Add transcription for: frames_zips/CGCircuit_RiggingCartoonRealistic_DownloadPirate.com.part3_week05 14 pose reader simple example creation_frames.zip
Browse files
transcriptions/frames_zips/CGCircuit_RiggingCartoonRealistic_DownloadPirate.com.part3_week05 14 pose reader simple example creation_frames_transcription.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"text": " In this video I wanted to create a post reader with you from scratch. So let's start by creating some joints here. Start by creating three joints for an arm for example. Then let's also create a cylinder. cylinder, here we go. Make this a little bit longer, a few more subdivisions, and let's Let's move this to the joints. With the joints what I want to do is I want to change the orientation, the orient joint tool. I want to make it the same as we always have. So I want to twist to be Y. So I'll choose that as my primary axis. And then for secondary axis I choose X. That's going to be my bend. and I want that to go up. So y plus, and then I'm the first one to hit apply. And what we get with this is we have, actually, something went wrong here, I think. Let's try this again. Apply. I think it did work with just on world mode here. That's why it doesn't look like it. But now from local mode, we can see that twisting is actually our y-axis and the x-axis is for bending. We'll select two joints, select two smooth binded, making sure that we are on selected joints still. And now we have that moving so we can bend, use twisting, y and side motion up and down. Let's also go in and create a new shader here, just so you can see that we're twisting a little bit easier. Select that whole column here. Design this new shader, and let's turn this yellow. Okay. Here we go. Let me quickly make all the geo layer for that. I can toggle it more easily. geo there OK. And now I want to show you a way how we can create a post reader in Vanilla Maya. There are a lot of ways how you can, or some people have written plugins to create that for you. But then you always need to carry around a plugin. And if you give it to somebody else, then they might not have to plug in or they have to install the first before it works. So I wanted to show you a way that works in Vanilla Maya without any additional plugins with using in constraints. And I've used it in pass quite successfully. So let's try to recreate that here. So what I wanna do is I wanna create a few locators. So I'll start with one. And let's group this, let's call this post reader, post reader main lock. Let's group that. Let's call this poseReaderGroup. And then we'll take the whole poseReaderGroup and we'll move the group onto this middle joint here. That's where we want to measure our distance or our rotations from. Then we'll duplicate that locator and we'll create a target. Let's call this target, post reader target locator, and we will move that all the way here in the front, maybe like to minus 10 units. And then the third thing that we need is an up object, so we'll take this main one again and move it here to the side, maybe minus 5. This is going to be our up locator. Now what we can do is we can actually think about this. Let's actually move this over here because our set is facing in this direction already. We have x facing in kind of negative one, but we could potentially also move this there, I think. But let's keep it here. And let's have this over here on this side. Let's call this 5. Put this to 5. Let's make this main locator a little bit bigger. So I'll use the local scale for that. Send it to 2. And the up locator, I'll make it a little bit smaller so that we can distinguish between the three locators. It's not necessary, just so we can see a little bit clearer of what's going on. Let's also make these locators color coded so that we can again see it a little bit easier. It's not necessary, of course, but set this to yellow, bright color, maybe. That one here as well. And then this one here too. Here we go. So we've got three locators, and now let's create those AIM constraints. first one. Aim. I want to aim towards this one, under the target here, and this is the main one that I want to aim, put the aim constraint on. So I'll select the target first, then the main, and then I want to just look at this locator here in world space and object space, although here object, local and world are all the same, but I want to be in local space of that locator here. And now I want to look how I want to aim, how I want to have these settings here, if we reset all the settings, I want to aim in minus x, right? So this is positive x. So we want to aim a negative x. So I have to put minus one next here, negative x for the up vector. I want to have it go in this direction. So that would be positive set. So then I just enter set one here and zero in y, zero in x, zero in y, just one in set for a up vector. And I also don't want to use the vector, I want to use the object up and I want to fill in the information here for post reader up. That's our up object, post reader up. Okay, that's what we called it. Now we hit apply and if we did everything correctly, we should see the aim vector created here, or aim constraint created, and zero rotations still, and also on the aim constraint itself, zero offsets. That means we did everything correct to this point. Now if we move this up and down, obviously nothing is happening yet, because we haven't parented this target onto this new joint here, or this middle joint we want to measure the pose from. So let's take the target and group it one more time. Let's call this pose reader target group. And then let's take the target group and parent it under, actually first before we do anything else let's take the whole thing and parent under the parent of where we want to measure the pose from. This is where we want to measure the post from. Let's take the parent, in this case it would be joint one, and take the whole post reader group parented under joint one. Let's take the post reader target and parent it under the joint where we actually want to measure it from. So that would be joint two here. So I'll take the target group parent under joint two. And now if we move this up and down, we should see the target following because it's in the target group which is parented under this joint. we should also see this one aiming towards that one here at the moment. So we can see we're getting a pose read from there. And also if we move sideways here or just bending it, then we can also see that the pose here is measured from that. However, if we now move too far, let's see if I can test that. If we rotate this beyond a certain point, we can see that we are getting popping here. It's popping from 0 to 180 degrees. So to avoid that, what we can do is we can take these two locators here, the post-reader main and the post-reader up locator, and move them backwards. I'm going to move them exactly the same amount that I moved the target here, but in the opposite direction. So the target we had at 10 or minus 10, so I'll put these two and move them in x, 10 units. And by doing that, now if I move this up and down, what we will see is that here we're only getting half of the rotation. So if I put this to 90 degrees or minus 90 degrees, for example, you can see now we're only at minus 45. And if I put this beyond 90, we can see that, again, we're only getting half of the reading here. But the benefit of that is that we're not getting any popping anymore between 0 and 180, as we saw before. When this is too close, I'll replicate it one more time. In a case like that, we would actually see popping here. Maybe that's a bad example. Here it actually does work. But if we go in, I think we saw it with the bend. If we go beyond a certain point, we can see this rotation of 180 degrees here. And by moving, let me under this here one more time, by moving these two guys away, 10 units, now if I apply the band, we can see that we're always staying beyond or below 90 degrees. No matter what we do, even if it comes super, super close here, 180 degrees almost. Even then, on this main guy, we're still seeing that we're beyond 90 so we will never get any flipping or popping of 180 degrees. And now the issue is that we have if we put this to 90 we're only measuring rotation of 45 but the great thing is now we can just use a multiply divide node and multiply it by two. And then we have the full pose here again. So let's actually start adding those and on this joint here I'm going to create some custom attributes for our pose reading. So first add a separator 10 underscores. And then I'll create a pose and pose twist and pose side. And let's hook those already up. So at the moment what we have, we have one for measuring our side motion up and down. So that would be rotate that. And then we also have one for measuring the bending, which would be our rotate Y. And if we're twisting at the moment here, nothing is happening. And that's actually good because we're going to do the twisting in a separate step. So let's first take our main, post reader main locator and the second joint and bring them up. I'll use the high push 8 but you might as well do it in the node editor. Bring both these nodes in here and then I'll connect from my locator onto my joint onto this bend inside. But as I said before, because we're only measuring half of the rotation here with this locator, locator, we have to plug it through a multiply divide node. So let's create that, multiply, then we connect the rotation from the locator, the main locator, into our inputs, x, y, y is set to set, and then we multiply everything with two here. And then the output is now going into these custom attributes. So here for side motion, sorry for bent, this is our bent for the elbow for example, that would be rotate y. So the y channel goes into bent. rotate y goes into bend. And then up and down that would be our side for the arm if you think about it, right? We have bending which would be x, we have twisting and then set would be side. So that would be on the locator, we would have set here. So the output of the multiply set, output set goes into the side. Now we should be able to already measure that. If we're using side here, you can see it's turning minus 180 or towards minus 180 and then it kind of like flips here. Obviously as we go beyond this locator, then it goes to 180 because then it's kind of the same thing as if you go to the other side or it would measure 180. But you can see we can measure the whole range here. And also the other thing is we're never going beyond. So if this is 360, we're still it's kind of the same as if it was minus 23. OK, and then the same thing for the bent here, of course. So if we apply motion to both things, then we can see here we're measuring a bent and a side. So that part is already working or should already be working. Now we just have to figure out how we can do the twisting, because again, if we twist, this locator is not doing anything. The twisting we can do in a second set of locators. So I'll take one of those, maybe this one, because it doesn't have the aim constraint on it yet, but we might as well also duplicate the main, but I'll take the up, duplicate it one more time, and I'll zero out all the translation zero, back to zero, so that is zero at the origin. This is going to be my twist main locator. I'll make this a little bit bigger because it's the main guy, so maybe two. And then here I also want to check how I have to place my aim and my up or my target group, target and up vector locator. So I'll create two more. And the first one is going to be my target, where this should aim towards. So if I want to twist this joint, that would be the target for it. So let's call this twist target locator. Set this maybe to a scale of one. Let's move it to round number here, maybe five. And then this one would be the up locator, twist up. And the twist up, we can scale with 0.5 for local scale and we'll move it backwards here towards the first joint, so maybe five. It doesn't really need to be exactly on that spot, but if I'm looking at that, this is kind of like the direction should be going in anyways. So whether we move it a little bit further out or further in, it doesn't really matter as long as it's kind of like in this range. We might as well set this to 7 here, and this one here too, maybe 7. Again, they also don't need to have the same distance. This might say at 6, or it doesn't really matter. So now we color code those two in a different color, just so we can keep them apart. So color code those maybe blue. And that one is also one of the twists. I'll call it more color coded blue too, and this one here as well. Here we go. So we have our blue locators and our yellow locators from before. And now we create another aim constrained from the twist. Let's hear it the center towards the, um, the twist, um, target locator, and then using the up here as well. So from the target, on to the, and then I select the locator that the aim constraint is going to go on. Aim constraint, and then I have to figure out the settings here just by looking at this locator. So our aim should be set here, set, then we'll plug in set one and x can be zero, set back to zero. zero and x, zero and y and just one for the vector in z. And then the a and the up vector would be x positive here. So I can just type in positive x, positive one, and then zero in z and zero in y obviously. And I also use an up object here, which is this one. So that would be the twist up, post reader, twist up, that's what we called it. Apply. And now we can again double check to see if we did everything correct. If on the twist main we have zero rotations and on its aim constraint we also have zero offset. So that all looks good to me. We just have to do some reparenting. We have to take the twist target and parent under the target group. It's the first thing that we want to do. Now if we rotate this up and down, we can see that both of the targets are actually under the place where we want to measure the rotation from, or the pose. And then we'll also take the two twist locators and we'll group them because what we want to do with this is that would be our pose reader twist group. What we want to do with the pose reader twist group is we want to rotate it the same amount that we're rotating this joint here. So if this join is, for example, minus 41, we want this post reader twist group to be also rotated minus 41. So that these three twist locators here are always kind of in the same orientation. Because then we can really start twisting this guy, and we can see that we should only get one channel here for the twist. So let's try this and connect those up. So we already have our rotation here, but this is only 20. So if we set this to, for example, 90 minus 90, we have to kind of like use the output of what we are getting here to apply that to the post reader twist group. So for example here, we can probably just take it from here from the bend and decide or go one step before, take it from the output of the multiplier. That's what I'm actually going to do. So here we have the output of the multiplier, which we have already go into our two bent and side poses here. And I want to apply the same output to my post reader twist group. So I'll bring that in here as well. Post reader twist group. And I'll connect the outputs, two of them anyways, I'll connect the output of the of the side motion, which was set to the group here, right? So that it goes up and down like that. And then I'll also connect the side motion or the bend motion here. So if this goes back to zero and we apply a band like so. Then we can see we get rotation here in Y. This is basically Y is measuring our twisting and so we want to apply that also to the Y of the of the group here of the twist group. So let's connect up set and Y. how to put that goes to rotate that. And alpha y goes to rotate y. And now we should have that same rotation here applied. So basically, when we now rotate this in any way, we should always be able to twist. And actually, here we're getting some additional rotation. But I think it is close enough. What we're really interested in is the twisting here. So rotate X. This is actually our twist from the twist main. So let's connect rotate X onto our twist post twist here. So let's bring in the post twist main locator and connect it to our join2. And we said it was x, rotate x. Rotate x. We connect that to our post twist. And that should, in theory, be it. So now we can test this out and hopefully it will work. Scale should be 1. Here we go. So now when we're applying twist, we can see that we're getting twisting here, right? 180 to all the way minus 180. Here we go. we're applying for now using Bend, for example. You can see we're seeing Bend here. Let's put this to 90. And if we're on top of that, now applying SideMotion, 90. Now what we can see is here we have actually Side and Twist applied. So it's no longer bending. It's no longer going to the side here, as we can see. There is only twist going on 90 degrees and side. So if we remove the twist now, set this to zero, then we can see we're just left with side. That's maybe not such an impressive example, but if we go down and remove the twist, Let's try to go as close as we can. We're left with just bent. Or if we go up, let's try a couple of more cases here. Maybe let's try to go to the other side. Let's try to rotate this outside. and trying to come up with a case where it actually does work. So here for example we can see on our joint we have some crazy values we have minus 70. actually if we put these two round numbers 90 minus 90 and 90. Then we can see in terms of the pose, this is really only twisted 90 degrees and this is exactly what our pose reader is telling us. So we have zero bend going on, zero side going on, but a 90 degree twisting. If we were to twist that out or over, then we can see now we have bend and twist. If we move this down, then we have only side motion going down and so on and so forth. So this is basically the pose reader that we were looking for that can tell us exactly which pose we're in. Let's try rotating this out here. So here we have minus 180, close to 180 and 90, but essentially this is the same as we would get when we would just use side up. And here we have close to minus 90 inside and close to zero and twist and in bend. So this is actually in fact working to tell us the pose that we're in as opposed to giving us just some rotations that the animator might have put in that specific joint. So I'm aware that this is maybe a little bit of a tricky setup to wrap your head around at first, but it actually does make sense once you think about it and just using these different aims. So what we have here, just to try to recapitulate, is that these all to zero. We created three locators for the main post-reader, so those are the yellow ones. Here we have one, and we moved it back 10 units so that we can actually measure the bend and decide without this locator getting popping. So we never go beyond 90 in terms of our rotations here on this aim constraint. And then we took that and we multiplied it with 2 and that was giving us our bend and our side poses. And then we have another set of locator these blue ones here. For some reason I can't deselect here anymore. Okay, doesn't matter. Those blue guys, those are the locators for the twisting. So So we have the twist target, which is part of the target group here. And the target group we said we parented under this joint we want to measure from. And all the other ones, all the other locators are parented under the post reader group. And then we have the post reader twist group, which contains the two locators, the main one and the target one for the twist that is going to measure the twist. those two, we put into an extra group that has the rotations that is basically coming out from bent and side, so that whatever the bent is doing, whatever the side is doing, that post reader twist group is kind of like replicating that. Okay, at least as close as possible so that we can really measure just when we're want to twist only. And you can see that here too. You know, if I rotate this joint, we can see rotations in all three axes. when in fact we're really only twisting here at this point and the post reader down here is actually telling us the truth about what is What is really going on here really only twisting at this point? And then if I if we take a quick look here at the hyper shade how we hooked it up Then we can see our three locators These are the main ones. And from the main one, actually, it has to be like this. Let me lay this out a little bit better. Let's go like that. So we have our up locator, we have our target, and then these are going, these informations are going into the aim constraint for the post reader main. that information goes under the post reader main locator that is measuring the bend and decide in half. So we have to multiply it by two. And that information goes on to actually two things here. Can see the group here at the moment. Let's bring that in. So it goes onto the joint onto these custom attributes, bend and side, And then it also goes on to the group for the twists that contains the two twist locators main and up. And those two are up here. And they're going into the aim constraint. And then from the aim constraint, it goes into the post reader twist. Sorry, actually, this is wrong. Wrong way around. Should be this way around. Here we go. So these are, it's our twisting system. So that's it. We can create a post reader and in the next video I would like to try that out on the actual character.",
|
| 3 |
+
"segments": [
|
| 4 |
+
{
|
| 5 |
+
"text": " In this video I wanted to create a post reader with you from scratch. So let's start by creating some joints here. Start by creating three joints for an arm for example. Then let's also create a cylinder. cylinder, here we go. Make this a little bit longer, a few more subdivisions, and let's Let's move this to the joints. With the joints what I want to do is I want to change the orientation, the orient joint tool. I want to make it the same as we always have. So I want to twist to be Y. So I'll choose that as my primary axis. And then for secondary axis I choose X. That's going to be my bend. and I want that to go up. So y plus, and then I'm the first one to hit apply. And what we get with this is we have, actually, something went wrong here, I think. Let's try this again. Apply. I think it did work with just on world mode here. That's why it doesn't look like it. But now from local mode, we can see that twisting is actually our y-axis and the x-axis is for bending. We'll select two joints, select two smooth binded, making sure that we are on selected joints still. And now we have that moving so we can bend, use twisting, y and side motion up and down. Let's also go in and create a new shader here, just so you can see that we're twisting a little bit easier. Select that whole column here. Design this new shader, and let's turn this yellow. Okay. Here we go. Let me quickly make all the geo layer for that. I can toggle it more easily. geo there OK. And now I want to show you a way how we can create a post reader in Vanilla Maya. There are a lot of ways how you can, or some people have written plugins to create that for you. But then you always need to carry around a plugin. And if you give it to somebody else, then they might not have to plug in or they have to install the first before it works. So I wanted to show you a way that works in Vanilla Maya without any additional plugins with using in constraints. And I've used it in pass quite successfully. So let's try to recreate that here. So what I wanna do is I wanna create a few locators. So I'll start with one. And let's group this, let's call this post reader, post reader main lock. Let's group that. Let's call this poseReaderGroup. And then we'll take the whole poseReaderGroup and we'll move the group onto this middle joint here. That's where we want to measure our distance or our rotations from. Then we'll duplicate that locator and we'll create a target. Let's call this target, post reader target locator, and we will move that all the way here in the front, maybe like to minus 10 units. And then the third thing that we need is an up object, so we'll take this main one again and move it here to the side, maybe minus 5. This is going to be our up locator. Now what we can do is we can actually think about this. Let's actually move this over here because our set is facing in this direction already. We have x facing in kind of negative one, but we could potentially also move this there, I think. But let's keep it here. And let's have this over here on this side. Let's call this 5. Put this to 5. Let's make this main locator a little bit bigger. So I'll use the local scale for that. Send it to 2. And the up locator, I'll make it a little bit smaller so that we can distinguish between the three locators. It's not necessary, just so we can see a little bit clearer of what's going on. Let's also make these locators color coded so that we can again see it a little bit easier. It's not necessary, of course, but set this to yellow, bright color, maybe. That one here as well. And then this one here too. Here we go. So we've got three locators, and now let's create those AIM constraints. first one. Aim. I want to aim towards this one, under the target here, and this is the main one that I want to aim, put the aim constraint on. So I'll select the target first, then the main, and then I want to just look at this locator here in world space and object space, although here object, local and world are all the same, but I want to be in local space of that locator here. And now I want to look how I want to aim, how I want to have these settings here, if we reset all the settings, I want to aim in minus x, right? So this is positive x. So we want to aim a negative x. So I have to put minus one next here, negative x for the up vector. I want to have it go in this direction. So that would be positive set. So then I just enter set one here and zero in y, zero in x, zero in y, just one in set for a up vector. And I also don't want to use the vector, I want to use the object up and I want to fill in the information here for post reader up. That's our up object, post reader up. Okay, that's what we called it. Now we hit apply and if we did everything correctly, we should see the aim vector created here, or aim constraint created, and zero rotations still, and also on the aim constraint itself, zero offsets. That means we did everything correct to this point. Now if we move this up and down, obviously nothing is happening yet, because we haven't parented this target onto this new joint here, or this middle joint we want to measure the pose from. So let's take the target and group it one more time. Let's call this pose reader target group. And then let's take the target group and parent it under, actually first before we do anything else let's take the whole thing and parent under the parent of where we want to measure the pose from. This is where we want to measure the post from. Let's take the parent, in this case it would be joint one, and take the whole post reader group parented under joint one. Let's take the post reader target and parent it under the joint where we actually want to measure it from. So that would be joint two here. So I'll take the target group parent under joint two. And now if we move this up and down, we should see the target following because it's in the target group which is parented under this joint. we should also see this one aiming towards that one here at the moment. So we can see we're getting a pose read from there. And also if we move sideways here or just bending it, then we can also see that the pose here is measured from that. However, if we now move too far, let's see if I can test that. If we rotate this beyond a certain point, we can see that we are getting popping here. It's popping from 0 to 180 degrees. So to avoid that, what we can do is we can take these two locators here, the post-reader main and the post-reader up locator, and move them backwards. I'm going to move them exactly the same amount that I moved the target here, but in the opposite direction. So the target we had at 10 or minus 10, so I'll put these two and move them in x, 10 units. And by doing that, now if I move this up and down, what we will see is that here we're only getting half of the rotation. So if I put this to 90 degrees or minus 90 degrees, for example, you can see now we're only at minus 45. And if I put this beyond 90, we can see that, again, we're only getting half of the reading here. But the benefit of that is that we're not getting any popping anymore between 0 and 180, as we saw before. When this is too close, I'll replicate it one more time. In a case like that, we would actually see popping here. Maybe that's a bad example. Here it actually does work. But if we go in, I think we saw it with the bend. If we go beyond a certain point, we can see this rotation of 180 degrees here. And by moving, let me under this here one more time, by moving these two guys away, 10 units, now if I apply the band, we can see that we're always staying beyond or below 90 degrees. No matter what we do, even if it comes super, super close here, 180 degrees almost. Even then, on this main guy, we're still seeing that we're beyond 90 so we will never get any flipping or popping of 180 degrees. And now the issue is that we have if we put this to 90 we're only measuring rotation of 45 but the great thing is now we can just use a multiply divide node and multiply it by two. And then we have the full pose here again. So let's actually start adding those and on this joint here I'm going to create some custom attributes for our pose reading. So first add a separator 10 underscores. And then I'll create a pose and pose twist and pose side. And let's hook those already up. So at the moment what we have, we have one for measuring our side motion up and down. So that would be rotate that. And then we also have one for measuring the bending, which would be our rotate Y. And if we're twisting at the moment here, nothing is happening. And that's actually good because we're going to do the twisting in a separate step. So let's first take our main, post reader main locator and the second joint and bring them up. I'll use the high push 8 but you might as well do it in the node editor. Bring both these nodes in here and then I'll connect from my locator onto my joint onto this bend inside. But as I said before, because we're only measuring half of the rotation here with this locator, locator, we have to plug it through a multiply divide node. So let's create that, multiply, then we connect the rotation from the locator, the main locator, into our inputs, x, y, y is set to set, and then we multiply everything with two here. And then the output is now going into these custom attributes. So here for side motion, sorry for bent, this is our bent for the elbow for example, that would be rotate y. So the y channel goes into bent. rotate y goes into bend. And then up and down that would be our side for the arm if you think about it, right? We have bending which would be x, we have twisting and then set would be side. So that would be on the locator, we would have set here. So the output of the multiply set, output set goes into the side. Now we should be able to already measure that. If we're using side here, you can see it's turning minus 180 or towards minus 180 and then it kind of like flips here. Obviously as we go beyond this locator, then it goes to 180 because then it's kind of the same thing as if you go to the other side or it would measure 180. But you can see we can measure the whole range here. And also the other thing is we're never going beyond. So if this is 360, we're still it's kind of the same as if it was minus 23. OK, and then the same thing for the bent here, of course. So if we apply motion to both things, then we can see here we're measuring a bent and a side. So that part is already working or should already be working. Now we just have to figure out how we can do the twisting, because again, if we twist, this locator is not doing anything. The twisting we can do in a second set of locators. So I'll take one of those, maybe this one, because it doesn't have the aim constraint on it yet, but we might as well also duplicate the main, but I'll take the up, duplicate it one more time, and I'll zero out all the translation zero, back to zero, so that is zero at the origin. This is going to be my twist main locator. I'll make this a little bit bigger because it's the main guy, so maybe two. And then here I also want to check how I have to place my aim and my up or my target group, target and up vector locator. So I'll create two more. And the first one is going to be my target, where this should aim towards. So if I want to twist this joint, that would be the target for it. So let's call this twist target locator. Set this maybe to a scale of one. Let's move it to round number here, maybe five. And then this one would be the up locator, twist up. And the twist up, we can scale with 0.5 for local scale and we'll move it backwards here towards the first joint, so maybe five. It doesn't really need to be exactly on that spot, but if I'm looking at that, this is kind of like the direction should be going in anyways. So whether we move it a little bit further out or further in, it doesn't really matter as long as it's kind of like in this range. We might as well set this to 7 here, and this one here too, maybe 7. Again, they also don't need to have the same distance. This might say at 6, or it doesn't really matter. So now we color code those two in a different color, just so we can keep them apart. So color code those maybe blue. And that one is also one of the twists. I'll call it more color coded blue too, and this one here as well. Here we go. So we have our blue locators and our yellow locators from before. And now we create another aim constrained from the twist. Let's hear it the center towards the, um, the twist, um, target locator, and then using the up here as well. So from the target, on to the, and then I select the locator that the aim constraint is going to go on. Aim constraint, and then I have to figure out the settings here just by looking at this locator. So our aim should be set here, set, then we'll plug in set one and x can be zero, set back to zero. zero and x, zero and y and just one for the vector in z. And then the a and the up vector would be x positive here. So I can just type in positive x, positive one, and then zero in z and zero in y obviously. And I also use an up object here, which is this one. So that would be the twist up, post reader, twist up, that's what we called it. Apply. And now we can again double check to see if we did everything correct. If on the twist main we have zero rotations and on its aim constraint we also have zero offset. So that all looks good to me. We just have to do some reparenting. We have to take the twist target and parent under the target group. It's the first thing that we want to do. Now if we rotate this up and down, we can see that both of the targets are actually under the place where we want to measure the rotation from, or the pose. And then we'll also take the two twist locators and we'll group them because what we want to do with this is that would be our pose reader twist group. What we want to do with the pose reader twist group is we want to rotate it the same amount that we're rotating this joint here. So if this join is, for example, minus 41, we want this post reader twist group to be also rotated minus 41. So that these three twist locators here are always kind of in the same orientation. Because then we can really start twisting this guy, and we can see that we should only get one channel here for the twist. So let's try this and connect those up. So we already have our rotation here, but this is only 20. So if we set this to, for example, 90 minus 90, we have to kind of like use the output of what we are getting here to apply that to the post reader twist group. So for example here, we can probably just take it from here from the bend and decide or go one step before, take it from the output of the multiplier. That's what I'm actually going to do. So here we have the output of the multiplier, which we have already go into our two bent and side poses here. And I want to apply the same output to my post reader twist group. So I'll bring that in here as well. Post reader twist group. And I'll connect the outputs, two of them anyways, I'll connect the output of the of the side motion, which was set to the group here, right? So that it goes up and down like that. And then I'll also connect the side motion or the bend motion here. So if this goes back to zero and we apply a band like so. Then we can see we get rotation here in Y. This is basically Y is measuring our twisting and so we want to apply that also to the Y of the of the group here of the twist group. So let's connect up set and Y. how to put that goes to rotate that. And alpha y goes to rotate y. And now we should have that same rotation here applied. So basically, when we now rotate this in any way, we should always be able to twist. And actually, here we're getting some additional rotation. But I think it is close enough. What we're really interested in is the twisting here. So rotate X. This is actually our twist from the twist main. So let's connect rotate X onto our twist post twist here. So let's bring in the post twist main locator and connect it to our join2. And we said it was x, rotate x. Rotate x. We connect that to our post twist. And that should, in theory, be it. So now we can test this out and hopefully it will work. Scale should be 1. Here we go. So now when we're applying twist, we can see that we're getting twisting here, right? 180 to all the way minus 180. Here we go. we're applying for now using Bend, for example. You can see we're seeing Bend here. Let's put this to 90. And if we're on top of that, now applying SideMotion, 90. Now what we can see is here we have actually Side and Twist applied. So it's no longer bending. It's no longer going to the side here, as we can see. There is only twist going on 90 degrees and side. So if we remove the twist now, set this to zero, then we can see we're just left with side. That's maybe not such an impressive example, but if we go down and remove the twist, Let's try to go as close as we can. We're left with just bent. Or if we go up, let's try a couple of more cases here. Maybe let's try to go to the other side. Let's try to rotate this outside. and trying to come up with a case where it actually does work. So here for example we can see on our joint we have some crazy values we have minus 70. actually if we put these two round numbers 90 minus 90 and 90. Then we can see in terms of the pose, this is really only twisted 90 degrees and this is exactly what our pose reader is telling us. So we have zero bend going on, zero side going on, but a 90 degree twisting. If we were to twist that out or over, then we can see now we have bend and twist. If we move this down, then we have only side motion going down and so on and so forth. So this is basically the pose reader that we were looking for that can tell us exactly which pose we're in. Let's try rotating this out here. So here we have minus 180, close to 180 and 90, but essentially this is the same as we would get when we would just use side up. And here we have close to minus 90 inside and close to zero and twist and in bend. So this is actually in fact working to tell us the pose that we're in as opposed to giving us just some rotations that the animator might have put in that specific joint. So I'm aware that this is maybe a little bit of a tricky setup to wrap your head around at first, but it actually does make sense once you think about it and just using these different aims. So what we have here, just to try to recapitulate, is that these all to zero. We created three locators for the main post-reader, so those are the yellow ones. Here we have one, and we moved it back 10 units so that we can actually measure the bend and decide without this locator getting popping. So we never go beyond 90 in terms of our rotations here on this aim constraint. And then we took that and we multiplied it with 2 and that was giving us our bend and our side poses. And then we have another set of locator these blue ones here. For some reason I can't deselect here anymore. Okay, doesn't matter. Those blue guys, those are the locators for the twisting. So So we have the twist target, which is part of the target group here. And the target group we said we parented under this joint we want to measure from. And all the other ones, all the other locators are parented under the post reader group. And then we have the post reader twist group, which contains the two locators, the main one and the target one for the twist that is going to measure the twist. those two, we put into an extra group that has the rotations that is basically coming out from bent and side, so that whatever the bent is doing, whatever the side is doing, that post reader twist group is kind of like replicating that. Okay, at least as close as possible so that we can really measure just when we're want to twist only. And you can see that here too. You know, if I rotate this joint, we can see rotations in all three axes. when in fact we're really only twisting here at this point and the post reader down here is actually telling us the truth about what is What is really going on here really only twisting at this point? And then if I if we take a quick look here at the hyper shade how we hooked it up Then we can see our three locators These are the main ones. And from the main one, actually, it has to be like this. Let me lay this out a little bit better. Let's go like that. So we have our up locator, we have our target, and then these are going, these informations are going into the aim constraint for the post reader main. that information goes under the post reader main locator that is measuring the bend and decide in half. So we have to multiply it by two. And that information goes on to actually two things here. Can see the group here at the moment. Let's bring that in. So it goes onto the joint onto these custom attributes, bend and side, And then it also goes on to the group for the twists that contains the two twist locators main and up. And those two are up here. And they're going into the aim constraint. And then from the aim constraint, it goes into the post reader twist. Sorry, actually, this is wrong. Wrong way around. Should be this way around. Here we go. So these are, it's our twisting system. So that's it. We can create a post reader and in the next video I would like to try that out on the actual character."
|
| 6 |
+
}
|
| 7 |
+
]
|
| 8 |
+
}
|