00104-00117880-00118448 or with one of dimash's performances because of this backstory uh
+q4Csg78XzWQ-00105-00118535-00119304 and and then the way he performs it just even just makes even more sense oh god
+q4Csg78XzWQ-00106-00119744-00120568 wow oh my god okay so guys i am not going to waste too much time i'm going to go
+q4Csg78XzWQ-00107-00120568-00121104 right away into the busto version of daybreak and i'm so glad that i
+q4Csg78XzWQ-00108-00121104-00121824 have an understanding of the meaning behind it and that will give even more uh to the
+q4Csg78XzWQ-00109-00121824-00122456 experience i think of watching this so let's go and check the one on dmacc's channel itself yeah
+q4Csg78XzWQ-00110-00122456-00123111 let's go into this one right away there are subtitles as well here oh my god guys i'm still
+q4Csg78XzWQ-00111-00123320-00124024 it's the first time that i really am a little bit perplexed uh because of emotions
+q4Csg78XzWQ-00112-00124135-00124783 in this one oh my god without any further ado let's just check this one out as well
+q4Csg78XzWQ-00113-00125904-00126183 head hanging down
+q4Csg78XzWQ-00114-00129928-00130383 is
+q4Csg78XzWQ-00115-00130896-00131784 my father's face for the last
+q4Csg78XzWQ-00116-00138048-00138352 i i'm gonna pause this for a moment guys i i mean
+q4Csg78XzWQ-00117-00138504-00139264 it's it's still i have to process this uh a little bit um it's so emotional and i want to comment on
+q4Csg78XzWQ-00118-00139264-00139888 everything that is going on as well but it's very hard to really get excitement from the whole scene
+q4Csg78XzWQ-00119-00139888-00140880 let's say because i'm so into the the song and the way he brings it it's it's incredible and i assume
+q4Csg78XzWQ-00120-00140880-00141528 here his voice is okay he didn't have any problems right there in this performance he sounds very
+q4Csg78XzWQ-00121-00141528-00142112 stable even though he sounded pretty stable in the other one but i think he's more in control in this
+q4Csg78XzWQ-00122-00142112-00142632 one like the fountains behind him that's probably projection i assume i don't think they have real
+q4Csg78XzWQ-00123-00142632-00144384 uh water or fountains in there uh probably yeah let's return a little and move on here we go
+q4Csg78XzWQ-00124-00144608-00145784 my
+q4Csg78XzWQ-00125-00146960-00147184 energy
+q4Csg78XzWQ-00126-00147688-00148224 this part really sounds and then in the last performance as well this part sounds so like
+q4Csg78XzWQ-00127-00148224-00148672 he's so frustrated placing himself in the eyes from the orphan i don't know the name
+q4Csg78XzWQ-00128-00148672-00149336 of the child back then but putting himself in his place being frustrated like i saw my parents die
+q4Csg78XzWQ-00129-00149336-00149840 why did this have to happen why did you leave me and leave all that frustration out which is
+q4Csg78XzWQ-00130-00149840-00151384 why he is going that high and giving that much power in this section of the song ah man
+q4Csg78XzWQ-00131-00151480-00152784 there was an ad guys sorry i will have to go back to another moment from here
+q4Csg78XzWQ-00132-00153992-00155584 oh is
+q4Csg78XzWQ-00133-00156104-00156984 oh my god
+q4Csg78XzWQ-00134-00158880-00159784 foreign
+q4Csg78XzWQ-00135-00160232-00161184 me
+q4Csg78XzWQ-00136-00162112-00162584 oh my god
+q4Csg78XzWQ-00137-00162944-00163984 he went even more crazy in this one than in the performance in the singer wow that is
+q4Csg78XzWQ-00138-00164408-00164536 i don't know what to say guys
+q4Csg78XzWQ-00139-00164760-00165384 the combination of the lyrics in the song the backstory behind the song
+q4Csg78XzWQ-00140-00165544-00165912 the struggle with his voice and both performances
+q4Csg78XzWQ-00141-00166144-00166512 i'm kind of i'm kind of speechless
+q4Csg78XzWQ-00142-00166744-00167120 i i mean in the last one he
+q4Csg78XzWQ-00143-00167368-00168088 he went even crazier with his voice probably as he had uh much more control in that situation
+q4Csg78XzWQ-00144-00168152-00168872 i assume but even the first one they're both so emotional uh and definitely like the first one
+q4Csg78XzWQ-00145-00168872-00169256 which gives that that struggle with him and the communication with the
+q4Csg78XzWQ-00146-00169256-00169776 original singer contacting him and the reason why she wrote that song that uh
+q4Csg78XzWQ-00147-00170072-00170528 i there are really no words right now that i can give to do justice
+q4Csg78XzWQ-00148-00170640-00171352 to this song and performance this is one that wolf would really like to react to i think
+q4Csg78XzWQ-00149-00171352-00172000 uh hopefully he watches uh my video i think he will get emotional for sure
+q4Csg78XzWQ-00150-00172072-00172968 in this one oh my god guys look i hope you appreciated my reaction to this one from
+q4Csg78XzWQ-00151-00172968-00173560 dimash it's hard to say like i hope you enjoyed my reaction but appreciated my reaction to this
+q4Csg78XzWQ-00152-00173560-00174416 one uh because it's not really joyful to have these sad situations and emotions but you can
+q4Csg78XzWQ-00153-00174416-00174928 have so much appreciation for the songwriter and for dimash for bringing it the way they
+q4Csg78XzWQ-00154-00174928-00175504 did so i hope you appreciate my reaction don't forget to subscribe like and share the video
+q4Csg78XzWQ-00155-00175504-00175960 also don't forget to check out my other reactions to dimash in the description down below
+q4Csg78XzWQ-00156-00175960-00176416 there will be some links here in the top of the screen as well i also have music videos
+q4Csg78XzWQ-00157-00176416-00177984 for you guys to check out and tutorials so thank you very much and see you guys next time bye
+q4Csg78XzWQ-00158-00179160-00179440 a lot of people suggested a lot of people should yet
+q4Csg78XzWQ-00159-00179600-00180456 uh with some so if you want uh there will be so without so let's check so let's first so let's
+q4Csg78XzWQ-00160-00180456-00181088 first check so let's first check so first let's so let's first check out the one on gloria wu
+q4Csg78XzWQ-00161-00181152-00181888 uh the one on glad when uh han which gave her the which gave her the
+q4Csg78XzWQ-00162-00182136-00183248 which give her even with uh his guys it's i'm really sorry that i have so i have a behind the
+q4Csg78XzWQ-00163-00183344-00183872 good or a good understanding on the school but yeah let's just
+q4Csg78XzWQ-00164-00184144-00185112 uh in in the in the of the of the why why did they have and giving this of the lyric and the both of
+q4Csg78XzWQ-00165-00185112-00185840 the both and and also don't forget i have other reaction video also don't forget
+q69W_5_xhoI-00000-00001652-00002071 I think when we're in China it's important to remember that that winning the battle
+q69W_5_xhoI-00001-00002071-00002248 may cost you the war
+q69W_5_xhoI-00002-00002248-00002566 and it's a natural Western reaction when dealing with conflict
+q69W_5_xhoI-00003-00002566-00002678 to want to win.
+q69W_5_xhoI-00004-00002678-00002872 If your opinion differs to mine,
+q69W_5_xhoI-00005-00002872-00003117 my opinion has to override yours.
+q69W_5_xhoI-00006-00003117-00003265 And you'll find that in China
+q69W_5_xhoI-00007-00003265-00003565 you may win the argument, but it will cost you in the long run. It'll cost you
+q69W_5_xhoI-00008-00003565-00003864 in terms of relationship because of losing 'face'
+q69W_5_xhoI-00009-00003864-00004201 Now as an example of this, two personal friends of mine
+q69W_5_xhoI-00010-00004201-00004632 are engineers and they deal with the oil and gas supply.
+q69W_5_xhoI-00011-00004632-00004870 One guy is a British-based engineer.
+q69W_5_xhoI-00012-00004870-00005196 The other guy's also British, but he's China based.
+q69W_5_xhoI-00013-00005196-00005577 A couple of years ago they'd signed a rather large pipeline deal with
+q69W_5_xhoI-00014-00005577-00005771 a State-owned enterprise.
+q69W_5_xhoI-00015-00005771-00006163 They provided the the gas meters
+q69W_5_xhoI-00016-00006163-00006398 for this pipeline.
+q69W_5_xhoI-00017-00006398-00006745 A month after the deal was signed and everything was fitted together
+q69W_5_xhoI-00018-00006745-00006869 they were called out.
+q69W_5_xhoI-00019-00006869-00007134 The complaint was that the pressure, they couldn't get the right
+q69W_5_xhoI-00020-00007134-00007368 pressure to pump the gas through  the pipeline
+q69W_5_xhoI-00021-00007368-00007807 And since the pressure gauges were their responsibility they called out the British
+q69W_5_xhoI-00022-00007807-00008014 engineers.
+q69W_5_xhoI-00023-00008014-00008212 The first thing they did was go to meet
+q69W_5_xhoI-00024-00008212-00008514 the State-owned enterprise Chinese engineers.
+q69W_5_xhoI-00025-00008603-00008899 And having checked the machinery, checked the pipeline, the pressure gauges and
+q69W_5_xhoI-00026-00008899-00009140 realising that nothing was wrong with it
+q69W_5_xhoI-00027-00009140-00009307 the next thing they had to look at was
+q69W_5_xhoI-00028-00009307-00009451 the mathematics.
+q69W_5_xhoI-00029-00009451-00009821 So they called the chief engineer who can along with his team
+q69W_5_xhoI-00030-00009821-00010044 and asked to see the numbers.
+q69W_5_xhoI-00031-00010044-00010334 When he presented the numbers the British engineer, the British-based
+q69W_5_xhoI-00032-00010334-00010544 engineer, was very quick
+q69W_5_xhoI-00033-00010544-00010703 to spot the problem.
+q69W_5_xhoI-00034-00010703-00010962 And the problem was that according to him that,
+q69W_5_xhoI-00035-00010962-00011186 the figure they were using for gravity
+q69W_5_xhoI-00036-00011186-00011650 was a different figure to the universally accepted figures for gravity.
+q69W_5_xhoI-00037-00011650-00011867 The Chinese engineer denied it.
+q69W_5_xhoI-00038-00011867-00012094 Said that these figures were correct.
+q69W_5_xhoI-00039-00012094-00012284 The British based engineer
+q69W_5_xhoI-00040-00012284-00012424 got more angry
+q69W_5_xhoI-00041-00012424-00012878 and started to lecture him, in front of the rest of his team, using a lot
+q69W_5_xhoI-00042-00012878-00013019 of sarcasm,
+q69W_5_xhoI-00043-00013019-00013439 asking him in terms of, "If I drop an apple in China does it drop slower than if I drop
+q69W_5_xhoI-00044-00013439-00013721 an apple in England? Is that what you're saying to me?"
+q69W_5_xhoI-00045-00013721-00013894 Getting more and more aggressive with him
+q69W_5_xhoI-00046-00013894-00014084 using aggressive body language,
+q69W_5_xhoI-00047-00014084-00014546 aggressive gestures in front of his team.
+q69W_5_xhoI-00048-00014546-00014845 Now the China-based engineer,
+q69W_5_xhoI-00049-00014845-00015220 also British, but been in China for a while, realised that
+q69W_5_xhoI-00050-00015220-00015557 the knock-on effect of this guy losing face
+q69W_5_xhoI-00051-00015557-00015708 could be rather serious.
+q69W_5_xhoI-00052-00015708-00016057 So he stepped in and took his British colleague to one side,
+q69W_5_xhoI-00053-00016057-00016555 explained to him that this aggressive and reaction was not the way to solve
+q69W_5_xhoI-00054-00016555-00016827 the situation.
+q69W_5_xhoI-00055-00016827-00017203 While he was talking to the British engineer, the British based engineer
+q69W_5_xhoI-00056-00017203-00017438 the Chinese engineer disappeared.
+q69W_5_xhoI-00057-00017438-00017727 He came back a couple of minutes later with a book
+q69W_5_xhoI-00058-00017727-00018162 which he quite angrily waved in the face of the China based British
+q69W_5_xhoI-00059-00018162-00018331 engineer.
+q69W_5_xhoI-00060-00018331-00018742 He opened it and pointed vigorously to the figures.
+q69W_5_xhoI-00061-00018742-00019068 My friend, the China based British engineer
+q69W_5_xhoI-00062-00019068-00019195 looked at the book
+q69W_5_xhoI-00063-00019195-00019365 and very clearly
+q69W_5_xhoI-00064-00019365-00019562 the number had been scratched off
+q69W_5_xhoI-00065-00019562-00019752 and in inked back in,
+q69W_5_xhoI-00066-00019752-00020061 so freshly that he could have smudged it.
+q69W_5_xhoI-00067-00020061-00020170 However
+q69W_5_xhoI-00068-00020170-00020358 being aware of the situation
+q69W_5_xhoI-00069-00020358-00020566 he was able to look at the engineer
+q69W_5_xhoI-00070-00020566-00021004 and say "Ah, now I understand where the mistake was. The book has given you the wrong
+q69W_5_xhoI-00071-00021004-00021179 information. However
+q69W_5_xhoI-00072-00021179-00021409 if we use these figures
+q69W_5_xhoI-00073-00021409-00021626 we will be able to pump the gas through.
+q69W_5_xhoI-00074-00021626-00021992 The engineer was able to turn to his team and show that
+q69W_5_xhoI-00075-00021992-00022349 it wasn't his mistake, it was a printing error.
+q69W_5_xhoI-00076-00022349-00022473 On top of that
+q69W_5_xhoI-00077-00022473-00022741 my friend the China based British engineer
+q69W_5_xhoI-00078-00022741-00023122 had gained a lot of rapport with this Chinese guy.
+q69W_5_xhoI-00079-00023122-00023595 Instead of a natural Western reaction that you're lying to me, it's an insult to my
+q69W_5_xhoI-00080-00023595-00023699 intelligence
+q69W_5_xhoI-00081-00023699-00023934 he realised that actually the Chinese engineer
+q69W_5_xhoI-00082-00023934-00024258 was looking for steps. He just needed to avoid
+q69W_5_xhoI-00083-00024258-00024364 losing face
+q69W_5_xhoI-00084-00024364-00024579 in front of his team.
+q69W_5_xhoI-00085-00024579-00024861 They were very quickly able to fix the situation
+q69W_5_xhoI-00086-00024861-00025302 and the China based British engineer formed a very close working relationship
+q69W_5_xhoI-00087-00025302-00025438 with the Chinese guide
+q69W_5_xhoI-00088-00025438-00025714 because he understood how best to work with him.
+q69W_5_xhoI-00089-00026700-00027000 Subtitles by the Amara.org community
+q7AdMZPNHN4-00000-00000136-00000863 If you have been losing your balance and suffering from dizziness and falls, Claritox Pro just
+q7AdMZPNHN4-00001-00000863-00001192 may be be the solution you've been looking for.
+q7AdMZPNHN4-00002-00001192-00001868 Claritox Pro is a creation from Jim Benson, a 67 year old from Alma, Tennesee who's passionate
+q7AdMZPNHN4-00003-00001868-00002310 about plants and their ability to support our health and bodies.
+q7AdMZPNHN4-00004-00002310-00002843 For the past 10 years, he has been researching the natural ways for anyone to support their
+q7AdMZPNHN4-00005-00002843-00003167 balance and prevent dizziness and falls.
+q7AdMZPNHN4-00006-00003167-00003664 What he discovered is that most adults suffer from balance and dizziness problems due to
+q7AdMZPNHN4-00007-00003664-00004468 poor circulation, anxiety, not absorbing enough minerals into your blood stream and dehydration.
+q7AdMZPNHN4-00008-00004468-00005102 Typical solutions are to lie down until the dizziness passes, drink plenty of water and
+q7AdMZPNHN4-00009-00005102-00005429 cut down on the coffee, cigarettes and alcohol.
+q7AdMZPNHN4-00010-00005429-00005811 However, these only work temporarily.
+q7AdMZPNHN4-00011-00005811-00006360 If you want to take control of your health and correct the actual causes of your problems,
+q7AdMZPNHN4-00012-00006360-00006865 you need to improve the overall circulation in your body and absorb more minerals into
+q7AdMZPNHN4-00013-00006865-00007061 your blood stream.
+q7AdMZPNHN4-00014-00007061-00007581 This is how taking Claritox supplements can help.
+q7AdMZPNHN4-00015-00007581-00008190 Claritox pro is made from a powerful formula, created using a special blend of minerals
+q7AdMZPNHN4-00016-00008190-00009029 and plants, including Zinc, Chromium, Green Tea, Berberine, Resveratrol, Milk Thistle,
+q7AdMZPNHN4-00017-00009029-00009726 Cayenne, Korean Ginseng and Banaba, which aid in maintaining a healthy blood stream
+q7AdMZPNHN4-00018-00009726-00010315 and boosting your bodies ability to absorb the important minerals it needs.
+q7AdMZPNHN4-00019-00010315-00010990 Once your body finally begins to properly absorb the minerals and nutrients it's lacking,
+q7AdMZPNHN4-00020-00010990-00011503 your balance problems, dizziness and falls will become a thing of the past, which is
+q7AdMZPNHN4-00021-00011503-00011982 exactly what Claritox Pro is specially formulated to do.
+q7AdMZPNHN4-00022-00011982-00012681 Not only that, Claritox Pro is made in the USA, at an FDA approved and GMP certified
+q7AdMZPNHN4-00023-00012681-00013034 facility, so you know it's safe to take.
+q7AdMZPNHN4-00024-00013034-00013644 Even better, it doesn't contain any dangerous, habit forming stimulants or toxins.
+q7AdMZPNHN4-00025-00013644-00014331 So, if you are looking for a 100% natural, chemical free remedy to maintain a healthy
+q7AdMZPNHN4-00026-00014331-00014987 balance system and prevent dizziness and falls, check out Claritox Pro.
+q7AdMZPNHN4-00027-00014987-00015443 It's important to note that everyone's body chemistry is different.
+q7AdMZPNHN4-00028-00015443-00015819 Supplements that work for most don't work for everyone.
+q7AdMZPNHN4-00029-00015819-00016453 However, the best part about Claritox Pro is that you can try it for a full 60 days
+q7AdMZPNHN4-00030-00016453-00016620 to see if it's right for you.
+q7AdMZPNHN4-00031-00016620-00017101 And if for any reason you are not happy with the results you are getting, there's a full
+q7AdMZPNHN4-00032-00017101-00017372 money back guarantee.
+q7AdMZPNHN4-00033-00017372-00017759 To learn more about Claritox pro, click on the link below.
+q7AdMZPNHN4-00034-00017759-00018531 With a 60 day, 100% no questions asked money back guarantee, you have nothing to lose.
+q7AdMZPNHN4-00035-00018531-00019308 And you can finally say goodbye to your balance and dizziness problems for good.
+qq5XCVcNQVu-00000-00000558-00000941 Health insurance is designed to help you and your family cover the cost of your medical
+qq5XCVcNQVu-00001-00000941-00001317 care, including doctor visits and prescription medicines.
+qq5XCVcNQVu-00002-00001317-00001798 Similar to car insurance, you are responsible for paying a specific amount, or premium,
+qq5XCVcNQVu-00003-00001798-00002086 each month, whether or not you use any medical services.
+qq5XCVcNQVu-00004-00002086-00002536 In return, your insurer will pay a portion of your medical costs, should you or a family
+qq5XCVcNQVu-00005-00002536-00002703 member get sick or injured.
+qq5XCVcNQVu-00006-00002703-00003157 Keep in mind that your plan will likely require you to pay a portion of the expenses using
+qq5XCVcNQVu-00007-00003157-00003259 your own money.
+qq5XCVcNQVu-00008-00003259-00003846 These out-of-pocket costs may include a deductible, copayments, and coinsurance.
+qq5XCVcNQVu-00009-00003846-00004362 Now, a deductible is the amount you have to pay out-of-pocket before your plan will pay
+qq5XCVcNQVu-00010-00004362-00004504 any expenses.
+qq5XCVcNQVu-00011-00004504-00004993 This can include both medical and pharmacy expenses, meaning the deductible must be met
+qq5XCVcNQVu-00012-00004993-00005310 before getting coverage for some or all prescriptions.
+qq5XCVcNQVu-00013-00005310-00005895 So, for example, if your deductible is $1000, your plan won't pay any of your health care
+qq5XCVcNQVu-00014-00005895-00006353 costs until you have paid $1000 in expenses out-of-pocket.
+qq5XCVcNQVu-00015-00006353-00006709 Once you have met your deductible, you may still have to pay a certain amount out-of-pocket
+qq5XCVcNQVu-00016-00006709-00006871 for services.
+qq5XCVcNQVu-00017-00006871-00007247 One type of out-of-pocket cost is a copayment, or copay.
+qq5XCVcNQVu-00018-00007247-00007796 This is a flat fee that you must pay for specific health-related services such as a visit to
+qq5XCVcNQVu-00019-00007796-00008281 the ER for an injury, a trip to the doctor when you’re sick, or picking up a prescription
+qq5XCVcNQVu-00020-00008281-00008410 from the pharmacy.
+qq5XCVcNQVu-00021-00008410-00008854 For some services or prescriptions, your insurance might require a different kind of out-of-pocket
+qq5XCVcNQVu-00022-00008854-00009192 expense called coinsurance.
+qq5XCVcNQVu-00023-00009192-00009564 Coinsurance means you must pay a percentage of total costs, rather than a fixed dollar
+qq5XCVcNQVu-00024-00009564-00009664 amount.
+qq5XCVcNQVu-00025-00009664-00010149 So, for example, a medicine might require a 30% coinsurance.
+qq5XCVcNQVu-00026-00010149-00010560 On a $1000 medicine, you would have to pay $300.
+qq5XCVcNQVu-00027-00010560-00010979 Because coinsurance is a percentage and can widely vary based on the cost of a service
+qq5XCVcNQVu-00028-00010979-00011357 or product, it can be difficult to predict and plan for.
+qq5XCVcNQVu-00029-00011357-00011688 All health insurance plans are different, so make sure you are aware of what you may
+qq5XCVcNQVu-00030-00011688-00012010 have to pay to see your doctor or access the medicines you need.
+qtOijA71XWy-00000-00001188-00001318 Hello, everyone!
+qtOijA71XWy-00001-00001396-00001620 I really hope you enjoy this track
+qtOijA71XWy-00002-00001712-00002088 I originally wanted to make a 'sad EDM/Electronic song'
+qtOijA71XWy-00003-00002232-00002608 I thought it would be interesting and was not very common
+qtOijA71XWy-00004-00002728-00003112 However the vibe of the song did not really seem 'sad' to me
+qtOijA71XWy-00005-00003322-00003794 It seemed more like a mixture of emotions hence the title, 'Emotional'
+qtOijA71XWy-00006-00004052-00004436 You can come back to this track whenever you want to let some emotions out
+qtOijA71XWy-00007-00004534-00004672 Sadness, pain
+qtOijA71XWy-00008-00004708-00004848 Joy, happiness
+qtOijA71XWy-00009-00004880-00005018 Any emotion
+qtOijA71XWy-00010-00005070-00005286 Or maybe just to feel..something
+qtOijA71XWy-00011-00005418-00005688 I just want you to know that..
+qtOijA71XWy-00012-00005720-00006034 whatever emotions you are feeling
+qtOijA71XWy-00013-00006114-00006360 just let them flow!
+qtOijA71XWy-00014-00006398-00006716 just allow yourself to feel them
+qtOijA71XWy-00015-00006786-00007068 It's OKAY to feel sad,
+qtOijA71XWy-00016-00007108-00007408 It's also okay to feel joy, you deserved it
+qtOijA71XWy-00017-00007548-00007934 Especially for my dudes..just know it's OKAY to cry!
+qtOijA71XWy-00018-00007973-00008306 People say we need to be strong and that's fine but..
+qtOijA71XWy-00019-00008340-00008656 what makes you think crying makes you weak :)
+qtOijA71XWy-00020-00008772-00009156 well anyways, just know that YOU'RE OKAY
+qtOijA71XWy-00021-00009192-00009612 and I hope I could reach a few people who really needed to hear that :)
+qtOijA71XWy-00022-00009652-00010122 I really hope this track helped you at least a little bit to release some emotions
+qtOijA71XWy-00023-00010154-00010450 or maybe you just enjoyed the music
+qtOijA71XWy-00024-00010544-00010992 In any case I'll shut up now haha...let me know what you thought of the song and these captions :)
+qtOijA71XWy-00025-00013213-00013358 Ohh I love this part x)
+qtbKf4BRFYM-00000-00000876-00001469 When disaster or emergency strike that cause major damage to your home ow work place such
+qtbKf4BRFYM-00001-00001469-00002078 as fire or flooding, naturally you want to restore you property to its original state.
+qtbKf4BRFYM-00002-00002078-00002645 So, if you are looking for the best professional, reliable and affordable fire or water damage
+qtbKf4BRFYM-00003-00002645-00002770 restoration repair company.
+qtbKf4BRFYM-00004-00002770-00002981 You've come to the right place.
+qtbKf4BRFYM-00005-00002981-00003429 We are proud to be the premier local restoration repair company and customer
+qtbKf4BRFYM-00006-00003429-00003750 satisfaction is our #1 priority.
+qtbKf4BRFYM-00007-00003750-00004091 We are fully licensed and insured and our restoration repair
+qtbKf4BRFYM-00008-00004091-00004674 specialists have years of experience dealing with every type of situation you can imagine.
+qtbKf4BRFYM-00009-00004674-00006475 So contact us now and let us help put your property back the way it was.
+qxyg3Z6gsh0-00000-00000000-00000418 Good morning.  My name is Mark Meyer. I'm one of the owners of Wigle Whiskey.
+qxyg3Z6gsh0-00001-00000816-00001488 We are a small, family-owned business, so we understand as well as anybody the challenges of the
+qxyg3Z6gsh0-00002-00001488-00002184 small business. We also understand, the financial pressures on workers today.
+qxyg3Z6gsh0-00003-00002598-00002828 It is tough out there.
+qxyg3Z6gsh0-00004-00002828-00003440 And so we think people should do what they can. And what we can do as a small, family-
+qxyg3Z6gsh0-00005-00003440-00004110 owned business is participate in the Pittsburgh Living Wage Program.
+qxyg3Z6gsh0-00006-00004461-00004766 We applaud the governor
+qxyg3Z6gsh0-00007-00004766-00005423 for bringing the issue to the table and we as a small business support the idea
+qxyg3Z6gsh0-00008-00005423-00006021 that Pennsylvania workers should get a small raise for their work.
+qxyg3Z6gsh0-00009-00006414-00006886 It is time to give Pennsylvania workers a raise.
+qzUIHGqBluo-00000-00000000-00000509 Christmas intro music 
+qzUIHGqBluo-00001-00000690-00001105 Narrator: Nosie Christmas at the Supermarket. 
+qzUIHGqBluo-00002-00001161-00001361 Santa claus: Ho Ho Ho...
+qzUIHGqBluo-00003-00001384-00002064 Santa claus: What's the secret of the cheapest supermarket in the world? 
+qzUIHGqBluo-00004-00002084-00002530 Santa Claus: It's me ofcourse, and the presents! Ho ho ho...
+qzUIHGqBluo-00005-00002537-00003096 Santa Claus: Shhtt... Don't tell anyone dear readers, it's our secret!
+qzUIHGqBluo-00006-00003176-00003484 Nosie: So, now I know your secret. 
+qzUIHGqBluo-00007-00003537-00003706 Santa Claus: Nosie?! 
+qzUIHGqBluo-00008-00003781-00004129 Santa Claus: B.. but how did you know I'm here? 
+qzUIHGqBluo-00009-00004195-00004395 Nosie: You know me, I'm the best!
+qzUIHGqBluo-00010-00004480-00004949 Santa Claus: However, please don't tell anyone about this, deal?
+qzUIHGqBluo-00011-00005000-00005410 Nosie: hmm... okay, only for one condition.
+qzUIHGqBluo-00012-00005437-00005781 Santa Claus: Ho ho ho... Merry Christmas! 
+qzUIHGqBluo-00013-00005818-00006018 Customer: Thank you Santa! 
+qzUIHGqBluo-00014-00006052-00006252 Nosie: I love it! 
+q-n85ii01ag-00000-00000016-00000523 In this video, I will show you how easy it is to migrate from ownCloud 10 to
+q-n85ii01ag-00001-00000523-00001069 Nextcloud 12. As you see I have ownCloud up and running here, including a couple of
+q-n85ii01ag-00002-00001069-00001510 applications. Like contacts and calendar, I also installed the audio
+q-n85ii01ag-00003-00001510-00002076 player app but for some reason it does not show up. So, let's start the migration!
+q-n85ii01ag-00004-00002076-00002509 So you simply go into the folder where ownCloud is installed and you copy-paste the
+q-n85ii01ag-00005-00002509-00003106 commands to download the "index.php" file, which is the migrator. And then you
+q-n85ii01ag-00006-00003106-00003535 move it into the "updater" folder, you see it gets named "index.php.1"
+q-n85ii01ag-00007-00003535-00003951 because there is already "index.php" file.
+q-n85ii01ag-00008-00004098-00004598 And then you simply visit it into the browser. So you go to "/updater/index.php".
+q-n85ii01ag-00009-00004598-00005257 And there is the migrator, it correctly detects
+q-n85ii01ag-00010-00005257-00005767 we are running ownCloud 10.0.0 and we can migrate to Nextcloud 12. So let's get going.
+q-n85ii01ag-00011-00005767-00006326 The system is checked, put into maintenance modes, backup is made, files are downloaded
+q-n85ii01ag-00012-00006326-00007222 the integrity is verified, the files are extracted.
+q-n85ii01ag-00013-00007222-00007880 And very quickly we are ready to continue. Now, we can let the web-based updater do it's thing, but I'm
+q-n85ii01ag-00014-00007880-00008266 going to use a command line tool to show how that works.
+q-n85ii01ag-00015-00008460-00009254 So the system is in maintenance mode and we run the upgrades. Oh, that's easy
+q-n85ii01ag-00016-00009254-00009896 and quick because that's done now. So now we need to turn off maintenance mode, but I always
+q-n85ii01ag-00017-00009896-00011029 forget the exact command. "maintenance:mode --off".
+q-n85ii01ag-00018-00011077-00011855 Now, let's check in our Nextcloud instance we've upgraded to a much faster and more
+q-n85ii01ag-00019-00011855-00012665 secure Nextcloud. Just a refresh and here we are, you see there's a problem with
+q-n85ii01ag-00020-00012665-00013325 the code integrity check, Nextcloud checks if all the files are secure so you see
+q-n85ii01ag-00021-00013325-00013661 number of files that have failed the integrity check and this is the "audioplayer",
+q-n85ii01ag-00022-00013661-00014033 the "calendar" and "contacts" and the "impersonate" apps. So these certificates
+q-n85ii01ag-00023-00014033-00014393 are not trusted they're still ownCloud certificates, they're not Nextcloud
+q-n85ii01ag-00024-00014393-00014879 certificates. So you wonder, what can we do about this, well, we're going to go
+q-n85ii01ag-00025-00014879-00015157 back to the App Store.
+q-n85ii01ag-00026-00015221-00015716 So we simply need to replace them with versions from Nextcloud. Well, you see
+q-n85ii01ag-00027-00015716-00016185 there are newer versions available so that is very easy. The audio player
+q-n85ii01ag-00028-00016185-00016827 doesn't have a newer version available, so we simply remove this one. So we
+q-n85ii01ag-00029-00016827-00017324 reenable the "calendar", "contacts" and "impersonate" apps. So you see them up here at the top.
+q-n85ii01ag-00030-00017324-00017811 And then we go and install the audio player app and there is also video calls!
+q-n85ii01ag-00031-00017811-00018612 Let's install that too! So now these apps are back!
+q-n85ii01ag-00032-00018612-00019122 So we go back to the integrity checks and let it do a rescan. And now you will see, that all files
+q-n85ii01ag-00033-00019122-00019851 passed the check. Note that you will not have to reinstall or reenable apps when
+q-n85ii01ag-00034-00019851-00020348 you do an update from Nextcloud because on Nextcloud 12 we automatically reenable apps
+q-n85ii01ag-00035-00020348-00020704 that were enabled before you run an update.
+q-n85ii01ag-00036-00020704-00021026 Thank you for watching!
+q_jWe7Mv_FQ-00000-00000000-00000488 The New Banner system isn't just a change in banners, and you'd probably thank mihoyo that
+q_jWe7Mv_FQ-00001-00000488-00000888 they did it. Although, there are some new things you need to think about before you
+q_jWe7Mv_FQ-00002-00000888-00001392 start spending your gems. Here's 5 things to consider when pulling after patch 2.3
+q_jWe7Mv_FQ-00003-00001496-00001896 Now the new banner system is subjective if it's good or bad,
+q_jWe7Mv_FQ-00004-00001896-00002448 but bottomline is this banner type is what we’ll be having from patch 2.3 onwards,
+q_jWe7Mv_FQ-00005-00002448-00003032 which means we can have a plethora of combinations from double dps banners like Xiao and Ganyu, or
+q_jWe7Mv_FQ-00006-00003032-00003632 best paired units like Hutao and Zhongli. We can also get double support like Albedo and Kazuha,
+q_jWe7Mv_FQ-00007-00003736-00004352 or even random combinations like idk… Yoimiya and Klee. So keep an eye out for Banner Announcements
+q_jWe7Mv_FQ-00008-00004352-00004784 because you’ll have to either think more on who you want better for your roster,
+q_jWe7Mv_FQ-00009-00004784-00005344 or just try your luck and get both because of synergy while pulling within their durations.
+q_jWe7Mv_FQ-00010-00005344-00005632 At number 2 is that new is that new characters will last longer.
+q_jWe7Mv_FQ-00011-00005920-00006392 I don't really know if this is confirmed or if I just misread something. But since reruns will
+q_jWe7Mv_FQ-00012-00006392-00007016 come in pairs, we can hope and be glad that new characters will have the full 20+ day duration. So
+q_jWe7Mv_FQ-00013-00007128-00007720 more chances to get the older characters, and more time to try and get that new 5 star character.
+q_jWe7Mv_FQ-00014-00007720-00008352 Granted is basically the same time as before but hey at least they won't half the new character
+q_jWe7Mv_FQ-00015-00008480-00008840 banners.More consistent reruns I've said this before but ill
+q_jWe7Mv_FQ-00016-00008840-00009320 elaborate on it more here. So not only do we get more choice when we save our primos,
+q_jWe7Mv_FQ-00017-00009320-00009880 we also get to use those primos that we save with less stress from saving them up. Unless of course
+q_jWe7Mv_FQ-00018-00009880-00010360 you're waiting for that favourite 5 star. you can pretty much be more relaxed when using your
+q_jWe7Mv_FQ-00019-00010360-00010936 primos, because the characters will be back after 2 patches or so. Similar to the Weapon banner but
+q_jWe7Mv_FQ-00020-00010936-00011552 you get a better rebound if you fail your 50/50. Let's say you lost in getting hutao, well that's
+q_jWe7Mv_FQ-00021-00011552-00012000 okay because depending on the next character, you can get zhongli the best tank ever, or
+q_jWe7Mv_FQ-00022-00012000-00012784 Ayaka or Ganyu the Best Cryo dps. Or you can wait until the next banner announces and use it there.
+q_jWe7Mv_FQ-00023-00013144-00014000 Again more choices for 5 stars, and more chances to get them.4 stars will be the same after 2.3
+q_jWe7Mv_FQ-00024-00014000-00014432 This is either a good or bad depending on the player. If you want more constellations for
+q_jWe7Mv_FQ-00025-00014432-00015080 your 4 star but cant quite get lucky enough per 10 pull, you might want to keep the banner that way.
+q_jWe7Mv_FQ-00026-00015144-00015728 For players looking to diversify their roster or new players looking for more characters to try out
+q_jWe7Mv_FQ-00027-00015728-00016696 this would be good. And depending on the 4 star you might wanna go for their c2, 4 ,6 instead
+q_jWe7Mv_FQ-00028-00016696-00018344 of c0. Possibly more new characters per patch This is just a theory, but since we can get more
+q_jWe7Mv_FQ-00029-00018344-00018832 characters, we could get more new characters after every patch. We’ll be seeing more of
+q_jWe7Mv_FQ-00030-00018832-00019424 both 4 star and 5 star new characters in the long run. Sumeru being just after a few more patches,
+q_jWe7Mv_FQ-00031-00019424-00020080 combined with this new system. I guess we can hope instead of coping for more characters,
+q_jWe7Mv_FQ-00032-00020080-00020792 and better chances at getting them. This video just sums up what we can expect in
+q_jWe7Mv_FQ-00033-00020792-00021408 the future, note really a theory video, but kind of a theory video in the form of more gameplay,
+q_jWe7Mv_FQ-00034-00021472-00022120 meta sense than story and lore. Hope you guys liked that video, be sure to hit the like button
+q_jWe7Mv_FQ-00035-00022120-00022664 if you found this helpful. comment below, say hi, or tell me what you think. Subscribe if you're
+q_jWe7Mv_FQ-00036-00022664-00023296 new to my channel i wanna hit 1k subs soon so yea click on it. And if you're interested
+q_jWe7Mv_FQ-00037-00023296-00023984 in more of my videos click the bell icon to stay updated to my videos. Ill see you guys later. bye!
+qB0J84LuLII-00000-00000002-00000142 When I was trying to turn
+qB0J84LuLII-00001-00000164-00000214 unconsciously
+qB0J84LuLII-00002-00000225-00000426 I had a slight runny nose
+qB0J84LuLII-00003-00000478-00000532 I saw it!
+qB0J84LuLII-00004-00000614-00000664 I saw it! [Coming out many Witnesses]
+qB0J84LuLII-00005-00000702-00000772 I saw it!
+qB0J84LuLII-00006-00000824-00000916 Please pay attention to listen to it
+qB0J84LuLII-00007-00000966-00001054 It was not gone (It : The snot)
+qB0J84LuLII-00008-00001132-00001224 I caught it
+qB0J84LuLII-00009-00001272-00001400 It didn't fly
+qB0J84LuLII-00010-00001440-00001514 Who's part?
+qB0J84LuLII-00011-00001556-00001742 Mill was here and I was there
+qB0J84LuLII-00012-00001764-00001814 Right side...
+qB0J84LuLII-00013-00001870-00001994 I had a slight runny nose
+qB0J84LuLII-00014-00002146-00002260 I hit by it
+qB0J84LuLII-00015-00002342-00002402 It didn't fly
+qB0J84LuLII-00016-00002450-00002524 I caught it
+qB0J84LuLII-00017-00002662-00002834 I threw it and it hit to Nine
+qB0J84LuLII-00018-00002862-00002944 I hit it
+qB0J84LuLII-00019-00002994-00003070 I didn't fly
+qB0J84LuLII-00020-00003094-00003164 I tasted it
+qB0J84LuLII-00021-00003454-00003515 Can you see?
+qB0J84LuLII-00022-00003534-00003614 Yes, I can see
+qB0J84LuLII-00023-00003636-00003686 I'm available to see it
+qB0J84LuLII-00024-00003714-00003906 Family name is Sap Given name is Salgae (Sap Salgae : Shovel)
+qB0J84LuLII-00025-00003984-00004079 How do you think about this hair concept?
+qB0J84LuLII-00026-00004186-00004294 This hair concept
+qB0J84LuLII-00027-00004320-00004438 Curtain hair
+qB0J84LuLII-00028-00004508-00004656 I was trying to practice hard
+qB0J84LuLII-00029-00004744-00004944 Please don't think I'm cute
+qB0J84LuLII-00030-00004966-00005128 I need you guys to focus more on the sexiness
+qB0J84LuLII-00031-00005150-00005294 Our group needs to emphasize sexiness
+qB0J84LuLII-00032-00005348-00005398 I want you to look sexy
+qB0J84LuLII-00033-00005544-00005940 I have been practicing and figuring out on how do we show the movements and angles of the choreography
+qB0J84LuLII-00034-00005940-00006168 I definitely see in front of the camera
+qB0J84LuLII-00035-00006168-00006332 Please don't make me cute
+qB0J84LuLII-00036-00006444-00006592 Please look forward to our sensuality
+qB0J84LuLII-00037-00006592-00006716 I love you all!
+qB0J84LuLII-00038-00006748-00006798 (Nine : Making a heart)
+qB0J84LuLII-00039-00006900-00006998 (Nine : Make a heart over your head)
+qB0J84LuLII-00040-00007092-00007178 JunJi : Like this?
+qB0J84LuLII-00041-00007190-00007240 Nine : No~ No~
+qB0J84LuLII-00042-00007322-00007386 Nine : Okay x3
+qB0J84LuLII-00043-00007409-00007508 Nine : Make a heart under
+qB0J84LuLII-00044-00007584-00007634 Nine : Okay
+qB0J84LuLII-00045-00007776-00007834 Nine : Making a finger heart
+qB0J84LuLII-00046-00008138-00008232 Nine : Again x2
+qB0J84LuLII-00047-00008324-00008382 Nine : I'm zooming right now
+qB0J84LuLII-00048-00008462-00008626 Nine :  1, 2, 3
+qBwYODQzi90-00000-00000300-00000600 ADG is an open source software
+qBwYODQzi90-00001-00000650-00001000 for automatic generation of technical drawings.
+qBwYODQzi90-00002-00001050-00001400 The idea was born in the nineties
+qBwYODQzi90-00003-00001450-00001800 with the aim to generate semifinished, finished and drilling draftings
+qBwYODQzi90-00004-00001850-00002300 of a thousand diesel nozzles.
+qBwYODQzi90-00005-00002450-00002800 Recording technical data into a database
+qBwYODQzi90-00006-00002850-00003200 allows to perform complex actions such as SQL queries on the quotes,
+qBwYODQzi90-00007-00003250-00003700 integration of project management software and batch processing.
+qBwYODQzi90-00008-00003750-00004500 This know-how granted the accurated shaping of the ADG requirements.
+qBwYODQzi90-00009-00004750-00005400 A specific feature is the adoption of two distinct spaces: local and global.
+qBwYODQzi90-00010-00006450-00006900 A chunk of the project provides a bunch of helper functions
+qBwYODQzi90-00011-00006950-00007400 for executing geometrical operations such as intersections, affine transformations
+qBwYODQzi90-00012-00007450-00007700 and the offset.
+qBwYODQzi90-00013-00008050-00008600 eNTiDi enters the market to develop ADG based applications.
+qBwYODQzi90-00014-00008650-00009000 A basic example (documented by the on-line tutorial)
+qBwYODQzi90-00015-00009050-00009600 is a bare database front-end that shows a table row in the canvas.
+qBwYODQzi90-00016-00009650-00010100 A more serious application is the demo distributed with the ADG
+qBwYODQzi90-00017-00010150-00010600 that allows the definition of new mechanical parts.
+qBwYODQzi90-00018-00010650-00011000 Its interactive use permits the creation of new parts
+qBwYODQzi90-00019-00011050-00011300 literally in seconds
+qBwYODQzi90-00020-00011350-00012000 minimizing in the meantime the mistakes and focusing efforts where needed.
+qBwYODQzi90-00021-00012050-00012400 It is also possible to add or remove machinings
+qBwYODQzi90-00022-00012450-00012900 up to radical changing a detail of the drawing leaving the rest untouched,
+qBwYODQzi90-00023-00012950-00013500 as already done for the tips of the diesel nozzles.
+qBwYODQzi90-00024-00013950-00014400 eNTiDi develops both desktop and web applications,
+qBwYODQzi90-00025-00014450-00014900 the latter are to be preferred whenever sharing and multiplatform support
+qBwYODQzi90-00026-00014950-00015250 are importants.
+qBwYODQzi90-00027-00015300-00015600 The online demo hosted on the ADG web site
+qBwYODQzi90-00028-00015650-00015950 performs the same operations as the official ADG demo
+qBwYODQzi90-00029-00016000-00016400 but it has been implemented on the web infrastructure.
+qBwYODQzi90-00030-00016450-00016800 This also means that the drawing can be manipulated with tablets and smartphones
+qBwYODQzi90-00031-00016850-00017500 anywhere an internet connection is available.
+qBwYODQzi90-00032-00017650-00018250 The source code will be released to clients in order not to bind them to eNTiDi.
+qBwYODQzi90-00033-00018300-00018800 Therefore, special attention has been paid to ADG documentation:
+qBwYODQzi90-00034-00018850-00019100 you can find online tutorials,
+qBwYODQzi90-00035-00019150-00019600 introductory material and reference manuals.
+qBEREVqdxSY-00000-00000010-00000086 >> Reporter: REPORTING IN 
+qBEREVqdxSY-00001-00000086-00000093 >> Reporter: REPORTING IN   
+qBEREVqdxSY-00002-00000093-00000153 >> Reporter: REPORTING IN  HACKENSACK, LAKELAND NEWS.
+qBEREVqdxSY-00003-00000153-00000160 >> Reporter: REPORTING IN  HACKENSACK, LAKELAND NEWS.  
+qBEREVqdxSY-00004-00000160-00000373 >> Reporter: REPORTING IN  HACKENSACK, LAKELAND NEWS. >> Dennis: 50 TO 100 VOLUNTEERS 
+qBEREVqdxSY-00005-00000373-00000380 HACKENSACK, LAKELAND NEWS. >> Dennis: 50 TO 100 VOLUNTEERS   
+qBEREVqdxSY-00006-00000380-00000547 HACKENSACK, LAKELAND NEWS. >> Dennis: 50 TO 100 VOLUNTEERS  FROM HACKENSACK HELPED OUT WITH 
+qBEREVqdxSY-00007-00000547-00000553 >> Dennis: 50 TO 100 VOLUNTEERS  FROM HACKENSACK HELPED OUT WITH   
+qBEREVqdxSY-00008-00000553-00000683 >> Dennis: 50 TO 100 VOLUNTEERS  FROM HACKENSACK HELPED OUT WITH  THIS YEAR'S EVENT.
+qBEREVqdxSY-00009-00000683-00000690 FROM HACKENSACK HELPED OUT WITH  THIS YEAR'S EVENT.  
+qBEREVqdxSY-00010-00000690-00000837 FROM HACKENSACK HELPED OUT WITH  THIS YEAR'S EVENT. THE SALVATION ARMY IN BRAINERD 
+qBEREVqdxSY-00011-00000837-00000844 THIS YEAR'S EVENT. THE SALVATION ARMY IN BRAINERD   
+qBEREVqdxSY-00012-00000844-00001010 THIS YEAR'S EVENT. THE SALVATION ARMY IN BRAINERD  IS HELPING EASE THE COST OF 
+qBEREVqdxSY-00013-00001010-00001017 THE SALVATION ARMY IN BRAINERD  IS HELPING EASE THE COST OF   
+qBEREVqdxSY-00014-00001017-00001141 THE SALVATION ARMY IN BRAINERD  IS HELPING EASE THE COST OF  SCHOOL SUPPLIES FOR FAMILIES 
+qBEREVqdxSY-00015-00001141-00001147 IS HELPING EASE THE COST OF  SCHOOL SUPPLIES FOR FAMILIES   
+qBEREVqdxSY-00016-00001147-00001394 IS HELPING EASE THE COST OF  SCHOOL SUPPLIES FOR FAMILIES  WITH THEIR ANNUAL BACK-TO-SCHOOL
+qBEREVqdxSY-00017-00001394-00001401 SCHOOL SUPPLIES FOR FAMILIES  WITH THEIR ANNUAL BACK-TO-SCHOOL  
+qBEREVqdxSY-00018-00001401-00001444 SCHOOL SUPPLIES FOR FAMILIES  WITH THEIR ANNUAL BACK-TO-SCHOOL BACKPACK PROGRAM.
+qBEREVqdxSY-00019-00001444-00001451 WITH THEIR ANNUAL BACK-TO-SCHOOL BACKPACK PROGRAM.  
+qBEREVqdxSY-00020-00001451-00001785 WITH THEIR ANNUAL BACK-TO-SCHOOL BACKPACK PROGRAM. ALL YOU NEED TO RECEIVE A 
+qBEREVqdxSY-00021-00001785-00001791 BACKPACK PROGRAM. ALL YOU NEED TO RECEIVE A   
+qBEREVqdxSY-00022-00001791-00002419 BACKPACK PROGRAM. ALL YOU NEED TO RECEIVE A  BACKPACK IS A PHOTO I AND PROOF
+qBEREVqdxSY-00023-00002419-00002425 ALL YOU NEED TO RECEIVE A  BACKPACK IS A PHOTO I AND PROOF  
+qBEREVqdxSY-00024-00002425-00002709 ALL YOU NEED TO RECEIVE A  BACKPACK IS A PHOTO I AND PROOF OF A STUDENT, SUCH AS A SOCIAL 
+qBEREVqdxSY-00025-00002709-00002716 BACKPACK IS A PHOTO I AND PROOF OF A STUDENT, SUCH AS A SOCIAL   
+qBEREVqdxSY-00026-00002716-00002779 BACKPACK IS A PHOTO I AND PROOF OF A STUDENT, SUCH AS A SOCIAL  SECURITY CARD.
+qBEREVqdxSY-00027-00002779-00002786 OF A STUDENT, SUCH AS A SOCIAL  SECURITY CARD.  
+qBEREVqdxSY-00028-00002786-00003293 OF A STUDENT, SUCH AS A SOCIAL  SECURITY CARD. THEY HOSTED A COMMUNITY RESOURCE
+qBEREVqdxSY-00029-00003293-00003299 SECURITY CARD. THEY HOSTED A COMMUNITY RESOURCE  
+qBEREVqdxSY-00030-00003299-00003570 SECURITY CARD. THEY HOSTED A COMMUNITY RESOURCE FAIR, PROVIDING INFORMATION ON 
+qBEREVqdxSY-00031-00003570-00003576 THEY HOSTED A COMMUNITY RESOURCE FAIR, PROVIDING INFORMATION ON   
+qBEREVqdxSY-00032-00003576-00003693 THEY HOSTED A COMMUNITY RESOURCE FAIR, PROVIDING INFORMATION ON  RESOURCES AVAILABLE IN THE AREA.
+qBEREVqdxSY-00033-00003693-00003700 FAIR, PROVIDING INFORMATION ON  RESOURCES AVAILABLE IN THE AREA.  
+qBEREVqdxSY-00034-00003700-00003883 FAIR, PROVIDING INFORMATION ON  RESOURCES AVAILABLE IN THE AREA. >> WE COLLECT UP DONATED 
+qBEREVqdxSY-00035-00003883-00003890 RESOURCES AVAILABLE IN THE AREA. >> WE COLLECT UP DONATED   
+qBEREVqdxSY-00036-00003890-00004033 RESOURCES AVAILABLE IN THE AREA. >> WE COLLECT UP DONATED  SUPPLIES AND DONATED FUNDS TO 
+qBEREVqdxSY-00037-00004033-00004040 >> WE COLLECT UP DONATED  SUPPLIES AND DONATED FUNDS TO   
+qBEREVqdxSY-00038-00004040-00004267 >> WE COLLECT UP DONATED  SUPPLIES AND DONATED FUNDS TO  PURCHASE BACKPACKS AND SCHOOL 
+qBEREVqdxSY-00039-00004267-00004274 SUPPLIES AND DONATED FUNDS TO  PURCHASE BACKPACKS AND SCHOOL   
+qBEREVqdxSY-00040-00004274-00004481 SUPPLIES AND DONATED FUNDS TO  PURCHASE BACKPACKS AND SCHOOL  SUPPLIES, A LITTLE BIT OF BOTH 
+qBEREVqdxSY-00041-00004481-00004487 PURCHASE BACKPACKS AND SCHOOL  SUPPLIES, A LITTLE BIT OF BOTH   
+qBEREVqdxSY-00042-00004487-00004731 PURCHASE BACKPACKS AND SCHOOL  SUPPLIES, A LITTLE BIT OF BOTH  THAT GOES INTO THE PROGRAM AND 
+qBEREVqdxSY-00043-00004731-00004738 SUPPLIES, A LITTLE BIT OF BOTH  THAT GOES INTO THE PROGRAM AND   
+qBEREVqdxSY-00044-00004738-00004764 SUPPLIES, A LITTLE BIT OF BOTH  THAT GOES INTO THE PROGRAM AND  GRANTS.
+qBEREVqdxSY-00045-00004764-00004771 THAT GOES INTO THE PROGRAM AND  GRANTS.  
+qBEREVqdxSY-00046-00004771-00005014 THAT GOES INTO THE PROGRAM AND  GRANTS. WE SERVED, I THINK, ABOUT 500 
+qBEREVqdxSY-00047-00005014-00005021 GRANTS. WE SERVED, I THINK, ABOUT 500   
+qBEREVqdxSY-00048-00005021-00005085 GRANTS. WE SERVED, I THINK, ABOUT 500  LAST YEAR.
+qBEREVqdxSY-00049-00005085-00005091 WE SERVED, I THINK, ABOUT 500  LAST YEAR.  
+qBEREVqdxSY-00050-00005091-00005408 WE SERVED, I THINK, ABOUT 500  LAST YEAR. EACH YEAR, IT VARIES DEPENDING 
+qBEREVqdxSY-00051-00005408-00005415 LAST YEAR. EACH YEAR, IT VARIES DEPENDING   
+qBEREVqdxSY-00052-00005415-00005625 LAST YEAR. EACH YEAR, IT VARIES DEPENDING  ON WHICH SCHOOLS MAYBE HAVE 
+qBEREVqdxSY-00053-00005625-00005632 EACH YEAR, IT VARIES DEPENDING  ON WHICH SCHOOLS MAYBE HAVE   
+qBEREVqdxSY-00054-00005632-00005889 EACH YEAR, IT VARIES DEPENDING  ON WHICH SCHOOLS MAYBE HAVE  OTHER SPONSORS.
+qBEREVqdxSY-00055-00005889-00005895 ON WHICH SCHOOLS MAYBE HAVE  OTHER SPONSORS.  
+qBEREVqdxSY-00056-00005895-00005952 ON WHICH SCHOOLS MAYBE HAVE  OTHER SPONSORS. >> Dennis: THE BRAINERD 
+qBEREVqdxSY-00057-00005952-00005959 OTHER SPONSORS. >> Dennis: THE BRAINERD   
+qBEREVqdxSY-00058-00005959-00006079 OTHER SPONSORS. >> Dennis: THE BRAINERD  SALVATION ARMY IS HOLDING 
+qBEREVqdxSY-00059-00006079-00006086 >> Dennis: THE BRAINERD  SALVATION ARMY IS HOLDING   
+qBEREVqdxSY-00060-00006086-00006136 >> Dennis: THE BRAINERD  SALVATION ARMY IS HOLDING  ANOTHER BACKPACK DISTRIBUTION 
+qCdC1YrN5ek-00000-00000600-00000700 Killer
+qCdC1YrN5ek-00001-00000800-00001050 Victim
+qCdC1YrN5ek-00002-00001250-00001400 Guilty
+qCdC1YrN5ek-00003-00001600-00001800 Innocent
+qCdC1YrN5ek-00004-00002400-00002700 The Forgotten
+qCdC1YrN5ek-00005-00004000-00004400 Coming Soon
+qCSe62n_SEk-00000-00000866-00001073 RONDO 2 has several indicator lights.
+qCSe62n_SEk-00001-00001128-00001398 These tell you information about your audio processor.
+qCSe62n_SEk-00002-00001534-00001696 When turning on RONDO 2
+qCSe62n_SEk-00003-00001696-00001988 the blue indicator light shows which program is selected.
+qCSe62n_SEk-00004-00002038-00002296 One blue blink indicates that program one is selected.
+qCSe62n_SEk-00005-00002348-00002633 Two blue blinks indicate that program two is selected.
+qCSe62n_SEk-00006-00002676-00002964 Three blue blinks indicate that program three is selected.
+qCSe62n_SEk-00007-00003026-00003288 Four blue blinks indicate that program four is selected.
+qCSe62n_SEk-00008-00003424-00003629 If the optional status light is activated
+qCSe62n_SEk-00009-00003658-00004154 a short blue blink will blink every 3.5 seconds to show that RONDO 2 is working properly.
+qCSe62n_SEk-00010-00004268-00004506 When using the FineTuner remote with your RONDO 2
+qCSe62n_SEk-00011-00004506-00004860 the blue light will blink shortly every time a FineTuner command
+qCSe62n_SEk-00012-00004860-00005018 is received by your processor.
+qCSe62n_SEk-00013-00005146-00005520 A continuous blue light indicates that either minimum or maximum volume
+qCSe62n_SEk-00014-00005520-00005760 or microphone sensitivity has been reached.
+qCSe62n_SEk-00015-00005868-00006262 A rapidly blinking blue indicator light shows that RONDO 2 needs charging.
+qCSe62n_SEk-00016-00006372-00006566 If the link check function is activated
+qCSe62n_SEk-00017-00006592-00006802 the green indicator light will blink three times
+qCSe62n_SEk-00018-00006802-00007022 when RONDO 2 is placed over the implant.
+qCSe62n_SEk-00019-00007062-00007280 This shows that RONDO 2 is working correctly
+qCSe62n_SEk-00020-00007288-00007440 and is communicating with your implant.
+qCSe62n_SEk-00021-00007572-00008016 If you have a C 40 or C 40+ implant, the green light will blink quickly once.
+qCSe62n_SEk-00022-00008062-00008284 This shows that your processor is working correctly.
+qCSe62n_SEk-00023-00008390-00008832 If the red indicator light blinks repeatedly RONDO 2 is positioned over the wrong implant.
+qCSe62n_SEk-00024-00008854-00009162 Ensure RONDO 2 is positioned over the correct implant.
+qCSe62n_SEk-00025-00009316-00009522 Short red blinks 10 seconds apart
+qCSe62n_SEk-00026-00009522-00009724 indicated RONDO 2 has powered off.
+qCSe62n_SEk-00027-00009770-00010086 Turn it off and on again and reposition it over the implant.
+qCSe62n_SEk-00028-00010180-00010682 Any other indicator light signals suggest that there may be a technical issue with your RONDO 2.
+qCSe62n_SEk-00029-00010720-00010986 Contact your CI Center for further assistance.
+qDvt7OTS-SQ-00000-00001002-00001757 On the first part of the journey We were headed off to Bree
+qDvt7OTS-SQ-00001-00001757-00002494 Tom Bombadil kept rescuing Us and Uncle Bilbo’s ring
+qDvt7OTS-SQ-00002-00002494-00003244 We got to the Prancing Pony Met a ranger with a hilt
+qDvt7OTS-SQ-00003-00003244-00003968 But another human, Boromir, Would take on a lot of guilt
+qDvt7OTS-SQ-00004-00003969-00004286 I’ve walked into Mordor bearing Isuldur’s Bane
+qDvt7OTS-SQ-00005-00004286-00004998 Though the burden was really a strain Couldn’t take an eagle or even a plane
+qDvt7OTS-SQ-00006-00004998-00005385 ’Cause the Eye of Sauron would notice a plane
+qDvt7OTS-SQ-00007-00005386-00006108 La la la la la la la la La la la la la
+qDvt7OTS-SQ-00008-00006108-00006809 La la la la la la la la La la la la la
+qDvt7OTS-SQ-00009-00006809-00007559 I was walking from the subway I wanted to get some food
+qDvt7OTS-SQ-00010-00007559-00008260 I didn’t want a pizza Or anything barbecued
+qDvt7OTS-SQ-00011-00008260-00008962 Those things are nice, they would suffice But I wasn’t in the mood
+qDvt7OTS-SQ-00012-00008962-00009604 I hit Happy Garden, got a quart of lo mein And a fortune that was not arcane
+qDvt7OTS-SQ-00013-00009604-00010004 It said “Be good to yourself if you want to be sane”
+qDvt7OTS-SQ-00014-00010004-00010368 I added “in bed” and I laughed like a drain
+qDvt7OTS-SQ-00015-00010368-00011096 La la la la la la la la La la la la la
+qDvt7OTS-SQ-00016-00011096-00011770 La la la la la la la la La la la la la
+qDvt7OTS-SQ-00017-00011770-00012450 I was strumming my ukulele And this song wouldn’t let me be
+qDvt7OTS-SQ-00018-00012450-00012776 “The heat was hot,” “there were...rocks and things,”
+qDvt7OTS-SQ-00019-00012776-00013466 It’s a poor excuse for poetry But the pattern gets into everything
+qDvt7OTS-SQ-00020-00013466-00014216 And your mind just knocks about And no matter what you want to sing
+qDvt7OTS-SQ-00021-00014216-00014505 This is what comes out
+qDvt7OTS-SQ-00022-00014505-00014960 I smell like Teen Spirit on a horse with Cobain— (no.)
+qDvt7OTS-SQ-00023-00015047-00015419 I called in a plumber and she’ll clean out the drain—
+qDvt7OTS-SQ-00024-00015522-00015846 I’ve gone to the silos where I keep all my grain—
+qDvt7OTS-SQ-00025-00015969-00016274 We’re off to the wizard, ’cause I don’t have a brain—
+qDvt7OTS-SQ-00026-00016347-00016632 I tried out for Hamlet, got the role of the Dane—
+qDvt7OTS-SQ-00027-00016632-00016930 I’ve been to the barber, now my horse has no mane—
+qDvt7OTS-SQ-00028-00016930-00016988 [frustrated scream]
+qDvt7OTS-SQ-00029-00017420-00018112 This two-chord song is stuck in my brain So I’m passing it on, such a shame
+qDvt7OTS-SQ-00030-00018112-00018800 This sequence of lyrics is a terrible pain And now all of you get to hear me complain
+qDvt7OTS-SQ-00031-00018801-00019484 La la la la la la la la La la la la la
+qDvt7OTS-SQ-00032-00019484-00020198 La la la la la la la la La la la la la
+qDThHAhNIcA-00000-00000042-00000240 Hello, this is Professor Steve Potter.
+qDThHAhNIcA-00001-00000324-00000936 Here I will show you the steps for carving a 3D Terrain map out of wood with a CNC machine.
+qDThHAhNIcA-00002-00001056-00001680 I've also written a detailed Instructable tutorial at instructables.com. I'll put a link below...
+qDThHAhNIcA-00003-00001889-00002130 These are the San Gabriel mountains, the backdrop
+qDThHAhNIcA-00004-00002130-00002760 of where I grew up in Altadena and Pasadena around here near Los Angeles, California.
+qDThHAhNIcA-00005-00003036-00003468 I've hiked and biked and camped all over these mountains with my father, whose name is Philip
+qDThHAhNIcA-00006-00003468-00004122 D. Potter. I decided to give him a carving for his 85th birthday. Terrain carvings are made
+qDThHAhNIcA-00007-00004122-00004938 from a GeoTIFF or digital elevation map (DEM), where satellite data is coded -- the altitude
+qDThHAhNIcA-00008-00004938-00005622 is coded -- in 16-bit grayscale. Dad worked for the Jet Propulsion Lab (JPL) and helped
+qDThHAhNIcA-00009-00005622-00006198 develop the side-looking radar that satellites use to create these elevation Maps. These data
+qDThHAhNIcA-00010-00006198-00006942 are freely available at the USGS website. But that was in 2017. I will show you how to use a
+qDThHAhNIcA-00011-00006942-00007500 web-based tool called TouchTerrain that makes collecting DEMs and creating a 3D model easy.
+qDThHAhNIcA-00012-00007566-00008238 I used my trusty handibot portable CNC carver to make that map of the San Gabriel Mountains
+qDThHAhNIcA-00013-00008360-00009342 The Handibot has a 6" x 8" work area, so to make this I had to move it along the workpiece eight
+qDThHAhNIcA-00014-00009342-00009942 times. You can see some of the imperfections caused by the joins between the different tiles
+qDThHAhNIcA-00015-00009942-00010470 there. I mark significant places with rhinestones: there's Caltech, where Dad went to school,
+qDThHAhNIcA-00016-00010470-00011088 and JPL, where he worked, and some places that we hiked and camped. I love the way this plywood
+qDThHAhNIcA-00017-00011088-00011862 looks in these carvings. The plies look a lot like contour lines on a topo map. But watch out
+qDThHAhNIcA-00018-00011862-00012534 for voids that you might see in the wood if you use this inexpensive plywood here. You might have
+qDThHAhNIcA-00019-00012534-00013128 voids or you might have overlapping plies and in fact, if there are those voids, the peaks might
+qDThHAhNIcA-00020-00013128-00013962 even break off. I wanted to carve larger works of art without tiling, so I built my own CNC machine.
+qDThHAhNIcA-00021-00014028-00014622 This is the Mostly-Printed CNC machine (MPCNC). I printed the plastic parts on my 3D printer.
+qDThHAhNIcA-00022-00014963-00015624 It's the Primo version, by V1 Engineering. I would say it has the best performance per dollar of any
+qDThHAhNIcA-00023-00015624-00016212 CNC machine, because it's very hackable. You can make it any size you want and it's quite
+qDThHAhNIcA-00024-00016212-00016818 powerful. I made mine portable. I have the control Electronics in this toolbox here and
+qDThHAhNIcA-00025-00016818-00017502 this is a portable cooling unit for the spindle. I fixed up the electronics to make this machine
+qDThHAhNIcA-00026-00017502-00018120 faster and more powerful than the first version. I made a whole 'nother YouTube video about that,
+qDThHAhNIcA-00027-00018120-00018690 which will be linked below. To make my CNC machine portable, I also added swivel wheels here, so that
+qDThHAhNIcA-00028-00018690-00019218 I can load it in and out of my car and take it all across Ireland where I teach maker workshops.
+qDThHAhNIcA-00029-00019218-00019728 So that's the kit that I'm using, but just about any CNC machine with a router or spindle can make
+qDThHAhNIcA-00030-00019728-00020370 terrain carvings. Start your CNC project with some good quality wood like this Baltic birch plywood.
+qDThHAhNIcA-00031-00020370-00021144 This is 18 millimeters thick. This Grade-B plywood costs about 180 Euros for a 4' x 8'
+qDThHAhNIcA-00032-00021144-00021984 sheet here in 2022. Look at the edges carefully and make sure that there are no holes. Also,
+qDThHAhNIcA-00033-00021984-00022638 you probably don't want to buy the top sheet, because it might have different humidity on
+qDThHAhNIcA-00034-00022638-00023052 the top versus the bottom of it, and when you bring it home it'll get warped. Once again,
+qDThHAhNIcA-00035-00023112-00023670 look carefully at the edges for voids, knots, and overlapping plies. It's fine for
+qDThHAhNIcA-00036-00023670-00024204 practicing with this cheap plywood, but it's not very good for making a nice work of art,
+qDThHAhNIcA-00037-00024204-00024738 so ideally find the A-grade plywood if you can find it. It's hard to find and very expensive,
+qDThHAhNIcA-00038-00024738-00025356 but it's really good stuff. This is my more recent carving using this birch plywood,
+qDThHAhNIcA-00039-00025464-00026100 and in this case, I also did a resin pour for the Irish Sea. These are the Mourne Mountains here,
+qDThHAhNIcA-00040-00026100-00026694 and this is the Cooley Peninsula. This is the town of Dundalk in the Northeast of Ireland.
+qDThHAhNIcA-00041-00026694-00027258 I've hiked all over these hills and mountains and so they mean a lot to me. This is a carving
+qDThHAhNIcA-00042-00027258-00027708 I made for my brother who recently moved to the state of Idaho. It also includes a
+qDThHAhNIcA-00043-00027708-00028158 fold-out map of the state. This is where my sister lives in Orange County, California.
+qDThHAhNIcA-00044-00028482-00029058 After cutting the wood to size, I glued two sheets of this plywood together using
+qDThHAhNIcA-00045-00029058-00029766 wood glue. I prefer this Gorilla Wood Glue, which dries pretty clear. You could
+qDThHAhNIcA-00046-00029766-00030264 also use TiteBond III Ultimate wood glue which is very strong and also waterproof,
+qDThHAhNIcA-00047-00030264-00030864 although we probably won't be putting our artwork out in the weather. Lots of detail is possible in
+qDThHAhNIcA-00048-00030864-00031595 these CNC terrain carvings if two things: (1) You get good satellite data like this,
+qDThHAhNIcA-00049-00031752-00032345 and (2) You use a fine finishing bit. I'll talk more about those in a minute.
+qDThHAhNIcA-00050-00032532-00033120 These small hills are the Drumlins in County Monaghan in Ireland where we live. They are
+qDThHAhNIcA-00051-00033120-00033792 only about 100 feet tall. Each one has its own local folklore and names. Now, I'll show you how
+qDThHAhNIcA-00052-00033792-00034554 to create a 3D model using the TouchTerrain interface. TouchTerrain was created by Chris
+qDThHAhNIcA-00053-00034554-00035136 Harding at Iowa State University and Franek Hasiuk at the University of Kansas. They created this
+qDThHAhNIcA-00054-00035136-00035946 wonderful open-source and free interactive tool to take satellite digital elevation models and turn
+qDThHAhNIcA-00055-00035946-00036534 them into 3D models, so that their students could print them and bring them on geology field trips.
+qDThHAhNIcA-00056-00037098-00037206 Okay, go to:
+qDThHAhNIcA-00057-00037206-00037890 touchterrain.geol.iastate.edu. Probably the first thing you'll want to do is to
+qDThHAhNIcA-00058-00037890-00038448 type in an area of interest to you so let's do Big Bear Lake.
+qDThHAhNIcA-00059-00038730-00039480 Okay, now probably you'll see terrain appear. If your (chosen region is) in the continental
+qDThHAhNIcA-00060-00039480-00040110 United States, this first one is chosen. If you're not in the United States then you need to choose
+qDThHAhNIcA-00061-00040110-00040788 one of these other ones. If you just see this map of the roads and stuff -- this slider can
+qDThHAhNIcA-00062-00040788-00041502 switch between showing terrain and showing the Google Maps underneath, or both of them faded out.
+qDThHAhNIcA-00063-00041646-00042318 This one has very good resolution: 10 meters per pixel. This one is a worldwide map made by JAXA,
+qDThHAhNIcA-00064-00042318-00042822 the Japanese Space Administration, with 30 meters per pixel resolution which is also
+qDThHAhNIcA-00065-00042822-00043344 very good. You might want to explore these other ones as well. The way that this looks
+qDThHAhNIcA-00066-00043344-00043878 on the screen here doesn't affect the way your model comes out. It's just for
+qDThHAhNIcA-00067-00043878-00044538 helping you decide what area to select. So, for instance, I may wish to zoom out a bit here...
+qDThHAhNIcA-00068-00044928-00045672 ...and grab these red corners and stretch them to cover a bigger area. If you don't see the
+qDThHAhNIcA-00069-00045672-00046116 red box you must click this "Re-center box on map". It's somewhere else in the world and it
+qDThHAhNIcA-00070-00046116-00046512 will come over to the middle of your map now. Once you've selected the area that you like,
+qDThHAhNIcA-00071-00046602-00047316 zoom back in on it and open this window that shows you the longitude and latitude
+qDThHAhNIcA-00072-00047316-00047940 of the area that you've selected. You have the option of using a polygon that you've created
+qDThHAhNIcA-00073-00047940-00048588 in something like Google Earth from a KML file. That's helpful if you, for example, wanted to
+qDThHAhNIcA-00074-00048588-00049068 carve something like a state or a country or an island -- something that's not rectangular.
+qDThHAhNIcA-00075-00049128-00049704 The next thing to do is go down to 3D printer options and -- in fact we're doing CNC options
+qDThHAhNIcA-00076-00049704-00050238 here so we'll follow the advice in this little pop-up window, which says choose
+qDThHAhNIcA-00077-00050238-00051024 one of these CNC options here. I'll choose the large size one and then for nozzle diameter
+qDThHAhNIcA-00078-00051024-00051617 you should probably choose High detail, which is approximately the same as 0.2 millimeter nozzle.
+qDThHAhNIcA-00079-00051816-00052236 And have a look at this box here called Effective DEM resolution.
+qDThHAhNIcA-00080-00052338-00052824 It will tell you whether or not you're getting close to what the best possible resolution is
+qDThHAhNIcA-00081-00052824-00053328 from the data set, which in this case is 10 meters per pixel. This effective DEM resolution
+qDThHAhNIcA-00082-00053328-00054084 is influenced by the size and the nozzle diameter that you've chosen. You'll notice that the maximum
+qDThHAhNIcA-00083-00054084-00054576 size you can choose is around 110 millimeters, which is not very big. Remember, this was made
+qDThHAhNIcA-00084-00054576-00055248 for 3D printers. If you're cutting a work of art that will be bigger than 110 millimeters across,
+qDThHAhNIcA-00085-00055248-00055788 you're going to be scaling this and this number may or may not be relevant to you.
+qDThHAhNIcA-00086-00055872-00056484 It will get scaled as well. I would choose the thickest model base thickness, unless for some
+qDThHAhNIcA-00087-00056484-00057005 reason you want a very thin base on your model. For Vertical exaggeration, that depends very much
+qDThHAhNIcA-00088-00057005-00057744 on how big of an area that you have in your chosen area there. If it's a big area like a country or
+qDThHAhNIcA-00089-00057744-00058338 a continent, you will want to use a very big multiplier, because the mountains are relatively
+qDThHAhNIcA-00090-00058338-00058998 small by comparison. If it's a small area like an individual property, you might even want to
+qDThHAhNIcA-00091-00058998-00059508 reduce it. You have many options and there are a couple of steps later on where you can change
+qDThHAhNIcA-00092-00059508-00060114 this if you don't quite get it right. I usually choose 3X. That's good for about 100-kilometer
+qDThHAhNIcA-00093-00060114-00060978 wide region there. You want to export it as an STL binary. In this box called manual settings,
+qDThHAhNIcA-00094-00060978-00061782 you have a number of options. The one that I like to use is called "lower_leq", which drops the
+qDThHAhNIcA-00095-00061782-00062352 areas below sea level down a few millimeters -- however many millimeters you specify.
+qDThHAhNIcA-00096-00062520-00062970 If you want to learn more about other manual options, you can go to the GitHub page,
+qDThHAhNIcA-00097-00062970-00063767 which is linked down at the bottom of this page. Scroll down here and you can see a list
+qDThHAhNIcA-00098-00063924-00064566 of all the manual options, including the one that I just mentioned, "lower_leq". Below that is an
+qDThHAhNIcA-00099-00064566-00065105 explanation of how to use each of those options. Okay, so once you think you have all this set up,
+qDThHAhNIcA-00100-00065105-00065814 before you hit Export Selected Area and Download File, it's a really good idea to do some screen
+qDThHAhNIcA-00101-00065814-00066294 captures with a couple of different settings of this transparency. One that has all the Google
+qDThHAhNIcA-00102-00066294-00066942 Maps stuff here -- By the way, you can also choose satellite if you prefer that for your Google Maps.
+qDThHAhNIcA-00103-00067152-00067650 And then you should do a screen capture with just the terrain, and you should do one with
+qDThHAhNIcA-00104-00067728-00068214 some sort of a blend of the two. You can adjust these parameters here to determine
+qDThHAhNIcA-00105-00068214-00069000 how the hill shading looks in this part here. If you do these screen captures,
+qDThHAhNIcA-00106-00069000-00069624 you might get to use them later on to create a fold-out map to go with your CNC carving.
+qDThHAhNIcA-00107-00069624-00070152 Okay, now we're going to hit Export selected area and download file, and if you're lucky, you'll see
+qDThHAhNIcA-00108-00070152-00070776 this locomotive chugging away here to say that their computer is processing all the data from
+qDThHAhNIcA-00109-00070776-00071454 wherever the data source came from, and turning it into a 3D model for you. If you're unlucky,
+qDThHAhNIcA-00110-00071454-00072012 a message will pop up that says "Data is too large!" or "Image is too large!" and you'll have
+qDThHAhNIcA-00111-00072012-00072408 to go back and change some of the parameters to reduce the number of pixels that you have,
+qDThHAhNIcA-00112-00072522-00073248 and hopefully get it underneath their threshold, which I think is around 700 000 pixels. That was
+qDThHAhNIcA-00113-00073248-00073704 very quick. You have an option to preview the STL in a browser, but I don't think it's worth
+qDThHAhNIcA-00114-00073704-00074178 doing that, because it takes just about as long to download it, and then it'll be quicker and easier
+qDThHAhNIcA-00115-00074178-00074742 to preview it on your own computer. However, before you leave here you should do two things:
+qDThHAhNIcA-00116-00074742-00075384 one is type in what you're going to do with this. I always type in a "CNC carving" because
+qDThHAhNIcA-00117-00075384-00075912 that way they'll know who's using this and for what, and they'll make the user interface more
+qDThHAhNIcA-00118-00075912-00076410 tailored towards people who are using this for a CNC machine instead of a 3D printer.
+qDThHAhNIcA-00119-00076518-00077184 The other thing you might want to do is to copy this text here, which is a URL that you can click
+qDThHAhNIcA-00120-00077184-00077814 on later on to get you right back to this page here that has all the parameters you've chosen.
+qDThHAhNIcA-00121-00077892-00078570 All the exact same parameters are in this URL here, and if you want to recreate the carving
+qDThHAhNIcA-00122-00078570-00079074 but with just changing a few of the parameters, it's very handy for that. Okay, so the next thing
+qDThHAhNIcA-00123-00079074-00079506 is to download the zip file. That might take a while, depending on your internet speed.
+qDThHAhNIcA-00124-00079596-00080172 Okay, once you unpack the zip file that TouchTerrain sends you, you will see
+qDThHAhNIcA-00125-00080268-00080712 three files. One is the GeoTIFF which is the grayscale-coded image.
+qDThHAhNIcA-00126-00080778-00081594 One is a log file which is a text file that has information about what you chose in the interface,
+qDThHAhNIcA-00127-00081594-00082140 and that's a good place to paste that URL that you copied from the web page here.
+qDThHAhNIcA-00128-00083244-00084024 Okay. Then you'll always have it in case you need to go back and redo things. But
+qDThHAhNIcA-00129-00084024-00084408 the thing we're really interested in here is the 3D model or STL file.
+qDThHAhNIcA-00130-00084612-00085164 Once you have created your 3D model with TouchTerrain, you want to check it and
+qDThHAhNIcA-00131-00085164-00085716 possibly scale it in some program that can look at STL files, such as PrusaSlicer.
+qDThHAhNIcA-00132-00085908-00086634 Okay, so we opened the STL file in PrusaSlicer. Here, you can have a good look at it and decide
+qDThHAhNIcA-00133-00086634-00087318 whether it has all the details that you want in your carving, and if not, you can go back and
+qDThHAhNIcA-00134-00087318-00087840 create a new model in TouchTerrain. You also want to have a look at the steepness of the mountains
+qDThHAhNIcA-00135-00087840-00088404 and whether the vertical scale is too exaggerated or too flat -- one or the other -- whether your
+qDThHAhNIcA-00136-00088404-00088998 CNC machine will want to carve that. The next thing I'm going to do is scale it to the size of
+qDThHAhNIcA-00137-00089064-00089868 the wood that I've cut. The width of the wood, or the X Dimension, is about 550 millimeters.
+qDThHAhNIcA-00138-00089964-00090660 I'm going to type 550 millimeters and it scales that up. So now, it's more the aspect ratio of
+qDThHAhNIcA-00139-00090726-00091284 the wood that I have. Now, you can see that it looks way too tall in fact, it's 148 millimeters.
+qDThHAhNIcA-00140-00091344-00092094 To scale the Z, I have to unclick this little lock here and then type in how thick I want it
+qDThHAhNIcA-00141-00092094-00093000 to be. So let's type 36 millimeters. Okay. That's more like it. This is much more like
+qDThHAhNIcA-00142-00093000-00093552 the real terrain, and much more like what it's going to look like when it's carved.
+qDThHAhNIcA-00143-00093660-00094290 If you're happy with that, then you want to save it. Export it from PrusaSlicer:
+qDThHAhNIcA-00144-00094386-00094998 Export plate as STL. And I'm going to note that this one is scaled.
+qDThHAhNIcA-00145-00095328-00096102 Okay, and we're done with that part. Once you have your STL file scaled and it looks good even when
+qDThHAhNIcA-00146-00096102-00096690 you zoom in on it, the next thing is to import it into your favorite CAM or CAD package. This could
+qDThHAhNIcA-00147-00096690-00097500 be Fusion360. I prefer VcarvePro, and I'll show you how to do that now. All right, you've scaled
+qDThHAhNIcA-00148-00097500-00098334 your model in PrusaSlicer. The next thing to do is to set up a job in Vcarve Pro. You should select
+qDThHAhNIcA-00149-00098334-00099066 Create a new file and then you get to this Job setup window here, where you specify the size of
+qDThHAhNIcA-00150-00099066-00099654 the wood that you have. In my case, it's 600 by 300 millimeters and about 36 millimeters thick.
+qDThHAhNIcA-00151-00099654-00100254 You should actually measure the wood, because it might not be what you think it is. So before you
+qDThHAhNIcA-00152-00100254-00100698 do this Job setup part, go back and measure your wood. After you've glued it together,
+qDThHAhNIcA-00153-00100860-00101454 then you want to set the Z zero position and the XY datum where you would like them. I always set
+qDThHAhNIcA-00154-00101454-00102204 the Z zero at the material surface. if you are going to carve the top of the stock away as
+qDThHAhNIcA-00155-00102204-00102846 you're carving your terrain, and there won't be any more material left to zero against, you're
+qDThHAhNIcA-00156-00102846-00103398 going to need another object that's the right height to zero with, so keep that in mind. I talk
+qDThHAhNIcA-00157-00103398-00104064 about this in my instructable. My datum is in the lower left-hand corner. Your machine may differ.
+qDThHAhNIcA-00158-00104142-00104861 Modeling resolution doesn't matter. For Material setting, it's a good idea to choose something
+qDThHAhNIcA-00159-00104861-00105432 that is not birch, because it has grain on it. That might cover up some of the features that
+qDThHAhNIcA-00160-00105432-00106080 you're trying to look at, so I just choose these homogeneous ones -- plastic ones -- and hit OK.
+qDThHAhNIcA-00161-00106308-00106811 All right, you've got your job all set up in Vcarve Pro. The next thing is to
+qDThHAhNIcA-00162-00106811-00107790 go to the Modeling tab here and import the scaled STL that you scaled in PrusaSlicer.
+qDThHAhNIcA-00163-00108444-00108978 If it's pointing in the wrong direction, you can rotate it with these buttons here. The next thing
+qDThHAhNIcA-00164-00108978-00109985 to do is to center it into the workpiece. It looks like I didn't scale the Y dimension to
+qDThHAhNIcA-00165-00109985-00110759 be small enough. My work area on my machine is 300 millimeters and I see that this is 311. So
+qDThHAhNIcA-00166-00110759-00111240 I'm going to scale this now to 300 and that will shrink the other dimensions too, a little bit.
+qDThHAhNIcA-00167-00111504-00112140 Now it will fit in fit within the workpiece. This is also another chance to scale the vertical
+qDThHAhNIcA-00168-00112140-00112709 dimension, if you need to do so. The only thing you have to do now is click Position and import.
+qDThHAhNIcA-00169-00112926-00113592 Now in here, there is a plane here called the Modeling plane, which is useful for determining
+qDThHAhNIcA-00170-00113592-00113994 where you're going to add extra features. Like maybe if you're going to carve some text
+qDThHAhNIcA-00171-00113994-00114588 or some other designs onto the terrain, then you can use that Modeling plane as
+qDThHAhNIcA-00172-00114588-00115098 a reference plane. I usually set it at the top. We can set at the bottom, in this case,
+qDThHAhNIcA-00173-00115098-00115452 just to get it out of the way so we don't have to look at it. And then we click Import.
+qDThHAhNIcA-00174-00115758-00116748 Okay, so here you have a 3D rendering. It looks very good. Go back to the drawing tab now,
+qDThHAhNIcA-00175-00116820-00117468 and back to the 2D view. The next step is to draw a box around the model. You could use
+qDThHAhNIcA-00176-00117468-00118188 a rectangle tool. Another way to do it is to select the model and to use this icon here...
+qDThHAhNIcA-00177-00118398-00118854 ...which traces a bitmap (draws a vector) around it. Choose Black and white and then
+qDThHAhNIcA-00178-00118854-00119304 slide this threshold all the way over so that it selects everything in there.
+qDThHAhNIcA-00179-00119496-00120168 For the corner fit, you may want to choose a loose fit, in case the edges are a bit pixelated and
+qDThHAhNIcA-00180-00120168-00120780 jagged -- it might smooth them out a bit. Then, you say Preview. That looks like a good outline
+qDThHAhNIcA-00181-00120780-00121626 to me. Hit Apply and Close. All that did was give you a line that you can select around the model
+qDThHAhNIcA-00182-00121626-00122274 here. Once you have your 3D model imported into Vcarve Pro, the next thing is to create toolpaths.
+qDThHAhNIcA-00183-00122628-00123282 All right, so with Trace bitmap you've created and selected this outline of your model. The next
+qDThHAhNIcA-00184-00123282-00123798 thing you want to do is create a pocket that goes around the model that you're going to use to cut
+qDThHAhNIcA-00185-00123798-00124404 it out. The easiest way to do that is with the offset tool here in the Drawing tab. We're going
+qDThHAhNIcA-00186-00124404-00125238 to go out about one and a half times the width of your cutout bit, so if the bit is 1/4" it's
+qDThHAhNIcA-00187-00125238-00126168 about six millimeters. We'll say, eight or nine millimeters outwards. So that will be -- that
+qDThHAhNIcA-00188-00126168-00126840 and this will be the two lines that define a pocket that we'll use to cut it out. Go to tool
+qDThHAhNIcA-00189-00126840-00127750 paths. Select both of those vectors and go to the pocket tool path -- not the profile tool path --
+qDThHAhNIcA-00190-00127992-00128808 and you decide on depth. We'll say 36, assuming that the wood is about 37 millimeters thick.
+qDThHAhNIcA-00191-00129348-00130008 For this, you can have the stepover be almost the entire width of the bit.
+qDThHAhNIcA-00192-00130206-00130782 Pass depth should be conservative: 1/10" is a good conservative number. The feed rate is
+qDThHAhNIcA-00193-00130782-00131033 probably too fast here, so I would turn that down.
+qDThHAhNIcA-00194-00131400-00131874 Here you have a choice between Offset and Raster again. You definitely want Offset,
+qDThHAhNIcA-00195-00131874-00132450 which will go around in circles. Raster will go back and forth in the groove there. You
+qDThHAhNIcA-00196-00132450-00132918 definitely want to select Climb cut, not Conventional, because Conventional tends to
+qDThHAhNIcA-00197-00132918-00133482 grab and drive the bit into the work there. Especially if you're using a down-cut bit,
+qDThHAhNIcA-00198-00133482-00133980 you want to Ramp the plunge moves to give the chips some pathway to get out.
+qDThHAhNIcA-00199-00134478-00134574 There's a tool path.
+qDThHAhNIcA-00200-00134772-00135306 All right, the next step is the roughing tool path. That's this little pyramid of circles here.
+qDThHAhNIcA-00201-00135480-00136182 Rough Machining. I use a six-millimeter endmill. You want to use a sturdy carbide bit.
+qDThHAhNIcA-00202-00136260-00137208 I use up-cut three-flute. Your stepover could be close to the width of the bit:
+qDThHAhNIcA-00203-00137208-00137964 80% is what I use. The maximum speed of my spindle is 24,000 RPM. And you can push
+qDThHAhNIcA-00204-00137964-00138468 the feed rate very as fast as your machine can go without breaking the bit, because this is
+qDThHAhNIcA-00205-00138468-00138912 just a roughing pass. You don't have to worry about details and tear-out, things like that.
+qDThHAhNIcA-00206-00139020-00139416 The Machining allowance is how much wood will be left behind after the roughing pass.
+qDThHAhNIcA-00207-00139416-00139896 That needs to be cut away by the finishing pass, so two millimeters is a good number.
+qDThHAhNIcA-00208-00139980-00140484 You can choose different strategies here. I think that the 3D Raster one is a little bit
+qDThHAhNIcA-00209-00140484-00140898 safer than the Z Level strategy. It will carve around the Peaks and sometimes they might break
+qDThHAhNIcA-00210-00140898-00141408 off if you're going that fast. Also, there's a lot more plunging and lifting going on in the Z-level
+qDThHAhNIcA-00211-00141408-00142038 strategy than in the 3D raster strategy. You can save a little bit of time by clicking Avoid
+qDThHAhNIcA-00212-00142038-00142650 machine areas, but again there's more plunging and lifting. You don't need to worry about Ramping the
+qDThHAhNIcA-00213-00142650-00143232 Plunge Moves, in this case, assuming your bit can do plunges. And that's about it.
+qDThHAhNIcA-00214-00143544-00144282 Okay, the next step is the finishing pass. That's this dome-shaped icon.
+qDThHAhNIcA-00215-00144456-00144984 I use a tapered ball-nose bit that is 1.5 millimeters in diameter.
+qDThHAhNIcA-00216-00145344-00146202 Six millimeter Shank and it's a two-flute up-cut. Stepover is important. If you do anything more
+qDThHAhNIcA-00217-00146202-00146694 than ten percent, you're likely to see raster lines, which you know, you might like those.
+qDThHAhNIcA-00218-00146694-00147174 Or you might have to sand them off, which is a lot of work. So it's better to go with a
+qDThHAhNIcA-00219-00147240-00147864 very fine stepover for a minimum amount of sanding. This one, you can move very quickly,
+qDThHAhNIcA-00220-00147864-00148356 because it doesn't have to take off very much wood with the 10% stepover.
+qDThHAhNIcA-00221-00148782-00149496 You want to select the vector that surrounds your model there, and you have a choice of Offset
+qDThHAhNIcA-00222-00149496-00149976 and Raster strategies. The Offset strategy will start in the middle and circle around
+qDThHAhNIcA-00223-00149976-00150702 in polygons until it gets to the outside edge, and the Raster one will go back and forth. Or,
+qDThHAhNIcA-00224-00150702-00151266 if you've selected some other than a zero-degree raster angle, 35 degrees would go about like this.
+qDThHAhNIcA-00225-00151668-00152196 They take about the same amount of time. One nice advantage of the Offset is you can specify
+qDThHAhNIcA-00226-00152196-00152724 only Climb Cuts whereas the Raster one will do Climb and Conventional as it goes back and forth.
+qDThHAhNIcA-00227-00152850-00153318 Climb cuts are less likely to cause big strings of wood to get caught up on the bit,
+qDThHAhNIcA-00228-00153318-00154206 and I think they're more efficient. So I recommend the offset strategy and climb cut.
+qDThHAhNIcA-00229-00154476-00154710 It'll take a long time to calculate this one, because
+qDThHAhNIcA-00230-00154824-00155088 it's going to be about eight hours of cutting here.
+qDThHAhNIcA-00231-00155616-00156306 Okay, that took my laptop a few minutes to calculate because there are a lot of lines here
+qDThHAhNIcA-00232-00156306-00156846 following the terrain. If you zoom way in on them, you can see that they're very close together.
+qDThHAhNIcA-00233-00157350-00157992 And if we look at the time it'll take to do that, it'll be eight hours of cutting,
+qDThHAhNIcA-00234-00158058-00158472 so it's a marathon run. But it's well worth it, to get a good finish.
+qDThHAhNIcA-00235-00158742-00159300 Once you think you have your tool paths right in Vcarve Pro, it's really important that you
+qDThHAhNIcA-00236-00159300-00159912 then run a simulation to make sure that the result is what you think it will be. Here's
+qDThHAhNIcA-00237-00159912-00160392 what you should do. Have a good look at the simulated carving in Vcarve Pro.
+qDThHAhNIcA-00238-00160446-00161394 Obviously, you want to save your work, and then go to the Simulation icon there. Preview tool paths.
+qDThHAhNIcA-00239-00161610-00162294 We'll Reset Preview here. You could simulate them one at a time. We can do the cutout. We'll
+qDThHAhNIcA-00240-00162294-00162894 do them in order that we will actually cut them. We'll say Preview visible tool path,
+qDThHAhNIcA-00241-00163002-00163368 and there you go. You can see that I've left about a millimeter at the bottom
+qDThHAhNIcA-00242-00163368-00163770 of the pocket, so that when I've got screws holding the wood down,
+qDThHAhNIcA-00243-00163770-00164268 it will still be holding the middle part down. Now we'll simulate the roughing one.
+qDThHAhNIcA-00244-00164844-00164958 Okay, that looks good.
+qDThHAhNIcA-00245-00165306-00165408 And finally,
+qDThHAhNIcA-00246-00165536-00165762 the Finishing tool path.
+qDThHAhNIcA-00247-00166656-00166716 Lovely!
+qDThHAhNIcA-00248-00167034-00167448 You want to zoom in on it. Have a good look at it. Make sure that it has the
+qDThHAhNIcA-00249-00167448-00168084 amount of detail that you want. If you chose a stepover that was too large,
+qDThHAhNIcA-00250-00168084-00168606 you will see raster lines on here. So the fact that we don't see raster lines,
+qDThHAhNIcA-00251-00168606-00169074 even until we get all the way in to the level of individual pixels, is a very good sign.
+qDThHAhNIcA-00252-00169326-00170106 All right. So, if the simulation went well, the last step is to export these toolpaths as G-Code.
+qDThHAhNIcA-00253-00170106-00170562 Before I do that, actually, you may want to have a look at how much time it's going to take to do
+qDThHAhNIcA-00254-00170562-00171192 each cut. Then, you might also want to create a Job sheet, which is handy to print out and
+qDThHAhNIcA-00255-00171192-00171840 have on hand. It will tell you which bit you're supposed to be using for each toolpath. Okay, if
+qDThHAhNIcA-00256-00171840-00172626 you did create a Job Setup sheet in Vectrix Vcarve Pro, it looks like this. It's an HTML file and
+qDThHAhNIcA-00257-00172626-00173364 it shows you what the overall picture of the job looks like, where the model lies within the stock,
+qDThHAhNIcA-00258-00173484-00174090 how it's set up, and where the origin is. A summary of the three different toolpaths, and
+qDThHAhNIcA-00259-00174090-00174720 then a detail about each toolpath, saying how fast is it going and what bit are you using and that
+qDThHAhNIcA-00260-00174720-00175182 sort of thing. It's very helpful to print that out and have it on hand while you're doing the work.
+qDThHAhNIcA-00261-00175356-00176172 You want to then export the G-Code. You select one toolpath at a time. Choose your
+qDThHAhNIcA-00262-00176172-00176886 machine. Make sure that you're using the right post-processor for the machine that you have. I
+qDThHAhNIcA-00263-00176886-00177612 have a grbl controller. Then you save the tool path. Go ahead and do that for the other ones.
+qDThHAhNIcA-00264-00178254-00178836 So you've got your tool paths created. The next step is to get your CNC machine ready to do the
+qDThHAhNIcA-00265-00178836-00179346 carving. Before you put the workpiece down, you should face the spoilboard. You want to
+qDThHAhNIcA-00266-00179346-00180186 make sure that the surface of your spoilboard is parallel to the X and Y rails of your CNC machine.
+qDThHAhNIcA-00267-00180846-00181512 I use two different ways to fix the workpiece to the spoil board: tape and screws. The tape
+qDThHAhNIcA-00268-00181512-00182256 I use is TESA 4298 strapping tape and I put some double-stick tape on that -- some very
+qDThHAhNIcA-00269-00182256-00182790 strong double-stick tape made by ThermoWeb. You don't need very many pieces of that
+qDThHAhNIcA-00270-00182790-00183372 to stick it down quite firmly. I also drill through-holes in the wood, and then use long
+qDThHAhNIcA-00271-00183372-00183978 wood screws to screw it down firmly when I'm doing the cutout toolpath. I leave about a millimeter of
+qDThHAhNIcA-00272-00183978-00184650 wood there so that the wood won't come loose. The screws will still be holding it down as I'm
+qDThHAhNIcA-00273-00184650-00185208 carving the other toolpaths. That way, I can carve the cutout toolpath first, and that saves wear and
+qDThHAhNIcA-00274-00185208-00185724 tear on the roughing and finishing bits, because they're going to be turning around in midair,
+qDThHAhNIcA-00275-00185724-00186366 rather than in wood. Here it is screwed down in the corners. You screw it down tight enough so
+qDThHAhNIcA-00276-00186366-00186858 that you could not slip the corner of a piece of paper anywhere underneath it. It should be well
+qDThHAhNIcA-00277-00186858-00187788 adhered also, with your tape. These are the cutout and roughing bits that I use, 1/4" endmills,
+qDThHAhNIcA-00278-00187842-00188400 up-cut or down-cut. You can see on the right that I tried to go a little too fast with a cutout
+qDThHAhNIcA-00279-00188400-00189090 toolpath and it started to dig into the workpiece. This was using a Profile toolpath. I have learned
+qDThHAhNIcA-00280-00189090-00189534 now to use a Pocketing tool path instead, one that's a little bit wider than the bit,
+qDThHAhNIcA-00281-00189534-00190068 and make it move in the Climb Direction, not the Conventional Direction. Conventional Direction
+qDThHAhNIcA-00282-00190068-00190722 tends to dig in more. Also, don't have the bits sticking out any more than you need to. The
+qDThHAhNIcA-00283-00190722-00191034 longer they stick out (from the spindle collet) the more they're inclined to leave the pathway
+qDThHAhNIcA-00284-00191034-00191766 that the G-Code is intending them to move along. Okay, now I will show some roughing in action.
+qDThHAhNIcA-00285-00191994-00192348 This is the Conventional Direction with 80% stepover.
+qDThHAhNIcA-00286-00192606-00192990 There's the Climb direction, and you can see that it's a lot cleaner of a cut.
+qDThHAhNIcA-00287-00193446-00194046 The roughing process produces a lot of chips and dust, and you want to use a vacuum system
+qDThHAhNIcA-00288-00194046-00194568 or a shop vac -- or a blower perhaps -- to get the chips out of the path of the bit. Sometimes
+qDThHAhNIcA-00289-00194568-00194994 they can wrap around it and cause problems. Of course, you also want to protect yourself from
+qDThHAhNIcA-00290-00194994-00195516 the flying wood. I use this powered ventilator face shield by Sundstrom. It's comfortable
+qDThHAhNIcA-00291-00195516-00196074 enough to wear all day long. It doesn't fog up. It fits my glasses. It also has hearing
+qDThHAhNIcA-00292-00196074-00196656 protectors clipped to it, to protect my ears. Here's an example of roughing over the mountains.
+qDThHAhNIcA-00293-00196908-00197466 What you notice is that it slows down a bit as you're doing the cuts that involve a lot
+qDThHAhNIcA-00294-00197466-00198006 of vertical motion. That has to do with the limitations of the Z-axis on this machine
+qDThHAhNIcA-00295-00198006-00198402 at the time I filmed this. It's a little bit better now that I put a different lead screw on.
+qDThHAhNIcA-00296-00199176-00199620 When you're finished with the roughing pass, it looks -- not surprisingly -- very much like
+qDThHAhNIcA-00297-00199620-00200292 the simulated version in Vcarve Pro. Vcarve Pro has a couple of different options for doing the
+qDThHAhNIcA-00298-00200292-00201030 roughing pass you could do the Raster, which is what I prefer, on the left, or you can do what's
+qDThHAhNIcA-00299-00201030-00201570 called the Z-level or Offset strategy, here on the right. but the Offset strategy produces more
+qDThHAhNIcA-00300-00201570-00202134 of these sharp right angles in the wood, where forces tend to get concentrated and occasionally
+qDThHAhNIcA-00301-00202134-00202674 break Peaks off. So it's a little bit safer to use the Raster strategy. It does take longer, though.
+qDThHAhNIcA-00302-00202674-00203214 This was estimated to take 2 hours and 11 minutes by Raster, and only an hour and 20 minutes by the
+qDThHAhNIcA-00303-00203214-00203784 Z-level strategy. These are the finishing bits that I use. The middle one -- my favorite -- is
+qDThHAhNIcA-00304-00203784-00204516 about 1.5-millimeter diameter ball-end. You would probably only want to use the top, a very sharp
+qDThHAhNIcA-00305-00204516-00205260 one for a very small terrain carving or perhaps carvings in metal or something like a medallion.
+qDThHAhNIcA-00306-00205260-00205823 Or you could use the larger bit, which is about three millimeters or 1/8 of an inch,
+qDThHAhNIcA-00307-00205908-00206382 for something where it's big enough that you don't mind that it might have a scalloped surface.
+qDThHAhNIcA-00308-00206646-00207419 Here I'm comparing a stepover setting of 20% versus 15%. If we zoom in on that,
+qDThHAhNIcA-00309-00207569-00208067 you can see that the 20% shows some rastering lines and the 15%
+qDThHAhNIcA-00310-00208067-00208769 doesn't. In fact, I actually use 10% most of the time, because even the 15% stepover
+qDThHAhNIcA-00311-00208848-00209688 still shows some raster lines. A 10% stepover -- that's about 150 microns per line if you're
+qDThHAhNIcA-00312-00209688-00210282 using a bit that's one and a half millimeters across on its tip -- so that's six thousandths
+qDThHAhNIcA-00313-00210282-00210780 of an inch per stepover. I'll show some examples of the finishing tool path in action.
+qDThHAhNIcA-00314-00212112-00214362 [Music]
+qDThHAhNIcA-00315-00217356-00217752 While you're carving you may want to use an IR thermal camera to check the temperature
+qDThHAhNIcA-00316-00217752-00218123 of the bit, to make sure that it's not getting too hot. If it were to snap,
+qDThHAhNIcA-00317-00218123-00218814 that could cause a fire hazard if it falls into wood chips. It also wears the sharp edges down
+qDThHAhNIcA-00318-00218814-00219450 quicker if it overheats and softens the metal. It's fun to watch, during the finishing process,
+qDThHAhNIcA-00319-00219450-00219966 how the details get revealed as the wood left behind by the roughing pass gets carved
+qDThHAhNIcA-00320-00219966-00220434 away. You can see about three millimeters of wood here, left behind by the roughing pass.
+qDThHAhNIcA-00321-00220554-00221238 Here's a case where I was trying to push the process finishing too fast and I lost Z steps
+qDThHAhNIcA-00322-00221238-00221826 in the stepper motor that controls the z-axis. In that case, when it digs into the wood like this,
+qDThHAhNIcA-00323-00221826-00222432 you probably have wasted this piece of wood. I have fixed this problem by using a different
+qDThHAhNIcA-00324-00222432-00223026 lead screw that allows the z-axis to move faster, and keep up with the X and Y axes at
+qDThHAhNIcA-00325-00223026-00223619 about 50 millimeters per second, which is fast enough for me. Here is a dramatic reenactment
+qDThHAhNIcA-00326-00223619-00224178 of cutting the edges -- the waste wood -- off the piece, after finishing the finishing pass.
+qDThHAhNIcA-00327-00224286-00224796 With a curved shape like this, it's very handy to have a bandsaw. You can also use a coping
+qDThHAhNIcA-00328-00224796-00225419 saw or a jigsaw if you have them. If you have a piece that just has straight edges, of course,
+qDThHAhNIcA-00329-00225419-00225966 you could just use a table saw to cut those edges off... and then sand them with a belt sander to
+qDThHAhNIcA-00330-00225966-00226523 make sure that any marks left behind by the cutter are smoothed out. Even with a very small stepover,
+qDThHAhNIcA-00331-00226523-00227028 there will be some fibers that need to be sanded off. Ideally, what you want is something like
+qDThHAhNIcA-00332-00227028-00227610 this: it's fairly smooth. I spend a lot of time with the Dremel and an abrasive sponge wheel
+qDThHAhNIcA-00333-00227706-00228119 sanding these. It usually takes five or six of these sponge Wheels
+qDThHAhNIcA-00334-00228264-00228756 to get a piece so that it's completely smooth all over. These ones that are
+qDThHAhNIcA-00335-00228756-00229158 disc-shaped are good for getting into small valleys and crevices in your terrain.
+qDThHAhNIcA-00336-00229223-00229871 After sanding, you want to coat it with some spray lacquer. The spray lacquer from Lidl is very good.
+qDThHAhNIcA-00337-00229871-00230508 I prefer gloss. It usually takes three to seven coats of lacquer, with light sanding. You don't
+qDThHAhNIcA-00338-00230508-00231210 want to spray a very thick coat. Only a thin coat for each coating. Sand it in between coats. Gloss
+qDThHAhNIcA-00339-00231210-00231732 lacquer gives a nice effect called chatoyancy, which is how the luster of the layers of plywood
+qDThHAhNIcA-00340-00231732-00232373 changes with the lighting direction. Here's an example of that. With clear glossy varnish,
+qDThHAhNIcA-00341-00232434-00233028 the chatoyancy of the grain really shows up nicely with the light coming from different directions.
+qDThHAhNIcA-00342-00234012-00234521 After you've done your sanding and finishing, some possible extras that you might want to
+qDThHAhNIcA-00343-00234521-00234971 consider are to make a resin pour, to create a frame for your artwork,
+qDThHAhNIcA-00344-00234971-00235458 to brand it, or perhaps to make a fold-out map to put behind it.
+qDThHAhNIcA-00345-00235788-00236219 Resin pouring is tricky and there are lots of other good YouTube tutorials about that,
+qDThHAhNIcA-00346-00236219-00236616 so I won't go into detail, but just to say a couple of important things. One is, to make
+qDThHAhNIcA-00347-00236616-00237252 sure that the workpiece is completely level before you pour it. I used, in this case,
+qDThHAhNIcA-00348-00237252-00238254 a masking tape dam to keep the resin from pouring over the edges. That has resulted in a meniscus
+qDThHAhNIcA-00349-00238254-00238764 along the edge here of about a few millimeters that I had to sand off. It's very hard to sand
+qDThHAhNIcA-00350-00238764-00239388 that off without accidentally scratching the top of the ocean there. So a much better tape to use
+qDThHAhNIcA-00351-00239388-00239958 than masking tape is this TESA tape, the very same tape that I used for holding the workpiece down,
+qDThHAhNIcA-00352-00239958-00240732 4298 strapping tape. You want to fold it over on itself. So this is a folded edge here. The resin
+qDThHAhNIcA-00353-00240732-00241200 doesn't tend to climb up on this tape the way it does on masking tape. So there's very little
+qDThHAhNIcA-00354-00241200-00241812 meniscus. The resin I use is by Ink Lab and it's great because it has no odor. It takes about -- it
+qDThHAhNIcA-00355-00241812-00242562 has about a 40-minute working time, or in my cold garage, maybe three hours. If you want it to set
+qDThHAhNIcA-00356-00242562-00243330 faster, you can preheat it, or work in a warm space. I colored the ocean with this pigment
+qDThHAhNIcA-00357-00243330-00244008 and sparkles: mica powder. It produces these wonderful effects that evoke the idea of ocean
+qDThHAhNIcA-00358-00244008-00244752 waves and currents. To decide how much resin to mix up, I used a large piece of cardstock.
+qDThHAhNIcA-00359-00244830-00245346 I embossed it along the coast Edge there, and then cut it out and weighed it,
+qDThHAhNIcA-00360-00245346-00245921 and compared that weight to the weight of the whole sheet, to calculate what this area is and
+qDThHAhNIcA-00361-00245921-00246576 multiply that by the depth of the ocean to get the volume. Sometimes, knots and grain -- heartwood
+qDThHAhNIcA-00362-00246576-00247032 grain -- show up as interesting features that evoke sunken ships or ocean currents,
+qDThHAhNIcA-00363-00247098-00247614 so you might want to make the resin transparent enough that some of these show through.
+qDThHAhNIcA-00364-00247764-00248267 Here it is just after pouring, while it's still liquid, and you can see that the mica has an
+qDThHAhNIcA-00365-00248267-00248723 interesting texture to it. You can change that by moving it around with a stick or something.
+qDThHAhNIcA-00366-00248723-00249450 After it has settled for three hours in my cold garage, it's much less chaotic. If you wanted
+qDThHAhNIcA-00367-00249450-00250086 it to include those chaotic features, you might heat it up to make it set a little bit faster.
+qDThHAhNIcA-00368-00250469-00251052 You might want to make a frame for your CNC carving. I used some mahogany molding,
+qDThHAhNIcA-00369-00251052-00251669 cut on a fine-tooth miter saw, and then glued together with this MitreFast CA glue and
+qDThHAhNIcA-00370-00251669-00252623 accelerator. I sanded it with a random-orbit sander, 200 grit or finer. I made a box frame
+qDThHAhNIcA-00371-00252623-00253206 to go underneath the frame (subframe). It's fastened together in the corners with screws,
+qDThHAhNIcA-00372-00253206-00253806 so that you can loosen them and then adjust the height of the model within the subframe to set
+qDThHAhNIcA-00373-00253806-00254376 the height of the carving behind the frame. I glued the subframe to the frame using a small
+qDThHAhNIcA-00374-00254376-00255006 amount of polyurethane Gorilla Glue, which expands to fill the gaps. You might favor a
+qDThHAhNIcA-00375-00255006-00255617 simple flat frame, or perhaps something made from more fancy molding like this.
+qDThHAhNIcA-00376-00255912-00256319 You might also want to create a brand. Here, I'm designing a brand in Vcarve Pro,
+qDThHAhNIcA-00377-00256410-00256973 which I carved on my Handibot out of a very thick brass hinge. I heat it
+qDThHAhNIcA-00378-00256973-00257321 with a butane torch and then I can brand the back of my artwork there.
+qDThHAhNIcA-00379-00257634-00258396 Another extra is to make a fold-out map. Here I'm collecting screen captures from Google Maps,
+qDThHAhNIcA-00380-00258396-00259062 with the terrain setting turned on. The contour lines on the terrain only show up at certain Zoom
+qDThHAhNIcA-00381-00259062-00259926 levels, so I had to compile 25 tiles together. I merged those with a graphics program called
+qDThHAhNIcA-00382-00259926-00260544 GIMP. I then printed it on some quality paper and laminated it, and created a tape hinge.
+qDThHAhNIcA-00383-00260646-00261264 That way, it can be folded out if somebody wants to know where a certain landmark is, or where
+qDThHAhNIcA-00384-00261264-00261821 they're looking. They can have a look at the map and compare that to the carving. If you're going
+qDThHAhNIcA-00385-00261821-00262416 to hang it on the wall, you want to make sure that you use some sturdy hardware: some strong picture
+qDThHAhNIcA-00386-00262416-00263070 wire and a well-anchored hook. Ideally, attach your hook to a stud behind the wall. If you want
+qDThHAhNIcA-00387-00263070-00263718 more details please comment below or go have a look at my Instructable on instructables.com
+qDThHAhNIcA-00388-00263916-00264102 and of course, as always,
+qDThHAhNIcA-00389-00264102-00264498 "Like" and "Subscribe" if you want to see more YouTube videos from me! Thank you!
+qED6UC58L6I-00000-00000032-00000180 I know you don't like to get drawn into
+qED6UC58L6I-00001-00000180-00000408 politics uh Guru but something which has
+qED6UC58L6I-00002-00000408-00000775 hit very close home is uh recently there
+qED6UC58L6I-00003-00000775-00001054 was a man who was injured in a explosion
+qED6UC58L6I-00004-00001054-00001276 that took place in Bengaluru now it
+qED6UC58L6I-00005-00001276-00001558 turns out that this man sharik on his
+qED6UC58L6I-00006-00001558-00001803 WhatsApp status had a statue of Adele
+qED6UC58L6I-00007-00001803-00002001 how do you read the fact that you're the
+qED6UC58L6I-00008-00002001-00002367 man who was inspired by ISIS literature
+qED6UC58L6I-00009-00002367-00002603 he apparently self-radicalized he had a
+qED6UC58L6I-00010-00002603-00002839 Handler he was caught with a in an
+qED6UC58L6I-00011-00002839-00003163 explosion and he had a picture of Ari
+qED6UC58L6I-00012-00003163-00003346 Yogi hmm
+qED6UC58L6I-00013-00003346-00003555 I think I should be glad that even he is
+qED6UC58L6I-00014-00003555-00003932 carrying adiyogi
+qED6UC58L6I-00015-00003932-00004119 that's how I would like to see it but
+qED6UC58L6I-00016-00004119-00004353 that's not the reality I know one thing
+qED6UC58L6I-00017-00004353-00004655 is uh he's trying to mask his identity
+qED6UC58L6I-00018-00004655-00005078 that is a clever thing to do uh for the
+qED6UC58L6I-00019-00005078-00005364 work that he is doing but the important
+qED6UC58L6I-00020-00005364-00005623 thing is that we need to understand
+qED6UC58L6I-00021-00005623-00006264 uh
+qED6UC58L6I-00022-00006264-00006487 any image of this kind you know what
+qED6UC58L6I-00023-00006487-00006709 happened in Afghanistan to the Buddha
+qED6UC58L6I-00024-00006709-00006993 statues this has happened across
+qED6UC58L6I-00025-00006993-00007268 Northern Plains that anything that was
+qED6UC58L6I-00026-00007268-00007458 considered idolatry was destroyed
+qED6UC58L6I-00027-00007458-00007700 because somebody firmly believed that
+qED6UC58L6I-00028-00007700-00007906 so now I think most people have settled
+qED6UC58L6I-00029-00007906-00008079 okay you do your thing I do my thing
+qED6UC58L6I-00030-00008079-00008293 it's okay I think we should bring that
+qED6UC58L6I-00031-00008293-00008482 so there are still some radical elements
+qED6UC58L6I-00032-00008482-00008650 going on like this
+qED6UC58L6I-00033-00008650-00008906 for them the most terrible thing is that
+qED6UC58L6I-00034-00008906-00009149 there is other Yogi statue standing
+qED6UC58L6I-00035-00009149-00009457 there which uh which hurts them in some
+qED6UC58L6I-00036-00009457-00009728 way because they think we are worshiping
+qED6UC58L6I-00037-00009728-00009954 we are not worshiping we are using is it
+qED6UC58L6I-00038-00009954-00010189 an inspiration to go in a certain
+qED6UC58L6I-00039-00010189-00010318 direction
+qED6UC58L6I-00040-00010318-00010590 but still that is the belief so this
+qED6UC58L6I-00041-00010590-00010761 education needs to happen to all
+qED6UC58L6I-00042-00010761-00011082 communities that
+qED6UC58L6I-00043-00011082-00011338 how you want to reach something Beyond
+qED6UC58L6I-00044-00011338-00011494 yourself see essentially let's see it
+qED6UC58L6I-00045-00011494-00011679 this way whatever your religion is I'm
+qED6UC58L6I-00046-00011679-00011880 somebody who's never read any Scripture
+qED6UC58L6I-00047-00011880-00012122 of any culture I have stayed away from
+qED6UC58L6I-00048-00012122-00012427 that whatever it may be essentially I
+qED6UC58L6I-00049-00012427-00012679 believe all religions were created to
+qED6UC58L6I-00050-00012679-00012941 help a human being in some way to
+qED6UC58L6I-00051-00012941-00013180 transcend their limitations and find
+qED6UC58L6I-00052-00013180-00013416 higher qualities within themselves
+qED6UC58L6I-00053-00013416-00013596 somebody will do it by prayer somebody
+qED6UC58L6I-00054-00013596-00013863 will do it by ritual somebody will do it
+qED6UC58L6I-00055-00013863-00014066 by all kinds of worship somebody will do
+qED6UC58L6I-00056-00014066-00014253 it by meditation somebody will do it by
+qED6UC58L6I-00057-00014253-00014469 knowledge it doesn't matter how
+qED6UC58L6I-00058-00014469-00014644 essentially in yoga we say there are
+qED6UC58L6I-00059-00014644-00014881 only four ways you can use your body and
+qED6UC58L6I-00060-00014881-00015256 do it we call this ah Karma Yoga you can
+qED6UC58L6I-00061-00015256-00015434 do your intelligence and do it we call
+qED6UC58L6I-00062-00015434-00015643 this Nana yoga you can use your emotion
+qED6UC58L6I-00063-00015643-00015865 and do it we call this bhakti yoga you
+qED6UC58L6I-00064-00015865-00016041 can use your life energies and do it
+qED6UC58L6I-00065-00016041-00016284 this is called as kriya yoga so whatever
+qED6UC58L6I-00066-00016284-00016581 religion you may come from in some way
+qED6UC58L6I-00067-00016581-00016754 it is a combination of these four things
+qED6UC58L6I-00068-00016754-00016995 in What proportion it may vary but
+qED6UC58L6I-00069-00016995-00017204 that's all a human being can do you can
+qED6UC58L6I-00070-00017204-00017470 use your body mind emotion and energy
+qED6UC58L6I-00071-00017470-00017695 there's nothing else for you to use you
+qED6UC58L6I-00072-00017695-00017909 can invent whatever you want
+qED6UC58L6I-00073-00017909-00018240 so this maturity needs to come I think
+qED6UC58L6I-00074-00018240-00018378 it's beginning to happen in the world
+qED6UC58L6I-00075-00018378-00018540 across the world we're reaching billions
+qED6UC58L6I-00076-00018540-00018898 of people every year in spite of that uh
+qED6UC58L6I-00077-00018898-00019120 what I see is there is small fringes
+qED6UC58L6I-00078-00019120-00019414 margin uh margins of people who still
+qED6UC58L6I-00079-00019414-00019767 firmly believe whatever is being told to
+qED6UC58L6I-00080-00019767-00020051 them so I think this needs to change it
+qED6UC58L6I-00081-00020051-00020331 is uh not anybody's business how an
+qED6UC58L6I-00082-00020331-00020515 individual person is pursuing his own
+qED6UC58L6I-00083-00020515-00020870 well-being so about ADI Yogi becoming so
+qED6UC58L6I-00084-00020870-00021145 popular that even Terror suspect carries
+qED6UC58L6I-00085-00021145-00021358 adiyogi I think it makes me very happy
+qED6UC58L6I-00086-00021358-00021762 do you feel though that he could have
+qED6UC58L6I-00087-00021762-00021991 been thinking of targeting uh that is
+qED6UC58L6I-00088-00021991-00022212 always there that is not new for us I
+qED6UC58L6I-00089-00022212-00022341 don't know I have not counted and on
+qED6UC58L6I-00090-00022341-00022605 average in a year how many death threats
+qED6UC58L6I-00091-00022605-00022772 come to me they keep coming
+qED6UC58L6I-00092-00022772-00023079 uh but still I am alive so it's okay
+qED6UC58L6I-00093-00023079-00023597 what is it a cause of worry that in the
+qED6UC58L6I-00094-00023597-00024043 place that you live and in other pockets
+qED6UC58L6I-00095-00024043-00024215 in Karnataka for example this this
+qED6UC58L6I-00096-00024215-00024465 gentleman was in Mangalore he traveled
+qED6UC58L6I-00097-00024465-00024662 to Tamil Nadu he traveled to coimbatore
+qED6UC58L6I-00098-00024662-00024845 he traveled to Kerala that there is
+qED6UC58L6I-00099-00024845-00025120 growing radicalization in a set or in a
+qED6UC58L6I-00100-00025120-00025337 group of people no you must understand
+qED6UC58L6I-00101-00025337-00025613 he's doing the same route that I did
+qED6UC58L6I-00102-00025613-00025838 I was in my store constantly traveled to
+qED6UC58L6I-00103-00025838-00026227 Mangalore then Mysore to coimbatore
+qED6UC58L6I-00104-00026227-00026527 this guy is traveling the same route
+qED6UC58L6I-00105-00026527-00026695 I know I am joking about a serious
+qED6UC58L6I-00106-00026695-00027075 subject but it's very important that we
+qED6UC58L6I-00107-00027075-00027335 learn to handle
+qED6UC58L6I-00108-00027335-00027544 the most significant aspects of Our
+qED6UC58L6I-00109-00027544-00027810 Lives we learn to carry it lightly
+qED6UC58L6I-00110-00027810-00027989 otherwise we will want to kill each
+qED6UC58L6I-00111-00027989-00028337 other let's not go there
+qED6UC58L6I-00112-00028337-00028477 foreign
+qED6UC58L6I-00113-00028477-00028616 [Music]
+qED6UC58L6I-00114-00028616-00028755 everybody
+qED6UC58L6I-00115-00028755-00028894 [Music]
+qFFx2xj8XF4-00000-00000003-00000076 [music]
+qFFx2xj8XF4-00001-00000076-00000273 Canvas Inbox. 
+qFFx2xj8XF4-00002-00000373-00000553 From the Canvas dashboard, 
+qFFx2xj8XF4-00003-00000553-00000830 go to inbox, 
+qFFx2xj8XF4-00004-00000873-00001070 use the toolbar to filter 
+qFFx2xj8XF4-00005-00001070-00001296 the inbox view. 
+qFFx2xj8XF4-00006-00001930-00002216 Compose a new message. 
+qFFx2xj8XF4-00007-00002516-00002790 Select the course. 
+qFFx2xj8XF4-00008-00002950-00003260 Select the recipient. 
+qFFx2xj8XF4-00009-00003333-00003660 Write and send your message. 
+qFFx2xj8XF4-00010-00003879-00003996 Thank you for being a
+qFFx2xj8XF4-00011-00003996-00004226 Los Rios community college student.
+qKfPOFqCSVg-00000-00000028-00000624 Hey everybody, Scott Bintz here, "Doubtful  Passenger" chapter is now available.
+qKfPOFqCSVg-00001-00000624-00001062 It's the omitted chapter from the book,  "Principles to Fortune," because it
+qKfPOFqCSVg-00002-00001062-00001470 didn't really fit in with the theme. It's about my journey on how I
+qKfPOFqCSVg-00003-00001470-00002039 overcame some of the labels & the self- doubt in my own mind and the world
+qKfPOFqCSVg-00004-00002039-00002786 out there. Talks about  putting context to all the
+qKfPOFqCSVg-00005-00002786-00003342 labels I've been given my whole  life. Some of the labels are; Red
+qKfPOFqCSVg-00006-00003342-00004352 Head, Unnoticeable, B*stard, Carrot Top, Genius, Dumb Sh!t, Troublemaker, Whiner,
+qKfPOFqCSVg-00007-00004352-00005437 Rule Breaker, Brilliant, Introvert, Extrovert, Persistent, Helpful, Crafty,
+qKfPOFqCSVg-00008-00005437-00006053 Stubborn, Math Wizard, and Slow Reader.  Anyway, you get all these labels that are
+qKfPOFqCSVg-00009-00006053-00006693 slapped on you. How do you put them into context and how do you not allow them to
+qKfPOFqCSVg-00010-00006693-00007179 dictate what you're able to do  in your life? It's a quick 32 page
+qKfPOFqCSVg-00011-00007179-00007947 read & it's kind of a gritty, down & dirty look at the inside of my mind &
+qKfPOFqCSVg-00012-00007947-00008513 how I kind of put perspective to  those things. I hope you enjoy.
+qM30mtnmdVu-00000-00000605-00000911 [Music]
+qM30mtnmdVu-00001-00002567-00003385 hey guezz, do you like this video.
+qM30mtnmdVu-00002-00003513-00003830 Like , Share and subcribe our Channal.
+qM30mtnmdVu-00003-00005129-00007593 [Music]
+qM30mtnmdVu-00004-00007815-00008309 [Music]
+qM30mtnmdVu-00005-00012538-00012958 [Music]
+qM30mtnmdVu-00006-00014982-00015307 [Music]
+qPGrKxnSC3E-00000-00000983-00001313 When you're creating a media plan, you'll probably want to know what media outlets are
+qPGrKxnSC3E-00001-00001313-00001537 available in a given geographic region.
+qPGrKxnSC3E-00002-00001537-00001880 You might also want to know how much advertising costs in that area.
+qPGrKxnSC3E-00003-00001880-00002316 For example, perhaps I am working on a media plan for a company located in Madison, Wisconsin.
+qPGrKxnSC3E-00004-00002316-00002786 I can use the resource SRDS to find media outlets, SQAD cost-per-point ratings, and
+qPGrKxnSC3E-00005-00002786-00002898 more.
+qPGrKxnSC3E-00006-00002898-00003229 So let's say that I’m interested in finding all of the local media outlets that exist
+qPGrKxnSC3E-00007-00003229-00003360 in Madison.
+qPGrKxnSC3E-00008-00003360-00003721 This includes TV channels, magazines, radio stations, and so on.
+qPGrKxnSC3E-00009-00003721-00004272 I can find this information by clicking Local Media by DMA from the left side of the page.
+qPGrKxnSC3E-00010-00004272-00004501 DMA means Designated Market Area.
+qPGrKxnSC3E-00011-00004501-00004837 In this case our DMA is Madison, Wisconsin.
+qPGrKxnSC3E-00012-00004837-00005178 We can limit to Madison along the left side of the page.
+qPGrKxnSC3E-00013-00005178-00005646 And after we make our DMA selection, the center of the page will populate and give us information
+qPGrKxnSC3E-00014-00005646-00006026 about the various media outlets that exist in the area.
+qPGrKxnSC3E-00015-00006026-00006608 I can find information such as circulation rates, costs for advertisements, and more.
+qPGrKxnSC3E-00016-00006608-00007120 Clicking a specific media option will give more information about that specific TV channel
+qPGrKxnSC3E-00017-00007120-00007302 or newspaper or radio station.
+qPGrKxnSC3E-00018-00007302-00007809 If I’m getting more information than I want, I can select the type of media I'm interested
+qPGrKxnSC3E-00019-00007809-00007981 in, such as newspaper, radio.
+qPGrKxnSC3E-00020-00007981-00008283 or TV on the left side of the page.
+qPGrKxnSC3E-00021-00008283-00008644 One thing I may want to know is the SQAD cost-per-point.
+qPGrKxnSC3E-00022-00008644-00009054 When you multiply this number by the Nielsen rating for the program that airs during that
+qPGrKxnSC3E-00023-00009054-00009472 time, you can find out how expensive an ad will be in a given city during a specific
+qPGrKxnSC3E-00024-00009472-00009578 time of day.
+qPGrKxnSC3E-00025-00009578-00009999 You can find this information for radio as well as TV in SRDS.
+qPGrKxnSC3E-00026-00009999-00010447 Just hover over the Search by Media Type option then select the media type you're interested
+qPGrKxnSC3E-00027-00010447-00010554 in, TV or radio.
+qPGrKxnSC3E-00028-00010554-00010707 I'm going to click TV.
+qPGrKxnSC3E-00029-00010707-00011178 In the center of the page, click SQAD Cost Per Point Level Reports, and then find the
+qPGrKxnSC3E-00030-00011178-00011583 city you're interested in and click the SQAD Report link.
+qPGrKxnSC3E-00031-00011583-00011956 Note that SQAD cost numbers are broken up by time of day.
+qPGrKxnSC3E-00032-00011956-00012385 Dayparts are defined as Prime Access being 6:00 to 8:00 pm; Prime Time 8:00 to 10:30
+qPGrKxnSC3E-00033-00012385-00012866 pm; Late News 10:30 to 11:00 pm; and Late Fringe 10:30 to midnight.
+qPGrKxnSC3E-00034-00012866-00013284 For more information on SQAD cost-per-point levels and advertising rates, check out the
+qPGrKxnSC3E-00035-00013284-00014105 How to Find Advertising Rates guide.
+qQVIXWr5xJI-00000-00000352-00001040 hello and welcome to the introduction to biospecimens in the ABCD study this is week eight
+qQVIXWr5xJI-00001-00001096-00001655 of this repronim course my name is Kristina Uban I'm an assistant professor in public
+qQVIXWr5xJI-00002-00001655-00002176 health at the University of California Irvine and I'm very honored and very excited about
+qQVIXWr5xJI-00003-00002176-00002616 biospecimens in ABCD and to have this opportunity to share this with you guys
+qQVIXWr5xJI-00004-00002864-00003592 so the learning objectives by the end you should be able to describe the range of biospecimens
+qQVIXWr5xJI-00005-00003592-00004424 in the ABCD study describe the rationale for collecting each biospecimen, summarize methods of
+qQVIXWr5xJI-00006-00004424-00005056 collection and storage for each biospecimen, and then I'm going to dive a little bit deeper into
+qQVIXWr5xJI-00007-00005056-00005616 detail with salivary hormones. so by the end I'd like you to be able to leverage salivary hormones
+qQVIXWr5xJI-00008-00005616-00006240 in ABCD to really illustrate the complexity of all the considerations that you need to make when
+qQVIXWr5xJI-00009-00006240-00006736 you're utilizing a biospecimen that's collected in the study in your statistical analyses.
+qQVIXWr5xJI-00010-00006904-00007440 And at the end, some general "do's" and one "don't" when you're analyzing biospecimen data
+qQVIXWr5xJI-00011-00007631-00008064 so of course we're going to start with an overview this is something that we published as a group
+qQVIXWr5xJI-00012-00008064-00008504 back when there was a biospecimen work group it's now expanded to become a physical health work
+qQVIXWr5xJI-00013-00008504-00009272 group and we include biospecimens in that group so this publication was back from 2018 the references
+qQVIXWr5xJI-00014-00009272-00009976 down in the bottom right hand corner and it really just talked about baseline in year one
+qQVIXWr5xJI-00015-00009976-00010592 of course you guys are working now with 3.0 data so you're actually gonna have some year two and
+qQVIXWr5xJI-00016-00010592-00011152 maybe even some year three biospecimen data in there a little bit of it even with the pandemic
+qQVIXWr5xJI-00017-00011152-00011760 context. so this is a little outdated so i just wanted to kind of show you that in year
+qQVIXWr5xJI-00018-00011760-00012440 two had this publication included it we would be collecting all of these biospecimens for all of
+qQVIXWr5xJI-00019-00012440-00012976 the participants in ABCD and this continues on in year three four all the way up to ten
+qQVIXWr5xJI-00020-00013088-00013816 so previously in week six professor Krista Lisdahl had presented more details
+qQVIXWr5xJI-00021-00013816-00014480 on some biospecimens those are specifically used in our ability to screen a participant when they
+qQVIXWr5xJI-00022-00014480-00015088 come into the laboratory to see if they currently have alcohol or drugs in their system which would
+qQVIXWr5xJI-00023-00015088-00015512 mean that we would not want to assess them that day while they might be under the influence
+qQVIXWr5xJI-00024-00015512-00016176 but also provides more long-term biomarkers of substance use for participants as well.
+qQVIXWr5xJI-00025-00016240-00016936 so this really you can see in baseline here this we didn't do urine at baseline we did start doing
+qQVIXWr5xJI-00026-00016936-00017376 that year one but if you go into this publication it'll tell you that we really only did about 10%
+qQVIXWr5xJI-00027-00017456-00018112 of the participants in baseline and in year one this is because they were 9 and 10 and 10 and 11
+qQVIXWr5xJI-00028-00018112-00018816 and we really didn't expect wide-scale substance use across the board however now that we're moving
+qQVIXWr5xJI-00029-00018816-00019496 up into some year three and some year four data when the participants are 12, 13, soon to be 14,
+qQVIXWr5xJI-00030-00019608-00020312 we actually do expect bigger and bigger percentages of the ABCD cohort participants
+qQVIXWr5xJI-00031-00020312-00020800 to be engaging in substance use themselves that means that the percentage of the participants
+qQVIXWr5xJI-00032-00020800-00021256 we are testing for the screening of alcohol and drugs will keep increasing
+qQVIXWr5xJI-00033-00021384-00021992 in future years as we expect more use okay so because Krista Lisdahl went into great detail
+qQVIXWr5xJI-00034-00021992-00022496 on these I'm really just gonna kind of skim over them. I did provide slides on them in this lecture
+qQVIXWr5xJI-00035-00022496-00023120 so you had all your biospecimens in one set of pdf slides but I didn't want to be redundant so
+qQVIXWr5xJI-00036-00023120-00023640 I'm going to spend most of my time talking about pubertal hormones that we do by saliva and then
+qQVIXWr5xJI-00037-00023640-00024200 I'm going to also introduce genomics epigenetics and timeline of exposures through baby teeth
+qQVIXWr5xJI-00038-00024488-00024808 so the first kind of key aspect is to know that
+qQVIXWr5xJI-00039-00024880-00025456 when we're screening for alcohol or drugs or we're giving saliva for pubertal hormones those
+qQVIXWr5xJI-00040-00025456-00026048 are things that every participant eventually will get for the screening every time they come
+qQVIXWr5xJI-00041-00026048-00026760 in for their annual visit for puberty hormones participants get this every participant gets this
+qQVIXWr5xJI-00042-00026760-00027288 every year so one snapshot of your pubertal hormones every year so this is something that's
+qQVIXWr5xJI-00043-00027288-00027920 going to be repeated throughout the longitudinal study all the way up to the end year 10. whereas
+qQVIXWr5xJI-00044-00027976-00028632 genomics and baby teeth for timeline of exposures back to in utero all the way up to the time of
+qQVIXWr5xJI-00045-00028632-00029352 the child shed their baby tooth that timeline of exposures you know that doesn't change so
+qQVIXWr5xJI-00046-00029352-00030008 really once we collect baby teeth from them we've got it and for genomics once we take preferably
+qQVIXWr5xJI-00047-00030008-00030704 blood definitely some saliva going on in ABCD as a backup or a compliment but once we take their
+qQVIXWr5xJI-00048-00030832-00031536 their their genetic sample that again is a stable measure we will also get epigenetics from that
+qQVIXWr5xJI-00049-00031536-00032000 eventually in the future there are no currently epigenetic markers for us to work with but it's
+qQVIXWr5xJI-00050-00032000-00032592 something that's going to be coming in the future probably not with 4.0 but with 5.0 and clearly
+qQVIXWr5xJI-00051-00032592-00033072 you know epigenetics can be quite stable throughout someone's life and some epigenic
+qQVIXWr5xJI-00052-00033072-00033696 markers are are very variable right so that's something that might with funding and the right
+qQVIXWr5xJI-00053-00033696-00034352 interest from investigators with a funding grant source could use these bio-banked blood samples
+qQVIXWr5xJI-00054-00034352-00035048 and use the pellet the DNA pellet and absolutely do repeated epigenetic assessments with that
+qQVIXWr5xJI-00055-00035048-00035352 but that is not something that's currently available so I'm not really going to be going
+qQVIXWr5xJI-00056-00035352-00036272 into epigenetics so in a couple of years this talk might have greatly evolved so stay tuned okay so
+qQVIXWr5xJI-00057-00036272-00036992 again Krista Lisdahl went into great detail with the biospecimens that we collect specifically to
+qQVIXWr5xJI-00058-00036992-00038008 assess the adolescent's own substance use so you can see there's a range of them that we collect
+qQVIXWr5xJI-00059-00038008-00038480 so we do saliva and I have the details here so you guys have that for your records but
+qQVIXWr5xJI-00060-00038480-00039168 i'm gonna glaze over these since we don't want to be too redundant with a week six lecture the one
+qQVIXWr5xJI-00061-00039168-00039736 thing that I did want to point out just in case anyone didn't pick this up in earlier weeks but
+qQVIXWr5xJI-00062-00039736-00040288 when we're doing screening for substance use with breath and we're using a breathalyzer for alcohol
+qQVIXWr5xJI-00063-00040416-00040824 anyone at baseline that had a positive screen
+qQVIXWr5xJI-00064-00040888-00041448 meaning that they you know very currently or very recently still had some alcohol in their system
+qQVIXWr5xJI-00065-00041592-00042216 that was actually exclusionary criteria so they did not become part of the ABCD cohort participant
+qQVIXWr5xJI-00066-00042216-00042856 pool and I wanted to point that out because I think that's a kind of a big characteristic of
+qQVIXWr5xJI-00067-00042856-00043456 ABCD that people should be aware of. It really goes back to the big question of the original
+qQVIXWr5xJI-00068-00043528-00044232 you know grant writers or the the original intent of the in NIH institutes that are invested in this
+qQVIXWr5xJI-00069-00044376-00044688 it goes back to the original big picture question of the chicken
+qQVIXWr5xJI-00070-00044744-00045552 versus the egg which one came first but with mental health problems and substance use so since
+qQVIXWr5xJI-00071-00045552-00046112 in adolescence we see the beginning or onset of mental health problems particularly mood disorders
+qQVIXWr5xJI-00072-00046224-00047000 the onset tends to happen in adolescence and substance use and so since this is a common
+qQVIXWr5xJI-00073-00047064-00047632 developmental time that a lot of these things come on board but then become very problematic later on
+qQVIXWr5xJI-00074-00047632-00048352 particularly in adulthood which one happens first is it adolescent substance use that alters you
+qQVIXWr5xJI-00075-00048352-00048896 know risk or resilience to mental health problems and and so is the substance use that unveils a
+qQVIXWr5xJI-00076-00048896-00049544 risk for mental health problems or is it mental health problems that changes adolescent substance
+qQVIXWr5xJI-00077-00049544-00050280 use therefore exacerbating the underlying mental health problems and of course we all know it's
+qQVIXWr5xJI-00078-00050280-00050736 probably sometimes one of those scenarios sometimes the other the chicken came first
+qQVIXWr5xJI-00079-00050736-00051528 and the egg came first or they are interacting and influencing each other so that really was a big
+qQVIXWr5xJI-00080-00051584-00052360 intent of the ABCD study is to capture children before they started initiating substance use
+qQVIXWr5xJI-00081-00052424-00053088 and I don't know actually that we had to exclude anyone if we did it was one or two people at best
+qQVIXWr5xJI-00082-00053088-00053864 so it wasn't it didn't end up being an issue but it is a important characteristic of ABCD that we
+qQVIXWr5xJI-00083-00053864-00054384 can actually look at children coming in before they embark on substance use as adolescents
+qQVIXWr5xJI-00084-00054591-00055296 so I know Dr. Lisdahl also went through the screening with hair so cutting hair for a
+qQVIXWr5xJI-00085-00055296-00056136 range of substances and in year one we started taking urine for 10% and again that will ramp up
+qQVIXWr5xJI-00086-00056136-00056888 and we look at cotinine—the main metabolite of nicotine—and this is a good indicator of any smoke
+qQVIXWr5xJI-00087-00056888-00057440 exposure or tobacco exposure and if it hits a certain level we can fairly safely assume
+qQVIXWr5xJI-00088-00057440-00057984 that the adolescent is actually using tobacco themself and not just getting it through their
+qQVIXWr5xJI-00089-00057984-00058576 environment but this is a really great biomarker I think that people might be interested in
+qQVIXWr5xJI-00090-00058776-00059455 okay so most of this lecture is going to is going to be focused on pubertal hormones so that was me
+qQVIXWr5xJI-00091-00059455-00059872 briefing over the screening biomarkers because you guys have already gotten that but the slides are
+qQVIXWr5xJI-00092-00059872-00060504 there the details are there and again there's more details in this 2018 publication that Dr. Lisdahl
+qQVIXWr5xJI-00093-00060504-00061224 and I and a bunch of other ABCD investigators worked on so going on to pubertal hormones again
+qQVIXWr5xJI-00094-00061224-00062000 this is something that is really my specialty my area of focus and part of it is because puberty
+qQVIXWr5xJI-00095-00062055-00062736 is going to be a really big deal for all the participants with whose data you're working with
+qQVIXWr5xJI-00096-00062855-00063464 puberty is a process that is experienced across mammals so it's something that is
+qQVIXWr5xJI-00097-00063464-00063984 not just unique to humans so for me that feels very powerful this is something that's preserved
+qQVIXWr5xJI-00098-00063984-00064824 across all mammalian species often with puberty in other non-human mammals you see huge spikes in
+qQVIXWr5xJI-00099-00064824-00065544 male aggression sometimes in female aggression and you clearly see a mammal go from not being able to
+qQVIXWr5xJI-00100-00065544-00066240 reproduce to being able to reproduce after pubertal maturation so in terms of evolution
+qQVIXWr5xJI-00101-00066240-00067088 and reproduction I mean this is this is key what's really interesting is that it causes all these
+qQVIXWr5xJI-00102-00067088-00068056 changes in behavior and so it's not shouldn't be a surprise that ABCD originators really felt like we
+qQVIXWr5xJI-00103-00068056-00068728 needed to have some metrics to capture individual differences in puberty since we are starting in
+qQVIXWr5xJI-00104-00068728-00069472 childhood where most people entering the study are pre-pubertal or early pubertal stages
+qQVIXWr5xJI-00105-00069576-00070248 and by now with the 3.0 data release we actually have most people in the middle of puberty
+qQVIXWr5xJI-00106-00070248-00070848 late stages or done so and still some people haven't started yet particularly those born
+qQVIXWr5xJI-00107-00070848-00071584 biologically male but all the data you're working with is a hundred percent in a pubertal context
+qQVIXWr5xJI-00108-00071640-00072160 so if you never really thought about puberty and it's not really your thing hopefully after
+qQVIXWr5xJI-00109-00072160-00072688 the next few slides I might convince you that you might want to at least use some
+qQVIXWr5xJI-00110-00072688-00073344 pubertal assessments from ABCD in your descriptive table at the very least if not in your analyses
+qQVIXWr5xJI-00111-00073576-00074224 okay so in the brain in the hypothalamus there are specific nuclei that release
+qQVIXWr5xJI-00112-00074224-00074816 the hormone gonadotropin releasing hormone or GnRH that stimulate the pituitary gland at the
+qQVIXWr5xJI-00113-00074816-00075400 base of the brain and that goes down to release luteinizing hormone, follicle stimulating hormone,
+qQVIXWr5xJI-00114-00075400-00076152 and ultimately it acts on lots of different tissue in our in our body but it also acts on our sex
+qQVIXWr5xJI-00115-00076152-00076824 organs or gonads so in males individuals born biologically male would be the testes
+qQVIXWr5xJI-00116-00076880-00077464 and in individuals born biologically female it would be the ovaries and in here
+qQVIXWr5xJI-00117-00077464-00077952 these hormones stimulate it to produce more sex steroids or gonadal hormones
+qQVIXWr5xJI-00118-00078056-00078824 so in biological males we see testosterone as the highest level occurring hormone after
+qQVIXWr5xJI-00119-00078824-00079440 puberty and in females who also have testosterone but we see highest levels occurring in estradiol
+qQVIXWr5xJI-00120-00079440-00080064 and then these hormones go back and feed back on all sorts of tissue so there's a whole feedback
+qQVIXWr5xJI-00121-00080064-00080872 loop once testosterone and estrodial are produced and this feedback loop absolutely
+qQVIXWr5xJI-00122-00080872-00081464 includes back to the brain and it's not just back to the hypothalamus although that's a key target
+qQVIXWr5xJI-00123-00081464-00081880 it's back these these hormones for estradiol and testosterone and other gonadal hormones
+qQVIXWr5xJI-00124-00081936-00082448 really are widespread throughout different brain regions so no matter what neural circuitry you're
+qQVIXWr5xJI-00125-00082448-00083128 looking at whatever MRI modality of data that you're bringing in I can guarantee you that if you
+qQVIXWr5xJI-00126-00083128-00083776 were to consider gonadal hormones that we assess in ABCD it probably would explain some percentage
+qQVIXWr5xJI-00127-00083776-00084304 of variance of what you're seeing because these kids are actively going through puberty
+qQVIXWr5xJI-00128-00084304-00084952 right so we know that they go back they change the brain they can change decision making, they
+qQVIXWr5xJI-00129-00084952-00085528 can change behavior, they can change emotional regulation, they can change working memory, these
+qQVIXWr5xJI-00130-00085528-00086048 hormones can change all the different cognitive behavioral outcome measures that you're looking at
+qQVIXWr5xJI-00131-00086048-00086784 so you don't have to include these certainly lots of people don't but I definitely am an advocate
+qQVIXWr5xJI-00132-00086784-00087344 just think that you deepen your investigation when you consider individual differences in
+qQVIXWr5xJI-00133-00087344-00087848 these hormones particularly in kids actively going through major changes in the hormones
+qQVIXWr5xJI-00134-00088016-00088712 this is a very popular netflix show it's all about puberty and they talk about
+qQVIXWr5xJI-00135-00088768-00089416 hormone monsters changing the way you think changing mental health mood disorders changing
+qQVIXWr5xJI-00136-00089416-00089992 your physical features becoming more from child to adult like and they label as the hormone monster
+qQVIXWr5xJI-00137-00089992-00090640 which is really those hormones acting on your body and your brain and changing it and that's a brand
+qQVIXWr5xJI-00138-00090640-00091344 new experience for these kids of this age that we have data on in our 3.0 data set so this hormone
+qQVIXWr5xJI-00139-00091344-00091872 monster is really different for each kid when it shows up in someone's life is different so
+qQVIXWr5xJI-00140-00091872-00092648 onset of puberty how long is it is it there from premature pre-pubertal to fully post-pubertal
+qQVIXWr5xJI-00141-00092760-00093528 fully mature that duration of of puberty is individually different the levels of hormones
+qQVIXWr5xJI-00142-00093528-00094208 that you need to make those changes are different the patterns of hormones here I am specifically
+qQVIXWr5xJI-00143-00094208-00094792 referring to circadian patterns so the pattern of hormone fluctuations throughout a 24-hour cycle
+qQVIXWr5xJI-00144-00094920-00095560 the three hormones that we're asking in ABCD do have circadian fluctuations but we only take it
+qQVIXWr5xJI-00145-00095560-00096232 one sample at one point in in the experimental testing day so we cannot ask questions about
+qQVIXWr5xJI-00146-00096232-00096872 circadian patterns in ABCD within an individual perhaps across individuals
+qQVIXWr5xJI-00147-00096872-00097360 and then sensitivity to hormones right so maybe you need a little bit of testosterone
+qQVIXWr5xJI-00148-00097360-00098152 to drive tissue maturation for pubertal changes or maybe you need a lot before
+qQVIXWr5xJI-00149-00098152-00098664 you have any outward physical changes so that kind of sensitivity to hormones can be
+qQVIXWr5xJI-00150-00098664-00099312 wildly different among individuals all of these different factors that are individually
+qQVIXWr5xJI-00151-00099432-00100392 variable across our ABCD participants and mammals in general you know lead to differences in what
+qQVIXWr5xJI-00152-00100392-00100904 funeral maturation looks like at any snapshot in time through baseline year one year two and
+qQVIXWr5xJI-00153-00100904-00101432 beyond in the body, in the brain, so the body and the brain are becoming more adult-like,
+qQVIXWr5xJI-00154-00101608-00102440 differences in how mental health problems, how how how at risk someone might be to developing
+qQVIXWr5xJI-00155-00102440-00103008 a mental health problem, and how at risk someone might be when they engage in substance use
+qQVIXWr5xJI-00156-00103008-00103488 and how does that substance use impact their mental health resilience how does
+qQVIXWr5xJI-00157-00103488-00104111 that substance use impact their cognition or their likelihood to develop problematic use
+qQVIXWr5xJI-00158-00104111-00104983 well hormones are part of that story and of course behavioral regulation inhibition aggression
+qQVIXWr5xJI-00159-00105064-00105783 decision making all of it mood regulation all of it it has some part to do with your hormones
+qQVIXWr5xJI-00160-00105880-00106256 okay so we really talk about a lot of these in these couple publications down here
+qQVIXWr5xJI-00161-00106320-00106608 and I'll talk a little bit more about those publications here in a second
+qQVIXWr5xJI-00162-00106768-00107311 okay so rationale and methods for saliva I hope the rationale is really clear now
+qQVIXWr5xJI-00163-00107416-00108000 and if it isn't it's it's repeated in both of these publications that came from our
+qQVIXWr5xJI-00164-00108000-00108608 abcd co-investigators that I have cited down in the right again all participants gave one sample
+qQVIXWr5xJI-00165-00108672-00109152 in the laboratory testing time so the time when they gave a sample as variable and they
+qQVIXWr5xJI-00166-00109152-00109800 gave one sample via saliva every year but this started in baseline and it's going to continue
+qQVIXWr5xJI-00167-00109800-00110392 every year until year 10. so even though we can't ask circadian rhythm differences in individuals
+qQVIXWr5xJI-00168-00110392-00110800 and we have some methodological noise which will be really obvious to you in a couple minutes
+qQVIXWr5xJI-00169-00110800-00111824 here we still get to see a snapshot of hormonal development in hormone levels across 10 years down
+qQVIXWr5xJI-00170-00111824-00112511 the road so in the 3.0 data you will see some coming from year two it should be fully it's
+qQVIXWr5xJI-00171-00112511-00112904 almost filled out in year one and it's fully filled out in baseline which is really exciting
+qQVIXWr5xJI-00172-00113111-00113704 we collected saliva through passive drool and passive drool is whole saliva it comes from all
+qQVIXWr5xJI-00173-00113704-00114408 different glands in the mouth contribute to that saliva and we they did it in a saliva collection
+qQVIXWr5xJI-00174-00114408-00115008 aid and it was sent from each site it was put in a cooler to stop any potential bacteria that was
+qQVIXWr5xJI-00175-00115008-00115511 also collected to grow bacterial growth can interfere with our ability to
+qQVIXWr5xJI-00176-00115680-00116072 accurately assess the hormone levels in that saliva with the assay because it can
+qQVIXWr5xJI-00177-00116072-00116568 interfere with the assay binding so we stop that bacterial growth by keeping it cool immediately
+qQVIXWr5xJI-00178-00116632-00117096 for some sites that was just putting in a minus 80 degrees celsius freezer right away
+qQVIXWr5xJI-00179-00117096-00117511 other sites that's putting it in a cooler with an ice pack and just keeping it cool enough
+qQVIXWr5xJI-00180-00117511-00118000 where bacteria won't grow and then putting in the minus 80 degree freezer by the end
+qQVIXWr5xJI-00181-00118000-00118704 of that testing day—always within that testing day and they sat in the minus 80 degree freezer
+qQVIXWr5xJI-00182-00118704-00119472 for anywhere from a day to two months and were shipped then over to Carlsbad California
+qQVIXWr5xJI-00183-00119472-00120320 and assayed by a company hired by abcd called Salimetrics and Salimetrics thawed the samples in
+qQVIXWr5xJI-00184-00120320-00121016 the wet lab so they've gone through one thaw and simultaneously assessed testosterone and DHEA in
+qQVIXWr5xJI-00185-00121111-00121856 biological males at the same time with the same thaw and in females they thaw the sample and
+qQVIXWr5xJI-00186-00121856-00122568 simultaneously assessed testosterone, DHEA, and estradiol so biological females in abcd
+qQVIXWr5xJI-00187-00122568-00123280 have a third gonadal hormone: estradiol. so the rationale for selecting these hormones
+qQVIXWr5xJI-00188-00123359-00123783 really came from the fact that once you've gone through puberty maturation
+qQVIXWr5xJI-00189-00123783-00124296 biological males have the most testosterone in their in their bodies and biological females
+qQVIXWr5xJI-00190-00124296-00125056 have the most estradiol for part of the month and then it spikes and then you know egg is goes down
+qQVIXWr5xJI-00191-00125056-00125808 the fallopian tube and as that hits near in the uterus then you actually see a switch and
+qQVIXWr5xJI-00192-00125808-00126552 progesterone usually is higher than estradiol for this right before and during menstruation and so
+qQVIXWr5xJI-00193-00126632-00127040 females really have kind of some fluctuating hormones but in general we think of estradiol
+qQVIXWr5xJI-00194-00127040-00127848 as being a really prominent hormone for females almost synonymous with testosterone. DHEA is
+qQVIXWr5xJI-00195-00127848-00128432 the precursor to all these hormones including testosterone and estradiol it's also known as
+qQVIXWr5xJI-00196-00128432-00129088 kind of the kickstarter of puberty so if you see here in this graph I just kind of highlighted
+qQVIXWr5xJI-00197-00129159-00129983 the ABCD participants the age ranges that they'll be throughout this whole study so you can see
+qQVIXWr5xJI-00198-00129983-00130752 DHEA it's the first hormone to really start rising before testosterone and estradiol and you can see
+qQVIXWr5xJI-00199-00130752-00131408 that as the ABCD participants entered our study they probably had already been expressing some
+qQVIXWr5xJI-00200-00131408-00131936 elevations in DHEA before they entered at least a little bit in the background unless they're one
+qQVIXWr5xJI-00201-00131936-00132456 of the individuals who will go on to have a very late onset impurity they might have been low at
+qQVIXWr5xJI-00202-00132456-00133024 the beginning but in general when they enter they have very low levels compared to adult levels here
+qQVIXWr5xJI-00203-00133120-00133776 but you can see that very quickly upon entering ABCD in this age range all the way from you know
+qQVIXWr5xJI-00204-00133864-00134600 nine all the way up to 11 years old we see this huge increase and then after that
+qQVIXWr5xJI-00205-00134600-00135000 huge increase in DHEA happens we see the increase in testosterone estradiol follow
+qQVIXWr5xJI-00206-00135120-00135504 so DHEA is really important for understanding pubertal onset
+qQVIXWr5xJI-00207-00135768-00136336 okay so i'm going in deeper about the salivary hormones and i'm presenting in three phases
+qQVIXWr5xJI-00208-00136432-00137472 the phases represent I guess my understanding of how to really leverage ABCD hormones in analyses
+qQVIXWr5xJI-00209-00137528-00138376 and then the third phase is really a bigger circle of you know brains thinking about it
+qQVIXWr5xJI-00210-00138376-00139032 and giving input on it so in phase one it was just a couple of us putting this out here's the site
+qQVIXWr5xJI-00211-00139032-00139680 it's the same citation from the biospecimen work group in ABCD and we know that methods
+qQVIXWr5xJI-00212-00139768-00140160 influence the hormone levels you look at so if i want to know you know how
+qQVIXWr5xJI-00213-00140360-00141032 number one compares to participant number two but participant one's sampled at 9am and participant
+qQVIXWr5xJI-00214-00141032-00141616 two's sampled at 4pm well that's that's quite a big difference I mean that if you think about the
+qQVIXWr5xJI-00215-00141616-00142136 circadian pattern they could look very different in hormone levels just as a function of when
+qQVIXWr5xJI-00216-00142136-00142704 I sampled them not as a function of their own actual hormone levels so you know how do we deal
+qQVIXWr5xJI-00217-00142704-00143472 with that and I'm going to show you what I how I deal with that right now but it's evolving and I
+qQVIXWr5xJI-00218-00143472-00143992 hope that that's really clear to you that this is um a collaborative process that is evolving
+qQVIXWr5xJI-00219-00144056-00144576 so the first one in 2018 we were looking at this and here highlighted in green you can
+qQVIXWr5xJI-00220-00144576-00145072 see that this is just the time of day that the participant gave their saliva sample for hormones
+qQVIXWr5xJI-00221-00145168-00145992 well we have some people giving it at like midnight and at 5:55 a.m and when you go and
+qQVIXWr5xJI-00222-00145992-00146576 talk to all the RAs at the 21 sites in ABCD they did not sample anybody at these times.
+qQVIXWr5xJI-00223-00146576-00147088 so what happened? well it's either a data entry error where am should have been pm
+qQVIXWr5xJI-00224-00147256-00148072 is basically what happened but do I know that this value at 1:30 a.m is really 1:30 p.m? can i
+qQVIXWr5xJI-00225-00148072-00148680 be sure? so that again those kinds of decisions on how to manipulate this data that is a method and
+qQVIXWr5xJI-00226-00148680-00149352 you know that it isn't accurate that there's an error in it or nothing is totally perfect
+qQVIXWr5xJI-00227-00149352-00149784 so i'm just going to kind of let you know there's some issues around the methods with saliva
+qQVIXWr5xJI-00228-00149856-00150232 but you can count them I mean I can count the number of people here where this occurs
+qQVIXWr5xJI-00229-00150232-00150992 i i think it's it's definitely well under 30 people in the sample of 11,888 right so how do
+qQVIXWr5xJI-00230-00150992-00151552 you deal with this you could either exclude these people and just not even bother with it if you
+qQVIXWr5xJI-00231-00151552-00152032 know having an accurate sample time is essential for your understanding which it probably is
+qQVIXWr5xJI-00232-00152088-00152648 so you can either exclude it, you can publish that you decide to manipulate am to pm—which is
+qQVIXWr5xJI-00233-00152648-00153272 what what I did, and the next the study i'm working on now I am excluding them,
+qQVIXWr5xJI-00234-00153272-00153760 so I am making different choices on how to handle this I think as long as you report this and you
+qQVIXWr5xJI-00235-00153760-00154296 share this in your publication it's going to be really important for reproducibility is how do you
+qQVIXWr5xJI-00236-00154296-00154960 handle these kind of more gray area methodological concerns but if you really look this is a normal
+qQVIXWr5xJI-00237-00155136-00155752 window of time in which we test ABCD participants across all sites and the A through T here is all
+qQVIXWr5xJI-00238-00155752-00156512 the different sites so the major majority of abcd participants times are accurate they're accurately
+qQVIXWr5xJI-00239-00156512-00157208 entered if someone came in at 2 pm in baseline usually they come in 2 p.m plus or minus an hour
+qQVIXWr5xJI-00240-00157208-00157816 and year 1 and so on so not when they come in but when they give the saliva sample so you know
+qQVIXWr5xJI-00241-00157816-00158384 across years when they come in could fluctuate a little bit but we are trying to preserve that
+qQVIXWr5xJI-00242-00158480-00158856 but these are all things that you should be thinking of if you're looking at hormone changes
+qQVIXWr5xJI-00243-00158856-00159480 across the years why would they go down for an individual when their pubertal maturation is
+qQVIXWr5xJI-00244-00159480-00160240 going up could be just when they gave the sample the time of day this is how long they took to spit
+qQVIXWr5xJI-00245-00160240-00161168 in the tube the duration of their saliva sampling across sites again just nobody took this long to
+qQVIXWr5xJI-00246-00161168-00161688 spit no one took over 20 minutes this but this talking of the RAs this is not an accurate time
+qQVIXWr5xJI-00247-00161688-00162112 so what happened is they probably did something else and then they remembered to check the time
+qQVIXWr5xJI-00248-00162112-00162688 and put it in or some entry error so how do you handle this with this many participants you could
+qQVIXWr5xJI-00249-00162688-00163280 probably just exclude them right if you don't have good there's no way to really know what
+qQVIXWr5xJI-00250-00163280-00163736 this methodological variable should be and we know it's important for hormone levels you could just
+qQVIXWr5xJI-00251-00163736-00164472 exclude them but if you already are looking at something that's a little more rare or uncommon in
+qQVIXWr5xJI-00252-00164472-00164920 ABCD and you're already dwindling down your sample size to a really small group
+qQVIXWr5xJI-00253-00164920-00165616 you may have some you know fancy things that you could do to keep these participants in your study
+qQVIXWr5xJI-00254-00165672-00166352 and remember what I'm showing you right now is all baseline data again the major majority of kids
+qQVIXWr5xJI-00255-00166352-00167024 all spit within six minutes across the sites and this is baseline this is the first time they ever
+qQVIXWr5xJI-00256-00167024-00167608 did it they're doing it every year they're getting very very good at it and so this is only getting
+qQVIXWr5xJI-00257-00167608-00168248 the methodology around saliva collection is only getting more and more sound as we move forward
+qQVIXWr5xJI-00258-00168440-00169280 so this is phase 2 so from the 2018 publication about ABCD salivary biosciences or gonadal
+qQVIXWr5xJI-00259-00169280-00169992 hormones this is a paper that Megan Herting at USC University of Southern California and I worked on
+qQVIXWr5xJI-00260-00169992-00170680 as a consortium paper with all other investigators in ABCD so this is something that just came out in
+qQVIXWr5xJI-00261-00170680-00171320 publication that you guys got to read for this course and this is what i would say is the next
+qQVIXWr5xJI-00262-00171320-00171896 level of evolution for understanding how to use this so here we published a decision tree
+qQVIXWr5xJI-00263-00171896-00172584 on what what did we do when we were up against a decision that was methodologically based so
+qQVIXWr5xJI-00264-00172584-00173080 I'm just going to walk you through this and I hope that this serves as an example of something that
+qQVIXWr5xJI-00265-00173080-00173552 can be really helpful when you do go work with other biological specimens in abcd so if you're
+qQVIXWr5xJI-00266-00173552-00174064 working with cotinine in hair or you're looking with looking at the genetic assay which I'll go
+qQVIXWr5xJI-00267-00174064-00174824 into a little bit here I really hope that our blood biomarkers that that this decision tree
+qQVIXWr5xJI-00268-00174920-00175624 is something that you could use in a publication to help other people understand how you decided
+qQVIXWr5xJI-00269-00175688-00176584 to handle ABCD data versus how they decided to handle ABCD data again if you find something
+qQVIXWr5xJI-00270-00176656-00177216 testosterone's increased with decreased you know default mode connectivity that's false
+qQVIXWr5xJI-00271-00177216-00177832 I just made that up and then the next person finds that testosterone increased with increased default
+qQVIXWr5xJI-00272-00177832-00178216 mode connectivity you guys find the complete opposite thing where you both looked at the
+qQVIXWr5xJI-00273-00178216-00178968 exact same data ah what happened it's probably a methodological it's probably a decision that you
+qQVIXWr5xJI-00274-00178968-00179560 made about how to handle these methods right it's not a difference in methods because you're using
+qQVIXWr5xJI-00275-00179560-00180048 the exact same data it's how you handle it so i think these decision trees are going to become
+qQVIXWr5xJI-00276-00180048-00180584 extraordinarily important to publish particularly around biospecimen data okay so this is how we did
+qQVIXWr5xJI-00277-00180584-00181368 saliva here me and Megan Herting in 2021 this is a collaboration with Marybel Gonzalez, Eric Kan,
+qQVIXWr5xJI-00278-00181368-00182200 and literally all other ABCD investigators, and Douglas Granger—who is not part of ABCD
+qQVIXWr5xJI-00279-00182200-00182936 but is part of one of my collaborators at UCI and also with Salimetrics. so first we asked ourselves
+qQVIXWr5xJI-00280-00183024-00183728 does the salivary sex match master sex so for master sex we're talking about
+qQVIXWr5xJI-00281-00183800-00184472 biological sex at birth so it's either biological male or biological female no consideration of
+qQVIXWr5xJI-00282-00184472-00185104 gender identity at all at this point just at birth—were you xx, xy, what did your gonads
+qQVIXWr5xJI-00283-00185104-00185808 match now salivary sex is referring to when the research assistant went to take the saliva sample
+qQVIXWr5xJI-00284-00185952-00186568 in a tube and collect it and then screwed it on there's a certain color cap that's
+qQVIXWr5xJI-00285-00186568-00186992 associated with biological males versus a colored cap associated with biological
+qQVIXWr5xJI-00286-00186992-00187440 females so there's one potential avenue for a human error to just grab the wrong color
+qQVIXWr5xJI-00287-00187504-00187992 and then back at Salimetrics when they receive it they also code it so it's possible that they said
+qQVIXWr5xJI-00288-00187992-00188608 the cap was blue when it was actually green right so there's two areas for human potential error
+qQVIXWr5xJI-00289-00188680-00189360 and what happens is that's how we code if the saliva sample at Salimetrics was thawed
+qQVIXWr5xJI-00290-00189424-00190120 and this assayed for testosterone, DHEA with other biological males on that plate or was it
+qQVIXWr5xJI-00291-00190120-00190640 assayed for estradiol, testosterone, DHEA with other biological females on that plate so if you
+qQVIXWr5xJI-00292-00190640-00191232 put a male sample on a female plate it's fine it's probably still a really good sample and
+qQVIXWr5xJI-00293-00191232-00191704 now you have estradiol on the male so you could probably choose to retain that so it's like a
+qQVIXWr5xJI-00294-00191704-00192208 totally appropriate decision to make but you need to report that you're making that because guess
+qQVIXWr5xJI-00295-00192208-00193384 what in this this publication in 2021 I decided not to include them and that excluded 33 people
+qQVIXWr5xJI-00296-00193880-00194808 so if the biological sex with how the sample was assayed at Salimetrics for gonadal hormones
+qQVIXWr5xJI-00297-00194808-00195456 did not match the biological sex reported at birth then we chose to exclude them in the
+qQVIXWr5xJI-00298-00195456-00196128 study and lost 33 people again their samples were still assayed and you get at least two
+qQVIXWr5xJI-00299-00196128-00196680 out of the three for females and it's not a huge deal that they were on a different plate
+qQVIXWr5xJI-00300-00196752-00197064 honestly it probably doesn't matter you could have totally chosen to retain them
+qQVIXWr5xJI-00301-00197064-00197608 but you should say so, so that if you find something different from me and Dr. Herting
+qQVIXWr5xJI-00302-00197608-00198440 we'll know that maybe this decision actually matters right okay so we went ahead
+qQVIXWr5xJI-00303-00198808-00199664 and asked the second question was a hormone sample collected okay so if it wasn't obviously they
+qQVIXWr5xJI-00304-00199664-00200432 were not part of our analyses so we had just gender missing again biological sex at birth
+qQVIXWr5xJI-00305-00200432-00200920 is has a blanket label of gender it's not it's actually biological sex so we just
+qQVIXWr5xJI-00306-00200920-00201496 didn't have that data on five people at this time I believe that's been filled in since then for 3.0
+qQVIXWr5xJI-00307-00201648-00201984 sometimes an RA just wasn't able to collect for various reasons,
+qQVIXWr5xJI-00308-00201984-00202280 sometimes the participant refused to give a sample for whatever reason,
+qQVIXWr5xJI-00309-00202384-00202816 spit kind of grosses some people out we have pretty good luck though with kids here in the US
+qQVIXWr5xJI-00310-00202944-00203424 or it's not collected for whatever reason and so we basically excluded an additional 81
+qQVIXWr5xJI-00311-00203480-00204152 participants. the third question we asked was has the sample been processed for hormone levels so
+qQVIXWr5xJI-00312-00204152-00204720 of course we did this—these analyses were done at a point where we really only had baseline
+qQVIXWr5xJI-00313-00204720-00205271 saliva data that had been sent and was under processing at Salimetrics then the actual hormone
+qQVIXWr5xJI-00314-00205271-00205871 levels de-identified are sent back to ABCD and eventually it makes it up into a public release
+qQVIXWr5xJI-00315-00205944-00206688 of data so at this time we actually only had like 45% of the sample with processed salivary
+qQVIXWr5xJI-00316-00206688-00207264 hormone data so this question probably won't be as important if you're doing baseline in
+qQVIXWr5xJI-00317-00207264-00207792 year 1 now but it is important still in your 3.0 release for year 2 and it will continue to
+qQVIXWr5xJI-00318-00207792-00208416 be important as we do longitudinal analyses with brand new data releases. the fourth question do
+qQVIXWr5xJI-00319-00208416-00209208 replicate values fall above higher limits of each hormone assay so testosterone DHEA and estradiol
+qQVIXWr5xJI-00320-00209208-00209992 the actual salivary assay enzymatic based assays that we use ELISAs they have lower limits of
+qQVIXWr5xJI-00321-00209992-00210832 detection and higher limits of detection so this and we also take one whole saliva sample but once
+qQVIXWr5xJI-00322-00210832-00211592 it hits Salimetrics when the aliquot it out they actually take two pulls from it participant to
+qQVIXWr5xJI-00323-00211664-00212344 so all hormones are assayed in replicates from each participant so they end up having
+qQVIXWr5xJI-00324-00212448-00213256 two baseline estradiol values for biological females or two testosterone replicates for
+qQVIXWr5xJI-00325-00213256-00213744 biological males and females so what do you do when you have two replicates what does
+qQVIXWr5xJI-00326-00213744-00214384 that look like so this first question is asking if they fall above higher limits of the assay
+qQVIXWr5xJI-00327-00214448-00214896 sorry i'm jumping ahead of myself I'll get to the replicates if there only have one in just a moment
+qQVIXWr5xJI-00328-00214952-00215464 if it falls above the higher level of detection so the assay the level of detection is really where
+qQVIXWr5xJI-00329-00215464-00216360 we're very confident that the specificity and accuracy are you know very reliable
+qQVIXWr5xJI-00330-00216432-00217208 in that range so if a value that we get from the assay a quantified hormone value is below
+qQVIXWr5xJI-00331-00217208-00217823 the lower limits of detection it's still a quantified value it's just we can't it's not
+qQVIXWr5xJI-00332-00217823-00218640 nearly as reliable when it falls out of this range if it's higher than the upper limits of detection
+qQVIXWr5xJI-00333-00218640-00219208 of the hormone assay it's still a quantified level but it's just not as reliable as if it
+qQVIXWr5xJI-00334-00219208-00219840 had fallen in the limits so how do you handle this again I think a lot of people have different ideas
+qQVIXWr5xJI-00335-00219840-00220592 and depending on how big of the subsample you're analyzing ABCD is you know how much you think this
+qQVIXWr5xJI-00336-00220592-00221223 matters if you could be highly highly conservative in these decisions and just say I only want people
+qQVIXWr5xJI-00337-00221223-00221912 with full data and only within this time frame and no methodological errors or concerns and just work
+qQVIXWr5xJI-00338-00221912-00222560 with that sample I mean ABCD is so big it's still probably some of the largest salivary biomarker
+qQVIXWr5xJI-00339-00222560-00223128 study data that we even have out there which is another thing that's really cool about ABCD it's
+qQVIXWr5xJI-00340-00223128-00223744 a hallmark study for us in salivary biosciences because this is literally the biggest group of
+qQVIXWr5xJI-00341-00223744-00224528 adolescents we've ever done saliva for non-genetic markers on at this age range and shipped it all
+qQVIXWr5xJI-00342-00224528-00225016 the way from all over the country and so this has just really shown the power of a
+qQVIXWr5xJI-00343-00225016-00225528 super non-invasive biomarker that could be done in anyone's classroom, home, anywhere they don't even
+qQVIXWr5xJI-00344-00225528-00226223 have to come in the lab to do this unlike a blood draw or MRI so that's the cool thing about saliva
+qQVIXWr5xJI-00345-00226223-00227064 but we do have these windows of reliability so if it goes above some people would argue that you can
+qQVIXWr5xJI-00346-00227064-00227760 still retain it but you might want to replace it with that highest level and you may want to see
+qQVIXWr5xJI-00347-00227760-00228288 how many participants fall above that higher limit of detect detectability and just give them all the
+qQVIXWr5xJI-00348-00228288-00228992 same value because we aren't positive that someone that's above that limit and is the highest person
+qQVIXWr5xJI-00349-00229071-00229616 is actually has the most hormone out of the whole ABCD cohort another person who's above
+qQVIXWr5xJI-00350-00229616-00230112 the limits of detectability but is the second to highest might have the highest
+qQVIXWr5xJI-00351-00230112-00230584 because when you are above that window of detectability it's just not as reliable so we
+qQVIXWr5xJI-00352-00230584-00231167 can't really compare those people as easily so you could either retain that data and just report that
+qQVIXWr5xJI-00353-00231167-00231808 as is or you can replace with that highest level of detectability which is what we did so
+qQVIXWr5xJI-00354-00231912-00232360 what we did is we went an extra step and if they were above this limit of detectability
+qQVIXWr5xJI-00355-00232360-00233152 we went to the research assistant notes on saliva and we asked do they endorse any concerns with the
+qQVIXWr5xJI-00356-00233152-00233640 sample so when they give the sample it could be a weird color because they were sucking on some you
+qQVIXWr5xJI-00357-00233640-00234184 know candy which they're not allowed to eat for an hour and they rinse their mouths 30 minutes before
+qQVIXWr5xJI-00358-00234256-00234552 to make sure that the oral environment is clean and there's no
+qQVIXWr5xJI-00359-00234680-00235360 high probability of food in the mouth so different things can cause discoloring a sample it could
+qQVIXWr5xJI-00360-00235360-00235776 have a lot of food in it it could be really bubbly so we're not sure we got the full volume
+qQVIXWr5xJI-00361-00235776-00236384 um which we are aiming for about 1.8 milliliters so there's a drop down menu that the research
+qQVIXWr5xJI-00362-00236384-00237023 assistants could select if they endorsed a concern and the participant sample was above
+qQVIXWr5xJI-00363-00237023-00237600 the higher limits of sensitivity then we did not use it no we did not retain the replicate
+qQVIXWr5xJI-00364-00237832-00238904 and if it was above it but there were no methodological concerns we did retain it
+qQVIXWr5xJI-00365-00238976-00239432 so this was such a big decision tree I actually just cut it in half and moved it over here to
+qQVIXWr5xJI-00366-00239432-00239912 the right so this is just picking up where we left here so after we asked if it was above
+qQVIXWr5xJI-00367-00239912-00240576 the limit we asked the exact same question but is a replicate too low for detection
+qQVIXWr5xJI-00368-00240712-00241208 so if the answer is below the lower limits of detectability if the answer is
+qQVIXWr5xJI-00369-00241208-00241919 no then we retain that replicate if the answer is yes then we decided to change it to
+qQVIXWr5xJI-00370-00241919-00242623 zero so this was a big decision here in this paper so if someone's hormone level for DHEA
+qQVIXWr5xJI-00371-00242728-00243312 was a quantified value in ABCD that you see you see a number for DHEA but it's actually
+qQVIXWr5xJI-00372-00243312-00243856 below the assay lower levels of detection which are all put up there with the assay info sheets
+qQVIXWr5xJI-00373-00243856-00244319 on Salimetrics websites for all the hormones they're also repeated in all of these publications
+qQVIXWr5xJI-00374-00244464-00245088 because this is only baseline data and we expected participants to just be starting that DHEA surge
+qQVIXWr5xJI-00375-00245088-00245744 to kick off puberty we just decide to replace with zero because they're so pre-pubertal had this been
+qQVIXWr5xJI-00376-00245744-00246384 adults we might have taken the lower level of detectability and zero and averaged it because
+qQVIXWr5xJI-00377-00246704-00247576 NDS so a level that is not in that optimal range of detectability
+qQVIXWr5xJI-00378-00247648-00248367 is that means that there is a saliva sample it was assayed and it detected the presence
+qQVIXWr5xJI-00379-00248367-00248904 of the hormone we were looking for but we're just not super confident about the exact
+qQVIXWr5xJI-00380-00248904-00249584 quantified value so there's hormone in there so you could replace it with you know an average
+qQVIXWr5xJI-00381-00249584-00250016 between the lower limits of detectability and zero or you can do something like that on the
+qQVIXWr5xJI-00382-00250016-00250632 top level too so there's a whole science and Dr. Jenna Reese one of my colleagues is part of a
+qQVIXWr5xJI-00383-00250704-00251200 textbook that came out on a lot of these issues and she offers a course at our school
+qQVIXWr5xJI-00384-00251200-00251856 UC Irvine on this so there's a lot of different decisions so this whole decision tree by no means
+qQVIXWr5xJI-00385-00251912-00252328 is should be like a bible and you always have to follow this this is a baseline data
+qQVIXWr5xJI-00386-00252416-00253040 for pre-pubertal adolescence and you could have justified different decisions the point is that
+qQVIXWr5xJI-00387-00253040-00253696 you should be intentional you should report it and you should publish your justification somewhere
+qQVIXWr5xJI-00388-00253840-00254528 okay so after we looked at each replicate is it too high above detectable ranges is it too low
+qQVIXWr5xJI-00389-00254528-00254944 we made these decisions we altered it we either replaced with zero or the highest
+qQVIXWr5xJI-00390-00254944-00255600 level of detectability which really didn't happen very often actually i don't think
+qQVIXWr5xJI-00391-00255600-00256128 it happened at all there were no yeses because they're all pre-pubertal but that will change as
+qQVIXWr5xJI-00392-00256128-00256688 the hormones go up and up in future years we then ask the question here number six
+qQVIXWr5xJI-00393-00256823-00257440 did we obtain a single value for each hormone type okay so remember
+qQVIXWr5xJI-00394-00257519-00258119 if a replicate doesn't exist or we made a decision to discard it because of methodological concerns
+qQVIXWr5xJI-00395-00258280-00258823 then now we're left to one replicate value for a person whereas all other ABCD participants
+qQVIXWr5xJI-00396-00258823-00259719 have two values for each hormone so we either one average these corrected replicates from
+qQVIXWr5xJI-00397-00259719-00260119 steps four and five into a single value so we took the average of the replicates
+qQVIXWr5xJI-00398-00260184-00260623 or in this paper we decided if only one replicate exists and it's
+qQVIXWr5xJI-00399-00260688-00261256 it went through all of our filtering that we would use it as the final value alongside all
+qQVIXWr5xJI-00400-00261256-00261680 the other participants averaged replicates and that was the decision we made here.
+qQVIXWr5xJI-00401-00261752-00262624 so we ended up you can see that at the end there was 36 plus four plus 26 so we had about
+qQVIXWr5xJI-00402-00262624-00263384 66 participants who ended up with one replicate representing all of their hormone level and that
+qQVIXWr5xJI-00403-00263384-00263856 varied on the actual hormone that we were looking at most people had two replicates that passed
+qQVIXWr5xJI-00404-00263856-00264408 this filtering decision tree and then we averaged them so we had one hormone value for each person
+qQVIXWr5xJI-00405-00264672-00265416 so the same paper wanted to point out that we collected some data ahead of time that we
+qQVIXWr5xJI-00406-00265416-00266144 knew would likely be confounding variables for salivary hormones this includes caffeine in the
+qQVIXWr5xJI-00407-00266144-00266840 past 12 hours physical activity rigorous physical activity in the past 12 hours this says time since
+qQVIXWr5xJI-00408-00266840-00267328 midnight but really it's the time of day your circadian pattern and what time you collect during
+qQVIXWr5xJI-00409-00267328-00267992 that circadian pattern in hormone secretion how long you take to spit at the collection duration
+qQVIXWr5xJI-00410-00267992-00268488 and then the time to freeze so collection duration time to freeze weren't significant in
+qQVIXWr5xJI-00411-00268488-00269048 baseline data they certainly time for to freeze is really just a mythological check it should never
+qQVIXWr5xJI-00412-00269192-00269784 show up in our in our hormone values because the standard operating procedure is so tight in ABCD
+qQVIXWr5xJI-00413-00269784-00270320 around this collection but it's always good to check collection duration may start mattering
+qQVIXWr5xJI-00414-00270320-00270896 more and more as they become older and certainly caffeine physical activity and the time of day did
+qQVIXWr5xJI-00415-00270896-00271672 matter even at baseline in super pre-pubertal or early people children ABCD these will only become
+qQVIXWr5xJI-00416-00271728-00272240 more and more important as the child ages and as there's more individual variability and caffeine
+qQVIXWr5xJI-00417-00272240-00272912 intake more kids taking caffeine more individual variability in activity before they come in to
+qQVIXWr5xJI-00418-00272912-00273520 give their saliva sample so all these will become more and more important in salivary biosciences we
+qQVIXWr5xJI-00419-00273520-00274360 know that the oral environment so braces cavities infections all of those things can matter for the
+qQVIXWr5xJI-00420-00274360-00275064 quality of your saliva sample and we unfortunately just were not able to collect that in ABD so that
+qQVIXWr5xJI-00421-00275064-00275664 is one limitation of using this data another one is flow rate because we weren't able to weigh
+qQVIXWr5xJI-00422-00275664-00276184 the samples and see the volume that was given in a certain amount of collection time which we do have
+qQVIXWr5xJI-00423-00276184-00276688 the collection time we don't have actual volume of the frozen saliva sample those two components are
+qQVIXWr5xJI-00424-00276688-00277312 used to calculate actual flow rate so the longer you go to spit sometimes you can over concentrate
+qQVIXWr5xJI-00425-00277312-00277872 a sample or you can dilute a sample depending on the assay you're looking at so it is actually
+qQVIXWr5xJI-00426-00277872-00278296 a really important mythological consideration that we don't have in ABCD so I just want to be
+qQVIXWr5xJI-00427-00278296-00278784 upfront that these are limitations that we should be reporting in our papers when we publish on this
+qQVIXWr5xJI-00428-00278928-00279488 we do have this new publication that shows that with increasing Peterson Development
+qQVIXWr5xJI-00429-00279488-00280120 Scale so physical maturation this is just biological girls increasing physical maturation
+qQVIXWr5xJI-00430-00280120-00280856 we see increasing the gonadal hormones for all three hormones and you can see a lot of individual
+qQVIXWr5xJI-00431-00280856-00281456 individual variability even some biological girls who have started menstruation denoted here in
+qQVIXWr5xJI-00432-00281456-00282080 blue still have lower levels than other girls who have yet to even begun pubertal maturation and of
+qQVIXWr5xJI-00433-00282080-00282624 course these hormones can fluctuate depending on your cycle so as menstruation becomes more
+qQVIXWr5xJI-00434-00282624-00283144 and more a factor as these kids age when you're looking at biological females you need to also
+qQVIXWr5xJI-00435-00283200-00283680 look at the menstruation questionnaires to see if you can kind of estimate when they came
+qQVIXWr5xJI-00436-00283680-00284160 in and gave their saliva sample even though all these cycles aren't necessarily regular
+qQVIXWr5xJI-00437-00284344-00285080 it's complex but it's totally doable the data's there okay so phase three this is really the final
+qQVIXWr5xJI-00438-00285080-00285784 phase—this is where we're at now this is not the final phase I think we're only going to get better
+qQVIXWr5xJI-00439-00285784-00286392 and better at handling a lot of these biospecimens that are collected in these this huge mass of the
+qQVIXWr5xJI-00440-00286392-00287336 ABCD participant cohort so this is now a whole group of really leaders in intellectual thought
+qQVIXWr5xJI-00441-00287336-00287872 and academic research around puberty and using pubertal hormones with physical maturation that
+qQVIXWr5xJI-00442-00287872-00288520 are not part of ABCD so they're not giving these repronim lectures they're you know all around
+qQVIXWr5xJI-00443-00288520-00289160 the world and they graciously invited Dr. Megan Herting and myself to be part of a commentary
+qQVIXWr5xJI-00444-00289160-00289904 on the other publication so they're—this is coming out probably in february 2021—and they
+qQVIXWr5xJI-00445-00289904-00290512 give you all of the specific variable names and the downloaded data that are used so really good
+qQVIXWr5xJI-00446-00290512-00291056 helpful tool it talks through the results that we found because the paper Megan Herting published
+qQVIXWr5xJI-00447-00291056-00291688 is a beast and so what I really recommend if you want to use salivary hormones to really look at
+qQVIXWr5xJI-00448-00291688-00292384 the three publications that I have outlined here especially the Chen et al 2021 is it will be out
+qQVIXWr5xJI-00449-00292384-00293224 here next month in february 2021 so you can see the evolution they all have you know really if
+qQVIXWr5xJI-00450-00293224-00293672 you read all three of them i think you'll feel pretty confident about navigating them in ABCD
+qQVIXWr5xJI-00451-00293896-00294616 okay so to finish on the biomarkers we also have collected biospecimens for genetics
+qQVIXWr5xJI-00452-00294824-00295208 so we have some saliva baseline for participants for genetic purposes
+qQVIXWr5xJI-00453-00295304-00295936 but mostly the goal is to do blood in everyone and this is really ramping up for 100% participants in
+qQVIXWr5xJI-00454-00295936-00296592 year two pandemic did hit at this time there can be some delays in when this is being collected but
+qQVIXWr5xJI-00455-00296592-00297216 the nice thing about genetics is it is a one-time snapshot these are stable metrics that we don't
+qQVIXWr5xJI-00456-00297216-00297896 expect to change so we are going to have this data in everyone who is willing to give it to us
+qQVIXWr5xJI-00457-00298016-00298528 so for genetics the data that's in your 3.0 is using a smokescreen genotyping array
+qQVIXWr5xJI-00458-00298600-00299072 and epigenetics have not been processed and they will be coming i don't think they'll come
+qQVIXWr5xJI-00459-00299072-00299624 for 4.0 is what I hear but they will be coming eventually down the road and they're certainly
+qQVIXWr5xJI-00460-00299624-00300240 in the biobanked area there is leftover you know things like blood and some saliva
+qQVIXWr5xJI-00461-00300240-00300856 and from the gonadal hormone saliva it's all being stored in San Diego in a biobank for the
+qQVIXWr5xJI-00462-00300856-00301792 blood it's being stored in Rutgers in a biobank and so both of them can be processed for genetic
+qQVIXWr5xJI-00463-00301792-00302280 epigenetic markers saliva spun down and has the same kind of pellet at the bottom
+qQVIXWr5xJI-00464-00302280-00302856 that can be used for genetics and epigenetics so lots of potential for getting funding getting
+qQVIXWr5xJI-00465-00302856-00303552 ABCD and NIH to buy in and to access these biobank samples to do way more and then with expectation
+qQVIXWr5xJI-00466-00303552-00304096 that you would share them in the next ABCD public data release when when the data is available
+qQVIXWr5xJI-00467-00304272-00305088 so I am not a geneticist but I asked the people in abcd who are and they said that that the details
+qQVIXWr5xJI-00468-00305088-00305896 on the genetic array that is available to you in the 3.0 dataset has very good detail in the
+qQVIXWr5xJI-00469-00305896-00306632 release notes and each of the variables has an accompanying README file so the here is a genetic
+qQVIXWr5xJI-00470-00306632-00307136 smokescreen array specifically around genetics that people think are important for substance use
+qQVIXWr5xJI-00471-00307136-00308024 disorders and if you're like me I like to go and see if someone else has used these biospecimens
+qQVIXWr5xJI-00472-00308024-00308656 in ABCD i like to go read their publication before I start to work with these variables so i did find
+qQVIXWr5xJI-00473-00308656-00309336 a trusted list of authors who have used this data before that you could go look at and if you're
+qQVIXWr5xJI-00474-00309336-00309832 interested in getting connected again just reach out to myself or any of the other ABCD leadership
+qQVIXWr5xJI-00475-00309896-00310384 I'm not a leadership i'm just an investigator but reach out to any of us and and you know we'll get
+qQVIXWr5xJI-00476-00310384-00311264 you connected with the right people so finally we're doing ABCD baby teeth and this is described
+qQVIXWr5xJI-00477-00311264-00311776 a little bit again in this 2018 publication this is still ongoing for participants in that you know
+qQVIXWr5xJI-00478-00311776-00312272 as they're shedding more and more teeth you know the parent can snatch it from the tooth fairy,
+qQVIXWr5xJI-00479-00312352-00312856 bring it to ABCD, but the idea is that you can scrape away the layers of the tooth just
+qQVIXWr5xJI-00480-00312856-00313384 like the rings of a tree show you about the different environmental exposures the tree was
+qQVIXWr5xJI-00481-00313384-00313944 exposed to in a specific year you get the same kind of layering across the lifespan
+qQVIXWr5xJI-00482-00313944-00314408 beginning in a second trimester in utero all the way to the time that you shed the tooth
+qQVIXWr5xJI-00483-00314464-00315144 and so there's a sub-study going on where they selected 500 participants in ABCD
+qQVIXWr5xJI-00484-00315144-00315616 because of the cost they couldn't do this across the board but they have way more teeth than 500
+qQVIXWr5xJI-00485-00315616-00316320 goes to Manish Arora's laboratory for analysis and a lot more details on this can be found with NIH
+qQVIXWr5xJI-00486-00316384-00316800 the grant lives with the PI Elizabeth Sowell at Children's Hospital Los Angeles
+qQVIXWr5xJI-00487-00316880-00317376 and this exposure will show you from second trimester to the time of the shed tooth
+qQVIXWr5xJI-00488-00317376-00317920 and it what they are specifically funding in collaboration with Manish Arora is weekly
+qQVIXWr5xJI-00489-00317920-00318680 averages of lead and manganese exposure but there are so many other things you could do too
+qQVIXWr5xJI-00490-00318680-00319072 and other teeth you could do this with so I think this is a really exciting area for a lot of people
+qQVIXWr5xJI-00491-00319224-00319808 okay and then also lots of people apply and do sub-studies with biobanked blood or they get some
+qQVIXWr5xJI-00492-00319808-00320632 sites to buy in and do new data collection but on a smaller scale so these are like sub-studies
+qQVIXWr5xJI-00493-00320632-00321064 and I actually preparing this lecture I found that there was adrenal corticotropin hormone or
+qQVIXWr5xJI-00494-00321064-00321728 acth from blood that got released with a 3.0 data set and this is the hypothalamic pituitary adrenal
+qQVIXWr5xJI-00495-00321728-00322208 access hormone which interacts with the one I presented to you the hypothalamic
+qQVIXWr5xJI-00496-00322208-00322832 pituitary gonadal axis hormones so I got very excited and you know you can tell that this is
+qQVIXWr5xJI-00497-00322832-00323568 just on 939 subjects at this point but you know more and more sub-study data will become available
+qQVIXWr5xJI-00498-00323568-00323992 so you know something to keep your eye out if you're interested in biospecimens
+qQVIXWr5xJI-00499-00324240-00324744 so i was asked to talk about biospecimens but i think a lot of you are probably that
+qQVIXWr5xJI-00500-00324744-00325296 are interested in this lecture are probably really thinking more like biomarker land and
+qQVIXWr5xJI-00501-00325456-00325968 I just wanted to make everyone aware that there's a whole array of physiological measures
+qQVIXWr5xJI-00502-00326032-00326592 that aren't biological specimens but they are a direct biological measure of physiological
+qQVIXWr5xJI-00503-00326592-00327184 functioning so there's still a biomarker and could absolutely in connection with the biospecimens be
+qQVIXWr5xJI-00504-00327184-00327976 used to make very powerful data analysis and so I just listed some of my top favorites here
+qQVIXWr5xJI-00505-00327976-00328576 um we don't measure sars-cov virus but we do ask participants if they've been exposed
+qQVIXWr5xJI-00506-00328576-00329040 anthropometrics physical development with the peterson development scale
+qQVIXWr5xJI-00507-00329040-00329752 a traumatic brain injury quantification again that would be a subset of participants multimodal
+qQVIXWr5xJI-00508-00329752-00330176 neuroimaging for central nervous system functioning birth metrics blood pressure
+qQVIXWr5xJI-00509-00330176-00330688 respiratory functioning menstrual cycle so all of these kind of physiological metrics could
+qQVIXWr5xJI-00510-00330688-00331112 be considered in the realm of biomarkers above and beyond biological specimens so
+qQVIXWr5xJI-00511-00331112-00331520 I would encourage you if you're doing biological specimens to also consider some of these markers
+qQVIXWr5xJI-00512-00331696-00332088 so some general do's and don'ts I've gone over probably a lot of these but
+qQVIXWr5xJI-00513-00332160-00332872 always plot out all your individual data for biomarkers and the collection variables so the
+qQVIXWr5xJI-00514-00332872-00333536 methods to collect those biomarkers plot every ABCD participant out so you can see every dot
+qQVIXWr5xJI-00515-00333536-00334400 look at your data look for outliers and first ask yourself are these physiologically plausible
+qQVIXWr5xJI-00516-00334552-00334968 and methodologically did someone really go in 2 a.m and give that sample or right so
+qQVIXWr5xJI-00517-00334968-00335648 there's there's methodological and physiological constraints on what data is believable so do a
+qQVIXWr5xJI-00518-00335648-00336336 good job QCing seeing before you do anything and then in any of these individual data points that
+qQVIXWr5xJI-00519-00336336-00337160 are questionable you know us on the ABCD side that do the quality checking we will try to
+qQVIXWr5xJI-00520-00337160-00337856 eliminate clear errors in biological markers for you guys when we put these data releases out
+qQVIXWr5xJI-00521-00337912-00338416 but there are a lot of gray areas you know how old someone was when they had a child
+qQVIXWr5xJI-00522-00338480-00338984 I mean some could be really young some could be really old and i mean it's biologically possible
+qQVIXWr5xJI-00523-00339072-00339656 maybe improbable I don't know but it's possible so we'll really leave a lot of these decisions to
+qQVIXWr5xJI-00524-00339656-00340328 you the investigator so quality check all the data up front and then when you do make these decisions
+qQVIXWr5xJI-00525-00340328-00340856 be intentional and report them because this is about us being able to reproduce it and if you
+qQVIXWr5xJI-00526-00340856-00341680 and i find different things I can go back and say oh they decided to you know include that one baby
+qQVIXWr5xJI-00527-00341680-00342192 that weighed a really small amount I didn't think that was even possible but they included it i did
+qQVIXWr5xJI-00528-00342192-00342792 it right and and so if there's a methodological it's not even methods if there's an analytical
+qQVIXWr5xJI-00529-00342792-00343208 decision that you and I are making that's different and we're getting different results
+qQVIXWr5xJI-00530-00343208-00343624 that's really important for us to know so we need to know that i need to be in our publications it
+qQVIXWr5xJI-00531-00343624-00344304 needs to be shared publish those decision trees share those scripts got github I think that's a
+qQVIXWr5xJI-00532-00344304-00344704 really nice thing that we're doing to support each other helps with reproducibility and I
+qQVIXWr5xJI-00533-00344704-00345016 think it's particularly important when you're working with biomarkers in ABCD
+qQVIXWr5xJI-00534-00345192-00345600 read all the previously published articles with the biospecimen you're interested in to see how
+qQVIXWr5xJI-00535-00345600-00346200 they handled it and feel free to make different decisions and see what you get I think that that's
+qQVIXWr5xJI-00536-00346200-00346616 really powerful you don't have to follow what we're all doing you just have to let us know
+qQVIXWr5xJI-00537-00346616-00347144 what you're doing this differently so we can see and interpret the results in that context
+qQVIXWr5xJI-00538-00347312-00347824 if you are embarking on a biospecimen that you have not had experience with I highly recommend
+qQVIXWr5xJI-00539-00347824-00348416 buddy up with the expert right to really make sure that it's a meaningful long-lasting use
+qQVIXWr5xJI-00540-00348416-00349008 of that biospecimen like the genetic array I think it's really cool I am not a geneticist so if i was
+qQVIXWr5xJI-00541-00349008-00349528 going to do those analyses i would absolutely buddy up with someone who um has a good track
+qQVIXWr5xJI-00542-00349528-00350296 record of working with those biospecimens please please something we care a lot about in ABCD is
+qQVIXWr5xJI-00543-00350296-00351080 be considerate of the socio-demographic factors of the participants and how you interpret that
+qQVIXWr5xJI-00544-00351256-00351816 sample bias cutting hair if we don't get hair in someone and you're only you know talking about the
+qQVIXWr5xJI-00545-00351816-00352328 people that gave hair well is it just did we miss a whole bunch of boys that like short hairs or
+qQVIXWr5xJI-00546-00352328-00353024 something else about them you know why are we not collecting hair in a certain subpopulation or why
+qQVIXWr5xJI-00547-00353024-00353544 is a certain cell population not giving genetics you know you have to consider those actual real
+qQVIXWr5xJI-00548-00353624-00354096 you know factors that that lead to sample bias when you're doing biological specimens there's a
+qQVIXWr5xJI-00549-00354096-00354768 whole history of distrust of medical research that in ABCD we worked really hard to you know address
+qQVIXWr5xJI-00550-00354768-00355392 before we even started recruiting and you know by giving your biological specimen over to science
+qQVIXWr5xJI-00551-00355392-00355968 is still something that feels uncomfortable to some families that participate so be considerate
+qQVIXWr5xJI-00552-00355968-00356392 of the sample bias with biospecimens and the methods around collecting them don't just throw
+qQVIXWr5xJI-00553-00356392-00356912 a biospecimen marker value in your analyses and then call it done like use all the methods
+qQVIXWr5xJI-00554-00357112-00357904 that are given to you in considering your analyses report effect sizes that's for everything you hear
+qQVIXWr5xJI-00555-00357904-00358480 that probably a lot in here in this course and my only don't is don't be scared you can do it
+qQVIXWr5xJI-00556-00358480-00359096 if you have not used biomarkers or biospecimens before you can do it just create your dream team
+qQVIXWr5xJI-00557-00359096-00359576 find the people with the expertise that you want to learn reach out to them ABCD is all about
+qQVIXWr5xJI-00558-00359648-00360144 teamwork and collaboration so if you want to do it there are people who will do that with you
+qQVIXWr5xJI-00559-00360208-00360792 this is the place to get going with that of course tons of acknowledgements to mentors Liz Sowell,
+qQVIXWr5xJI-00560-00360792-00361528 Doug Granger, colleagues that I've worked with heavily on the publications and in ABCD,
+qQVIXWr5xJI-00561-00361528-00362120 all of my lab members at Developing Brain Laboratory at UC Irvine, huge thanks to the
+qQVIXWr5xJI-00562-00362184-00362760 the wonderful people that created ABCD repronim course and all my colleagues at the institute for
+qQVIXWr5xJI-00563-00362760-00363424 interdisciplinary salivary bioscience research at UCI they're all listed there where we really work
+qQVIXWr5xJI-00564-00363424-00364160 to better understand and replicate biomarkers from saliva all right thank you very much
+qRFwNgKGwRc-00000-00000004-00000295 Sadhguru: Silence does not mean
+qRFwNgKGwRc-00001-00000298-00000469 shutting your mouth silence.
+qRFwNgKGwRc-00002-00000753-00001193 What is being referred to as silence or nishabdh.
+qRFwNgKGwRc-00003-00001653-00002017 One cannot do this.
+qRFwNgKGwRc-00004-00002019-00002173 One can become this.
+qRFwNgKGwRc-00005-00002580-00003444 See if you can articulate the same thing that you're saying with half the number of words,
+qRFwNgKGwRc-00006-00003446-00003790 suddenly you will become extremely conscious of everything.
+qRFwNgKGwRc-00007-00004558-00005056 Sadhguru: The wind makes its own sound,
+qRFwNgKGwRc-00008-00005059-00005818 so does the water,
+qRFwNgKGwRc-00009-00005820-00006531 so does the earth, in fact so does the sky.
+qRFwNgKGwRc-00010-00006534-00007154 All the celestial objects are making their own sounds.
+qRFwNgKGwRc-00011-00007156-00007576 There are sounds of pain, sounds of pleasure,
+qRFwNgKGwRc-00012-00007578-00008216 sounds of misery, sounds of joy
+qRFwNgKGwRc-00013-00008218-00009073 for everything there is a sound. With human life it’s very true
+qRFwNgKGwRc-00014-00009077-00009856 sounds of life and there are sounds of death.
+qRFwNgKGwRc-00015-00009858-00010595 And there are even sounds of silence.
+qRFwNgKGwRc-00016-00010598-00011520 So this complex amalgamation of sounds, different kinds of sounds
+qRFwNgKGwRc-00017-00011522-00012339 creates different intent, different levels of compulsiveness,
+qRFwNgKGwRc-00018-00012341-00012876 different levels of bondage
+qRFwNgKGwRc-00019-00012878-00013707 and different possibilities of entanglement.
+qRFwNgKGwRc-00020-00013709-00014347 In all this why
+qRFwNgKGwRc-00021-00014350-00014807 one chooses to become silent?
+qRFwNgKGwRc-00022-00014809-00015800 Silence is not a way of denying yourself life.
+qRFwNgKGwRc-00023-00015819-00016484 It is not even about denying yourself articulation.
+qRFwNgKGwRc-00024-00016487-00017207 Silence is about moving from compulsiveness to consciousness.
+qRFwNgKGwRc-00025-00017209-00017557 All sounds represent
+qRFwNgKGwRc-00026-00017560-00018237 different forms and different dimensions of compulsiveness.
+qRFwNgKGwRc-00027-00018239-00018695 When one chooses to become conscious,
+qRFwNgKGwRc-00028-00018697-00019445 not be entangled in the compulsive nature of the existence,
+qRFwNgKGwRc-00029-00019448-00019936 he naturally chooses to become silent.
+qRFwNgKGwRc-00030-00019939-00020223 Silence does not mean
+qRFwNgKGwRc-00031-00020225-00020984 shutting your mouth silence, plugging your ears silence.
+qRFwNgKGwRc-00032-00020986-00021692 You can sing in silence, you can dance in silence,
+qRFwNgKGwRc-00033-00021694-00022357 you can even speak in silence, you can even scream in silence. You don’t try.
+qRFwNgKGwRc-00034-00023388-00023695 That does not mean when everybody is silent you scream,
+qRFwNgKGwRc-00035-00023697-00024498 that is not screaming in silence (Laughs),
+qRFwNgKGwRc-00036-00024500-00024985 because sound is of the surface,
+qRFwNgKGwRc-00037-00024987-00025475 silence is of the core.
+qRFwNgKGwRc-00038-00025477-00025846 It is only the surface which makes the sound.
+qRFwNgKGwRc-00039-00025848-00026512 See now the water is hitting the surface and making the sound.
+qRFwNgKGwRc-00040-00026514-00027018 Deep down there is no sound.
+qRFwNgKGwRc-00041-00027022-00027447 That is so with life. On the surface,
+qRFwNgKGwRc-00042-00027450-00028145 it is a very complex mesh of sounds,
+qRFwNgKGwRc-00043-00028147-00028695 a maze of sounds happening,
+qRFwNgKGwRc-00044-00028697-00029298 but at the core there is no sound.
+qRFwNgKGwRc-00045-00029300-00029722 There is no word for no sound in English language.
+qRFwNgKGwRc-00046-00029725-00029993 We call it nishabdh.
+qRFwNgKGwRc-00047-00029996-00030603 That means absence of sound, a total absence of sound.
+qRFwNgKGwRc-00048-00030606-00031125 Absence of sound means absence of reverberation;
+qRFwNgKGwRc-00049-00031127-00031563 absence of reverberation means absence of life;
+qRFwNgKGwRc-00050-00031565-00031952 absence of life means absence of creation;
+qRFwNgKGwRc-00051-00031954-00032676 absence of creation means absence of the Creator too.
+qRFwNgKGwRc-00052-00032679-00033080 So a space which is beyond creation and Creator,
+qRFwNgKGwRc-00053-00033083-00033695 a dimension which is beyond life and death,
+qRFwNgKGwRc-00054-00033697-00034187 that is what is being referred to as silence or nishabdh.
+qRFwNgKGwRc-00055-00035111-00035862 One cannot do this. One can become this.
+qRFwNgKGwRc-00056-00035864-00036401 Commonly there are two words employed to describe this
+qRFwNgKGwRc-00057-00036403-00037296 one is moun, that means you’re practicing silence, you’re not silent.
+qRFwNgKGwRc-00058-00037298-00037663 You shut your mouth and
+qRFwNgKGwRc-00059-00037665-00038158 you remaining in moun, this is not nishabdh.
+qRFwNgKGwRc-00060-00038160-00038534 Moun means you’re practicing silence.
+qRFwNgKGwRc-00061-00038536-00039076 If you’re practicing something, obviously you’re not that.
+qRFwNgKGwRc-00062-00039078-00039646 Are you life or are you practicing life?
+qRFwNgKGwRc-00063-00039649-00040257 I’m asking you, are you life or are you practicing life?
+qRFwNgKGwRc-00064-00040260-00040651 Practicing life? No, you are life.
+qRFwNgKGwRc-00065-00040653-00041259 May be not very well practiced but you’re life (Laughs).
+qRFwNgKGwRc-00066-00041262-00041695 Similarly there is a difference between
+qRFwNgKGwRc-00067-00041698-00042250 practicing silence and becoming silence.
+qRFwNgKGwRc-00068-00042255-00042984 Practicing silence starts as a certain restriction
+qRFwNgKGwRc-00069-00042987-00043563 but slowly as one practices silence, one
+qRFwNgKGwRc-00070-00043567-00044112 can possibly become silent.
+qRFwNgKGwRc-00071-00044114-00044608 One can become silent in intense activity
+qRFwNgKGwRc-00072-00044610-00045081 but nobody can keep up intense activity forever.
+qRFwNgKGwRc-00073-00045083-00045643 Moments of intense activity are only moments, nobody can sustain it.
+qRFwNgKGwRc-00074-00045646-00045992 You’ll expend yourself.
+qRFwNgKGwRc-00075-00045995-00046552 Only if you can achieve it in inactivity, it is sustainable.
+qRFwNgKGwRc-00076-00046554-00047016 If something is possible only in intense activity,
+qRFwNgKGwRc-00077-00047018-00047495 it is never sustainable.
+qRFwNgKGwRc-00078-00047498-00048056 In a very intense and ecstatic moment of singing,
+qRFwNgKGwRc-00079-00048059-00048582 somebody may become or taste a moment of silence,
+qRFwNgKGwRc-00080-00048584-00048891 but how long can you keep it up?
+qRFwNgKGwRc-00081-00048896-00049467 Your throat will tear and
+qRFwNgKGwRc-00082-00049469-00050215 your neighbors may kill you (Laughs). Possible!
+qRFwNgKGwRc-00083-00050218-00050549 So this possibility
+qRFwNgKGwRc-00084-00050551-00051083 of coming to silence or becoming silent
+qRFwNgKGwRc-00085-00051085-00051494 also means to become still,
+qRFwNgKGwRc-00086-00051498-00052031 to become utterly still.
+qRFwNgKGwRc-00087-00052034-00052991 Stillness can be achieved in so many ways. If you watch a snake,he’s still.
+qRFwNgKGwRc-00088-00052995-00053376 If you watch a tiger which is on a hunt
+qRFwNgKGwRc-00089-00053378-00054173 absolutely still, charged but still
+qRFwNgKGwRc-00090-00054175-00054850 These are all different dimensions of stillness practiced for survival.
+qRFwNgKGwRc-00091-00054853-00055325 If a snake moves too often, he’ll get killed.
+qRFwNgKGwRc-00092-00055328-00055858 If a tiger does not know how to be still, he’ll never get his food.
+qRFwNgKGwRc-00093-00055862-00056345 These are all acts of survival.
+qRFwNgKGwRc-00094-00056347-00057024 A human being for survival he has to be active.
+qRFwNgKGwRc-00095-00057026-00057794 Only when his vision falls upon something which is beyond survival,
+qRFwNgKGwRc-00096-00057797-00058159 then he will think of stillness.
+qRFwNgKGwRc-00097-00058162-00058827 Then he will understand  as he observes himself, as he notices
+qRFwNgKGwRc-00098-00058829-00059351 the complexity of one’s compulsions,
+qRFwNgKGwRc-00099-00059356-00059960 he understands the only way to solve it is to become still.
+qRFwNgKGwRc-00100-00059962-00060536 To train this body, to train this mind
+qRFwNgKGwRc-00101-00060539-00061513 to become still,takes some effort.
+qRFwNgKGwRc-00102-00061515-00062051 If one is simply blissed out, he will become still,
+qRFwNgKGwRc-00103-00062053-00062786 but otherwise one has to work bit by bit, one has to work
+qRFwNgKGwRc-00104-00062788-00063116 to bring this body to absolute stillness.
+qRFwNgKGwRc-00105-00063118-00064078 Most people cannot hold their body in one place for any length of time,  they’re (Gestures)
+qRFwNgKGwRc-00106-00064080-00064766 all the time in movement.
+qRFwNgKGwRc-00107-00064768-00065001 As it is said,
+qRFwNgKGwRc-00108-00065004-00065749 ‘A man is ill only because he does not know how to be still,’
+qRFwNgKGwRc-00109-00065752-00066679 because in stillness you’re in touch with a dimension way beyond creation.
+qRFwNgKGwRc-00110-00066681-00067347 Once one is in touch with that, there is no such thing as illness.
+qRFwNgKGwRc-00111-00067349-00068038 If I sit here and the body rots and falls down, still there is no illness. It’s just
+qRFwNgKGwRc-00112-00068041-00068316 there because…
+qRFwNgKGwRc-00113-00068318-00068714 In fact, when one becomes utterly still,
+qRFwNgKGwRc-00114-00068716-00069163 he may not be able to hold on to the body unless
+qRFwNgKGwRc-00115-00069166-00069681 he learns the tricks, the mechanism of the body.
+qRFwNgKGwRc-00116-00069683-00070202 ‘So, why should I become still? Will I lose my body?’
+qRFwNgKGwRc-00117-00070206-00070536 Anyway you will lose your body.
+qRFwNgKGwRc-00118-00070541-00071145 When the time to lose your body comes, if you want to lose it gracefully,
+qRFwNgKGwRc-00119-00071148-00071617 then you must know at least a little bit of stillness.
+qRFwNgKGwRc-00120-00071625-00072324 If you want to put your body down like you can put this piece of cloth down,
+qRFwNgKGwRc-00121-00072329-00072796 with the same ease, with the same gracefulness,
+qRFwNgKGwRc-00122-00072798-00073375 then you need to have some stillness in you, otherwise
+qRFwNgKGwRc-00123-00073378-00074185 you and your body you have to be torn apart which is an ugly thing to do.
+qRFwNgKGwRc-00124-00074187-00074930 When you’re being torn away from your body it is an ugly and painful thing to do.
+qRFwNgKGwRc-00125-00074932-00075355 Unless you can keep it down,
+qRFwNgKGwRc-00126-00075357-00075704 there’s no grace in life
+qRFwNgKGwRc-00127-00075706-00076083 So if one has to attain to that grace,
+qRFwNgKGwRc-00128-00076086-00076743 at least a certain amount of stillness has to enter your life.
+qRFwNgKGwRc-00129-00076746-00077586 For this to happen, a certain level of silence has to enter your life.
+qRFwNgKGwRc-00130-00077589-00078258 If right now it is not coming, at least practicing a little bit of moun
+qRFwNgKGwRc-00131-00078261-00079244 In an hour if you’re speaking 1573 words right now, how many maa?
+qRFwNgKGwRc-00132-00079246-00079678 How many words do you spit out in a minute ?
+qRFwNgKGwRc-00133-00079681-00080433 1573 you’re actually doing that many? I thought that’s impossible (Laughs).
+qRFwNgKGwRc-00134-00080435-00080854 Okay, I believe you (Laughs).
+qRFwNgKGwRc-00135-00080857-00081448 So whatever number you’re doing, see if you can
+qRFwNgKGwRc-00136-00081451-00082031 articulate the same things that you’re saying with half the number of words.
+qRFwNgKGwRc-00137-00082034-00082507 Suddenly you will become extremely conscious of everything.
+qRFwNgKGwRc-00138-00082510-00083032 Right now a oral diarrhea is happening
+qRFwNgKGwRc-00139-00083034-00083644 If you’re doing 1573, it’s diarrhea, okay (Laughs)
+qRFwNgKGwRc-00140-00083650-00084317 Now if you have to hold back half the words and still say the same things
+qRFwNgKGwRc-00141-00084319-00084811 now you have to be very conscious, otherwise you cannot say it.
+qRFwNgKGwRc-00142-00084813-00085628 Just try this and see. Tomorrow one day, hmm?
+qRFwNgKGwRc-00143-00085631-00086169 In an hour normally what number of words you utter, make it half.
+qRFwNgKGwRc-00144-00086171-00086845 Don’t curtail your activities, still say the same things but with half the words.
+qRFwNgKGwRc-00145-00086848-00087779 How wonderful the whole world will feel, you know (Laughter).
+qRFwNgKGwRc-00146-00088362-00089015 World will be a wonderful place,, I’m telling you, tomorrow, if everybody does this. Hmm?
+qRFwNgKGwRc-00147-00089017-00089698 Many marriages will be saved (Laughter/Applause).