Add transcription for: frames_zips/CGCircuit_RiggingCartoonRealistic_DownloadPirate.com.part5_week08 06 fore arm twist_frames.zip

transcriptions/frames_zips/CGCircuit_RiggingCartoonRealistic_DownloadPirate.com.part5_week08 06 fore arm twist_frames_transcription.json

@@ -0,0 +1,8 @@
+{
+  "text": " In this video I would like to take another look at the twisting forearm twisting that was also requested by somebody. So let's take a look at that. So forearm twisting. We kind of have that working here already, but I just wanted to show again like how we can deal with that and then maybe also bleed into the bendy limbs as the second part. So let's remove that and show our low-risk geo here. Again, you know, I'm working with the low-risk geo, but if you don't, it's still kind of like the same. Only you have to paint the weights a little bit differently. For me, it's going to be quite easy because I have the low-risk cage here. But the same technique applies even if you don't use the low-risk cage, even if you skin directly your final geometry to those joints here. But if you remember what we did with the arm, we added these partial joints here to deal with the volume, help with dealing with the volume of, can actually show it here better probably, to help dealing with the volume of the arm, right? That we don't lose the volume here when we are rotating this, so we have partial joints in there, three of them. The red ones here, one, two, three, and then this one rotates, you know, half as much and then we also, I also added some translation here. And then this one maybe also rotates a little bit depending on how much this joint is getting rotated. But now, actually here we also have a portal joint for the wrist so that we can say, okay, if the wrist, for for example, bends down, up and down, this one gets 50% of the rotation for that. And then here we have, I call this partial too, but in fact it's actually almost more like a twist joint that helps with the twisters distribution. So on the lower arm, so when we rotate this, for example, 45 degrees, or that's rotate, 90 degrees, here we go. Then what will happen is, depending on how much you're rotating this one here, this joint will rotate half as much. I don't know what this is, 36 here. Let's take a look at the multiplier. For some reason, it's multiplied by minus 0.4. That's true if it did it on purpose or not, but let's set this to 0.5. And then it will rotate a little bit more. So now we're getting half of the rotation that's coming in into the rotation. So this one will rotate 45 degrees if the hand is rotating 90 degrees, right? Or rather, it is 90 degrees. This is 45, but you get the idea. So half of the rotation will be applied to this middle joint or to this twist joint. So that's actually named as twist. Might make it a little bit easier to see. Twist joint. We can actually also, if we wanted to, add a twist joint in here as well. Maybe I'll do it later on. But let's give that twist joint a different color so that we can clearly see it's not really a partial joint. It has a different purpose. This is only for helping with the twist. That's why we only have rotate y connected here. So let's give that a different color. Let's make this bright green. Here we go. That is our twist joint. And I haven't added that before, but maybe it might be a good idea to add another one in here as well. Just because right now what we can see, if I turn on this view, you can see it's a little bit, it's getting a little bit ugly here. So I think what would probably be better if we had a joint, right now what I did is I skinned this joint or I skinned this loop here and that loop to the twist joint. What would probably be better if this loop here was rotating 45 degrees or half as much as the wrist and this one was actually rotating 100% with the wrist. Right? The moment you kind of like get this diagonal lines it would be better if it it was straight and then diagonal here. It will probably help a little bit if we're looking at the final GU. I mean we can't really see it that much so it's not it's not that bad but I think we can improve it a little bit more if that was not rotated so much or more rather. So let's try adding another twist joint in here, let me reset this here. We can even take this twist joint that we already have in there, duplicate it. What we could even do is, actually that's just duplicate it. Duplicate it and multiply it with 2, multiply equal 2, that should put it exactly where the rest is because we originally had put this half way in between. If we're not sure where this needs to go, then we can simply look at this end joint, or should be able to look at this end joint, and we can see it's 20 at zero, and then this number 20.754 and then zero. So we can just take this and copy that onto this twist joint and zero those two out, zero, zero, should exactly be in the same spot here. Okay, this is just a distance from the elbow joint here or this mid joint all the way to the end. But now this joint is not going to rotate at the moment when we're rotating the twist. It's hard to see here, so maybe it might be better if we're adding a little cube in there. Apparent. You can see where everything out. And then we have this little cube box. So now when we're rotating this, we can see nothing happens at the moment with the cube. Let's call this here low arm-low twist 50. And let's call this other one arm-low twist 100%. And here what we're going to do is we're just going to connect the rotation, so rotate x directly to the, should be, rotate y here. Probably we have to multiply with minus one, I imagine. If we set this to 90, then this one here should rotate in this direction. So it would be minus 90. So we actually do need a multiplier in between there. So let's select both of these things, bring them into the connection tool of our choice. Node editor, hyper shade, if we prefer. We'll bring them in here. And then I'll add another multiplier. We have to multiply it with minus one. So we could go from the rotation. believe it was x here, that we had, for the hand. Rotate x into the input, doesn't really matter, the first one. And then output x goes into rotate y for the twisting for the joint, because the joint is a different orientation. And then we're putting this to minus 1, multiplying it with minus 1. And now we should see that here we have 100% of the twisting. But if we bend it up and down, then nothing should happen. Up and down, nothing happens. Left and right, nothing happens. But if we twist, then it follows. And if I add this little box sky here, another one, duplicated to this one, then we should see this rotates half as much. Here we can see. And then when we're bending, as I said before, actually nothing should happen. There might still be some rotations, that is because we're at the moment just using the direct rotation of these joints and they have different orientations plus, probably because they have different rotation orders, but ideally I would probably not use these direct rotations to connect to these twist joints. I would use a post reader, so that we get the real rotation, not the values here, but rather that we can see how much is it twisted, how much is it bent, how much is these different rotations actually. That's weird. Oh, I think it's because this is rotating in more than one axis here, we're probably in a weird space I guess. If we just use one axis at a time, then it's actually doing the right thing. Still, I think it would be better off if there was a post reader in there for this hand control to see how much it actually does rotate. For now let's just work with this because it's simpler. And I don't need to see these boxes necessarily anymore. I mean maybe let's keep them in for now. But what I now want to do is I want to add this new twist joint. I already have this twist joint here as a influence object. Okay. But now I want to add this new one here as well as an influence object. So let's add like that twist joint 100 and the skin and add that as an influence with locked weight 0, all good. And then we can start painting the weights here a little bit differently. So see what it's doing now before we paint the weights. So as I said before, we can see that we're getting kind of like this shearing here happening. So I think with the twist joint, 100% twist joint, it will look a lot better if we're painting this instead to the new twist joint. So if I go our arm star, here is our new twist joint and we can see the moment I've painted the whole thing here white, so these two rows. So now what I'm going to do instead is I'm going to paint this row here to the new twist joint And now we should see it kind of like pop over. Here we go. OK. And then we can maybe let's see what this one is weighted to. It's probably to the hand. Is it a partial hand joint, I guess? So let's add that to that one also for now. We have to figure out what we're going to do with the hand partial joint. Okay. So I think the twisting will look a little bit better now because the distribution is actually happening along the whole line here, right? Now it might be that the up and down might not look so good anymore, because we haven't painted anything to the hand partial joint anymore. So if we bend in this direction, we can see this is 100% or not really rotating with it. But now we can actually take this, let's see here, how are we going to do that? I think this one should still go to the partial joint. But let's get this back. So it should be the hand, our hand, partial, that we want to have this one here weighted to. But what we want to do with the partial joint, we also want to twist it. We can find it. Here we have our twist join. Let's make this a little bit bigger and we can clearly see, okay, these are the twist joins and these are the partial joins. So here with the partial join, what we can see if we rotate this, let's say minus 90, the partial joint is only rotating half, or is it half? It's not really half. have it's 0.75 so let's rotate this 100% let's rotate this to minus 1 and then we get 90 coming in and 90 going out so that means now this should look a little bit better so just in comparison if I undo here this is how it looked like before with 75% of the rotation but if we set 100% of the rotation through then we can see that it's now rotating exactly the same amount or it's twisting rather exactly the same amount as our hand is twisting. Okay and now we should be able to rotate this. Maybe not. This is exactly the reason why a post reader would probably be better because now we have more than one axis going. If we just use this it will actually look okay. We'll get half of the rotation in this way. We'll get half of the rotation for up and down. So that's all okay. For twisting also looks good, more or less. But now we have the problem if we use more than one axis. Actually this still looks okay. But this one here is doing something wrong. And that has to do with the rotation order. So if we set this to 90 we can maybe try to figure out if we can change the rotation order to make this work correctly. On the partial joint let's try the different rotation orders here out and see if one works better. I think maybe it's this one. It has to do with because they were probably facing a little bit a different way. So you can see the hand, set this back. The hand up and down is Z and twisting is X while for this partial joint that we are driving with it. We actually have the twisting BX and up and down BY. Is that the same? No. Is it? Actually, sorry, I was in the wrong space, I think here. Let's go to local mode. Then we can see the twisting here is y, while on the hand it is x. So different rotations that we're connecting here. That's why we need different rotation orders, unfortunately. But let's see if this actually worked. The first one that I tried, which is not the default. Here, so instead of x, y, z, we're using y, z, x. and see what we get. So this works. I mean we have to probably use it in combinations of twisting and that seems to be working fine. And up and down seems to be working fine. Let's do some more comparison or more, try some more combinations. I think that might actually do the trick. I have to test it a little bit and see anything weird is happening. This might do the trick. I still think it would be better to have a post reader in there, but obviously it takes some time to set it up. This might do the trick here. And now in terms of the twisting, if we look at the final geo, and as I said, you know, whether you have that low res in there or not, doesn't really matter. Then if you don't, it just means that you have to work a little bit harder on the painting and getting the fall of right and everything between those joints. But in theory, it's exactly the same. You can use those same joints and everything. And you can see that doesn't seem to be losing any volume, so it actually looks pretty good. The other way around, can also twist it quite a lot. I mean at one point I will still start losing volume and it might pop. If we rotate it far enough, you can see it's actually thinning out. Here we're getting still the candy wrapper effect, but on the other hand, I mean this is twisting it quite a lot, so I don't even know if you have to twist it that far. If we do it a little bit less. You might argue, okay, this is maybe a post that you should be able to hit. You can try it yourself, right? You also have to consider that the elbow is not really rotated at this point. So if we were to go, for example, here in IK mode, and we rotate this in a little, you can see that the elbow is still facing out. So try to do that yourself. to point your elbow outside and try to see how far you can twist your hand. Probably not that far as this character is doing at the moment. So what will happen is you will actually move your elbow down, which will then alleviate, and you can see it's quite slow here because I'm using this smooth preview. or perhaps something like that. And if you feel that this is not enough or you still feel that it's losing too much volume, well you could either use corrective shapes to help with that, or you could potentially use more twist joints. So you could potentially use not only 1 at 50 and 100, you can also use 1 it 75%, 25%, which will also help a little bit. Just like the partial joints help with bending, these twist joints actually help with the twisting and avoiding the candy wrapper effect. So you could add two more twist joints in there and kind of like do a similar thing what we were doing here instead of multiplying with 0.5, you would multiply it with 0.75 minus 0.75 the rotation or twisting that the hand is doing. And here you would multiply with 0.25 or minus 0.25. Obviously then here in the lower sketch example, we would have to then add more resolution here. There we can then actually skin to these partial joints. Or if you're just skinning it directly here, I actually have already resolution. So we could skin those directly to those loops. This is how I would deal with twisting, forearm twisting. And as I said before, we haven't added it yet. But we could potentially do the same thing for the upper arm. Also add a twist joint here at 50%. But then again, the upper arm does not really twist that much. I mean, what does twist is the elbow, the whole arm. And we have that already working. So I don't know if we necessarily need the same thing for the upper arm. It should be enough to just have it under the lower arm.",
+  "segments": [
+    {
+      "text": " In this video I would like to take another look at the twisting forearm twisting that was also requested by somebody. So let's take a look at that. So forearm twisting. We kind of have that working here already, but I just wanted to show again like how we can deal with that and then maybe also bleed into the bendy limbs as the second part. So let's remove that and show our low-risk geo here. Again, you know, I'm working with the low-risk geo, but if you don't, it's still kind of like the same. Only you have to paint the weights a little bit differently. For me, it's going to be quite easy because I have the low-risk cage here. But the same technique applies even if you don't use the low-risk cage, even if you skin directly your final geometry to those joints here. But if you remember what we did with the arm, we added these partial joints here to deal with the volume, help with dealing with the volume of, can actually show it here better probably, to help dealing with the volume of the arm, right? That we don't lose the volume here when we are rotating this, so we have partial joints in there, three of them. The red ones here, one, two, three, and then this one rotates, you know, half as much and then we also, I also added some translation here. And then this one maybe also rotates a little bit depending on how much this joint is getting rotated. But now, actually here we also have a portal joint for the wrist so that we can say, okay, if the wrist, for for example, bends down, up and down, this one gets 50% of the rotation for that. And then here we have, I call this partial too, but in fact it's actually almost more like a twist joint that helps with the twisters distribution. So on the lower arm, so when we rotate this, for example, 45 degrees, or that's rotate, 90 degrees, here we go. Then what will happen is, depending on how much you're rotating this one here, this joint will rotate half as much. I don't know what this is, 36 here. Let's take a look at the multiplier. For some reason, it's multiplied by minus 0.4. That's true if it did it on purpose or not, but let's set this to 0.5. And then it will rotate a little bit more. So now we're getting half of the rotation that's coming in into the rotation. So this one will rotate 45 degrees if the hand is rotating 90 degrees, right? Or rather, it is 90 degrees. This is 45, but you get the idea. So half of the rotation will be applied to this middle joint or to this twist joint. So that's actually named as twist. Might make it a little bit easier to see. Twist joint. We can actually also, if we wanted to, add a twist joint in here as well. Maybe I'll do it later on. But let's give that twist joint a different color so that we can clearly see it's not really a partial joint. It has a different purpose. This is only for helping with the twist. That's why we only have rotate y connected here. So let's give that a different color. Let's make this bright green. Here we go. That is our twist joint. And I haven't added that before, but maybe it might be a good idea to add another one in here as well. Just because right now what we can see, if I turn on this view, you can see it's a little bit, it's getting a little bit ugly here. So I think what would probably be better if we had a joint, right now what I did is I skinned this joint or I skinned this loop here and that loop to the twist joint. What would probably be better if this loop here was rotating 45 degrees or half as much as the wrist and this one was actually rotating 100% with the wrist. Right? The moment you kind of like get this diagonal lines it would be better if it it was straight and then diagonal here. It will probably help a little bit if we're looking at the final GU. I mean we can't really see it that much so it's not it's not that bad but I think we can improve it a little bit more if that was not rotated so much or more rather. So let's try adding another twist joint in here, let me reset this here. We can even take this twist joint that we already have in there, duplicate it. What we could even do is, actually that's just duplicate it. Duplicate it and multiply it with 2, multiply equal 2, that should put it exactly where the rest is because we originally had put this half way in between. If we're not sure where this needs to go, then we can simply look at this end joint, or should be able to look at this end joint, and we can see it's 20 at zero, and then this number 20.754 and then zero. So we can just take this and copy that onto this twist joint and zero those two out, zero, zero, should exactly be in the same spot here. Okay, this is just a distance from the elbow joint here or this mid joint all the way to the end. But now this joint is not going to rotate at the moment when we're rotating the twist. It's hard to see here, so maybe it might be better if we're adding a little cube in there. Apparent. You can see where everything out. And then we have this little cube box. So now when we're rotating this, we can see nothing happens at the moment with the cube. Let's call this here low arm-low twist 50. And let's call this other one arm-low twist 100%. And here what we're going to do is we're just going to connect the rotation, so rotate x directly to the, should be, rotate y here. Probably we have to multiply with minus one, I imagine. If we set this to 90, then this one here should rotate in this direction. So it would be minus 90. So we actually do need a multiplier in between there. So let's select both of these things, bring them into the connection tool of our choice. Node editor, hyper shade, if we prefer. We'll bring them in here. And then I'll add another multiplier. We have to multiply it with minus one. So we could go from the rotation. believe it was x here, that we had, for the hand. Rotate x into the input, doesn't really matter, the first one. And then output x goes into rotate y for the twisting for the joint, because the joint is a different orientation. And then we're putting this to minus 1, multiplying it with minus 1. And now we should see that here we have 100% of the twisting. But if we bend it up and down, then nothing should happen. Up and down, nothing happens. Left and right, nothing happens. But if we twist, then it follows. And if I add this little box sky here, another one, duplicated to this one, then we should see this rotates half as much. Here we can see. And then when we're bending, as I said before, actually nothing should happen. There might still be some rotations, that is because we're at the moment just using the direct rotation of these joints and they have different orientations plus, probably because they have different rotation orders, but ideally I would probably not use these direct rotations to connect to these twist joints. I would use a post reader, so that we get the real rotation, not the values here, but rather that we can see how much is it twisted, how much is it bent, how much is these different rotations actually. That's weird. Oh, I think it's because this is rotating in more than one axis here, we're probably in a weird space I guess. If we just use one axis at a time, then it's actually doing the right thing. Still, I think it would be better off if there was a post reader in there for this hand control to see how much it actually does rotate. For now let's just work with this because it's simpler. And I don't need to see these boxes necessarily anymore. I mean maybe let's keep them in for now. But what I now want to do is I want to add this new twist joint. I already have this twist joint here as a influence object. Okay. But now I want to add this new one here as well as an influence object. So let's add like that twist joint 100 and the skin and add that as an influence with locked weight 0, all good. And then we can start painting the weights here a little bit differently. So see what it's doing now before we paint the weights. So as I said before, we can see that we're getting kind of like this shearing here happening. So I think with the twist joint, 100% twist joint, it will look a lot better if we're painting this instead to the new twist joint. So if I go our arm star, here is our new twist joint and we can see the moment I've painted the whole thing here white, so these two rows. So now what I'm going to do instead is I'm going to paint this row here to the new twist joint And now we should see it kind of like pop over. Here we go. OK. And then we can maybe let's see what this one is weighted to. It's probably to the hand. Is it a partial hand joint, I guess? So let's add that to that one also for now. We have to figure out what we're going to do with the hand partial joint. Okay. So I think the twisting will look a little bit better now because the distribution is actually happening along the whole line here, right? Now it might be that the up and down might not look so good anymore, because we haven't painted anything to the hand partial joint anymore. So if we bend in this direction, we can see this is 100% or not really rotating with it. But now we can actually take this, let's see here, how are we going to do that? I think this one should still go to the partial joint. But let's get this back. So it should be the hand, our hand, partial, that we want to have this one here weighted to. But what we want to do with the partial joint, we also want to twist it. We can find it. Here we have our twist join. Let's make this a little bit bigger and we can clearly see, okay, these are the twist joins and these are the partial joins. So here with the partial join, what we can see if we rotate this, let's say minus 90, the partial joint is only rotating half, or is it half? It's not really half. have it's 0.75 so let's rotate this 100% let's rotate this to minus 1 and then we get 90 coming in and 90 going out so that means now this should look a little bit better so just in comparison if I undo here this is how it looked like before with 75% of the rotation but if we set 100% of the rotation through then we can see that it's now rotating exactly the same amount or it's twisting rather exactly the same amount as our hand is twisting. Okay and now we should be able to rotate this. Maybe not. This is exactly the reason why a post reader would probably be better because now we have more than one axis going. If we just use this it will actually look okay. We'll get half of the rotation in this way. We'll get half of the rotation for up and down. So that's all okay. For twisting also looks good, more or less. But now we have the problem if we use more than one axis. Actually this still looks okay. But this one here is doing something wrong. And that has to do with the rotation order. So if we set this to 90 we can maybe try to figure out if we can change the rotation order to make this work correctly. On the partial joint let's try the different rotation orders here out and see if one works better. I think maybe it's this one. It has to do with because they were probably facing a little bit a different way. So you can see the hand, set this back. The hand up and down is Z and twisting is X while for this partial joint that we are driving with it. We actually have the twisting BX and up and down BY. Is that the same? No. Is it? Actually, sorry, I was in the wrong space, I think here. Let's go to local mode. Then we can see the twisting here is y, while on the hand it is x. So different rotations that we're connecting here. That's why we need different rotation orders, unfortunately. But let's see if this actually worked. The first one that I tried, which is not the default. Here, so instead of x, y, z, we're using y, z, x. and see what we get. So this works. I mean we have to probably use it in combinations of twisting and that seems to be working fine. And up and down seems to be working fine. Let's do some more comparison or more, try some more combinations. I think that might actually do the trick. I have to test it a little bit and see anything weird is happening. This might do the trick. I still think it would be better to have a post reader in there, but obviously it takes some time to set it up. This might do the trick here. And now in terms of the twisting, if we look at the final geo, and as I said, you know, whether you have that low res in there or not, doesn't really matter. Then if you don't, it just means that you have to work a little bit harder on the painting and getting the fall of right and everything between those joints. But in theory, it's exactly the same. You can use those same joints and everything. And you can see that doesn't seem to be losing any volume, so it actually looks pretty good. The other way around, can also twist it quite a lot. I mean at one point I will still start losing volume and it might pop. If we rotate it far enough, you can see it's actually thinning out. Here we're getting still the candy wrapper effect, but on the other hand, I mean this is twisting it quite a lot, so I don't even know if you have to twist it that far. If we do it a little bit less. You might argue, okay, this is maybe a post that you should be able to hit. You can try it yourself, right? You also have to consider that the elbow is not really rotated at this point. So if we were to go, for example, here in IK mode, and we rotate this in a little, you can see that the elbow is still facing out. So try to do that yourself. to point your elbow outside and try to see how far you can twist your hand. Probably not that far as this character is doing at the moment. So what will happen is you will actually move your elbow down, which will then alleviate, and you can see it's quite slow here because I'm using this smooth preview. or perhaps something like that. And if you feel that this is not enough or you still feel that it's losing too much volume, well you could either use corrective shapes to help with that, or you could potentially use more twist joints. So you could potentially use not only 1 at 50 and 100, you can also use 1 it 75%, 25%, which will also help a little bit. Just like the partial joints help with bending, these twist joints actually help with the twisting and avoiding the candy wrapper effect. So you could add two more twist joints in there and kind of like do a similar thing what we were doing here instead of multiplying with 0.5, you would multiply it with 0.75 minus 0.75 the rotation or twisting that the hand is doing. And here you would multiply with 0.25 or minus 0.25. Obviously then here in the lower sketch example, we would have to then add more resolution here. There we can then actually skin to these partial joints. Or if you're just skinning it directly here, I actually have already resolution. So we could skin those directly to those loops. This is how I would deal with twisting, forearm twisting. And as I said before, we haven't added it yet. But we could potentially do the same thing for the upper arm. Also add a twist joint here at 50%. But then again, the upper arm does not really twist that much. I mean, what does twist is the elbow, the whole arm. And we have that already working. So I don't know if we necessarily need the same thing for the upper arm. It should be enough to just have it under the lower arm."
+    }
+  ]
+}