Animate a Professional Looking Jump in Maya. Includes polishing!

1. Introduction and Preview: Jumps are one of the key blocks of human locomotion. Mastering jumps gives you the ability to unlock more complex body mechanics. Seems. Hi, my name is Amedeo Beretta and I am a 3D character animator with a generalist background. I started working in 3D CGI in the year 2003. And alongside production, I kept training the students and professionals ever since. Many of the problems we encountered as animators can usually be easily solved by understanding the basics of locomotion and employing the right workflow. By the end of this course, you will be able to animate human jumps with confidence, interpret reference footage for animation, and adopt the same techniques professionals employ to produce their animations. The course we'll touch on working with video references, identifying and designing the key poses of a jumped, troubleshooting, posing, and understanding weight shifts in locomotion, converting animation between different spaces and setting up priorities first, blinding, The course will run in Maya, but those principles will be applicable to any software. I designed this course for animators who want to reach that extra level of polishing and control in their work, but struggle to achieve and retain convincing posing while delivering the high-quality is blinding and polishing typical of high-end professional level animation. I'd like to thank you very much for watching this video. You can learn more about this course by browsing the extra material and I hope to see you soon on board. 2. Understanding the Story Behind a Jump: The story of a jump is the story of a character or an object that in order to reach a certain position or overcome an obstacle, has to resort to explosive energy and cover more ground that they would normally do during a walk or a run. In order to do so the character we'll have to behave like a spring that will first compress so that the moment they released the compression state, the full extension of their bodies will generate enough momentum for the body to travel the ford and upwards at least to a certain degree. How far we go greatly depends on the push. And in general, you can jump farther away higher if you start from a dynamic motion already like Iran, for instance, instead of standing jump, in the case of our jump, we really want to animate a standing jump up, animating a jump that others around them would be a bit too complicated for these kind of tutorial. We were saying that after being fully extended, we would have generated enough momentum to keep travelling forward and upward without any additional push. And of course, the arms will play a part in it because they will help generate that momentum. Now the push forward will be gradually decelerating and coming to a halt while the AP movement will be counteracted by gravity itself. As the character of a jump is about to land, the legs will extend to make sure that the touched the ground before anything else. And they can act as springs during the landing, so they will just cushion the impact after touching the ground, the crucial work progressively slow down the body to a prevents the body from falling forward and down and be preventing the bones in the articulations to receive excessive stress after the body is fully down and the cushion, as reached its bottom point, the body will tend to naturally stabilized upward again as a spring. Now that we understand the unfolding of the story of a jump, we can think about what's the minimum amount of poses we need to tell the story. And we will do that in the next chapter. 3. Working with Video References for Animation: Now that we have learned the story of a jump, we understand that a jump would need to start from a pose of some sort. Maybe you're standing posed in this case. And in order for the current to start flying, that needs to be a kickoff pose. Which poses should we look for? We should always be looking for the first and end bows and all the extremes say they pose in which the character is at its top height or bottom height. All of these extreme poses are very useful to find because they help us identify the poses we will also do in 3D. And every other pose somehow will be some sort of mix between these key poses, these extremes. So let's go find this Bose's, we're just going to use these jump as some sort of blueprint to understand poses. And if you look at it before the jump even begins, you would essence startup static pose. So put a cross in here just to mark on the timeline where I have my static bows and you can call it a rest pose maybe or a startup books. As I said, they're on their own. The body's supposed to compress like a spring. One thing that I would like to attract your attention to is that the body compresses. You see that the line of action starts the band like a spring, if you think of it, but the arms are moving to the back because they will help, of course, the jump as the body extends forward. Now this is the moment of maximum compression of the body if you think of it and all the R's being at the back. So they will swing forward this way and they will increase the momentum moving up and forward. I would say this is definitely our anticipation, Posner and I will put another cross. Then, as we said, after compressing the spring, we need to fully extend this spring. And in fact, any year day you'll have the full extension of the spring. It will be easier to mark down the pose in which the feet are last in contact with the floor. This bows, I'm going to call a kickoff and I'm going to put a cross in there. And I know it's a kickoff because it's the last pose in which a fruit, he's on the ground. Please notice that there can be no jump. If there is no push know, kickoff, you cannot quite jump if you haven't extend your body and if you're not kicking off the ground, so to speak. So there needs to be one pose in which both feet are about to leave the floor. And this is the one posed, the kick-off. You can also call it a contact each of inquiry because if you play it in reverse, that also acts as a contact. So I might do mistakes as I explain stuff because very off their efforts. This drink is eye contact. She's a bit confusing for people, but in practice, contact and kick-off are more or less the same thing in terms of principles, but let's call it a kickoff. Now as we said, these post generated enough momentum for the character to fly upward and forward. And in fact, you see there it works. Now one thing to notice and we will look into these were more detail later on is that the legs are moving forward and they are trying to, again help us propelled forward, which is very useful. Now you see that the body is at the top of its movement and then it will start going down due to gravity. So I think it would make sense to find the extreme up. Please notice that in these pose, the character has reached its maximum distance from the floor. So these will be. Definitely my up. Once we have reached the, suppose this is when gravity will start to do its job. So the energy that we generated by jumping up will be counterbalanced by gravity pushing down. So this is the moment where the character can no longer go up, but it will start traveling down. Of course, the cocktail will keep travelling forward until the initial inputs forward is depleted. And as mentioned earlier on after the op, we would have a full extension of the feed, which in this case you see it's not really a complete extinction. But in animation it would be nicer if he were. And this is done so that the character can touch the floor there and starts cushioning the lending. Now, this is definitely a contact pose. So another mark in there on the timeline will be really, really useful. Again, after the contact, the body will start going down and the legs job will be to cushion that impact and prevents the body from traveling in an unwanted direction, which is face first into the floor, probably the body will go down and you see that the legs are doing their job. They start to cushion the landing until you see that body reaches the bottom position, which is about here, I guess. So again here this will be down position after the contact. These are exactly the same poses we add in a walk if you think of it, except maybe we didn't quite think in terms of key cost. We walk, now, we are down. You will see that the character was low, the lift up to try and find a final position. So I will just say put a cross in here before the clip ends so that I know that suppose, so now if you think of it, we have identified very few poses that we will need to describe the bulk of the motion, just like we did in my previous tutorial for the walk. If you have watched it, that is, if you haven't watched it, you will find that discounted link in the description. So we have a starting pose, then we have a downpours or dissipation. If you think of it, it's just like a spring. Remember compressing before the jump, then we have the full extension. This is the last frame in which the character we'll have one foot on the floor. And then we have the up. And is the moment where gravity tends to win over the jumps to the jump was lying up until gravity sort of exhausts the jump energy and starts to bring the character down again and that character flying down. We'll have to prepare for lending. And they would do so by expanding their legs. And the moment we touched the ground we have contact pose again and then we compress and there will be the lowest point of the lending, that's the down pose again. And then we would have our follow-through which is opposing which the cart they're tries to find a rest pose. These seven poses would be enough, trust me, to make our initial blocking and then we will work our way into other poses. The next step I suggest we take would be to have a look at the characters that we can sort of understand, which kinda jump we can design, which kind of poses we can design. As I anticipated there, your own, we're not going to trace a reference. We're going to use a reference as a blueprint and you're going to design our own POS is based on the current that we elect for this tutorial that I would really like to try Rocket Girls from London mirror Markov. I think that's the name I the now to pronounce it, I will leave it to you guys be rig itself. Rocket girl is downloadable from gum road, following the link in the description of the video and needs for free, I believe this read is fairly flexible and appealing and the character itself is fairly unique. So I thought it could provide a good playground for us. 4. Thumbnails: Sketching the Poses: Now that we know our character a little bit better, remember that I said that we do not want to trace the poses we have in these reference that we use to evaluate the jump itself. But we want to create our own. The only thing we needed from this reference in this case was an idea which poses we would actually need to animate the character and this jump. So I think we can make our own. And I think a good way to do that would be not to assume that everybody following this, we'll be able to draw. So we'll do that using stick figures so that we do it in the easy way just to show that it's also possible to plan animation without being very good at drawing, sketching out that poses, even just with stick figures, is going to make posing in 3D a lot faster. The workflow for which you start your posing straight into 3D without planning your scene. First is called window shopping by many people because it consists of going into the software and just see what the software froze at you or what you can get out of the rig. The problem is that window shopping is a very time consuming way of working. And very often, especially if you work on TV series, you do not have that luxury and working straight in 3D. And the current state of technology is still not very convenient. In my opinion. I will make my character a bit smaller than this guy. Otherwise, I want that space by pressing T like Tango on your keyboard while on since sketch, you can deem the image reference, and you can also click on this button here, that sort of small grid here. And you can even further, so I can set it to say 75. Now when I press T, I'm switching from full-on colors to seventy-five percent of color. So this way I will be able to draw on a screen and at the same time see the reference down there. So this is my stick man that he would do in here. So now the first thing probably want to think is that maybe I will put one food have been more to the back in there because I do not want to have the ball feet parallel. And then I guess a will have the character stand there with the legs almost straight, but not quite that much straight. So this will be my beautiful character there. This is the pelvis. And then maybe we can start with the character sort of leaning in that direction and looking down and checking what goes on in there and the arm should be ready for action. Maybe you went one here they go. This is the arm and maybe the second arm again in there. So the guy is now ready for action. Now as you see, we just want to plan that emission and design our own poses. This is going to be our starting pose. Then we want to do the anticipation of the post. So this guy would be the anticipation. If you click on the gossip down here on sings sketch, you can decide how many drawings, and I can consider only the sketches. So I'm going to do onion skinning only with the sketches. Even if I go very far away from the drawing, I will still be seeing the drawing very easily there. So as an anticipation there, we said we could possibly have the character bend down like a spring. Let's go down and design this pose. So as a first thing, the food would be still flat on the floor and maybe I can just bend a leg a lot more. The weight will be a lot lower there. And this guy here, I guess I could have the food flat on the ground at the beginning of maybe be lifted the back and then the leg would be super bent down there and going up and this will be our pelvis there. So we go down quite a lot. I would say probably you can have the spine Even a lot further down and the guy would be probably looking down somewhere over there. And then to give it a nice dynamic pose, perhaps we could have the arm to the back ready to swing. Maybe in a feast, Read the swing upward this way and help the body liftoff. The other arm could be other back like this for instance, but much depend on the pose weekend doing 3D pose that we could perhaps have would be to have the almost frontally in here and with the elbow leading up. Maybe that could be at pose. A more academic pose would be with the arm and the bag down here. That's the moth academic pose, but I'm not sure we can have that with a 3D model. So in here we'll probably put the arm here in front of the character. Then, as we said earlier on, we have the stretched right? So the stretch Posey is actually a contact. Now, a nice way to have an idea of weight on the contact is the, put the tip on the floor and just you see bent the food a lot on just after the tip, then more band you have in there and the stronger the sense of push. And then the character will have its leg straight forward there, but the other leg will be probably a bit more Benton there. I could have the lag a bit more towards the last pose, which would make a lot more sense. I think I would like to have some sort of line of action going up that way. And that would be also useful to create son conscious if you have a line of action like this one, let's say this is the head and the next one say thus the opposite line of action. So the opposite arc, what you will get in there is a very clear contrast and is Int2Pos Is happened to be fairly close to one another. So he imagined the jump where we have a line of action like this one where we are about to jump. So it's so zig-zag if you think of it and it's all compressed. And if we have the next pose in which the character is still with most of the feet on the floor. It dips down the legs stretched up in the air, and they land of action in the opposite direction with the opposite arc, you will get a lot stronger feeling of push. You see that? So when we create poses, we always want to play with the line of action. And if we have an arc that goes one way in oppose, maybe on the other, we tried to go the opposite way. And you see that that makes it a lot more readable. C, What happens if I keep the same line of action and it just move up the character, let's say this is the line of action I have in the post before the jump. And let's say that in next books are just translate up this line of action. And you see that this becomes very boring. While if you compare it with the previous line of action, we created the previous line and oxygen is a lot more dynamic. So whenever we make poses, we are thinking of line of oxygen and contrast with the arcs are these line of action is really, really important because that's how good animation comes into being. So again, would be nice to have a curve up there. And the head will be somewhere over here looking down, I suppose, at the destination. Now again, I'm doing this blocking. We stick figures just to demonstrate that you can perform very efficient sketch for your animation even if you don't know how to draw properly. And then of course, I could add the right arm moving up. That's the one that is going to prepare our character up in the air. And then the other arm could be equally up there. Maybe you could have the elbow lead and the hand actually still down there. I will try that. I'm not very happy about this drawing. So maybe they could have the elbow leading up this way and the arm almost, almost about to change the direction and be fully extended. So this would be the left arm. So now if I disable the ghost one, a heavy is the standard Post Dana compression and then the stretch moving up and jumping up. I have decided to keep the SS arm down because they want to start to break the symmetry or the two arms so that they are not parallel to one another. But it's possible that in the end I will have that arm up as well. I just don't want these jump to look a bit afflicted as in, here's how you'd jump. I would like it to look like a jump, but not like a demonstration on how a jump should be working. So afford the contact. One thing that I want to consider is which food leaves the floor first and which foods will lend first. So we'll disable temporarily the ghost. If you hit G on seeing sketch, you can disable enable costing. One thing that tends to happen with a jump is that the food which is the closest to the gop you want to skip over is the one that tends to leave the floor last. So I expect that since the left foot or the character is the one closest to the gap down there. It will also be the last one to leave the floor. So maybe I could make it so that it's also the last one to touch the floor, in which case it will be there, right. Food which is going to touch the floor, which means they're probably the hip will be favoring it while the left leg will still be up somewhere over here as you see, I'm being super rough with my sketch. I just want to go very quickly through it just to show that you can also be rough we are planning. Then in terms of line of action, you see this guy's pretty much with the body already forward, otherwise it will fall down. In a cartoon, you could do a line of action that goes the opposite way. So maybe I could think about just keeping the land of action to the back. And I would imagine that at this stage, the head would be looking down at the destination and the arms would still be up. This arm would be a bit lower, so that what we have in here is non parallel limbs. So now we have the contact we can probably think in terms of up. So we'll hit G on my keyboard to show the onion skin in they're going in the middle of the poses. This will be the rest bows and they oppose. We probably have the right leg a bit higher up just because he was the first one to leave the floor, sir, probably you will be a bit faster while the right leg will probably be somewhat behind. So maybe I can put it down here. I feel like this. And I guess that the line of action will tend to go forward just like the reference because we still want to shift the weight forward. And I guess that you will still be looking down at the destination for any type of dynamic action, the head tends to stick onto the target. Imagine a cheetah running after an antelope very, very quickly. Although they run happens at incredible speeds and there's a lot of motion in the spine of the cheetah. We will find out that the head is more or less stabilize. Otherwise the sheet, I would see a lot of motion blur in there. It will see everything shaky and it will be very difficult to aim. Will tend to keep the head more or less stable on the destination point somewhere over here. So it doesn't matter in which pose I am the character will always be focusing onto the destination point. And in fact, in this animation, we will have a bit more of facial animation is compared to the woke tutorial for instance. So now we have the upward should the RCP doing well, I guess one of the arch will be somewhere over here and the other May 1 be again a bit lower there. These are myths become super long. In the meantime, we're probably sharpen it a tiny bit. So after the contact we need a down and then we need to settle down. And, and for the settled down, I don't want this pose we have here because fairly static if you think of it and we want to use the reference video in this case as a source of poses so that we know which poses we need, but we do not want it to become the pose. We have the coups, thes guy he's showcasing for us the ideal mechanics of annihilating standing jump. In our case, we're going to take some artistic leeway there. And after the contact is food certainly will be flat on the floor. He is lag will be bent forward and the weight will be very low in their hip would be somewhere over here. And I guess the calculator will have a very bent over line of action there and it will have stabilized looking forward, I suppose somewhere over here, there you go. And then there are, maybe we'll stop the character from falling over and the left arm maybe could be in some sort of superhero pose. The other leg will probably be higher up in there and a bit detached from the right food. And these will be the post. Now we need the down, but if you think of it, the down could be with the right foot. Exactly this imposition of a baby. The lag a bit more bent and the wait a bit more down and a bit more to the back and the head somewhere more to the back and a bit lower, somewhere over here. And the arm the right arm will probably be about to touch the floor with the, with the fingers. We will figure it out in 3D, I think. And the left arm will be almost in the position it had their Iran, I hit g to disable the onion skin and I forgot to draw the left leg if you think of it. And now I want to hit g. And if you want to see that poses for a bit longer, you only need to click on this icon in the same sketch UI. And you can say hold frames new sketched since sketch is going to hold the frame until the new sketch. And now as a press Play, you'll see that my drawings there define all the poses a need to have for a job. They're not looking really good, even with some very basic stick figures and a super cheap old wacko. And from the nineties, I think you can still designed these and sketches that tell you which poses you will need in 3D at this stage, the only issue you are going to have is that the reference is going in slow motion. But that's not a big deal because on, since sketches you see you can play at different speeds. So I can maybe speed up the scene considerably. And you see the Now my character works really well already. I can also go a 100%. We've traced capacity. And now what I have in there is my opposes their ego. We could also read time this thing so that the app stays for a bit longer. If that can be done by clicking on this button on the sink sketch platform, you see it says activate drag markers. And now if I want opposed to last a bit longer, say this one in here, I'll need to drag a bit earlier. And the same goes for the stretch bows, the kickoff bows. And now as a press play, you will see that they go that looks a lot better. You will see, will do a lot more of these kinds of staff later on in 3D. There you go. So now we have our poses sketched out and we have an idea what we can do with those. And we've seen sketch, we can also download these poses, this as a grease pencil so we can import them onto Maya if we need them. It is generally speaking, a good idea to understand the story. Have some references had hand and sketch down a few poses, the void, just tracing the reference. So now that we have this poses, we can maybe go finally into our 3D package where they're Meyer blender or whatever, and start blocking down that poses. So on. Since sketch, I will go and click on the download button and then scroll down until I find download my agrees pencil file as I clicked sink sketch will think about it for awhile and then it will lead us download it. There you go. And we will get, you see a ZIP archive down there. And in the next video we are going to see how to import these sketches into Maya and how to set up the rig so that it's easier to operate for our animation. 5. Converting the Sketches to Maya's Grease Pencil Drawings: So now that we have the key poses of our animation and the archive that contains the grease pencil drawings from seeing sketch, we can finally open Maya, I'm going to use my a 20-20, but this tutorial we work also for Maya 2019. So don't worry about that too much. He might 20-20, you would like to set up a project. So we'll go File Project window and I will make a new project which I will call jump underscore t. My k stands for tutorial. It doesn't really matter where you store your projects for as long as they're not stored in a path that contains weird symbols like accents and brackets and so forth. Because in certain conditions, these symbols Campos a bit of a problem. I click on Accept, and if I go and check the project folder, you see I have jumped the And then in here I have the scenes folder. And in a new scene you go panel's perspective and we have the perspective of year will make a new camera which we're going to call our shopped come. So that's the camera we are going to use to animate our shot from right now. It doesn't matter where the camera is pointed at because we have nothing in the scene yet. And into view port, we want to click on the grease pencil tool here on the top left side or the viewport and the grease pencil box shows up. Need to click on this button with the new document and the arrow pointing to the right. That's the input Greece frames tool. We click on it and we go find the zip file that we downloaded from sin sketch. As you click on import Maya, we'll think about it for a bit until eventually it will show you the frames. However, as it seemed when we press play, there's nothing there. And that's because in the original file you see that the frames happened to be at frame 1985 or so. So let's go back into my f for a second and I will make the timeline a lot longer safe, doesn't have 500 frames. And now as I scroll through the timeline, you see that the frames happened to be there. They're just a bit too late, but they are there. Otherwise, normally they don't appear on the timeline as keys by default go figure why although they are there. But if you try to draw an a frame anywhere with a grease pencil, you see that they appeared those little tics in there. So I can just grab them. And with the middle mouse button, I can shift middle mouse button, drag them to the left until they happened to be at the beginning of the timeline. So all we need to do in here is some good old retiring. So will shift, select the keys, middle mouse drag the first key to frame six, and then I will start moving. The other is a bit closer to one another so that this the keyframe. I don't really want to have audience skinning on. So here is the rest pose. Then we have the down which we're going to keep for about, I don't know this kid for about four or five frames there. We want to have my timeline even shortly thereafter o, then we jump up. And this one we can keep for shorter and we can keep the kickoff pose for a shorter time and just have the oppose last a bit longer and then the landing would be a lot faster. So you see right now and not even checking which timing I have in there and just eyeballing it. So now we want to check and you see that that starts to look like something. Maybe I can keep the anticipation a tiny bit longer just to make the kick-off a bit more explosive. And maybe then the kickoff can stay for a bit of a shorter time. So we'll make the transition between kickoff and up a lot faster than you. And then maybe the transition between contact and down and follow through would be a bit faster. So contacted down a lot faster and then follow through. I press play. Okay, feel like the AP is a bit too long, so we'll shorten the up by a couple of frames. And that starts to look like something. You see that now it might seem to you that a couple of frames here and there make no difference. But if you think of it, if we oppose that lasts for about four frames a, we make it a couple of frames shorter. That means we are making this post last 50% shorter. And that's a big difference in timing if you think of it. So a timing that to the layman looks negligible like a couple of frames to the animator might well be the difference between an emission that works and one that doesn't. And now that we have degrees pencil sketches in our view port in Maya position, the viewport anyway we like because after rural At the end of the day, it doesn't matter where the camera is because the sketches are always relative to the camera, which is an advantage in this case and the disadvantaged in others in this case, that's pretty good to have. I am going to save this scene before moving on to the next video. So safe CNRS, again, it will be redirected automatically to my scenes folder, and I will call this thing capital and m stands for animation, jump, that's the jump version 001 RES, pencil, safe. That's it. So now that we have our sketches there in the next video, we're going to set up the rig so that it's easier to animate. 6. Setting Up the Rig: So I've got my sketches in Maya and the rigor archive. The rigor archive contains the Maya file for the rig itself in a couple of textures. Now the Maya file is to go into the scenes folder, while the two texture files need to go into the source images folder. And now there is a fourth file in there, the post library, which you can use. We have studio library, we are not going to use it in this case. So I will close this one and I'm ready to work between the two videos. I accidentally lost one of the sketches. I could re-export the sketches from sin sketch, but this is a good opportunity for us to see one of the many bugs or the grease pencil tool in Maya, very often you get this problem where the screen becomes a single color there, there used to be a sketch. There is no longer there. That tends to happen when my loses the file or you do something wrong with the file. I might have done something wrong in there and we will never know, but we know for a fact that now we don't see it anymore strain here. You shift them on the frame, which is the problematic frame. You right-click and you delete it. And now you see you're back to having your sketches in there. So that's better than nothing. As I said, it's not a big problem if I lost the sketches because in the sketch review, I still have those. So that shouldn't be a big deal for me. And I can always re-export them if I really need them. But right now, I think the only pose that I'm actually missing is the starting pose. Everything else seems to be there. And you see I'm only missing the first pose, which is not bad, I can handle it. So I've got these sketches in Maya and I've got the rigor archive in here. The rigor archive contains the bluster file, which is the garden of the girl, and it contains the rig itself. Right now we can just obstruct the rig itself. So I'm going to leave the rig into the scenes folder. And maybe even better, I'm going to make a rig folder into the CSS folder and leave the rig in there. I do not need the archive anymore for now. So the rig to ever regain a scene in my ego file, go into the rig folder and go rocket girl reference. If you want to learn more about the importance of referencing break before you start the animation, please have a look at my YouTube tutorial, which I will link to in the description before starting the animation. As usual, I want to make sure to set their EEG in such a way that is convenient for me to animate with it. And also, if you remember, when we were finding the poses, we've noticed how they handled the person is jumping here. That's the stay stabilized and always look in the same direction. So it would be good if the rig were behaving the same way to begin with. So back into Maya, I'm going to make sure they're the head of the character is going to behave the same way. So I want to have to count their animated. So if I rotate the chest doesn't troll for instance, you see that the head is following the chest. That doesn't work for us because in our animation they had behavior is different. In fact, also wanted to make sure that I always select controls and not the character by mistake. So up here in this election mosque of mine, if you don't see it, you have to click on these R0 with the triangle. You see I want to have everything disabled apart from the curves system there. So now I want to make sure again that the head is relating independently. It you see it's not the case. So I select the head control. You will find that in most rigs. You can select which rotation space the neck is at the hearing to. In practice, this could be called the global parent, local space, rotation space, space, which things which they all mean the same thing. The control is asking you, who should they follow? And in here is following the neck. But if Excited to either know torso, Now let's see what's the behavior there. You see that as I rotate the chest, they had stays global as I rotate the spine, the head follows the spine. So that's quite handy. I think I can set this thing to world so that even if I rotate the center of gravity down here, which is called body control on this rig, you see the head stays global. Let's have a look at the eyes. And as I rotate the head around, the eyes are following. You see that they always look in front of the head. We could set the i controller not to follow the head so that as they head moves around space, the eyes, they in the same position. You see that the character will keep staring at the eyes. But in practice for a scene like this one, I think it's easier to animate if we have the highest follow just because the character will be traveling across a fairly big distance. If you think of it, if we keep the ice-free, then now we will have to animate the control across that distance due. And that would be another control we have to keep track of which we don't really want to do. The other thing I want to consider is that the character is going to travel across space, as I just said. So it will be good to think about how we're going to move this character through space in Maya, I think I could potentially grab the body control center gravity, also called sometimes, and just move it around and make sure that the feet are following as well and so forth. And maybe the other solution could be to move the character around for space with the master so that for as long as the character stays on the spot and doesn't really jump, we are going to use the body control to create poses and move the character around at the moment the character kicks off, we're going to use the master to control the position. So I think this will be my approach and I suspect that we'll have to use some scripts or some workflows to convert animation from the master control to the lock of controls at the certain point, possibly when I will have to polish this scene. Also, if you think of it, you see how the hands are sort of moving with the body at all times. It would be great if in Maya, when I moved there route around, the hands were also moving around. And you see they do. But you see that as the roots are rotating, their hands are also rotating. That's not very good because in our reference you see that the hands are sort of independent. They actually start to jump before the body even can start to jump. So they produce a lot of the thrust that we have for the jump. If we are to move the body around and the hands are moving with the body, want one, then we will end up counter and emitting the hands quite a lot, which can be time-consuming. So let's see if there is a better way to set up these hands on other finger will want to consider for the hands is that I will need pretty precise placement of the hands throughout the animation and quite quick changes of positions of the arms throughout the animation. So while FK is very handy because it gives us nice arcs for free very easily. I think that in terms of precision of movement for this particular shot, I would really like to use IK, And I was showing in my previous tutorial about walk cycles how Muslim rigs have a control which is near the wrist and it controls usually lets you switch between IK or F k. In this case, you see this little key control there, some sort of a circle with a straight line popping out. You see that indeed it contains the IK to FK switch, which I'm going to enable. So now the hands should be IK is you don't know it. Ik stands for inverse kinematics and it essentially means that you can affect a whole chain of joins just by grabbing the last one, that's IK, as opposed to FK, wish to position or non for instance, you will have to rotate each single control individually, which has some advantages and disadvantages in general, although many years ago I mainly used F k. Now I grew accustomed to choose whichever I think it's best for a shot. There is no right or wrong in here. The best solution is always to have a tool that lets you convert IK animation to have gay animation. If you work in fairly advanced productions, you will have such tools. Otherwise you will be pretty much on your own unless you find some tools online that is. So in here I'm going to set the hands to IK And I think I will want them to follow the roots so we'll translate the root around, but you see they don't follow the route. That's going to be very inconvenient because as the character jumps around, it means I will have to manually position the hands after positioning the body. And that's going to be a lot of work for nothing if you think of it, if you watch our sketches, you see that although it's true that the hands of the beginning lead the action at the same time as the body travels across space, they move with the main body and then they need to become somehow independent on the end when the right hand has to touch the floor and stay there. So I think that for the most part, we would want to have a behavior that follows the body more or less. And this is possible if you think of it by using space switching so we can select both controls in there. And then under parents space, we said that parents space to body. And now as they move around the main body, you see that both hands are following. We will need to find the solution for when the right hand touches the ground. You see that as we prepared the rig, we evaluate our sketches. You see how important they are because every decision we take on the risk depends on those few lines in there. That's why even if they are just stick figures, they are already providing us with an immense amount of support. We have to start from somewhere as far as the feet are concerned for No, I think it's okay if they are independent from the rest of the body. I think also that the moment the character will kick off the ground, I will use the master to move the whole character around. As I said just earlier on, as far as the spine is concerned, I think it's a good idea to keep it as it is. In summary, eggs you can pick between i k and f k. And here the spine is essentially by default or Nike's bind that can stretch and squash. I think that will work for us in general. I find it easier to animate IK spines because I can decide where the chest is a bit quicker than with the FK spine. I think that's about as far as the rig is concerned. I think I can just select all controls and make sure they are 0 doubt before I start the animation. And now I would probably need a floor in their right for this process of the floor and they'll want to see the grease pencil tool, so we'll just hide it. And I will go create polygons cube and I will make a cube as a Florida, you will make sure maybe that pivot is up there and I will make sure the floor is actually on the floor, maybe a fake go to the side view. This is going to be easier. And then I'm going to duplicate this and move it to the side. Maybe to make this scene a bit more appealing, we can't lie a checkerboard to the cube so that they don't look laying gray. So let's maybe put a checkerboard in there. I hit six, they go if you want the checkerboard to repeat itself, more, just increase the repeat movies. And there you go. That's my scenario. Now let's go establish the position of our camera and go bundles perspective shotgun. And in there I want to position the camera and maybe want to show the resolution gate by clicking on this button with the green dot and the white square in there they go. That's a bit better because now we can see what we are indeed framing. And I'm going to maybe position the camera a bit low and in such a way as to frame the gap in there so that we can better appreciate the extent of the jump the girls doing and something like this, I guess we'll do I want to maybe show a bit of the floor so that I can see how to position the feed. These I'm doing more for the tutorial then for the beauty of the scene, let's say now I am going to lock the camera by clicking on the camera with the lock icon on the top left side are the viewport. Now it doesn't matter anymore if I want to zoom in or out or pan, the camera would be locked. And other the weak and the scenes are set up and we have some decent planning in place. In the next video, we're going to see pose to prioritize, to make the best of the interpolation system of the 3D software, we want to make sure that we don't spend time doing poses that maybe the software can help us build, for instance. So I will see you later. 7. Weight, Balance, and Line of Action for Animation: We are good to start with the very first pose of the action portrayed in our reference, which is the rest bows because that will inform the anticipation and the subsequence kickoff. If you go back into Maya, you will realize that if we drag a selection in the view port, we're going to select accidentally geometry and controls alive, but he's not good for animation. So we want to make sure that our selection mosque up here, and if you don't see it, you click on this bar and you expand it. We want to make sure that this election mosque is set only to select curves. So if we click on the triangle on its left and set old objects to o and then set only the nerves curves do on now that we drug some items into view port, we can only so that control, if you select the master, it will appear as if you're selecting the geometry, but really, that's not the case. These geometry is only highlighted as a result of hierarchy. We're going to perform these selection Moscow operation pretty often every time we open up the scene, for instance. So it would be good to store it into our shelves. That's easy. You can just click on the triangle next to this election Mosque and then select save to shave from the dropdown menu as you do so your preference will be storing to the shelf. And now you see if I select any other item up there, when I click on my go back to animation, as a matter of fact, I already worked that way. And you see that I can swap between the main selection mosque and the animation selection mosque pretty easily. So for the initial post, we could use the pose. We haven't the reference, but we are not here to trace the reference, but we are here to interpret it. So I just don't want this jump to become an athletic feats. I don't want to make the jump we see in the reference video in here. So we'll start with our recipes in there, which she's a bit more like the character looking down and checking out what's going on in the, in the gap and then he's going to prefer to jump. So let's go and set up the body. I think I will move the character a lot closer to the cliff. May be moved left fruit, or be more forward and the right food at tiny bit backward. And to decide something like these, I suppose. And you see right now the weight is right on top of the legs. When do we know that the way it works? Well, for instance, in here you see that the weight seems to be a bit off if he's like the concrete is about default, The same thing would happen if we move the weight backward or forward. How do we know if a character is falling down or if they are balanced? So this tower, for instance, which is very famous in Italy and around the world, the reason why it did not fall down to the floor, if you think of it, is because its center of gravity is still falling within the base of the tower. So for as long as the center of gravity falls within the weight of the tower, the tower want fold. That's more or less how it works. And that's the same way it works for people and older people you see in here with some rare exceptions, they had one thing in common. Their center of gravity is falling way off their feet, their base, if you think of it, and they are all falling down because of that reason. Now the center of gravity is an imaginary line that starts at the center of the shoulder blades and goes down vertically. Whenever these imaginary vertical line falls out of the base of your character, that means your character is falling. Maya will let you do the pose anyway because my dad doesn't know that the character is falling and plus, why should he care? But you have to be careful about that because otherwise your characters will look off balance even if they're not supposed to be off bonds. One of the exceptions in this picture is this guy. This guy is not falling down because if you think of it, is center of gravity falls straight vertically down on the base of these feet, so he cannot really fall. There are some notable exceptions to this rule. Neo from the matrix for instance, is another bill exception except 0s using wires, so ofcourse, he's not falling down. You see center of balance is a bit of his, you incubate. Another exception is a run. So if you're doing a run like this one aneurysm there balanced, we'll be off. But the whole trick in here is that you're pushing the body forward and then your legs will move under the center of balance before you fall down. That's why you don't fall down in these cases and see this in therapy policies really off in this case. But again, the language is forward is going to reach forward before the weight will fall to the floor. Keeping these in mind, let's go back to our scene. When you create your opposes in 3D animation, it is very tempting to post stuff from the camera. So you would do a pose like this one for instance. And you would think, well, it doesn't quite feel like the character is falling down. She's standing, right? And then you go into 3D and your character is all over the place. May be a static pose like these 1B spotted in camera. But as you will start spline in your animation, the motion will feel less convincing and it will be very hard for you to understand why it's less convincing. I know because I spent years animating in camera and spending hours while it's blinding, trying to figure out why things within work and I couldn't quite pinpoint why things would not work until one day I realized that I had to look at them in 3D. And if they work in here, they would be most likely working in camera as well. So I will position my character in 3D instead. I will make sure that the center of gravity is well-placed, may be on the left fruit like so, maybe a bit more forward and a tiny bit lower, something like this. Suppose I will also re-center the right food a tiny bit. In body mechanics, one of the basic rules is can't quite leave the food if there is weight on it. So for instance, say the weight was on the right food. I couldn't quite lifted because the character would just fall down, you see? So I would first need to move the weight onto the left leg and then I could leave the right one. You should think this way all the time, then animation would be a lot easier for you. Now let's have a look at the arms. And for the arms, they just for now I just want to move them down. And one thing you notice we've IK arms is that as you translate them around, you see that the race doesn't rotate and it's very easy to feel like the wrist is broken there. So one of the main things that you always have to keep in mind as you place an IKR is to make sure that the race is oriented in such a way that makes sense with the rest of the forearm. You don't want to have situations like this one or this one. There was even an old trailer from Avatar in which you could clearly see NIK breached the breaking. It was then corrected in the second version of the trailer. So we'll pose the left arm as well. For now, I will just make a post that is rather natural. I want my character to be arched a tiny bit for words that we can tell the story of someone who is checking the GOP and the landing spot, either or both. So maybe it will select the free controls of the spine, rotate the spine forward and give me that arched line of action that I have in the sketches. So you see that I will have to reposition the hands in there, which usually tends to indicate that it's a good idea to start first positioning the body. And once the body is ready and only then you start fixing the leaves that are extending from the body like legs and arms and head for instance. So going body first is usually a good way to think about it. And I will see that the character is a lot more involved in the jump itself. There's a tiny bit of line of action there that she's almost about default if you think of it. Maybe it's a good idea to just grab the root and move it back ever so slightly, the more you bend forward and the more your center of gravity will fall outside the food and the more you will fall down if the heat release. So if you want your character to lean forward quite a lot, it's a good idea to move the character to the back so that the character one fall because we want that famous center of gravity to fall onto one of the feet. So now she's a lot more involved in the jump itself. By the way, if you press alt one on your keyboard, you can swap visibilities for nerves in Maya 2020. In Maya 2019, This does not work the same way, so that's a tiny bit better. However, one thing that you tend to notice in people is that people have the weight on one leg, then the other leg tends to be bent. This does not seem to be the case for our character. And why is that the case when people have their weight on a single leg, what happens is that the leg becomes straight in order to make sure that the weight is supported by just the bones, not the muscles. This way the body has to do less work to counteract gravity coming from above as a result of it, the side of the body which does not have the support of a leg, tends to go down. And what you perceive in the resulting pose is that one side of the hips is going to be higher than another. And the legs, which is under the lowest side of the hips, we'll be bent. So let's go back into the scene and apply what we have just learnt to our world. Go grab their hip controller in here and I will rotate it higher on the language is holding the weight. You see that automatically we get that bent that we were looking for and the pose looks a lot more natural now, similarly, we're going to grab the chest and counter animated. If the chest were positioned, say this way, and the character will be clearly falling down on the right-hand side of the character itself. Very often, if the left hip is up, the left chest will be done and there will be you see opposing the whole time. And if you think of it now we start to get to oppose that starts to make a lot more sense and it certainly looks a lot more aesthetically pleasing here. You could also grab the hips and rotate the right hip a bit more to the bank because they right leg is a bit more to the bag that will let you move the food a tiny bit further back. Very often as you duties they lag, holding the weight will become stretched so you want to grab the root? So I want to go down and make sure that the leg starts bending a little bit, but still feel straight for the most part, I will maybe grab the first control of the spine so that it's a bit more in line with the hip. And then I will grab the second to create that Benton there so that it looks a bit more interesting. I do not need to rotate the hip forward that much. Nagel, when people lean over a cliff, they tend to risk to fall down. So usually they take some precautions. In general, there is 1 fourth, which is closer to the cliff, which is the one holding almost all of your weight. If unlike this gentleman, you have no want to support you, you will find that you will often have mostly a weight on the leg standing forward this way that leg is in control and should anything happen, it can bring you back to safety. And you do these even if you are rich and famous. So let's implement in Maya what we've just learned. So Becky, Maya, I want to apply what I've just learned and I will go in the perspective and maybe I will move off the root and the food backward. I want to use the world axis. And to show this menu, you have to hold down W and left mouse button together. And now you can pick whichever axis you want. I want to have the world axes on there. Because if I use object, you see that the root will also translate up and the food is moving in another direction. It's moving according to its own object axis, which is not what we need. So I will make sure the axis is set to weld and it will move both the root and the right food back a tiny bit. Maybe I will lower the root tiny bit more. And now it looks a bit, she's a bit more careful about not falling down and then I will position the hands. Are there already was a bit premature to position the arms because really the body wasn't really ready. So I will position the arms now a bit better just in case there's wind or Augusta and the wind pushes the character forward. The character wants to have the arms have big more to the bank. I want to go and check these posts from the camera point of view. And you see that in here the capillary is pretty much in a profile pose which is not very appealing in general. But if you watch any animated movie, you will find that must be shots we, as an audience perceive as profiles shots are indeed three quarters shots. So let's try and make the same in here. So we use the backslash glass right mouse button in Maya to zoom in onto the locked camera. This does not affect the camera is like a 2D zoom on a canvas and the middle mouse button with backslash again to just pan around. This way you see even if a character is away from the camera, I can have a good look at the pose. And if I hit backslash again, you see I can toggle between the bookmarked view and a regional one. Let's maybe grab their head controller and rotated a bit more towards us so that the character will look a bit more interesting because we can finally see above the Ising there. This way they pose will be a lot more appealing already we can see both eyes and we know that humans have two eyes. Usually better if we show both of them in a shot from the camera point of view, we miss the second arm, which is a bit sad because it makes for an incomplete silhouette. I would be really happier and I think the post will be more appealing if I could see also the left arm imagined that this will be the first time we see these factors. So we want to introduce the character as a whole so that the audience, should there be a next shot, would be able to recognize it. So I will go and grab the left arm in there and I will just move it a bit more forward. And this will be just the seed in the camera. No other reason really maybe can twist the upper body a tiny bit this way, counterclockwise from this view, that means that the right arm will also have to move you that further. I can help their right arm moved further back by moving the clavicle as well. And these would let me translate their left arm a tiny bit more forward. Maybe I need to have the pole vectors a bit further out in there. Probably will look a bit more natural. For now. The hands, they look a bit like puppet hands. And that's because we didn't impose the fingers yet. 8. Understanding Eyelines: If I look at the silhouette now you see that we manage through successfully have both arms visible in their media, can put a bit more of a gap between the arm and the body and the character steel has a balanced bows, however, doesn't look like she's looking somewhere specific. So I would want to go back to the 3d view. We want to make sure that the character is looking at the right position in space. So we want the character to be looking down at the destination point. So that's the first thing we do. And you will find that as an attribute named eyes tweak in here, this guy, and we enable it. And now with the translation, you can position the eyes. For now it's very difficult to tweak the individualized without knowing how they look like in camera. So my suggestion would be to go tear off the camera view ports so that we can have a zooming onto the face of the girl and have that perspective visible. And they usually, you get an idea on how to treat the eyes. So if you look at the right eye, it feels a lot like the right eye is looking to camera right or the left eye is looking at bit to camera left and higher. So in these cases we just grab one of the eyes a bit closer to camera, right? And now above the eyes will be a bit more consistent. Obviously, facial expression will help. But if you think of it, these Eileen is a lot better than what we had earlier on where the character was looking into nothingness. The trick in here really is not to leave much white on camera, right? If we see that an Irish person or a cat there has no white of the eye in, on one side, we start to assume that that person is looking in that direction. And now that we have the whole character posed in opposed that he's somewhat natural and believable. In the next video, we're going to see how to create an appealing pose in the fingers. See you later then. 9. Creating Appealing Hand Poses: Now that we have a basic pose for the body, I think it's time for us to fix the fingers which right now look a bit like Forks. Daily seal. So you have watched my previous tutorials about human walks. You will know what I mean by that. I might use the main controllers a year, these little keys coming out of the knuckles, this controller is usually lead us progressively curl the fingers. However, one thing I like about currently the fingers using main controllers is that each phalanx curled by the same amount. That's not very natural. A more natural approach would be to have the second phalanx, this one a year rotate a bit more than the others. So I will grab the main controllers and then it will enable finger tweaks. These attributes may have a different name on different break spot. More often than not, you will find rig supporting controls for each individual filings. So as I said, I want to rotate the second phalanx of each finger a bit more. You go, and this is already should give me a bit of a nicer pose for the fingers. I want the post to be Relativity relaxed. So maybe it will grab the top controller for each file lungs. Maybe you rotate them a little bit on the inside. But in general, Luthien and dance to happening in fingers is that they index and the middle finger, they tend to be a bit more rotated outward and the other fingers, so we'll keep them outside. And also from the side view and year very often what happens is that middle and ring fingers Dan to stand a bit closer to one another while the index sticks out a little bit. The pinky is the fingers that usually is curl the most. So maybe I could kill this thing even more. And other thing you want to check is the model. If the palm model is a bit too flat, you can grab a cap controller, which usually you find in this area of the hand. And you can, and you can cut the palm versus likely to make it bit curve here that you go. And these looks a bit more natural as a curve of the palm. And other thing that tends to happen in real life is that fingers dent to aim at the center of the palm. So you see I know it's not anatomically correct it to his fingers this way. But in real life they tend to be a bit twisted and aim towards a unique point in there. That's going to be my pose for the phalanxes. Maybe I can stick out. They index a little bit more and banned the second falling a little bit more. I want the post to look relatively relaxed with not too much for the thumb. I can maybe you're the last phalanx of a little bit in the same thing goes for the second phalanx, that will be it. Now in general, the thumbnail tends to be oriented almost parallel to the palm so I can rotate this vague a little bit. You have to be careful with the thumb because it's very common to find rakes when the thumb is oriented this way, but that's not really anatomically correct. Usually the thumbnail is not lined up with the top of the pile but rather with the side of the pond. And once you see the anatomically correct pose in a drawing or in a picture, it's very hard to forget it. So there you go. That's going to be my poses a bit more natural than it was earlier on. To be honest, the modelling year was quite good already to begin with because they hand was looking a bit less regular than other models. And if we go back to check the camera for a second, but we can maybe grab the wrist controller and rotate it until we get to a more appealing silhouette. Makes us understand that there is a bunch of fingers down here and a thumb sticking out. This will be a bit more interesting as a silhouette, you should check the siloed as it was earlier on. You see that that doesn't look like there's a thrombin there. But the moment we rotate the handler tiny bit, we get the thumb in there and to see where it looks, a lot more appealing value. And now that we have the fingers set up in the next video, we're going to see how to mirror that both so that we don't have to manually do the finger pose on the other hand as well. 10. Mirroring Poses with Studio Library: Now that we have posed one set of fingers, it would be great if we could seamlessly mirror that posed to the other side. But you're in for our surprise because if I grab all the fingers controls, enable the finger tweets and then grab all the fingers controls on the left hand side, set a key, copy the key, grab their fingers controls on the right-hand side and right-click, paste, paste. The post does not look the same. And that's because my requires the selection or there to be the same left and right when you copy and paste keys. So if I started by selecting the first phalanx of the thumb and then the second and the third. I would need to do the same in here on the right-hand side and so forth for all the 15 phalanxes that we have, which let's face it, it's not a very smart way of working. And to be honest, I would expect a modern workflow to automate pose mirroring. And this is when studio library comes into the picture. The tool lets you create mirror setups, safe poses, apply mirrored poses, Save Animation, import and export on innovation. It is really, really useful. I would like to use Studio library to mirror the poses because they think it's a powerful and flexible tool. However, with this particular rig, the mirroring doesn't work quite as well as it does with other rigs. And that's probably because of the way there equals built, right nines to your packet is a suite of tools that is very useful for animations. It performs similar tasks to those of studio library, although I think the library is more character animation oriented to know more about how to install it, please check the link in the description. Once you have installed it, in order to launch it, you click on the icon and a new menu shows up in Maya, you have the animation toolkit. This is the Toolkit we will use, the mirror poses. It's not as flexible as the one from studio library, but it has some very useful tools. Let's see how to mirror these poses. First of all, we want to import a mirror table I prepared for this tutorial and you will find it included in the project files. To import the mirror table, you need to go open the red nine menu and find mirror setup. You click on it and the mirror setup Windows shows up. Now you need to select all the controls of the rake and then import the mirror table that I provided. Select all the controls. All you need to know is first of all, understand what's the naming convention of this rig? And if you select any control, you will find that the control carries the name of the namespace. Namespace is usually the name of your file, followed by the column. And then we have the name of the control itself followed by a token which is underscore capital CTRL. You will find that whichever control you select that token, capital CTRL is always going to be there. So I think we can safely assume that all the animation controls and these rig and with these token. Hence that means that in order to select all the animation controls, we could just ask Maya to select whichever node ends we've underscore capital CTRL. Let's see how to do it. Up here at the top right hand side of the top bar, Maya, you will find this select by name tool very often, however, when you open my a, you won't be seeing gate, but you will be seeing the absolute transform tool. And sometimes you won't even be seen that like in this case, for instance, if you don't see either of those, you click on this line with the triangle that will expand. If you still don't see the Select By name, you click on the crosshair in here and you go under Select by name. Note that the naming convention includes the namespace, the column, and the controls. So maybe we could tell Maya select whatever and whatever is translated into star in a software usually is followed by a colon symbol, and then has whatever name followed by underscore capital control. Now by specifying this syntax, Maya will cycle through whatever has a namespace because it will find the colon in there. So we'll cycle through every single item in the scene which has a namespace in there. And then it will cycle through which ever note ends with the symbol underscore, followed by the capital letters CTRL, that will copy and paste this line into this leg by name, hit Enter. And you see that if I look into the outliner, I selected all the controls. Not bad. So now to load the mirror tables, I can click on Load mirror configs. Go find the mirror tables I shared with you, which you will recognize by the mirror map extension name. I click on it and load it. Now that the mirror tables are open, we can see if this works for real mirror oppose. You need to often read nine and emission toolkit and click on mirror pose I, you see pose works. It's successfully mirrored. Now let's undo. To mirror the fingers you need to select all fingers controls. We do not need the risk controls and we do not need the leg for sure. And then you can symmetrized the bows from left to right if you think of eight, so oppose left, right. And now we have the same pose on the other side. And whenever we need to mirror poses now, it's as easy as selecting twin controls and hit mirror pose or symmetrized pose, really, really useful. In fact, now that we are decomposing the fingers in the next video, we'll go back to the face and create an appealing pose. 11. Designing Appealing Facial Expressions: A body mechanics shot like this one, facial expressions are not so important. However, a little bit of emotions in the phase goes a long way in any short, in this particular shot, I think it would make sense to have the character relatively interested in either the gap or the landing spot. In general, the neutral pose of a rig is never very appealing, but you can really make it very appealing just by bringing down the eyebrows ever so slightly. And that can already create an interesting beause if you think of it. If you compare these pose with the standard one, you see that we have a whole different emotion going on. Now one thing that could happen when you pose your eyebrows, for instance, is that you create a pose like this one. Now this is not a very good Posey my opinion, because there is no clear communication between the two eyebrows and the two eyebrows, they seem to be disconnected from one another. Having the intersection of both eyebrows talk to one another usually lead to more appealing poses. So that will create a more continuous line of action where has just a few weeks ago the expression look a bit broken. It is really, really important to think in terms of line of action, even when doing eyebrows expirations, if not spatially, because if you have a bypass, the audience will focus a lot on the eye and on the eyebrow mask. You really want these expressions to be well-curated. Usually the eye shape is trying to adapt to the direction in which the iris is looking at. So he's, the iras is pointing towards the inner side of the nose. The eyelid will tend to be a bit more open towards the inner side of the nose as well. However, you see that in this case, I don't seem to be able to really achieve that pose. Let's see if maybe when I grab the controller for the IEA, him, this guy in here, let's see if I have something to tweak the eyelids and you see in there I have eyelids tweak so I can maybe set it to on you see that now I can not only close the eyelids a little bit more, but I can also open up the eyelid where the iris is. Do you don't want to go past, say, the free quarters of the pupil, otherwise, your character will look a bit intoxicated. So usually what I do is I bring down the eyelid right above the pupa, either stop at the pupil or you go higher up in depends on this situation. And a very brief Google search makes you understand how the eyelids very rarely wants to cut through the pupil, because the pupil is the window through which light reaches the retina. You don't want to put something in front of it, otherwise, you won't be able to see stuff. So you see that most of the islands you see around stopped rightward the pupil starts. The relationship between eyelids and pupil is also dependent on the amount of light in the scene. So in general, the brighter the scene that tiny near the pupils also, please notice that in most pictures, you don't really see any white above or underneath the eyes, you will start seeing white. Nifty. I for instance, if you're having a top-down camera for which the character is looking up. You see that now you start seeing the white under the eyes. So maybe let's try and create a bit more of a serious boats with this factor. And I will close the lower eyelid a tiny bit. I will remember to open it up a bit more in the direction of where the irises and I will do the same with the left eye. I will close both eyelids a tiny bit more and I will open them up a tiny bit more where the iris is. You see that now she looks a lot more interested. I would make sure to set a cookie and every single control for post-war soon now that character looks a bit more interested. Maybe we can make it look a bit more determined. Maybe a little bit of a smile who would help build these, the termination. When you do facial expressions for a shot, it's a good idea to start maybe in 3D and then again, checking camera if everything works the way you want, and if the character is very small on screen, then it becomes difficult to evaluate the facial expression. So we can hold down backslash and right mouse button and zoom in and pan around until we see the face really well in there. We can actually tear off these camera. That we can keep the camera in one view and perspective in another. Normally, you would have two monitors when you animate, you can either manipulate the controls from the camera or from the perspective, whichever is faster. Expression is sort of possible, but it looks a bit neutral. Let's push it a little bit more. So for instance, if we close the eyelids a tiny bit more, we will give more of a sense of focus. If we crunch down the eyebrows a lot more, we will give more of a sense of determination. Remember to keep the inner sides of the eyebrows talking together. You see these expression is OK, but then we could push it a little bit more. You can do these streaks on frame one as you follow along, but I will do them on the second frame, just so that later on we can cross compare the poses. If we close the eyelids a bit more, we give more of a sense of focus and intent. So we'll just close down the eyelids a little bit more in there. I will maybe it makes sure that the corner that is facing camera becomes a little bit sharper. If you're squinting with the lower eyelids, then it makes sense that the cheekbones will raise. The problem with this rate perhaps is that as we move up the cheekbones, we are changing the silhouette of the helmet around so maybe we can move them out as well as they move up. And this will also simulate a little bit of squashed in the face if you think of it. And then if you feel you're losing volume, you can use the PAF control. They go to add a little bit of roundness to the face. If you want the eyebrows to look a little bit more determined, you can crunch them down quite a lot. The important thing to remember is key, talking to one another. So even this one, the inner side of the left eyebrow will follow along, will go down and we'll build a continuous line there. And maybe I can break these straight line and just rotate the center control a tiny bit just to give us a bit more of a sense of continuity with the line of action. Now, another finger could do is I could maybe increase this Merck. There you go quite lot. Please notice that I'm using object axes so that I am using the controls in the way hopefully that they were intended to be used. And now I can compare the pose I had just a few minutes ago with a new one. If you think of it just a few seconds ago, we believe that that fissure the pose was sort of okay, it wasn't too bad. Now let's go back to it and see how it looks like. You see the little tweaks we've just performed, made these original posts were made look like a neutral pose. We created the line of action between the two eyebrows and the right eyebrow. The character is pushing down quite a lot more if compared to the left one. So maybe you can push even more. So maybe we can consider that that may be pushing down the eyelid a lot more. So maybe can close these Eileen and squint on this side a tiny bit more if you think of it, I don't want to cut the lower pupil too much, but I think I could push a little bit, just a little bit. There you go. And maybe I could try and close the eye even more here close to the corner we're building oppose in which the right-hand side of the character, which is camera left in this case, is crunching the phase is compressing the face a lot more than the left-hand side. So on camera left we have a compressed eyebrow, a closed eye. So the facial expression opens towards the landing point and you could create an even stronger impression if you were able to rotate the chin sideways. You see in here however, we can only translate it. However, maybe we can grab these mouth controlled and rotate it a little bit. You see just to help open the face in the direction of the gays. And again, if we go back and check, can you see how neutral this verse pose was if compared with a little bit of a push. And now you see that we have a nice open facial expression towards the landing point. You can even open it a little bit by rotating the nose just ever-so-slightly You see how straight the nose was pointing down in this direction, but with a tiny bit of rotation, we make the nose take part in this line of action. It is not my intention to spend too much time on these facial expression, but I just wanted to give you the very basics of facial expressions that you start to see things in a bit of a different way, hopefully. And this kind of reasoning happens all the time in feature films when there is an acting shot. So I will do a very brief comparison between these bows, the first post we created and the neutral pose. You see that the neutral pose looked really scared and pretty neutral if you think of it. The first pose we created looked a little bit more self-assured. And the tweaks we made really brought the Emotion to the next level. She's now leading towards the badass area. Of course we could push even more, but our aim is to do the jump. Let's have one last look from the camera view. This is the current pose we have, which seems to be pretty confident. This is the pose we devise a regionally, which comes across as a bit perplexed. And this is the pose we had by default in the rig, which comes across midway through between spaced out and outright frozen. And now that we have some emotion in there, in the next video, we are going to see which pose to prioritize for blocking. 12. The Anticipation Pose: Now that we have some emotions in there, we have to have a look at which poses to make next. So we'll make sure to go back to the shotgun. I have a initial rest pose, then I have an anticipation on the way down, a kickoff on the way up and oppose up in the air. And a contact pose or the bottom of the arc at decompose followed by a follow-through pose. So which pose wouldn't it make sense to do next? One way of approaching this task would be to go straight from the initial pose to the last boat. Dissuade interpolation of Maya will take care of the movement for a space. However, they issue with that would be that as we go down into the dissipation through the following kickoff pose, we would find ourselves living forward through space. I think it will start from the anticipation. Also, I think I need to decide how to move the character for space because I have a few options in there. One would be to move the route forward and animate the feed everything in world space. You see that posing a character away from the master this way becomes a bit more convoluted as a process because you have to move a lot of controls around individually. So I think that the moment the character will move away from the ground, I will use the master to move the character for space. They hope is that by doing so, I could position the character anywhere in space and I could easily pose it, you see, and opposes are easily consistent. I don't have to move around a bunch of controllers and trying to make them work. So let's do the anticipation. And I've put an anticipation at around 16 a will of course have a look at my references before even starting. So you see that the anticipation has the legs bent, the body forward, and the hands at the back. Think of it in terms of line of action. You can imagine that these line of action is already straight to begin with. And then for the anticipation, the main sense of line of action is this one. When I look at this transition between the rest opposed and the anticipation pose, the first thing that comes to my mind is a spring. And in fact, if you think of it, that's exactly what's going on in there. We'd go from a rest state to a compressed fade so that we can explode straight again with these kickoff pose. So you see we are acting exactly like a spring and I will go back into Maya and I will try to apply the same thing. I will first lowered the root quite a lot in there. You see that as we compress the buttocks move to the back in here we wanted to do the same. I will want them to be a bit more to the back and will rotate the body forward and maybe we'll want the legs to be a bit more open. Otherwise, she will look a bit awkward. If you do something like days, it doesn't feel like she's about to jump it. So I will move the legs a tiny bit more out using the pole vectors. Either any one Manumission to look exactly like the references. I want the references to inform my animation, but I do not want to keep exactly the same pose. However, I do want to preserve the nature of those poses and the nature is rest pose, for the most part, straight line of action. Anticipation predicates exactly line of action very compressed like a spring kickoff straight line of action. Instead of having it straight, we will prefer maybe a bit of an arc, but overall straight. So back in my, I maybe want to move this arm up. We know that when the arm moves up, the Clinical should be helping, right? It should raise as well. But if the clinical moves these much to the back, then probably that means that the chest also needs to move to the bag. I could move the arm back like in the reference, but as I said, there will be a bit too much. It will look very athletic. But as in professional sports men, we went there to look athletic, but also quite cool. So let's start by fixing that, that right arm and then we will find a solution for the left one. If I moved their right arm to the back. Right arm will be able to swing up during the kickoff bows to give a stronger sense of preparation. I will maybe move the food autonomy more to the back and maybe a bit more out. And maybe I will use the heeled pivot to open the tip a tiny bit so that it will feel a bit more like the girl is preparing for the jump. And then if they're right leg goes backward a little bit, maybe I can rotate forward the left hip a little bit, and maybe I will bring up the shoulders quite a lot more, both left and right. And I would really like to make sure that the place The girl is looking at the beginning of the shot, which is where my camera is right now, more or less, where you want to make sure she keeps looking at that place. You see that right now at the beginning she looks a bit higher and at the end she looks a bit lower. She loses defocus character animation. Whenever you have a character we've faced, it's really important to keep that focus on the action which is at hand. So in here I will make sure that it actually lowers the weight of the body. She will keep on looking in the same direction. This way, she will feel committed to the action. You see that I am starting to create that line of action that have identified in the references. The left arm needs a bit of love. As I said, I don't want it to be at the back. The first reason being deposed, we'll look a bit to symmetric and this will becomes just a regular athletic jump. The other thing is that if I go back to the camera I have by bringing the arm behind the body, you see that there's a silhouette and if you press seven, you will see the silhouette, your viewport. You see that in terms of cigarette, we lose the arm and now this becomes a lot less clear. You always have to remember about how you introduce these conductor to the audience. Remember the honest will have only a handful of frames to understand the story, the cochlear, the context, everything. So we introduce this character as a character with two arms and a main body and two legs. Why we go down to the anticipation, these character becomes a calculator with a main body, two legs, and only one arm. And anticipation is opposed. We will keep for quite a long time, which take quite a while in anticipation pose before we can actually spring away from it. Here we do not want anticipation to look like a different character, so we will have to do something about this silhouette. Remember slit is one of the principles of animation. He said, We'll try and move the hands forward. I know it seems counter-intuitive, but I want to see if I stand a chance there. Will that give me a better pose? Maybe if I could show the elbow will look a bit better. So maybe it can raise the elbow and recognize shoulder there you see? So I want to raise the shoulder and every rig we'll behave different from one another. So you always have to check which kind of poses at Rijk can-do and a big part of pre-production of a movie. In fact, the strength, understand whether your reagan, your character can support a certain poses. And for sure if they can't support the poses that you will have in the movie. Maybe if I bring the arm a tiny bit more to the back in there and to decide and I keep the elbow relatively high up and they're always remember to adjust the conical, whichever motion you Ramdas from the top view. These Phil seems to make some sense. It still seems like disarm overall is not that far forward, so it could still be used to propel the body. And from the camera point of view, these could work as well. It has a continuous line of action between the two arms. We don't want maybe the two angles to look exactly the same, this one and this one man, maybe this could work. However, one of the issues we have is that the hands are in a very relaxed position right now. Maybe we want to underline the fact that this woman is committing energy to the jump. We want that big more tension, the fingers as well. And the first thing I can think of is a feast bows just do underlying the tension there. So I will grab all of the fingers controls, maybe not this global controls wave the line protruding out of the circle. And if you're a date them on a single axis, they go, You will have automatically a feast for the most part. You will still have to go there and tweak the post quite a lot to make sure there are no gaps between the fingers because a feast won't have any news when you see maybe some modelling issues, some areas that could be improved in the model, but nice through late you have the modeling, you have to work with it. I will cap the palm even more in there, and I will make sure that the first phalanx of the index sticks out a lot more. The second one will closing a lot more. If there are a few intersections is not a big deal for as long as we don't see them. And then remember in a face to the first, second phalanx of the families, tend to cover the second to third phalanx of the index. If there is a bit of an intersection, we do not care too much about that. And then you go, we have our posing there and it looks definitely like a fist now will be good to have the same posed on the opposite side. So well, that should be easy. I select all controls without the W3C controls, I go to the opposite side and select the same controls in there. And then I click on the red line to go into the retina and menu animation toolkit. And here I want the post to be symmetrized from right to left. So I will click on symmetry pose right to left. If I go back to the camera pose, you will see that just by changing the fingers, these new posts makes a lot more sense. It's not grade, could be improved of course, but it makes a lot more sense. So maybe could grab the chest and the second control of the spine and just roll them forward quite a lot more. And you see that by doing that, look at how much more involved the character becomes the jump here. She's happy to jump. She's confidence she's about to jump with these new pose. She's a lot more invested in it. If we compress even more, we give a lot more of an idea of a spring. All of a sudden that pose just works a lot better. These arm, of course, has been left behind. So you see how much more dedicated to the jump. She's now. There you go. And you see that even now the left-hand works a lot better for that very reason. So sometimes you'll find yourself trying to find a solution to unarmed post. But really what you didn't consider was the body. You should always start and fixing the body first. But I find it very difficult to nail the posts at the very beginning. But you see that in my sketch, I had this boss is just that I did my sketch with stick figures to prove that you can sketch Bose's even if you don't know how to Joe. And in practice, the more drawings you will make of you opposes, the faster the posing process will be in 3D. But maybe I could drop the head controller and tilt the head ever-so-slightly. You see if I tilt the head, the poses a lot more interesting. This is cool. This is a nice books, but this one shows a different kind of interests towards the action. It's a lot more interested in the action. This one is interesting. This one is more interested and your job isn't animator is to make things look more appealing. There is an intersection between the head and the color here. If you look at it from the camera, that's not there. So we don't care too much. There's also an intersection between the arm, these sorts of bracelet. But in practice we can't do anything about that. So we will just make sure it doesn't show too much. It would be nice if they NCL didn't look like a solid sticker juncture coming up vertically and then breaking in one direction. It would be nicer to have a more continuous line of action flowing towards the need. So to do that, usually some rigs allow you to change the rotational near the heel on this one, if you grab the food control, you have something called a lean control. And you see you can lean forward and look at the difference it makes. It's very tiny by you see how stiffened straight these food woes. And now that we lean a bit more in that direction, it looks a lot more dynamic. If you notice that your tip goes into the floor, you can maybe use the toe. We go to count their unmade the position of the dough. You want to look like it's going into the floor to honesty was going to floor a bit earlier as well. We'll have to live with this intersection. You've got to try and do the poses you need to have to deliver good animation. I want to make you think it where the weight will be. We moved the right foot backward. So maybe we could shift the weight onto that fruit. As I do that, it would be a good idea to also shift the equal vectors because otherwise they will be left behind. There you go. That means that the pelvis will be a bit higher on the right leg. In here, I could maybe build a stronger impression of a spring. If I move the chest that bit more forward, I can lower the chest a tiny bit, some maybe, maybe, maybe I will cheat with a translation a tiny bit and move the chest a bit forward and downs that is closer to the thigh. Obviously there will mean fixing this guy. You see that now we have a better sense of compression. And if I check the line of action from the references as zigzagging line of action in here, that's exactly what we get. Our line of action is clearly zigzagging, so it's so broken up and tens so that by the time we extent, we will get a lot more contrast Out of the pose C that the action is already clear, even if we just make 9s. So if we make the character follow the lines, I expect animation would look very functional. 13. Posing the Face in the Anticipation: A way to improve this pose even further, it will be to exaggerate the facial expressions. So the fingers are tens in a face to the body's tense, all compressed, ready to jump. It would be great if even a phase was participating in it. And the simplest way to make the face participate in it is to crunch the eyebrows a lot more as much as we can without breaking the face. In this situations, I find it useful to have the wireframe on top of the geometry so that I can tell when I'm pushing the rig a bit too much so I can lower the I rose quite a lot, making squint more with the cheekbones. You see I can push them up quite a lot. Maybe I can see what happens if I puff them and if I move them outside a little bit and you go and move them outside, I should give us a sense of compression of squash. And then with the chamber became press their lips and then move the face controller a bit higher. I want to go into 3D view. If we are squashing in this direction, if you're going up this way, then the mouth will extend sideways. Whenever you compress on one side, you stretch on the other. So I am going to compress the lips quite a lot. And that means that will justify my extension sideways. Maybe I could even play a trick in here and move up the nose ever so slightly and make it a tiny bit wider so that he will feel like it's squashing as well. Usually, advanced breaks have the ability to squash the head. We have an attribute called head tweak. Let's see what it does. But you see doesn't quite squash is just scaling vertically, but it doesn't really scale anything sideways. So it's not very useful for a squash and stretch, but it could be useful should we want to give the phase a stronger curvature, stronger curvy or line of action there. And that will be done by grabbing the controls and just rotating them this way. And you see that a little bit on rotation goes a long way there, we can make a comparison. So this is with the curvature of those additional controls. And this is without, with very little rotation, recreate a much more flexible face there. It would be good to have a key set on these controls, even at one, so that we keep it 0 doubts because they weren't visible earlier in Iran. And see if we can spot the change in the phase between frame 116. You see that in framing 16, the face is compressing. A lot more will try to see how much it can push it without breaking it. So we'll close the eyelids a lot more. Maybe I can play with the pupil size and just make it a tiny bit smaller. That will give me a bit of leeway and I can close the eyes a lot more for as long as we don't spot this trick, we can play that trick. This starts to become a bit too closed. Remember, I want the eyebrows to talk to one another to create a nicer line of action all the time. Usually you have a control the moves, the eyebrow shape your made up and down so that it's easier to move it around. In this case, we have to move each single control individually, which is not as efficient as one would like it, but it's still good enough. So you see that we are making the face work a lot towards the aim. And if you check the beginning, at the beginning of suppression was rather self-assured, but it wasn't that involved. Now that we are into the anticipation where we are using all the energies we have to jump forward. You see that the phase takes part in this action and it makes for a more compelling post, again, to further prove my point, I will show you the anticipation with and without the new facial expression. You see that the dissipation with the original facial expression for the most part works at the moment we crunch. Phase is taking part in the action as a last touch to the pose, we could maybe Raul day food a tiny bit higher. They're just not to have it entirely on the floor in there so that the weight is mainly on the bowl of the food. And maybe we could push it even more to the back to make sure we get a bit more of a difference between the two legs. And I wonder if I could push down the route even more. Remember one of the principles of animation is exaggeration. It should exaggerate opposed as much as you can without of course, breaking the character. There you go. And now, if we examine the different line of action between the two poses, you will see that we have a much more relaxed line of action for the starting post. And then as we go into the anticipation, the line of action becomes a lot more springy there. See that? And then I have action alone is telling us what's going to happen in the next post and in the next video we're going to design the lending pose. So hopefully we will benefit from my as interpolation when designing the other poses. If you have enjoyed this course so far and you think you found it useful, please consider leaving a review that would help me because it would confirm that people find my work useful. And these will give me more motivation to create more material in the future. Thank you very much. 14. The Contact Pose: Now that we have that anticipation, I think it's time for us to go for the next poses. Let's have a look at what we should go for animation with progress front anticipation to the kickoff pose, to the oppose. And then finally, the contact posed it down pose and a follow through of some sort. So those are the minimum requirements for this animation to work. Now if I start it by doing the kickoff post these bows here, what I will find out is that I am making the character jump in one direction, but I'm sort of guessing the direction in which the character should go. In the case of this animation, the guess wouldn't be that wild at the end of the day, we sort of know what the character is going. But if these were more complex animation, I might work for a bunch of minutes or hours on this pose just to find out that I'm directing the character wrong direction. So maybe a better approach would be to go and do the lenny pose, which I sometimes accidentally call a kickoff, just because it looks like a kickoff in reverse if you think of it, anyhow, if I do that contact posed land impose first, then they oppose and they kick off. Both will be at least partial interpolated by Maya automatically. And then I could use what Maya gives me as a way to build the poses. So at least I would know that the interpolation is working in the right direction. I wouldn't have to make up from scratch old opposes. So in my opinion, starting from the contact pose the land imposed in here, it's a good idea because it will help reduce the amount of work I do in the previous three poses. I could even go as far as doing the downpours and then reverse engineer the rest. But in practice, I feel like the action actually starts at the kickoff and ends at the contact. And then I could treat the down as a consequence of the contract. So let's do the contract. I will go grout the master in 3D. I will go find on which frame the contact is in my sketches at about 38. So we'll move the master at 38 forward before doing anything else, I think it would make sense to select or controls and set a key at 16 where I have the anticipation and at one where have the default post. He said that suppose I have in here is very similar to the down pose we designed in our rudimentary sketches. See this guy in here. So maybe I could reuse this pose and see if that works at any rate, I want to keep it. So sulci at 44 where the post has been drawn, I will go back and we'll first of all make sure that I lend a bit closer to the edge of the cliff and that deposition of the master stays the same afterwards. So I can just delete this key and that's going to be my starting position. Then going back to check the contact, you see we wanted a fairly straight line of action, slightly arched to the bag, probably hands up in the air, one food touching the floor with the heel and the other leg a little bit delayed on the way out. At any rate, I think it would be unwise to start doing the cardiac post without having at least check the references. And you see that line of action I drew in my sketches isn't really the same we have in the video. However, I said I wanted the line of action to be a bit more extremes, so I'm fine with it. The gentleman in the video references is lending with his heels. However, if you have ever watched anybody doing parkour, you would have noticed that the landing happens always on the bows of the feed. And that's because if you land on the ball and defeat the bow, the feed can absorb part of the shock and then by rotating in time, it transfers the wave to the heels, which in turn will transfer the weight you see to the knees. So the whole lending operation, it's a technique device to make sure that the shock is absorbed by soft tissues like muscles instead of hard joints. So if we were to land with the heels, we would be putting a lot of stress on the heels. Now these gentleman is landing with the heels because he's landing on sand. So the lending zone is super soft, in which case you can actually afford to land on the heels, but usually you wouldn't want to land on the heels. Now the thing you will find while looking at bare cool footage is that the landing always happens on two feet at the same time. Because by landing on two different platforms, you're effectively halving the amount of weight that each platform mistaking. So say these bodies 65 kilos for instance, what will happen in there is that each food will take over about 30 to 35 kilos. So it will be a lot easier for these limbs to absorb the shock. If you were to lend on a single leg like in this sketch, you would have the entire weight of your body on a single leg, particularly so on the heels, so on a heart joint. So the risks of injury are a lot higher than they would be if you were in London, the Parker way. However, as I discussed, I do not want this animation to look like an athletic demonstration, so I will delay one of the fetal little bit we tried to lend with the bowl of the food. So let's take care of the posts. I will move up the root until I realized that the right leg is extended. If I am landing with the right leg, I expect the hips will be rotated to allow for a further extension of the leg and probably rotated down this way counterclockwise from the camera to allow for further extension. You see that by doing so, I can now raise that weight even further. And you see that the fruit is still preserving the posts from the anticipation we made earlier on. So I don't quite like that kind of stuff. So let's have a look at what we can do it here to fix it. I will check the heel pivot there. Right now. We don't really need to keep the heel pivot active so I can disable it. I don't really need that thing. And even the doorway goal, I may need it later, but for now I don't really need that kind of stuff. And the lean, I can set it to 0 for now again, so that's the position I'm going to land in. I think no, maybe maybe baby, I can have the food a bit straighter there as I land and let's see if there's anything else we can do. I said I wanted to land on the bowl of the food, which I think I am doing to a certain degree. But maybe we can imagine that the contact posts will happen with the TPP first. So maybe I can try roll these a little bit more and use the stories to give the impression that we're touching with the tip of the toe. And I think that would start look a lot more interesting now, maybe I will need to move the route a little bit more forward and higher just to create a straight leg line. And I don't quite like the bumpy knees, ankles. So by moving the route forward, I reduce that bumped that. I don't quite like they're from the camera view. I feel like the leg is in full profiling their Medea can try and see what happens if I use the tilt of the food. Ac, by using the tilt of the food, we have much better posed because he feels you're not just lending with the tip of the toe, but also we wanted the size of the toe and the whole line of action of the food becomes a lot more harmonious, which I much prefer. Take care of the left leg, which I will bring up. Even with this leg, I would prefer to land with the boulder food. I'm not quite happy about the fact that the pole vector moves with the rotation of the food because it feels that whenever I rotate the food, I will need to counter-rotate lag in there. For now. I want to just place their food, their delay going down with the bowl of the food thirst and you go something like this, I guess will work as opposed ESA disarm has been lagging behind. So I will first move the arm forward and then I will rotate it this way counterclockwise from the side view. So the pound would be directed towards the face. And I don't want the arm to stretch too much. One thing that you will notice as you raise your arm is that the elbows deigned to face forward. We'll go grab the pole vector of the arm, which in the meantime has gone into the arm and I will move it outwards and a tiny bit forward there. There is a problem with these Reagan, which sometimes the arm flips. So you will have to be careful with that kind of stuff is not a unique problem. Other rigs have similar problems. The important thing is that you get to your post without triggering the problem, they go. Now let's remember that we wanted the arc of the spine to be inverted, which is not very realistic because if you wanted to avoid falling down into the cliff, you would want the weights to be forward like it is now our main reference. You see the weight tends to go forward there because the bomb is backward. You want the top of the body to go forward there. However, I feel that I would create a stronger sense of contrast. If I rotated the body to the back, I will drop the body control rotated a bit more to the back and then I will grab the secondary control and equity to the back and I will do the same with the chest and the other controls until I get to create a pose in which the line of action is arching a bit more to the bank. I will do this in 3D however, because they feel like doing it from the camera at this stage is a bit too early. You definitely want to cheat position of things based on camera, but you want your initial posts to work in 3D because eventually you will have to spline this thing free d, That means that interpolation will happen in 3D. Movement does not make sense in 3D would be a lot harder to interpolate it. So now I have my startup arch back with the spine. If the Belize is pointing down this way, then I can maybe opposed the rotation of the pelvis with the chest. Maybe move away their rockets for a second, just to make sure I understand the pose of this body, it's very difficult to pose the body while you are seeing the arms breaking all over the place, you have to use some imagination there and just imagine the pose, the body's hanging there. And then once you have the body as usual, you fix the arms. It's always the same procedure at the end of the day or the first and leaves next. So the left arm again, I can't keep it forward and the pole vector is ready for word in there, as you land is in your interests to keep staring down at the destination point. So we'll make sure that the same happens with this girl. You see that the neck is all over the place here. So I will grab the 0 it out just to see where it goes. And then I will neck control and I will 0 that out is just to have a clean start imposed. And then I will rotate both neck and head forward, maybe the neck a little bit less than the head and then I will grab the chairs and counter-rotate it forward a tiny bit so you see, it's difficult to keep that line of action Archean bag if you have to look down. So I will have to find a compromise in there. If you have a character with two eyes and a phase, it's usually a good idea to show both eyes at all times because we tend to recognize faces by the fact that there are two eyes and a mouth there. These pose, generally speaking, tends to look more appealing than a food profile. So I, in general, I avoid profiles and I'm not the only one I think, unless they're really necessary. Unless, I mean, unless the movie requires a profiler shot, in which case I can only abide to what the movie needs in there. And then I will make sure that the eyes are staring at the destination point in 3D space. Now, these posts doesn't look really appealing yet. And if we raise the arms, usually what we do is we want to raise the clavicle. You see how much better the shape becomes once we raise the clavicle earlier, RNA looked a bit like a broken action figure. And now if we raise the clavicles, everything looks a lot better and maybe we can move the clinical a bit more forward as well. There you go. That looks a bit more interesting. And I will want the two arms not to look to symmetric. So I will move one or the elbows a bit more in the left one and I will make sure that their left arm is not in the same exact pose and orientation of the right one. Maybe even the right elbow could be a bit closer to the body. Going back to the camera, I want to maybe tweak the pose a little bit in camera. And I wanted to see if I prefer to see the fist from this angle or this angle. In practice, I probably should be seeing the inside of the bomb in that Bose. I think I will go for a bit of a compromise. I will show me the thumb and a bit of the palm in there. Now, yes, it would be great to have some stretching the spine. So maybe we'll grab the chest in there, the two arms controllers and the pole vectors. That is to say all the controllers that define the upper position of the body and in world space, I will move them a tiny bit higher and the back, I can try rotate the pelvis to favor the general line of action. And you see that gives us a much better shape. So we have our posed, I'm pretty sure, improved quite a lot, but at the same time, I am mindful of time and we really want to get the basics going there. We're not really interested in fine tuning the animation too much right now, looking from the back view, I feel like the character is a bit unbalanced as if she's falling down this way. So maybe I can balance that pose a little bit better and move the body control and the pole vectors for the arms a little bit more to the right, and that seems to work a lot better. Maybe I can also grab the left food in there and move everything a little bit more to the side. Maybe I could increase the curvature from the back, the curvature of the line of action. So I will try and move the chairs and the arms a little bit more to the right. And you see that by doing that, I create a bit of a more balanced pose. It's hard to say how much of a difference that will be in the camera, to be honest, but I'm thinking always in terms of interpolation, once i will be splicing, I want to make sure that at least the basic poses makes sense in 3D. I will know that at least that part is taken care of the physics of it are taken care off to a certain degree, at least to make the pose a little bit more readable. I could grab their main body control and I could twist the body a little bit to camera to favor a better, less profile both. You see a bit more of a free quarter Posner. And now I can grab all controls and set a key on 38. Don't quite care about the rocket right now if I probably wanted to go even higher and break the connections in here, because I think that the rocket is making it more difficult to evaluate the poses of the spine. And I could either hide the geometry or just move it up a little bit. If I go and check now i had the starting posed anticipation and lending and if I scrub the time then between 1638 you see that I already have some interpolation there, which hopefully will make my life a little bit easier when designing the next bows. And in the next video we're going to see how to make the kickoff post. See you later. 15. The Kick-Off Pose: We have the contact and the anticipation pose. Let's do the kick-off. So we'll go to the place where I place the kickoff at round frame 29. I think I want the master to stay more or less aware he was a friend 16 because the feet are supposed to be there. So we'll go back to the perspective pose and I will duplicate the pose of the master to 28 so that the master you see want and I can move the pole vectors and the body controller forward. Now why you see that I already start designing my kickoff this way. Now, which foods should be the first one to leave the floor? If you're doing a Parker jump, you will tend to leave the floor. We bought feet, then on the floor with both feet. And that's because that's the most efficient way of doing it in practice. In our case, however, we are not going for efficiency, but for a more ice statical approach, we want it to look nice to look appealing. This pose we have in Park2 is very efficient, but in practice looks a bit robotic right now. So we would prefer to have something a little bit less robotic. If I go and check footage of people running and jumping, you will find out of course that the food which is the closest to the cliff, this guy here for instance, is the one that will leave the floor for last, let's have a look bag. Now that I know I can go back and I can maybe check which food was closest to the gap and you see it's the left food. So at 28, I will duplicate frame 16 for the left food. And now I'm on the floor and look at the eyes, they're bad. If you have this kind of behaviors, you can grab the master, move it and undo, and everything will go back to normal. If we go back to the camera, you see that in my drawing the bodies fully extended. So I really want to grab the left food, so we'll roll the left foot up as high as I can and I will make sure that the pelvises favoring that extension now will move Domain Controller apps that we have a nice straight line of action going up. Maybe the pelvis will need to rotate backward and maybe and maybe the route control can rotate forward a little bit. This way, I will have the weight More on the left fruit because the right food has already lifted off in these boats. So in fact, I will grab the left food. And what I will do is I will 0 out all of these attributes that I used when he was on the floor and just rotate it down. This way. I want the pole vector to stay up there somewhere in front of the character. And maybe I want a little bit of rotation in there, quite like the fact that full vector moves when I rotate the food, you see now the legs are fully extended forward trying to jump. Super BZ field doesn't quite feel like that from the camera. So maybe it can exaggerate that a little bit more. So a little bit more forward as a jump, I will check this food and here I want the light to be a bit bent. And then for the spine First I will maybe 0 out the spine. I want to know what the polls would be. Spine was geodata. You see that the spine is pretty much to the back in there. So maybe you can rotate the main body controller a lot more forward there and then count their arch, that Chez and I can even stretch the chest ever-so-slightly there. So that should give me a nice, stronger, straighter line of action. We were a little bit of a band hopefully in the middle. And that should give me the line of action I wanted from my reference to begin with. I can even try and see if I can push the pelvis forward even more. Now you go and create a more solid line of action in practice, I think that the pose I will have on 38 with the arms will work more or less for these pose at 28 as well. Maybe I will just duplicate the post manually middle mouse drag from 38 to 28 and hit. As for the arms, I would want the same post that I have on 38 more or less for now. So I will move the risk controls high up. I want to check the reference. You see that one arm is quite high up there and other back almost somewhere over here and will help with the clavicle as much as I can. And then of course I will make sure the arm doesn't stretch. You see that in here, the elbow, however, it's pointed to the back that looks a bit broken. So we'll grab the pole vector, move it forward and u, and that gives me a much nicer pose. I want it to be ideally perceived that straight, but it was a little bit of abandon their suffering like this, for instance, I think that could work. Same story with the left arm raised up o vector forward in a bit more open maybe for the left arm, you see that I was doing something different in there. I wanted the elbow to lead, but I am trying to maybe delay the arm a little bit and create opposing which the elbow is about to leave the arm there. You see I will rotate the wrist down and we'll have to lift up the shoulder as much as I can as well. And in here, I want to give the feeling that the elbow is leading the movement up there. Here, the position of the pole vector is crucial to make sure that the elbow stays where we wanted to be. So I will move it to the side of the arm and a bit higher up. Now if I move the wrist, you see that the elbow stays oriented that way. That makes it a lot easier to pose the arm or maybe keep it a bit closer to the body. So that will be my post one arm extended and the other one almost extended. Any will be extended on the way up with the elbow leading on the way out. I want to see if that pose makes some sense. It could work. I mean, I think it's a bit of a stretch, but I want to see if that's a possibility. Maybe I can relax the fingers a little bit. And for the left arm, I could open up the fingers a little bit more just to create a variation in the fingers in there easily to the post now is a bit more natural with the fleeced I didn't quite buy into the damn much meat you can try. And on the right hand as well do the same. Reduced a little bit the intensity of the feast in there. If I open up the fingers, I would probably have to move the thumb away because remember the thumb was resting on those fingers. And this will give me a bit more of a variation there and a bit more of action. I could even extend the fingers altogether now that I think of eight, but for now I'm going to keep the pose as these for the head i want they had to keep looking and staring at the same landing pose of earlier on. Remember that on 38, we rotated the head and we rotated the eyes, but we didn't quite fit the visual expressions that if you're animating a jump, what you're really trying to do is to make sure that the physics of the jump look believable. And then we can make sure that the faces look appealing. We just need to establish the general mood of the faces which we did interested at the beginning evolve into dissipation. And from then on, always involved. I came maybe moved the whole post forward, a tiny bit in there so I can select maybe the body control the pole vectors and move it onward and upward until I get a straighter leg for the kick a finger could even create. You see, abandon the lag there. And I think that would look a bit nicer. We wouldn't really jump with this book. At least it's not a symmetric beause. If the jump is pushed by the left leg, I would expect the weight of the character to be falling onto that left leg. And you see that right now the character seems to be falling to camera left. But if I move the weight to camera right now, it feels a bit more stable. There is something wrong in the leg in there. Let's find the pole vector and see what's going on in there. So where's the pole vector on the left leg gone, you see went into the leg itself. So of course it's going to flip the leg, we're going to bring it forward. I was saying now that we have the pose more or less sorted out, these aren't feature is experimental, at least now we have the idea that a character wants to jump over the gap and we are respecting the line of action for the most part. We have in the drawing. So you see how important it is to have the drawing. Because if you start doing window shopping, if you start going into my eye or the 3D software and figure out, I want to try this or the other pose is going to take you hours literally. And you're not necessarily going to come back with anything relevant. Because if in 3D, in any software, because if you decide to work on opposing 3D, and that's true for any software without having a plan first or even a stigma and figure for a design of a post. You're going to spend hours before you even get there. You're trying to reach a pose and you don't know what's the function of that post. And you will know if you're going in the right direction relatively late if compared to a plant. Now I have my opposing there and I can hit asked, We have a key on every control for that post. Every time you look at your animation or oppose after you're done it, you spot more mistakes, more errors, more things that you want to fix. Maybe now, I wanted to go even more forward there. I like it more. This was earlier on. This one is a bit more out of balance and going towards the jump. Now that we have the kickoff pose, we can start thinking about the future again. And the future, in my opinion, is the oppose which if you look at the interpolation we have in Maya, it's sort of making itself already. So it seems to me that the oppose should be the post we're going to do next. And I will see you in the next video. 16. Understanding Parent Space and Space Conversion: First of all, let's make sure that on the kickoff pose we have a key for each controller so that when we make another pose at a later time, we're not going to accidentally aided the kickoff bows. And here we have the AP photos. If you think of it, the oppose is quite interesting because the master is already doing the job of translating the character back and forth, which is not a bad thing in my opinion. So we'll go to the frame where in theory I want to place the oppose and thereafter the sign out to make it. I think it would be fantastic if I could rotate the whole character around its center of gravity as opposed to the bottom of their feet in there. So let's see if there is a way to do so. So we're going to use the master as a device to move everything around from the center of gravity. The problem is that that mastery is on the floor down there. So ideally we would like to move the muster up here to where the root is and then automate the master to do the upper part of the animation. But the problem is that when we move the master around, the character moves around as well. So in here, I'm going to need a way to keep them animations it is, while moving the master opt to do so, we are going to employ Morgan loses World bake tool. You will find it on his website. I will put a link in the description of the video, of course. And for the installation threes referred to his installation instructions, which are fairly clear. One thing we have to keep in mind is that Morgan is asking us in the requirement section to download one of his scripts from his GitHub page. So please remember to do it. Otherwise, the world bake to want work just so that you know, you can do this thing with or without the world bake two, except using this tool makes your life a lot easier if you're doing creature animation. These won't bake tool or a similar tool that does the same thing. It can really up your game. So in here, let's try it. First of all, we know them when we move the master, we lose the position of the lady. We will need to find a way to store that position. The first thing you need to understand is that whenever you grab a controlling Maya, unless that's the master, Of course, the values you see in the channel box here are not absolute values, but they are relative to the parent of the control. So for instance, let's say that I have a cube in here, the cube as, as a translation value, some translation values, which if I 0 out, we'll make the Cube go back to the 0. In other words, the space of this cube is the grid. And in fact, if you check the outliner, the cube is there in the main route of the seed. So the parent of the cube right now is the scene or the grid. That means, as I said earlier on, that whichever value you see up there in the channel box is going to be a value relative to the grid. And when you 0 things out, that Q will go back to its parent, which is the greed. Now we'll make a torus, or it doesn't matter which shape you make. This is my donot. If I move it away from the great N a 0 out, the translations you see goes back to the center of the grid because these guide is storers still has his apparent the scene just like the cube. Now if I make the torus, the child of the cube C, that the torus is the child of the cube. If I move the torus around and 0 out the translation, the torus will go back to the queue. Let's do an experiment. I will move the torus around now and I will move the cube away. So you see that as I move the cube away, the torus is falling as well because the torus is the child of the cube. If I now grab the torus and 0 the translation out, where do we expect it to go? Well, in theory, if what I said is that all the values you see up there in the channel box are relative values. That means that the values we see in here are relative to the cube. And a good way to understand this is to animate the cube. So I will give the cube a couple of keys that Cuba will start from 0. And as a scrap the timeline, you see that the values up here in the channel box are changing. That's pretty interesting. That's normally, we expect that, right. Values are relative to the Greek because the cube is the child or the grade. You see when the tube is at the center of the greed, the values are 0. Now when I select a torus, I expect those values not to change. I expect that, but not all people expect that. So you see that as the timeline changes, the torus is not even animated. It's values are not changing. And that's because these values are relative to the cube. If I now 0 out the translation values on the torus, you see that these storage stays there where the cube is. In other words, these stories is moving in the cube space. The cube is the parent, the torus is the child. And the cube behaves exactly like the center of the greed of maya. For the doors, you see that as I rotate the cube, the torus doesn't record any variation in relation values. And that's because the torus does not think it's being rotated. It's just the space above the torus that is rotating. This is very, very important to keep in mind because when you start an animation, you tend to believe that these values you see in one control are relative to the grid. That means that you might be tricked into believing that you could simply grab the chest, copy the keys of the chest, and then grab an arbitrary object like a cube and paste those keys there. And you would expect the chest and the cubit to be aligned. But you see that's not the case. You see that the cube has some motion that reminds us of the chest, but it's not really doing the same thing because the chest is moving in the space, probably are this control while the cube is moving in the world space. So the chance does not really translating through space. It's, these guide is translating for space. Chess moves along it just like the torus was moving along with the cube. That means that you cannot simply copy and paste values in the channel box for two controls are two objects to move the same way in a 3D software, if you want to store the emotional, some controls over some other item in a 3D software. We say you have to convert that motion to world space. And that's exactly what this tool does. The World Bank from Morgan Lewis does exactly that. The way it works is you select a control. You see you can specify which space to bake do and we're going to use world. If I click on bake selection to Locator, the tool has made allocator, which I will make it bigger for your benefit, that crossing there is a locator is a tool we use to locate things in the world. And you will see that the locator has the same keys of your raced in there and the same position in space. If you select the locator, you will see these values in here. If you select the breeze, you will see different values because the locator moves according to the space of the grid while they move in a space of some other control. This way, I know for a fact I have stored the keys of the research, which is a great thing to have because now let's say that I moved the master up. I will still know that the recent needs to be there. I will still need to find a way to bring the risk there. But you know, at least I know where the rest was supposed to be. Maybe we can perform these world bake operation on a bunch of controls. Let's delete the locator for a second and let's decide which controls we need to store the position for. Well, in practice, you will need to store the position for any control that the master is moving around, the Master and only the master. That means that if I grab the body control and move the body controlled around, you see that the body control is moving around the whole upper body and the hands. It's not moving around the pole vectors of the arms. And in fact, you see that the left arm loses his bows. So that means that those controls need to be stored as well as a position you see now as a move. The body control and the pole vectors, the post does not change in the upper body. However, you see that in the lower body the feeder are not moving, so we'll need to select them as well. And now if I move around the feet, the body control and the pole vectors for the arms, you see that the whole post translates in space and everything works just fine. So this is the selection set I think I will need to employ to convert the animation of the girl from the local space of their controls from the local hierarchy space to the world space. In fact, I can maybe go create set, quick select Set, and call this convert. And now if I click on the shelf, every time I click on this convert icon here, I will get to select those controls. Now, I would like to place the master at the location of the hippo, the character, however you see whenever he moved to muster the hips of the character move as well. So that's a bit of a bummer. But if I think of it, I could go create locator and that gives me, guess what, another locator. Then we located selected. I could add to the selection, the body control. And then I can go and modify much transformations and i will match the translation. And now that I have much that translation, if I scale up the locator for you, you see that we are exactly with the body control is we can grab the master control, the locator to the selection and go and modify much transformations, much translation. And now the master is right there where we wanted it to be, which is not a bad thing. I have a selection set for all my controls. I can store their animations because the next step is going to be moving the master around, so losing that animation. So I'm going to select the controller described earlier on, click on bake selection to locators. The tool is going to convert this animation. Be careful because it's going to convert only the visible timelines. So you need to have the whole time visible to do this operation. And now I can go to 28 and I can align the master up to where the body was. The body control was where these locator wasn't here. So at 28, I will grab first the master control, then I will add it to the selection, the body controlled, and it will go under Modify, match transformations, match translation. And that's where the body control was earlier on. Right now doesn't seem to make sense. But if you think of it once we wouldn't be done with placing the master control, we will be repositioning the rig where it's supposed to be for now, we just need to place the master control. So at 38, Do the same thing. I select first the master, then the body control locator, go under Modify, match transformations, mesh translation. Then let's go to 44, and I will do the same thing. Then what happens after 38 is not ready, irrelevant to us. In effect, we can grab the master control and make sure that it doesn't move after 48. Also, we do not want the master control to move before 28. Otherwise, it's going to move the feet of the floor, which we don't want to do. I will delete all the keys between 116, so we only have 2838 as keys. Now we want that emission look exactly like we did earlier on, but with a master in here. But if you think of it, we have all the educators that are storing the animation where we needed. So all we need to do is go onto the World Bank to go into bake from locators. We have selected the locators in your C, all of them, and we click on bake selected locators back to objects. I do that the tool does its math and now you see that the master doesn't move at the beginning, but an emission is the same. And between 1628, the girl goes into the kickoff pose, and from then on, the Master moves the girl around. And after 38, The Master stops moving and animation goes back to the local controls, which is super useful because now that means that if we are midway through the air and we want to control the orientation of the girl. We can do it from the master. So this is super useful as a technique, and not just for the jump, but especially for creature animation and complex body motion. This is really a useful technique to learn. 17. The Up Pose: If I check the camera, you see that I stored my oppose at 32 so I can go there and set a key on the control. If you think of it, the master is translating mainly horizontally. But now with this control, we can move the whole character around. And if we need to ever rotate it, you see we can do it, which is great. It's a great advantage we have right now. So the doctor was jumping up then lifting up in the air and good to exaggerate that. It's falling down again. Maybe I can move it a bit closer to the destination because the initial push would be pretty quick and then it will lose in intensity. And I'm going to check the poses and we have the starting pose, the anticipation, the key cough, and the oppose, and then the landing. Maybe that poses a bit too far forward and you see that it's so easy now to adjust that it's really a piece of cake to adjust the position with that. And this is one of the key, and this is one of the keys are working in 3D animation. You need to have tools that make the work easier. Otherwise, you're just a victim of the software and its shortcomings. So I've made my animation there and I want to make the oppose exactly like I was drawing it up there. You see arms up and one of the legs up and the other one behind and go to the side, which leg is going to be behind? And since the right leg is the one that kicked off earlier, I think at 32. And we want the right leg to be forward. Maybe I can 0 out the Tories though we go all of that stuff that I don't need them while I'm up in the air and now I have done my Posen there. I, as I said, I'm not super happy. As I said earlier, I'm not super happy. I've, I mean dipole vector follow the food. But hey, I can always convert the position or the locators to the world position. If you think of it now that I know how to do it, I can easily do conversions. Now the left leg is the one that has just done the kickoff. So it will be a bit delayed, I guess. And I can 0 out again, football and all those fancy attributes. And I will reach there rotated pose using only the rotation. That's going to look a lot nicer. Maybe I can do a bit of tow wiggle there and that's going to be the cause maybe I will make it less straight. If one leg is forward, the pelvis wheel rotates forward to favor the extension and I will have to invert the line of action if, if you see my drawing, the line of action is now inverted. So I will probably have to rotate forward the whole body, maybe u in this control. And maybe why not even the route a little bit, that will give me a line of action which is more similar to the one that I was drawing. His arm is a bit broken, so maybe a world Keep it up there and he or she is kinda scratching the back of her head. So maybe maybe maybe in these arm will have gone up as well. It's kind of a good pose. I think it could work and maybe I can have the lag a bit less extended at the back there and we have the NEA a bit further out it that is, if I find the pole vector of the Ni Ru, the pole vector or the need no, I, you know, they got this guy in here. See why I don't like the feet moving the pole vectors because then the collectors keep moving around in space, which is not a great idea. Now, I want to make sure that they had keeps looking down at the destination which, which it is doing. And this is my opposed C was pretty easy to make it if you think of a interpolation, did most of the work for us, maybe I can't separate the two legs because I feel like the silhouette is a bit mixed up in there. So maybe I can push this leg forward. There you go. So now I get a bit of a better silhouette there. Perhaps, maybe it could delay the tip of the toe. There they go. Something like this. Suppose you go now that's my oppose. And I want to select all controls, set a key, y'know, controls that I don't accidentally aided that pose anymore. We were a bit faster than doing it the oppose, but it seems to be working anyhow, we have older poses for the jump and we can possibly even check if the blocking works. One issue you will encounter is if you're enabled us that preview, you will see that your character changes position. And that's because not all characters are compatible. We stepped preview. We can just have a look at the amount of difference we have in the poses. And you see that there's quite a difference in the boson there. That means I am unable to easily evaluate blocking without converting the tangents to step, which is not a big deal. I can select all controls, opened the graph editor, select all keys and convert them to step. I do not need this pose at 60, that's suffering there. World Bank tool made for us either renewed that. I'm going to hide grease pencil for now. I don't need it. And I'm going to press play. I want you to notice this is the first time I press play to evaluate my animation. Until now, I didn't have a press play because at the end of the day I didn't even know if I had the old deposits I need. There you go. And you see that animation works for the most part. If you forget the rocket flying away, that is actually if you think of it, that rocket is going to become a bit of a problem. So I'm going to go to the end of the timeline, select all channels and right-click break connections that needs that the rocket back will be staying there and moving around with the spine now because he was at 0, press play. And this is my blocky. One thing I don't quite like about this blocking is that it feels like the chest is moving backward, then rotating forward. They're rotating backwards again. This is a big too much, there's going to be a bit too much motion going on in there. So I am going to rotate the chest that bit more forward on these contact posts. I feared that are a bit too many inversions of motion on the chest in there. So I'm not very confident about the outcome of this blinding. So I will just tweak. Every time you tweak the chest, you're going to have tweak their poses are the arms anyway, they go now, I did that in the wrong position, but it's not a big deal. I can just middle mouse drag and hit as to copy the pose, there is still irritation of the chest forward and then backward. But now it's a bit less and not quite happy about these curve in there. I feel like I have rushed it. So maybe I can rotate the spine counterclockwise from the front. And then the second controller is buying could go clockwise. So maybe that will give me a bit of a better curving there. I will go to the outposts and maybe I will rotate this bind forward a little bit less. So that should give me less chest rotation, counter rotation. You see that chest is lot more stable. And if we go and check the references, you see that the most octave portion of the chest rotation is at the beginning when the chest is used to create momentum on the way up and onwards. But then after that the chest stays more or less inert. That's also why I wanted to reduce the rotation of the chest in there. I do not want to cover the eyes with the arms at any stage because they want to make sure those facial expressions are still going to be readable. And now that we have the oppose, I think we are ready to animate the end of the animation. And now that we can finally evaluate the blocking, it's probably time for us to do the end of the animation. Right now we have the same pose we had in anticipation. And although it doesn't quite read the same way, I think I should probably do something about that. So in the next video, we're going to do the DOM bows. 18. The Follow-Through Pose: Once you look at the logic behind the downpours, you will realize that decomposes the result of the body coming down, the legs compressing and the hand touching the ground to stop the jump forward, it's more than reasonable to expect that after the hand has touched the ground, the body will for a certain amount of time still go down before the arm can actually stop them motion. And then finally the body will lift up a tiny bit. So I think it's worth thinking if we want to do the down or maybe the end puzzle, the animation, and then reverse engineered the down from debt. I think I will go for the NFO isn't animation, which could be called a down as a whole. But in practice, we need to work into these bows disposes, made some motion anyway. So you see that we can identify some key poses like for instance the down. But then at the same time we realize once we look into the details of that motion, we realize that that pose in itself is a small universe containing sub poses, some other key poses. So let's do the endpoints of our animation, this sort of down. But then let's find a way to get to it. More believable weight. So the down has the right hand on the floor. Left hand is at the back and the feet are on the floor as well. I think we get it. So first of all, I need to grab the hand and move it down to the floor. You will find that the moment you put your hand down on the floor for now, ignore the fact that it's super stretched. It's going to take you a while to position these handwrite is not really going to be a straightforward. It always takes a while to position hands precisely. But then if you move the body control around, you see that the hand is moving around as well. So this is probably not a very efficient way of unmade in these last post, I think it would be more efficient to pose this hand is the hand weren't following the body control. In order to do so, we need to change the paren space of the hand. So I will set it to world. The hand goes up there, so maybe I want to keep the position there. Think of it. We have the World Bank tools, so maybe we can first of all create allocator for this hand. And then once we have the locator for that hand, we can change this barren space too well, and then we can select the locator and bake from locate. And all the previous poses of the hand will still be in the same position. So you see it's very easy with a two like this one to change this pace in which you operate. By the way, there is another tool that does similar things to the world space and a lot more. It's called uninvolved and it's really, really useful. Unembellished is an excellent tool to improve your animation workflow. And if you work in companies, you will find out that very often as you walk through the door, you will find the animators will already have these installed. The reason why I don't show an embodied in tutorials is because, and what is not for free. So adding it to my tutorials would mean that the users of the tutorials will have to pay that additional price, but it's a tool which is really worth the money back into Maya, we were making the downpours, remember? So in here I'm going to lower the weight, a little bit more raw food accordingly. And that gives us a nice decent posts. I probably want the hand to land a bit closer to the center of the body, like here, for instance. And for now I don't want to think about fingers too much. I just want to place the hand down there. You see that the arm is a little bit stretched so I can grab the chest controlling their and probably even the second control of the spine and it can rotate them. And you see that as I rotate the chest and the spine, the becomes a lot more believable automatically. Of course, the chest is going to help. The arm extend. So this is kind of cool if you think of it, the moment you involve correct physics in your animation, things start to work a lot better. So maybe it will try to rotate the whole body a bit forward as low as I can without breaking everything in there. Help a little bit with the clavicle, keep rotating, the chess keep twisting. And when they start seeing that the arm is no longer stretch, that's when I feel I'm in the right position and then you go, that's my pose. For the left arm, I could do multiple things. I could just keep the arm up here in the poses quite cool already, isn't too bad at all. The problem is that you see that we don't see that coolness factor in the boas from the camera. So we will have to do something about that. And in my sketch, you see that the left arm was left behind on purpose this way hopefully make it visible. So I'm going to rotate the hand, move the arm maybe a bit closer to the body, even a bit hires who are breaking the elbows. So probably a good idea to raise the whole vector. They go and we are in fact probably breaking the arm all together. But the point here is that we want to see that arm silhouette. And this is going to become very difficult if we have the rocket pack, the way they're, maybe we could cheat the position of the rocket pack and translate it a little bit more towards the right, they should give us you see a bit more visibility or the silhouette of the arm there. And in here, we are probably going to need to show a lot more of the back of the palm in there. We're going to need to have the position of the arm with the comical. So the clavicle would probably move to the bank a little bit to help the extension of the arm of the bag. There you go. It's a bit of a shame that we don't see the connection between the shoulder and the arm in there. I have to say that those boosters are a bit of a bummer really, because they make reading that pose a lot more difficult. Now in terms of the head, the head isn't a decent position. There is a bit of intersection there. We have to understand if that's something that we want or we don't. The one thing that probably we don't want to have is a full profile shot in general, as I said in a previous video, if you have a face of a character with two eyes, you want to make sure that the audience sees both the eyes all the time. So you see all the rotate the head ever-so-slightly the camera. So that would give me you see a much better pose. We've bumped the ice visible and she looks a lot cooler than she did just a few seconds ago. And look at this this one. Yeah, it was okay. But if you think of it, you couldn't quite tell what the expression was on the left-hand side of the phase. If we go back to the new beause, it's very clear why she's thinking and what she's doing there. So we're going to keep it that way. But in practice for as long as it's perceived as a profile from the camera, we are going to be happy with it. And for now on when you watch an animated movies, understand which shot is a profile and we really ease a cheated profile. You will find out that in most of the animated movies we watched profile shots actually are free quarters, which we perceive as profile. There's a bit of intersection there between the color and the head, and I am not finding an easy way to fix it. So I will just move the head a little bit more forward, trying to keep the intersection to a minimum without breaking the bows and not very happy or this curvature of the neck. Maybe I could try lower the head a little bit and no luck. Anyway, maybe it could go a little bit more profile. So I would really that control normal in a regular control here to control the color. So right now, I think I'm stuck with this bows. I could perhaps rotate the chest a little bit higher there, and that improves the situation a little bit. So we'll stick to this Bose. Let's see if I can rotate this thing forward. Something that gives we'll do she looks cooler when the head is not parallel to the floor, so maybe I will keep her not to the floor. And I you and that's my typos. Now, after I am done with the main posts, so body and then limbs and then had I can check the fingers. It didn't really make sense for the fingers to be done earlier because we weren't quite sure of how we wanted the whole thing to be. It doesn't really make sense to spend time figuring out the position of 15 fingers controls when you still don't even know which kind of pose you're going to have for your character. So you see that I'm going to do the position of the fingers only once I'm done with the position of the body and I find that it's the only sensible thing to do really, if you sink the fingers a little bit into the ground, is not going to be a big deal for as long as we still see the tip. And if you want to give the impression of weight, you can rotate the second Finance down and then counter-rotate this deferred filings up and that will give you the impression of wait, maybe this is a bit too much, so let me reduce it a little bit. Something like this will do one thing that happens to fingers as you land and you distribute weight onto your hand is that the fingers they tend to spray it. So I can use the control of the bomb in here to spread the fingers ever so slightly to give the impression of the weight being distributed across the fingers, I would let me try to keep the middle and ring fingers a bit closer to one another just to avoid the rake poses or the fork Bose's. And this is going to be my Posner day ego. That's going to be the post. I feel a bit for key, I don't like it. So maybe we're bringing the fingers a bit closer to one another, at least the ring and the, at least the ring and the middle finger. And I think now I get a nicer pose in there. Maybe try and open up the thumb even more. Something like this, I guess could work. I could spend more time on them to be honest, but I just don't want to spend too much time fine tuning details now, now a select all controls and I will set a key, and that's going to be my down bows. And in the next video we're going to see how to get to that downpours. Because remember, this is the dark push with the silent we will have at the end. But really we need to go from the contact pose to a downpours which is a bit lower than this one and then come back to our final pose. So I will see you in the next video. 19. The Down Pose: I do not need these key that the World Bank tool created so I can delete it at 60 and then I'm going to duplicate the pose. I have a 44 and place it. I didn't know I had 51, so I have two poses which are the same. Now, I would really like to treat the post 44 as a contact posts for the hand and a pose at 51 as a follow-through of that post. So let's do it is the contact is already there for me. One thing that could happen during the contact is that the chest is a lot more rotated to favor that contact is a lot more rotated down towards the hand and the body is not entirely rotated down yet. So the idea is that the cortical, the food translates into a contact with the hand and then the weight keeps pushing the character forward, you see? And then I guess eventually they will go a little bit back up. So this is what I will want to do. Another thing I could do is help even more with the shoulder and rotate the shoulder down even more there to help with the contact. The more I rotate the shoulder down, the more I can counter-rotate the spine up, and the more in the next book, we will feel the weight pushing down. See that await pushes down the body and the shoulder is pushed up by the arm. That's the kind of motion we want to transmit in there. Let's have a look at the bottom of the character in there. And maybe we can exaggerate a little bit the rotation in here, just to suggest that the weight is still on the right leg in here. And then later on, you see the weight could stabilize between the two legs. They go something like this will do, for instance, so we keep that body moving. Now, let's have a look at their head. The head may be a bit delayed on the way down, so it's not going down one-to-one with the body, but it's being a little bit delayed. I can use a bit of translation at the head and a little bit of rotation of the neck. Let's have a look at the camera so we go down. The head is delayed and then it moves a little bit more. Not very happy about the pose yet. To help contact with the arm, I can possibly rotate the body clockwise from the front view and then I can twist the body over itself you see to help contact the floor a bit more easily. Now you go. So the action we want to see in the next book is a bit of counterterrorist, possibly even with the chest, it would be really cool to see some counter twisting going on there. Chess could go a little bit lower. And now we see the chest being pushed up on the right-hand side of the character by the fact that the hand is staying stable on the floor. Right now the hand is rotating because we have some stretch occurring. There is this IN hearing in which the arm is probably stretched. Let's have a look and actually it's kind of okay, that's cool. Actually, there is this moment you see at the on 51, We are a bit overstretched there. So maybe I could rotate this body control a bit lower the weight a bit more onto the right arm in there. I really want the clavicle to be pushed up, the chest to roll down even lower if possible. Let's have a look and see if this makes some sense. So we touched the floor, we are pressed, you see against the floor by gravity the hand helping staying up. You see the hand is pushing up the body which then twists. I'm not very convinced by this motion I made for the head. Let's have a look at what they had done. So they had is up there. We go back down with the body and the head is delayed, but you see it's all over the place. So maybe I will first of all 0 out the head to seaway should be. And you see that by doing that, we get the head in a decent pose already. We have a lot of twists going on. So maybe we can grab the neck controller and tried to see if there is a way to fix that. Probably the twist by, you see, kind of difficult. We can't quite fixed the twist in their notes ever looked at it poses against we contact with the hand and then we compress down due to gravity. And I think. We need to take care of their arm on the left. So the arm on the left will be still traveling back probably as we contact with their right hand. So we'll still be somewhere on the Arc leading to the next pose. I think I want to have the elbow a little bit more on the inside. That is, if I can find the pole vector for the arm, that ego, it's not there. So yes, something like this will do a lot better if you think of it, we have a much better line of action now between the two arms. And I know that we don't see that from the camera point of view. So we should probably make it our concerned to do nice posing in. There you go. So the left hand is still traveling to the back and it starts to make some sense. We are still involved in downshift inquiry. There's still work to do silicium 44 for everything, one at 51 for everything. And now we have the contact for the food at 38, the contact for the hand at 44. And then we compress down and we would probably want to compress back up again. The cheapest way to compress back up again. Now a cheap way to devise a last bows will be to convert all the tangents to auto. And we've all control selected. You can grab a POS which is in the middle between 4451, say something like 49 for instance, and just drag it to the right so the character will go down and then come back up. One thing we have to remember when doing the father fruit is that the body control was going downward and onward. And that means that after touching the floor you will probably have still some upward momentum. So that means that coming frame 51, we probably want to push a little bit more forward. And then at 56 it will come backward. So there will be a little bit of an arc in the body control as well. It would be nice to think about overlap for the head as well as so maybe at 44, we can rotate the head backward a little bit, then at 51, little bit more. For now I'm going to ignore that intersection. And if 56 is going to be coming down a little bit, there is a bit of overlap in there. We will need probably more pauses. There is rotation on the palm occurring there, so there is some stretching going on. So we'll go grab the chest, lower it and that's not changing the orientation of the hand. If I go to 44 and a lower chest ever-so-slightly, you see that the hand stabilizes. So usually if a limb supports stretching in a rake, your additional the extremity of the leading lives should not change that standard behavior usually, but every rig is built a bit different than what matters in year anyway, is that if I lowered the chest ever-so-slightly ice stabilize that hand. Now we compress the floor and then we compress backup. Now you got so now we have worked out our decomposes. I'm going to select and two hundred forty one and forty four. And if I now check, they transition there. It's kind of cool except for a BEC arm which are not very fond of. So I'm going to grab the back arm in there. And what I want to try and do is to make sure that the arm is following some sort of arc on the way up, maybe reaching an extremity. So maybe overshooting its intended bows on the way up and then coming back down later on, something like that, I guess could work. So we swing up and then settled down later on. Now let's select all controls and have a look at the blocking as it is so far. In the meantime, I keep saving of course, will select all control go tangents. And now I want to check the animation here. Press play. You see that now the animation is a lot clearer. I don't need any more degrees pencil tool visible in fact, because all their poses are there and not very happy about some pauses there is in particular the kickoff pose, this one with the other. I don't reelect that arm, so I wish I could place it somewhere a bit more aesthetic, but maybe I can raise it a little bit and see if the camera is happier with that choice. That's already better. Oops, wrong frame. I need to do it at 38. There you go. Let's have a look at it again. Yes, it's a bit less annoying now. So these blocking, we could potentially go to a supervisor or a client and ask them already, Hey, can you have a look at this thing and tell me if that's the right way to go, if that's what you want, just by looking at these, the client or the director or the supervisor would be able to tell us if it's worth spending more time on it. So in general. So in general, regardless of whether you work for yourself, for a client, for a director or supervisor, you want to make sure you get to a decent place in urine animation from which you can tell exactly what's going to happen in animation at a pretty early stage without spending too much time, if you think of it, this tutorial wasn't very long so far. And if you think of it, I also had to explain every single step and the decisions behind those depths. So those explanations made the tutorial longer. But in practice to get to these poses, he didn't really take too long. So with this in mind, I can decide, yes, that's the way it should be working. No, it should have been lending and other food. I wanted her to do a spin in the middle. I don't know whichever you can get feedback on the animation and then you can move on. So in the next video, and in the next video we're going to do a bit of retiring or this animation just to maybe make it feel a bit better, a bit more fluid. So I'll see you later. 20. A Bit of Re-timing and Tweaks: Another, we have the key pose is done. You can maybe call them the story poses that ease the poses without which you can't really tell you a story. And we can maybe read time those, for instance, if I press play, one thing that I notice is that it feels like we touched the floor. We want food that we go for the contact of the hand and then there's a little bit of a break, and then the character moves forward, which in my opinion seems to be breaking the connection between the momentum originated from the landing and the subsequence down. So maybe I can move that pose a little bit closer to the first contact pose or the hand. And I wanted to see if that makes some sense. Again, you see that now it makes a bit more sense. There is a bit of a glitch in there. You see this guy. But now we know that if you move the master and undo, that is fixed. And you see that now the connection between the contact of the hand on the floor here and the next book is a lot tighter and other thing possibly do is we could maybe keep the oppose a bit longer and in there. So maybe you can grab all controllers. And once we find the opposing there, we just grab the following poses and just move them a couple of frames to the right-hand side. And we just want to see if we are going to enjoy that molecule break a bit better. And I think it's time for us to have a look at the change of island in the character. Not going to be very visible if you think of it, but we still want to know what we're going to do there. And you remember that during the kickoff, we changed the I align, but we didn't really tweak the facial expression. So in there I want to move the eyelids down to favor the gaze towards the floor. And you see that the facial expression looks a lot better. Again, I will compare the current pose with the previous pose. You see that in the previous pose we actually can't see down but with the new post just by adopting the islets to the new y-direction, we get a much better bows down there and get that a lot more effective. I made the pose at 45, but really, these bows on the kickoff should have been at 40 right now. So delete 45. So there you go. That's the pause that I go to the next post. I want to see, I want to know where the character is looking at. An official expression is sort of a K and the character is looking straight in front of her, that's fine. In here. It will look better if the character, as it went, Dow kept looking in the same direction. Instead, she looks a bit higher here than she does down here. Maybe at 50, we are going to raise the eyeliner and she looks a lot cooler already looked at that. And we can grab the islets and raise the eyelids ever-so-slightly. They go a lot nicer than as she comes back up. We probably can keep the same pose. So we can probably duplicate frame 50 or 55 rook. Let's move the master and undo. Yes, I think 50 to 55 is a good choice. So now the character will be focusing for the most part on what matters to her name. She stays issue stays looking in the same direction, which is what we want. Nice. So now we retired the animation. We fixed a couple of things in the facial expressions that we didn't quite like. I will save my file. And in the next video I am going to add a couple of poses that I think I need before going into blinding. So I'll see you later. 21. Floaty Splining? Add Breakdowns!: Once you have all the key poses, I think it's easier to think that you could go into this blinding already. What we have in here is what you would call maybe a first possible blocking where all the key poses are there and only the story pauses are there is a little bit of additional posing at the end when we have the contact with the hand the down and they follow through. But that's about it, that there isn't much more than that. And it's a bit risky to go into splicing Once we have only the key poses. And the easiest way to realize that is just grab all the controls, convert everything to auto, and press play as nothing as easy as pressing play on your blocking. We're blind to see if you're an emission is working. And you see that right now, more or less we have something in there, but you see that the timing is all off, nothing really quiet working there. So you see that it's clearly too early for us to go into this blinding and hope to finish this animation without bloodshed. And look at the right-hand end here. Look at that elastic girl. What's going on there? And I know what's going on, but you see there are plenty of issues that we still need to solve before going into splicing, I think it's a good idea to add all the poses that we want to keep for longer. Because remember one thing we noticed when we were looking at our reference is that the gentleman was going down to the anticipation and then he was holding that this vision for quite a while. Because if you remember, when we were looking at the references, I made a mental note to myself as a reminder that the anticipation in the reference lasted for quite a while before the gentleman starting jumping up, if you think of it, that is not the same case for our blocking. Did we go from one pose to another? And then one frame later we already leaving. So everybody have their own workflow. But in practice, in my opinion, it's a better idea to make sure that even in the blocking, you already have those poses that you want to keep for longer. So if we haven't anticipation, in my opinion, we should keep that anticipation for longer before springing into action. And the only way we have to keep that anticipation for longer is to make another pose. Similarly, if you want to keep the character up in the air for a bit longer to exaggerate that sort of moving hold that we have once we are at the top of a motion, I wouldn't need opposing there. And similarly for the down pose in here at frame 50, we would probably want to keep that down pose for a bit longer before going back to the end post. So you see there is some more detail that we need to put in here before we could go into splicing. So let's do it. First of all, I would want to sort out the anticipation pose. So I'm going to maybe go down to the discussion a bit earlier in order to create additional policies quickly, I will need a tool named between machine. You will find it online for free along with installation instructions. I would want to keep the anticipation for a bit longer as I said. So I will go to maybe frame 11. I will load the twin machine there and I will move this slider to the right. Now what happens when you move the slider to the right in practice, one way to look at these UI is to think of it as these middle box that you see here, just underneath this. Neither is the current position in time. The box fire right, is the next post, the next key you ended on the timeline and the box far-left and represent the previous key frames. So this box is frame one. These boxes deposit forensics. And this is where we are now. This neither decides how much the current post leans towards the next or the previous one. Moving this neither to the left will give me oppose. That is more similar to the initial pose. You see she's standing and moving either to the right, give me, gives me upwards, which is more similar to the anticipation. So we'll go for a post which is a lot more similar to the anticipation there. And now that's my second bows. And you see that by doing so, I can't avoid those freezing poses there. So typical of beginners animations. So that's quite good already because that means that anticipation is going to work for me and stay there. Now between the anticipation and the kick-off pose, I probably will want to have a breakdown. And in my opinion, the breakdown should be the kickoff pose for the right food. Because right now we don't really know when is the right food leaving the floor. And that's kind of a bummer because as far as I'm concerned, this is suppose I need to keep track of. The other thing is if you track the right arm, we would expect their right arm to do this kind of arc, right? But as you watch the animation unfold, the right arm goes all over the place. Right now I'm starts breaking the clavicle and going the opposite way. So we want to take care of that post first. So maybe it will go, say free frames before the kickoff pose for the left food. And we will use the three machine to put the right food and position. And you see, we are quite stretched in here, but for now I don't care too much. I want to see if maybe I can help a little bit with the pelvis. And also I want to see if I can run the food and remove that stretch. And you see it's quite a tough and Dave or I can try The TO pivot there to help as much as they can and also the Tor raise. But you see, it doesn't matter what I do. We're still too far away with the root, so maybe we'll grab the key for the food and move it a couple of frames to the left to see if that gives me a nicer pose. And you see in here the pole vector, the leg is all over the place, this guy here. So I will just move the pole vector forward and moving the product and the foot to frames to the left, you see it's too much, so we'll go and find again 25. It seems like 25 could be a decent posed for the kickoff of the right food. But I think at the end of the day, I will need to move these key to about 24 and that seems to be about right mimicking the tip a little bit less, enroll the food a little bit less. That's going to be the kickoff for the right food. I don't quite need the left food to be that rolled up, so I can maybe keep it down quite a lot more. And you see, it works pretty well already. Lemma, we can push on the route a little bit more just to make sure that the right leg is fully extended. Designing the kickoff was for the right food is not going to be as easy as we think. One of the issues is that the speed at which we leave the floor is considerably lower than the one we use to fly through space. So it seems to me that these action between 1628 should take place in a considerably shorter amount of time. So I will select all controls, grab all the keys between 2855, use the middle gizmo here, the two triangles there and middle mouse drag them to the left until there is much less time between the anticipation pose in the kickoff posts. I want this action to make some sense, intentional speed. It doesn't need to be super precise, but at least it needs to be somewhat believable. And I think this kind of jump now makes a lot more sense, glitchy. But you see that between 1622, we don't really travel that much. A cheap way to try and understand what goes on with the weight of the character could be very simply to create a There. And if you have watched my previous walked tutorial, you will know where N going with this. I'm going to assign this fear and read material. Now that I have the red sphere, I'm going to select first the sphere, then the body control. Then I'm going to hit be like pepper. And that's going to parent this fear to the body control. You see now the sphere is part of the week. I would like the sphere to be aligned to the body controls. So with the sphere selected, first, I add to this lecture on the body control, and then they go on their modify much transformations, much translation. There you go. And now as the cathode goes around, you'll see that the sphere goes around as well. So quite possibly I can now grab the Geometry Group of the character, hit age like hotel to disable it. And now I'm going to press play. And you see that the sphere in there going a bit all over the place. I mean, the timing is sort of okay, but the arc is all over the place. Look at that, it doesn't quite make sense. So maybe can go under the animation toolkit, visualizing got selected. And that's going to show me the trajectory of the sphere. And you see a trajectory of the sphere isn't really that nice. Maybe one thing that I will have to do in here is to work a bit more and refine that arc. So we'll show the geometry of the rake back. I'm going to grab the controller of the body. But you see now this fear is a bit in the way, so I'm going to temporarily scaled it down a little bit there so that I can still select the controller of the body. There. There you go. Every time you grab the controller, it will look like a sphere is selected because the sphere is a child of the controller. Remember, so now I can better design and intermediate posed for these characters. So maybe I can start going up instead of just down. And you see that the arc now will make a lot more sense. But you see that after post 22 we start flying up quite a lot there. So maybe in here, I imagined I could lower the muster a tiny bit just to see what happens to the trajectory. And the trajectory in there is a lot better in my opinion already. However, now that I have somewhat fixed the timing, I have another question for myself. While it is true that I need that kickoff pose for the right food, chances are that difference in kickoff timing between left and right food is going to be a couple of frames only. So I wonder if I shouldn't be focusing instead on fixing the body first and then finding out where the kick-off of that control is. After all, that kickoff pose is going to be a byproduct of where the body is. So it probably follow the same workflow I usually follow, which is body first. So first I will fix the body and then it was fixed a feat. So instead of going for a kickoff, posing there, so instead of aiming for the kickoff, I'm going to aim for a good body trajectory. Also, there is the fact that the legs are going to be pushing, but it's the arms that are going to start the momentum probably. So I will have to keep that into consideration in order to have better tracking of the trajectory of the body control, I can make sure I mean to the animation tool kit in Maya and go under visualize, create editable motion trail. And that's going to give me the trajectory of the body. And you see that so far so good. Maybe I can delay a little bit the body at the beginning so that the body will really pick up speed once the leg stars to be really, really extended at the beginning is going to be harder to move a, a body from a static pose. The closer we are to the anticipation, and the harder it is to move this body because of course the action just started, then we will speed up. And that speeding up that we do at 22 is going to increase our momentum. And it's going to lead us fly through the air and then land down. So I'm going to check if the speed makes some sort of sense. And you see that it starts to make sense. It's not exactly what they want, but it starts to make sense. That means that this food will be on the floor. They go and you see, don't we love the automatic pole vectors moving with the food. So I will need to find where the pole vector is gone for that leg. Are you the pole vector? No. And the pole vector as magically gone up here. So I will move it from forward. So you see this blue vector is starting to become a bit of a nuisance, to be honest, not only that, but if you look at jump references, you will find that the arms move first because they start to introduce them upward momentum. That upward momentum is going to help the legs extend. This is interesting because in our animation we didn't quite think in terms of that. So I think all in all what we think could have been a kickoff posts for the right food really is going to be something else. It's probably good to act as an extension of the arms on the way forward rather than any figures. So I am going to think in those terms. So I'm going to go to frame 19. I'm going to delay the body because at the end of the day, the body will move just ever so slightly because the arms are doing their motion first. That's going to be the first movement we are going to see. And that means that these are nice to go down here. You should think of it because it needs to follow an arc, a downward arc then will go on the way up. I will need to find the pole vector for that arm. This guy's going to be fun to interpellate that. And then I'm going to pause the arm there. I really want the fingers to be likely we're in anticipation so close they go. If they aren't went down and forward, then the clavicle will be down and forward as well. You see now the arms are actually helping the body quite a lot. We can maybe rotate the body tiny bit more forward. The chess could be oriented counterclockwise in here to help the arm extend. Maybe delayed the rotation of the face. And I can certainly keep both feet on the floor like they were. I would have to go and search for the pole vector of the NEA, which keeps moving all over the place because of the fact that it's following the food. I can roll the food a little bit and I go that looks OK. Create a little bit of a line of action that can then inverse with the kick-off. So the line of action could be this way and then I'm going to reverse it for the kickoff. For the left arm, I am going to have the elbow lead on the way up. So you'll see the elbow is also helping propelling the body up a little bit, not that much related. It could delay the hand on the way down here. Maybe that will work. Let's have a look at the action there. It starts to look like something. You see that it's the arm that is pushing forward and upward. Mets, exaggerated even more. I will need the clavicle to be up. If I blend to move the arm that high up, I will need to have better control over the elbows so we'll keep the pole vector a bit closer to the arm there. So there you go. So now the arm is pushing up. Maybe I can help with the effort and twist the head lower a little bit. This is too much, sorry, they go something like this. Just twist it lower. I want to have a look at the shot count to see what goes on in there. And not quite happy with having a key pose in which we covered the face entirely. So we'll move the arm up and I will maybe see what happens if I have. The left arm following an inward arc on the way up, so it will start moving up by you see from the insight this way, then the elbow will start the lead. They're not quite sure this is going to work, but I'm I'm very curious, so I want to try it. I just don't want it to arms to move together. And we probably need a breakdown somewhere in the middle here, because you see that the arm is just linearly going to the next bows, then we extend. Now I want to see what happens if I set a key and every control on 19. And if I go back to step and I expect that first part of the jump to be clear. You see that it starts to become a lot clearer now you can clearly see the arms leading the auction. They go, that's quite cool. I think that maybe we could go a bit more backward, 11. And this way we will go down with the body a bit forward as we go down. So that when we move forward between 1619, the forward motion won't be a surprise because we're already in a forward arc between 1116, although a very shallow one. Admittedly, now that I've added this breakdown pose at 19, I can select all controls, go back to auto and see which other transition needs my support, or rather the support of more poses really. So in here we have the kickoff pose for the left food. I think I will end up having the kickoff pose for the right food at around 20 or 21 in the end. Now 24, I have a bit of mixed feelings because one of the issues I have with this animation is that we don't quite spend enough time up in the air. It feels to me that we are just way too fast. So I will maybe want to add one more opposing. There's, so first of all, I will try and generating opposed by having all control selected and this like the master and the body control. And maybe at 28 I duplicate the pose. That should give me a pose which is identical for the limbs but somewhere else in space because these two controllers are moving right now, they are not, but that's a glitch. If I move the controller a little bit and undo, you see that they keep moving. The general arc stayed the same, and now we have twice the same pose in time if you think of it. So now we'll add a breakdown between the kickoff pose for the left food and the AP photos that we have in here. So what I would really like to achieve in here is to get the audience to notice the stretched posed on the way upward and forward for at least a couple of frames. But the problem is that at 23, you see we are already bending, or a 24 is Well, you already bending all over. I think I would want to try keeping the line of action straighter at 24 by holding the food there. That food, of course, and maybe delaying the tip of it. The ankle is a bit broken there. So I probably want to rotate the way that looks a lot more interesting. And maybe to the back that should give me a stronger impression of push if you think of it and it should keep the line of action stronger and straighter. Number that works quite well as far as the right leg is concerned, I can maybe bend it a little bit less than 22 and then 24. I can rotate the tip downwards this way. And you see that gives me a more dynamic pose, especially from this angle, it looks quite cool for the legs now the left arm was moving up quite quickly, so I want it at 24 to be essentially fully extended already. So instead of having just a partial extinction in there, I want it to be already up and helping. Otherwise he will feel it too slow. I think I will remember to put the pole vector a bit more forward in there. And in here I want to check the post. I think this could work and then maybe I can keep the right leg a bit higher for a bit longer there on 28 doesn't move quite as much the right leg, but the food keep rotating, thus giving us more dynamic impressions, so to speak. I always want to check the pole vector in there. You see that this pole vector keeps moving along that place. I will really have to do something about it because I think it's a bit too active there. So at 26, I definitely wanted to be a bit more inwards and even the food, and you see that the food keeps moving and it keeps traveling towards its destination down there. So that's quite cool. I want to see what the left foot is doing on 26, sorry, on 28. So the left food was extended out 24. That's sort of okay. So now we have to extend poses that this is really great, that you see that we get at 26 is already retracted. So maybe we could make sure that at 26 it's not fully retracted yet. And then you see it starts to accelerate at 28, a bit better, a euro, I think that could work. I will set a key on all controls, on every key just to make sure I don't break they poses in there. And then as usual, I will convert these dangerous the step and I will check the resulting stepped. Now please notice how I press Play and animation only to check if the timing works more or less and if the posing works more or less, but not really checking the spline or anything. I'm just trying to understand. Instead is this story told we've all deposits a knee. And so far it seems to me that the story is going in the right way to think of it, it seems to me that now the job has the detail it needed to be told. Whereas earlier on I didn't I didn't quite have enough detail in there. And another way to check this would be to select the tangents again and go back into auto. And if I did a decent job, I expect to see this blinding work more or less already at the beginning of the shot. You see that at the beginning of the shot, it sort of works in there. There's a funky behavior or the arm, but I don't really care about it for now. So now that we have sorted out the poses at the very beginning of the jump between the anticipation and the actual oppose. I think we can start figuring out how do we get down to the dissipation? Because if you look carefully at it, this is just a linear transition. There's no priority or anything. The whole character is just moving all the way down the food, sliding the arms opening up everything together. So in the next video, we are going to add the additional poses we needed to make sure that the story, the transition between the initial post and anticipation pose is properly told. 22. Breakdowns Leading to and from the Anticipation: So let's have a look and see how we can improve the transition between the initial posts and they anticipation pose. So first of all, one thing that we need to understand is that we can't quite move, uh, food for as long as the weight is on that food and right now, the weight is definitely on that foot. We can give the first boasts a bit more time, so we'll duplicate it a frame for, for instance. And then I will use the twin machine to make oppose that may be a little bit more similar to the next post. So you see that we started already moving down, but not quite as much as just a little bit of movement there. I can't quite move the right food in there. So we'll keep the poses it is because the weight is on that right food. So as I go down, I will help the movement on the right foot by moving the weight a little bit more onto the left one. And as I do so as I do so I would probably have to counter the chest counterclockwise from this angle as the weight translates onto the left leg even more, I will have to help with the pelvis a little bit more. Now as we go down this way, that means that these foods can now lift. So in fact, I could maybe increase the football ever-so-slightly as we go down. And that will probably justify the next transition, which is with the food moving down. So this is going to be my first fixed to that pose. Then maybe it will keep the right are more or less in placing there. Whereas we translate between 14 at the end of the day, not much is going on there, so maybe I can delay the arm a little bit so it doesn't move one-to-one with arrest, the lifetime as well can be delayed. So you see that now the body is going down, but the arms that are a bit late in relation to the body, maybe I can also delay the rotation of the chest forward as we go down so that the bulk of the action will be played within the feet and hip area. I want to make sure that they had keeps looking at where it's supposed to be looking. If your weight is shifting on one leg, the head will try to compensate and go the opposite way. So you see we keep the head bit more camera left in this case. So now it becomes a bit more of a continuous action. You see that from the front view, the right arm is moving one-to-one with the translation to the right. Maybe I could delay that and transition a little bit more. You see that now it looks a lot more realistic. Similarly, I could grab the left arm and delay the left arm as well. That looks like something. Now as we go down, I can go at ground frame seven and I can maybe move the weight a little bit more onto the left leg again some more. So that should let me baby moved the right food finally. So we'll try and move the right food to the back and not quite happy with the fact that the tip has moved up. So something like this. And now let's have a look and see if that works. So first we shift the weight and then we start moving down and the right foods can finally move and settled down. You see it's already settled down by the time we are down, which is great. Let's have a look at the body. Remember body first. So maybe I will delay a little bit and even stretch a little bit the spine on the way up there so that we bend the spine a bit later once we are down in the contact. And maybe you can even move the body a bit more forward. And let's have a look at the arm in there at seven. I think at the beginning the arm should maybe be delayed a little bit. And then as we go down into the dissipation, the arm will start moving to the back following a nice arc. And then it will start propelling the body forward. That seems to be reasonable at the beginning as there are moves forward, I can lower the clavicle a little bit and move it forward a little bit. I want to check the left arm and maybe the left arm can move a little bit higher as we go back down. Let's find the middle ground. There they go. And maybe the elbow or the left arm can already be active and start to open up. That could work. I don't quite like the fact that the post seems to be fairly static from the front and there, she looks a bit like she's holding an invisible motorbike in there. So let's have a look at the post. You'll see that poses fairly static even from the camera. So maybe I will do something about this silhouette. You see that now that we have to create breakdown poses and we don't have any sketch to resort to. Things become a lot more time-consuming because we didn't make too much planning. And this is something to be learned as well. The more time you devote to planning and the last time you will be spending in the 3D software, fight to understand what is there I pose to do at any given time. In here, I'm breaking the symmetry between the two arms and I will make the right arm go first forward, m may be downward and the clavicle will help, and then it will start rotating to the back, you get something like this might work. Maybe you can delay the head ever-so-slightly. I wonder if I can rotate the whole body forward and then counter at it the chest backward, maybe that could work if their right arm is moving forward a medic and twist the chest to help with that rotation, maybe the left arm could go even more backward if you think of it, the right leg is forward, which means that typically they left arm will move backward with counteract there. The only thing I don't like is that we lose the silhouette entirely from the cameras. I would want to make sure that I still see the ARMA BET better. So we go down and we reach the posing their nice even now, it's not very clear how we get from these bows to the anticipation. If you ask me, as usual, I want to set a key on controlling that pose. I want to convert the tangents to step. Now that I press play with the step tangents, I start to see how the action unfolds and that seems to be going in the right direction. I just feel that the right arm is maybe moving a bit too fast forward on frame seven, and then inverting a bit too fast as well. It's still moving very quickly on 16, and then it's already changing direction quite quickly on the next post, which for me it's a bit problematic, so I would probably need a breakdown somewhere in here to keep that right arm a bit higher up at the back before it changes direction. So one thing that I could possibly do in here is to make this transition between 711 a little bit faster, even just by one frame, then I will make sure that we are back into auto tangents so that we can benefit from the interpolation of Maya. I can go to frame 13 and at oppose in there. So on frame ten, what I want to do is maybe delay the chest ever-so-slightly, so that between 1013, the chess, we'll do a bit more of rotation. There you go. And you now can tell their additional chest is doing and maybe I will keep the chest rotating at 16. There you go. And then I will also want to delay the head a bit. So we go down to ten and I want the head to be ever-so-slightly higher up. You see that now between 1013, there is lot of head movement going on there. And again, little bit more head movement between 1316. So you see we keep these character alive. Remember the whole point of these new pose was to keep the right arm bit longer, closer to the top so that we had time to invert the motion earlier on. So we'll lifting the right arm quite a lot so that the main motion that we will see between 1316 is not going to happen at this level of the arm is going to happen and deliver the forearm and maybe the priest, there you go. You see that it's the forearm that is doing most of the job in that transition. I will maybe delay the arm on the way back. A little bit more, going to be tricky. Possibly work. And then in here, this transition at the arm is just way too fast. I think I will like to make this a bit slower, so, uh, we'll probably have to move all these keys to the right by one frame or so, so that I can go in the middle between the two key poses and make sure that the wrist is doing something that makes sense. You see that the rotation at the wrist a bit all over the place in there. So one thing you want to do before you add more poses, in my opinion, is running what we call a new layer filter or I think in Europe you would call it an Euler filter. If you go under graph editor, you would select any key from the curves are the raised. And please notice that we do not have a rotation key on 18, we just have translation. And notice how all of the curves are changing direction in here, all of a sudden. So we want to make sure we select a key from each curve. Any key will do and go curves. You Euler filter or Euler filter, depending on where you're from, you click on it and Maya would try to recalculate the rotation. You will see what really changed the poses. If I go to this bows and I go curves, you will or filter. The post does not change, but the way we reach it in interpolation does change. You see it's different there. It has to do with what we call the rotational order, which is a jargon term for something which is very important in animation. But right now, for this tutorial is maybe a bit too advanced. I will maybe make sure that in 3D, the arc of the hand make some sense in terms of translation. So maybe the hand will move a bit further outside to come back in and it will keep moving back in as well as it starts inverting. It can maybe come back out little bit, will slow down the hand moving up a little bit, elbow and there is a big gorilla like so maybe I want to route the pole vector. Whoa, maybe I want to grab the pole vector there and make sure that the pole vector is helping but not giving us that gorilla pose. I want to maybe delayed irrotational, raised a little bit. So you see that now the Arc of the arm is a bit more taken care of is see there's crazy interpolation going on year. And that's because if you go gimble, you see in here the interpolation sucks. So maybe I can go into the graph editor. You see we have that crazy variation of rotation. Again, I can go curves, you there filter and you see that now the interpolation is much better and the curves are more continuous. So the Euler filter is really one of your best friends in here. I will maybe delayed the arm a little bit more and I think I need even more clavicle in there at 20, even if the clavicle for one frame goes into the color for as long as we don't see it, we should be OK in here. Probably we're going to need to lower the clavicle as the arms wings lower, which will give us a bit more work to do on the ResDAC. Now we can move it even lower. That should sort it. Now let's have a look at the opposite arm and see if he needs any help there. The opposite arm could maybe do we have a new layer as well? It's a look and see if that fixes stuff that you're much better now already. So maybe I don't need to do much for the other arm. If I said a key on 1800 controls, I can click on stamped. And you see that now if I play the blocking as it is in step, it's not clear that the right arm is used to propel forward and upward. It's a lot clearer. In fact, we are getting somewhere and now that we have sorted out the first poses that lead to the anticipation there. Now that we have sorted out the posing for the first part of the animation, we can sort out the second part of the animation, which is from the up to the final pose. So in the next video, we're going to see in which way we can improve the look of that part. Because right now if you go back into water tangents, you will find out all sorts of crazy stuff going on in here and we don't really want that to happen. I mean, that doesn't look really good. She probably add samples, is there right. If we were to spline this thing, I think we will get more headaches than anything. So I think it would be good for us to think about next step. So I will see you in the next video. 23. The Left Contact: Now that we have the initial poses jotted down, we want to fix the second half of the animation. If you consider that we are looking at an emission in step tangents, you can tell that the innovation is very clear at the beginning, but towards the end, that becomes a bit less clear. You see that there's a bit of a hold at around 29, you see? And I think at frame 35 as well. So before I'd imposes just like we did earlier on, we want to maybe speed up these animations. So I have about four frames from the AP. So maybe I can keep the app a bit shorter, say to free frames shorter. So I wanted to check what happens if I go in just a frame to the extended pose. That's probably too fast anyway. And then I want to move this reframes again to the left. Now, wanted to press play just to see what's the speed? Well, actually you know what the speed seems to be about, okay. But I think we could possibly keep the bows up here for a bit longer. It seems like overall it works, but it feels a bit too fast. So maybe I can slow down a bit. The transition between 2931, if you think that the transition between 2729, I had this little amount of spacing and then between 2931 we almost double the speed. It seems to me that maybe we could just move everything a couple of frame to the right and give the transition between 2931, couple more frames. And I think something like this could maybe work a little bit better. It's more in line with the timing we had so far. If we have this little gap in just two frames, then it's reasonable to expect almost twice as much between 2933, which is twice as much the time on the way down. I think we might need a breakdown between 3336, a breakdown as in an intermediate posts that helps us. They'll How do we get from key pose a to keep us b? And I'll make sure we are back in 02 tangents there. Any year you see that we have these little a problem which is the hand going the hunt very quickly to the floor down there. One of the issues you're going to encounter as you go into spline in here is that if you check the parents space or the hand, you'll see that it's set to body and eventually it becomes world. Not all rigs can deal with this transition graceful, especially in this case, you see that the transition between body and world happens at, around these position in time. So I think we could probably fix this post to begin with. And then before going into splicing, SET that space to be always body instead of transitioning. So this way we should have an easier time blinding. Now I will move up the arm there and move down their relative pole vector. If this is the contact with the right food, I expect the next post to be flat on the ground. Let's do it. I will just duplicate the key post 235, and then I will roll down the telomerase, roll down the food troll so that the food is flat on the floor and 0 out the teal. I will probably have to lift up the controller because it seems to me that it went a bit too deep into the ground there. So this is going to be the transition I expect. And then of course, the food will roll back forward. So we'll probably fix this position a tiny bit to prevent the tip of the food from sliding. You see the tip of that food is lighting quite a lot. Their fingers I healed BY what things will use the healed BY what on 35. Still, I will probably have to go in there and manually and just tweak the post so that it will feel like it's sliding less, they go something like this, I suppose. So that's sorted. So now we have the DOM posts for the food which we really, really needed. I guess we could also decide the arc for the body controller. So we'll just create a nicer arc on the way down, right in the middle, maybe a bit closer to the top because we are going to accelerate as we go down as a press played and emission should play nicely. Auc thus play nicely. Now it's kind of cool actually. So that sorted. Now let's have a look at the left food. What should the left food be doing? So we have the contact on 33, maybe we can have the left foot contact a couple of friends later. So first of all, I will use the Posey 37 when the food is on the floor and copy it over 35. So this way, the food is on the floor. If I add these artifacts, remember move undo and that sorts it. Contact should happen on the bowl of the food. So I'm going to raise the bolded foods, but you see, I don't really need to raise it as much. They wait is probably going to be a bit more onto their right leg right now because it's the first one that touched the floor. So I'm going to lift the hips onto the right leg a little bit more and I'm going to bring the food a bit more outside that, it's already going in the direction of the lending post. Maybe you could use the bulk p, but to add a little bit of rotation, you see on the floor, I didn't know which direction to go, to be honest, maybe it will keep the heel out and then have it coming later on. So now we have even some more rotation on the food there, which is going to look hopefully decent. Let's have a look at what the spine should be doing in this transition. So the spine is probably going to help the hand reached on the way down. So I will twist the spine ever so slightly and then I will reach down with the arm. So maybe I will first lower the clavicle and bring it forward to help with the extension of the arm. And then I would try and bring the arm more or less where the arms should be going or radius, so it will be ready straight, I guess. And maybe we can extend the fingers quite a lot in there just to give this feeling of trying to reach for something. Every single file line seems to be rotated more or less by the same amount. So we'll probably want to create some variety there and exaggerate the index rotation on the way up will exaggerate the rotational, the second phalanx of ring and middle finger. Then I wanted to try and see what happens if I make a Spiderman pose there. Let's see if it looks decent and then I will move out the quite a lot. In fact, I should give me the idea that the Handy's trying to reach for something that you, I guess something like these could work. I always want to check that the recent is not broken and you see that it is quite broken because if you check the elbow, the elbow is pointing out quite a lot in there. It's not really promising. So we'll move the elbow controller but more to the back. And there you see now the rest is a lot less broken. Let's see from the 3D view where the hand is going. So it's going towards the center. So maybe I can and point towards the center already so that it will go from the outside down towards the center. And then hopefully diagonally, I wonder if I can lend a bit more centered in a bit closer to the body there, let's have a look at what the left arm was doing. So that Don was actually, you see, was winging down. These architects were using that aka the left time is simply going to linearly translate to the new position. And I don't want that. I want to control the arc in there. So we'll just rotate the hand to follow the arc in there. And we'll go find the pole vector which has gone down here and position it so that it makes some sense, something like these I suppose. And what I really want to do is the arm to start this wing on the back. So we'll delay maybe the face. See it starts to swing back. By swinging back, it prevents the body from falling forward further later on. And now we have some sort of decent are going on there as the arms wings down, I want to lower the clavicle and not very happy about the landing spot in here with the left food. I feel like I could have landed a lot more forward. I think if we lend a lot more forward, that leg could help quite a lot to prevent the body from going forward. The more and look at it and the more I feel that left foot should lend a lot more forward, something like this, for instance, I will try that will reduce the foot role to 037. I have gotten rid of all the keys after 37 for that food. I don't really need them for now. And I want to check from the back if the character is falling, you see that the risk is that the character may be falling there. So one thing I could do maybe is moved the food a lot more to the outside so that the character won't be falling anymore. Anything dispose would look probably a bit cooler than the previous one I had. It it can tilt ever-so-slightly there. Yeah, I could work. It could work. Now let's have a look at the transition as the left food travels forward, I can maybe keep it a bit higher after all. It's going to land a lot later as him to friends later. And I'm going to keep the knee closer to the body and maybe at 35, I'll try to contact with the left food. That means that I could either use the twin machine to create that pose. You see it's pretty far forward if you ask me, let's see if we can get there. So maybe I can rotate the hip to favor that extension. You see wooh, it's long way home. Long way home. Maybe it's going to take us a little bit longer than just two frames to go from this bows to the contact. I can check this bows and on 37 helped with the pelt is rotation. In this situation, I maybe they want to use world rotations that I can rotate the pelvis parallel to the floor and they should help the extension of the leg quite a lot. So let's try sort out these bows. Here is, simply speaking, it's too early to contact with the tip. I think if anything, probably we want to contact at 37 now or at 36, if we're lucky, maybe you could just move the food a little bit to the back there and make that lending posts and it will look a bit more. And then a trained 35, I guess I can try and see how far out I am from the leg being stretchy and maybe settled for that poses that that means that the contact will probably take place at 36 if I am lucky, let's see. I think at 36, If a duplicate the pose you see I can finally contact with the floor. They would lack the contact pose to be extended. So I will try and 35 to extend forward as much as again, even if that means threshing leg, I would much prefer to land with the teapot there, but I think it's more of a dream than anything else. Let's try and see. There's no way I can then with the bowl of the food in there. So we'll just try at 35 to keep the leg extended. That 36, keep it a bit more extended. Roll up the food, rolled out a $100. We goes that it doesn't look too strange at 37, Copy that poses at 36 and withdrawal back to 0 and we go back to 0, they contact will happen at 36, more or less, will have to reframe a little. This is an experiment. I didn't think of it when a test round these tutorial, I didn't think of doing it this way when I am looking into it and I think it could be cool to have that pose or normally also I don't animate frame by frame, but in here I wanted to figure out that contact posts. Now let's go at 35. We did the arm extended down at 35, so that's sort of cool. They're not sure it's a good idea anymore to extend the arm because they will make the transition between 3335, a stretch for both the right arm, left leg. So maybe you want to stretch their right arm a little bit less so that it will feel like Maslow emotion comes from the left leg. That means that I will have less rotation with the chest. I will keep the head a little bit closer to the body there, and I will keep looking at the destination with the head. Now I want to check the left arm. Again. I didn't think of the legs and now everything is changing. So this is the cost of improvising. But you see very often as you travel, you make your plans. And then once you are in the place where you want it to be, you realize that you're interested in something that you didn't think you would have been interested in. So right now I disposed that I created, which is changing the pose I had in mind the pose F-test run for this tutorial. And now I feel that probably a preferred the new post if compared to the old one. Also, I don't want to have the feast in there anymore for the left legs. So maybe can grab the finger controls, go open retina, right? Nine animation toolkit. And maybe at 35, I can symmetrized the pose right-left. Gotta select the controllers right to left. Maybe you could do something which is after weigh between open and close lines pose there. So we'll just copy 3435. So that will give me, You see, a face that is a bit open but not so much for this glitch. Move the controller. And now maybe I can open up the index a little bit more. And the last phonics and the indices a little bit more just to create some variation. And that should give me a more natural post, which probably I preferred to show the audience this way rather than frontal. And also I don't want the arm to be perfectly flat horizontal to the floor. So that's going to be maybe, maybe my post. I'm improvising a bit too much near and now I have that breakdown that I was looking for. I still don't have a breakdown at around 31, but it kinda think that I might get away with just explaining it. 24. Blocking the Follow-Through Motion: Now we touched the floor with the left leg and then we want to touch the floor with their right hand. But maybe we can make a contact with the right hand that is not entirely flat on the floor to begin with. And then as the weight goes onto it, we press with the hand on the floor. So this is going to be a bit difficult to be done without tools. If we had uninvolved installed, we can use an embodied to rotate the hand from these pivot. Right now. We kind of can't do that, which sucks in many ways because I can't really show up paid for two Lynette tutorial like this one. So I will raise the hand in there, then I would lower it down through rotation. So the Kotok will have the hand there and then you see will go flat on the floor. And then maybe I will invert the curvature for the fingers so that the transition we see each finger inverting the curve there is when you need IK fingers really ego something like a vase, I guess I'm making it a bit more extreme. Maybe they could even push up this guy that starts to look like something. This kinda stuff takes forever. If you don't have the right finger set up. And it will certainly want to invert the curve on the pinky that you associate another fingers are taking a bit more weight on. Even the thumb will go down that way. And then as we press against you see we'll open up to the floor. And the other thing you can do is the floor with the fingers a lot closer to one another. And to do that, I will rotate that in world so that they will always be on the floor. You see, I will make the fingers and even these control here a lot closer to one another to begin with, as the weight goes onto the fingers, you see that the fingers spread open and that's going to look a lot more believable. So now even from the distance, we should be seeing the fingers taking care of the weight is kind of cool if you think of it. Don't really need any key after 39, that's it. I need the fingers to stay there at 39. There is a bit of motion there you see? And that's because the arm is probably stretched. If I move down the clavicle a little bit, the arm, It's going to be that stretched anymore and the hand is going to stay there. Now let's have a look at which other poses we need to do in their handled. Like suppose I have at 35, It looks like someone is about to Danes. Maybe I should reverse the line of action of this binder. I would try that trick. See who works. And I think it looks better than before. Earlier on it looked Redis different vertical. Now it's a tiny bit better as still unlike the position of the neck there, I will try to 0 it out to see where it's supposed to go. And even for the head that will try to 0 out the translation to see where is that supposed to stay. And I think that position looks better ready? I could cheat a little bit and delay the translation of the head and delayed even here, I would grab all controllers and Sadducee at 351, at 371, at 39, and then 141. I will go back to see the animation in steps. And I would want to press play to see what's going on there. And I think the front leg is kind of cool and dynamic and like the weight slides into position, I feel that the left-hand needs some love. The NDC, it lands into position but it starts to stays there. And I feel there's a need for one or two poses at the end after we touched the floor, I can feel how cheap. My choice of poses was at the end. Remember I was playing that cheap trick there as the weight goes down, or probably tried to have the weight go a bit forward between 3741. So we'll lend a tiny bit more forward at 39, I will go even more forward and lower if again, then at 41, I will move even further forward and then I will come back a little bit. And you'll see down there, I'm making a little bit of an arc and I wanted to see if that looks a bit better. And I think it starts to look a little bit better. I feel like I should be delaying the upper side of the body so that it doesn't move as much as the root. And then as we go into 41, that's when I can maybe start rotating a bit more and then I will settle back, probably come up ever so slightly at the end. But that tells me that between 4146, I probably need a little bit of love there and sort of pose. So if we reached the bottom at 41, maybe as we come back up towards 46, I could create a bit of a delay in the upper part of the body and the head. So we'll do that and we'll select all of my controls again. Please notice haven't touched the single curve yet, I am still posing the character really. Now, as they go past 41, you see that the whole character is linearly sliding into the next post. So maybe it could go to frame 43, set a key. And in there, I want to try using the twin machine to see what I get out of the box for the upper part of the chest, I want to get the pose, which is a lot more similar to the previous pose, but you see very little going on is going on up there. So I think I will have to think about it instead of using a tool. So we want the chest to go down at 41. And as we come back up, I want the chest to keep going down. So now you can feel that the bomb goes up thirst and there is a little bit of a reverse of curvature in this spine that's going to look cool I think. And then finally, they chest will come back up. Now I want to see what happens in there. It's probably a bit too fast. I will need more pauses because as the chess comes back up at 36, there's probably going to be energy for another bounds. We'll duplicate 36 to say 50. And at 36, I maybe want the head designed to be delayed. So if the chests were delayed on the way down at 41 and it keeps going down at 43, then there had certainly must be delayed as well. So I can maybe rotate the head upward a little bit, the lead frame 41 and see what the head is going to do. It's going to come down at 41. I probably want to try and cheat and just translate the head a bit higher up. Tough game recurs. There are intersections, their head is going to be delayed. Then at 43 is going to come back down. It can delay the rotation as well. And 43 is going to come back down. And at 46, as we go back up with the body, we want to delay the head quite a lot in rotation and also with the neck, so that now we are introducing a different motion you see between head and chest. So the head is a bit delayed. It will be the last thing that settles. And finally, maybe at 50 as the chest goes up, I want the chairs to keep going up ever-so-slightly. I don't really want the body control to just die Amin here. So maybe we'll drop 45 and copied over 46. That means that now I reach 46 and keep going a little bit with the body so that I don't kill the character right there. I don't freeze it. I did everything correctly. I should get a bit more softness, bit more alive. At the end you see we start seeing the head bobbing independently. I think the problem here is that we reached the bottom of the motion, but then we immediately come back. So maybe what I could do is reached the pose at 43, but stay there a bit longer. So I will move the keys couple of frames to the right and then I will drop 44 and paste it over 43. Now we get down the chest and neck, keep going down or to do probably is to keep going forward for a little bit longer. With the body control, they go, maybe something like this will work. And in the end, we don't go back that fast with a body if I can't hide their head for a second, see that that's a tiny bit better. I will have to fix the head. Now, whenever you play with the head that you starts translating it around, it's very easy to disconnect accidentally the head from the body. So I will want to make sure that's not the case in here. Who will start deleting 411 of the problems is that we are going down pretty quickly with the head, but there is no rotation. So I will maybe delayed the rotation on the way down. And I can even delayed and knows little bit so that we go down with knows a bit more camera left. And then as we reach the bottom, you see that noses rotating forward and then maybe can keep rotating forward a little bit and upward as too much. Maybe you don't want this Bose either. I'm trying to have too much grander like controlling here. Now I want to check the shot gum. It's sort of OK. Now are not quite happy about the focal length of the camera because I feel like it's distorting the character quite a lot, especially closer to the camera in here. So I will unlock the camera temporarily and I will set the focal length to say a 60 meals. We get a bit more of a zoom effect and they're not very happy about the neck of the girl. And there, if I 0 it out, it's still too long. So I will probably have to do something about that sooner or later. Let's have a look at the poses. The problem is this pose at 43. Wonder what happens if I can leave it a bit higher up for a bit longer as well. In here, your animation is probably going to be very different from mine thing. This works a lot better when they had goes down and then it comes back up. Now the weight goes down and here is a bit too much, so we'll reduce that quite a lot. And even here I think you can reduce this both quite a lot. I think I should be going up, I shouldn't be really be going going down in here. Let's check. Yeah, I think it should be going up and you go and that's sort of okay. There's still a bit of this connection between the head and the body there. Yeah, you see it's a bit disconnected. Probably this last rotation in the end. I don't think we really need it and neither do we need that much sideways rotation, I will check again and this is something that will have to cleanings blinding because you see the motion is a bit uncertain for the head. So now we have done older poses. I think let's have a look. Set a key on all these bonuses before we can go step and we'll save go step tangents. Now we want to check the blocking. Well, why are we not instep tangents? Go step tangent. And now I can have a look at the blocking as I wanted it. And you see that now it looks a bit cooler as an animation. I have all the effects I need at the follow-through. I will need to clean up the interpolation of the arms, of course, because it's not very convincing yet, but only know the animation is there. And I really like the way the left food is lending a while later than the right one. And it's used to prevent the girl from falling forward. Thing that's really cool. The left arm is hard on the back there. Figured this stage I could have probably kept the left arming FK after all is not contacting anything. So maybe one thing you can do is I could delay a little bit of the arc to the back with the left arm. Yeah, I need to find that all vector, which is this guy a guess I needed to be a bit further away. So I'm going to play a little bit of the left arm. So that means I can maybe try and delete these Basel together. And I really need to be noted tangents right now because I want to see the interpolation that Maya gives me when I delete keys and I want to fix that interpolation, I think I can definitely reach this. Bose may be a bit slower. These could maybe work then I want to overshoot it. I want to go past that pose a little bit and then I want to fall back a little bit as well here I don't want the arm to bend too much now I want to check that army motion. You see that the army is now a lot softer up there at the back. Nice, nice and soft. There's a bit of gyri motion at the top with the armed with their right arm moving back and forth in there at something and can probably fix his blind. And they should think of it. By the way, we're looking at these animation for which adjusted the blocking. I just didn't even check the curves. You see that in practice, we're sort of done. I mean, display name could already work. Let's say that a client, that director or supervisor came to me asking to see the animation. I could already show them a preliminary version of this in which we worked on the blocking that was approved. They're long revolt opposes. We added intermediate poses and we already know that this blinding we work more or less back at when I was showing you how this blinding was working when we just had the key story poses. I'm pretty sure you will spot the difference between that's blinding we got earlier on and this one. You see how I prefer to call it the blocking only once I am very clear about how do we get from pose to pose B, I don't really want to go into splicing if I don't know how I am going from pose a to pose B because they find it that it takes me forever to design a pose in 3D if I don't have a proper plan in place already. So now I can save this file. And in the next video we're going to start this blinding. 25. Preparing the Rig for Splining: As seen in the previous video, we have all deposits we need to start blinding if you think of it, the animation is pretty clear already. The timing seems to be okay. I think it could be better. Of course, everything could be better in bed with the time we had at our disposal. I feel like we did a decent job. Now we need to set up the scene for lining so that we are going to do as little work as possible. That's our planning there. So let's execute our plan. First of all, I feel that the hand is a pretty good position, the right hand. But you see that every now and then I feel like the right hand is sort of moving in a strange motion. It it moves up vertically and then accelerates quite lot forward and then stopped and accelerates again. It's a bit erotic if I go back into splicing. So if I go into auto tangents for all the controls, you see I get this glitch, but we know how to fix it, moved the master and the master and so forth. If I go back and check the animation, you can clearly see that there is a double movement in the right hand. There is a bit strange to look at. I just don't quite understand it, but there could be a lot of stuff going on in there. For instance, it might be that this control is rotating a bit erratically and you see it causes their hand to move as well. There there are many things that could be at playing there and right now I don't know what they are. Also, if I look at the right hand, you see that there is this crazy long bows down there. Tried to go between parents space body and time-space world. So that's going to be a problem as well that I would really like to solve. There's also the fact that for some of the time, we will have the master moving the controllers around and for some other time we have the body control doing the same job. That means that if you go towards the center of the animation and you open up the graph editor, you will find that the body controlled counts to a mysterious halt in there. Not so mysterious once you realize that it is the master, indeed dead stars moving. So not incidentally, if you compare the curves of the body control to data the master, you will find that when the body controls starts freezing here, when the Master starts moving there. So that's something we really would like to fix. I also would like to make sure that the master doesn't change the speed forward to offer an abruptly. So maybe we'll try and see if I can just keep this curve with the Master. And I really want to see what's the effect on the trajectory there. I want to see how does the animation feel and it feels a bit weird. So maybe we'll try and delete only these key to see what happens in there. And I think that's going to make it look more interesting. Let's look and see. That makes it look a bit more. Okay. I think that if we track the motion of the pelvis forward, we will find out that the pelvis is moving a bit faster forward between 2023 if compared to 2325 and so forth. So maybe in a same, maybe we can speed up the whole jump in the middle on the way forward. And I think I will try something that might be a bit difficult to understand if I select all of my controllers there, I noticed that at 33 I have a key and that key is the last key which the master is moving forward. After that, the body control takes action. If I moved every pose from 33 on one or two frames to the left, I wouldn't be creating a big problem there because if you think of it, the master itself you see, is going to just compress the speed at which it goes. That means that maybe I could delete the keys for the translate forward the master We've all control selected. I could move their whole animation from 33, a couple of frames to the left. And now let's see if that makes it better. And I think it's sort of makes it better. Maybe now I and speeding up too much. So we'll maybe do a single frame to the left. And have a look at what happens there and I think it's still works. Ok. Now the issue and seeing is that really between 2325, I see as low down or the character, it feels like the cat that is translating forward quite a lot for the kickoff and then start the counts to heart there you see? So let's have a look at the master curve. And if you look at the master curve, you will notice that after the kickoff pose, the Master starts low on the way forward, but really it should be going at full speed. So I can break the tangents and move the tangent up to make the master go at full speed. And you see that it's going to change quite a lot, quite dramatically. The position of the jump, another jump looks a lot more realistic. So really the problem was the fact that the master was moving slow at the beginning. So they kickoff would speed up things and then the master would slow them down after 23. And then you see creates a bit of an issue because now it means that we have to somehow match the acceleration of the master forward with one of the main body. As it comes to heart, that's going to be super tough. So maybe we can think of something else. I believe we could think of transferring the animation to the main control and let the master be starting this way, we will have a single curve and we wouldn't have to second guess. The curves will need to employ to make sure that transition is nice and smooth between the master and the main control. And I think it would be doing exactly that. The thing that I will probably want to do is that we want to make sure that the pole vectors or the feet are not following the feet anymore. Because remember that gave me grief in so many situations. I don't really want that to be a problem anymore. So just like I did earlier on, I will select all the controls. I need to convert the animation back to the log of controls. But I also want to select, in this case, the all vectors in here. And I want to select often. The reason why I wanted to select the breasts as well in this case is because I fear that one of the reasons why the wrists are a bit erotic, there might be the rotation of the main body control. So I will want to control the breasts independently. So once again, I've selected the body control, both race controllers because I want them to move in absolute world space so that they are not affected by the main control, the pole vectors for the arms, because I need to select them earlier on anyway, that both vectors for the feed, because we don't want them to depend on the feet anymore and then defeat. Because remember, if I stop moving things using the master control and I use the main control, the feet won't follow. And I think I have all I need probably. I think I want to create a selection set just for safety and we'll call it cc, and it's up there and it's there on my shelf. And now I'm ready to try and convert the space, probably wanna save before doing that just in case anything goes wrong. And I'm going to click on world bake. And first of all, we saved innovation to the educators. They go there, we have it, and then we need to set up the scene. So first of all, I want these vectors are said to be independent from the feed. To do that, I will need to grab both pole vectors and set the parent space not to food anymore, but to world, so that they will move based on the world. So that will stay in that position I put them in. I want to break the connection because this way the parents BaseOne change back to body. Now 7y will break, but hopefully when I will restore the animation that will look exactly the same except there will be in world space. Then I want to make sure that the priests are also in world space. I think it would make sense to perhaps decide a strategy on how to clean this animation before we even convert anything. Remember that usually we prefer to clean the body first because everything else depends on the body. As a said, we probably don't need to worry too much about the fact that the pole vectors of the legs are going all over the place once a, set them back to world. Now we want to delete the animation on the master because that's the thing we don't really need anymore. So we'll just break connections on the master channels and other characters you see it's all over the place, stays there and that everything becomes crazy. She becomes a spider there. It's the big time we want to see what happens if I grab all the locators and go bake from locators and bake that emission back onto the object. Now we should have exactly the same poses or hopefully, and they look like the same with exactly the same province. Except now innovation doesn't happen anymore for the monster, but it happens through the body control in there. And now I only need to tweak the curves of the body control if I want to control the movement forward for the lady, which is a lot easier. I still haven't fixed the problem with interpolation of the hands. You see there suffering because of these transition from body to world. But for now I want to keep things as they are, because as it will be tweaking the movement of the body control, the hens will be affected. For now all we want to keep the hands to move with the body. Once I will know that the body will be working nicely, I will convert the hands, the world. If I already converted their hands to world now, I might have a perfectly decent pose for the hand in there. But as they adjust the body control, you see that I'm breaking the pose of the arm and that's something I can't really risk right now because he will mean more work for me. So for now, I'm going to keep the hands attached to the body. So now that we have the animation where we want it working the way we want it. In the next video, I'm going to start cleaning the body. 26. Splining the Torso: In the previous video, we converted animation from the master to the body control. Now it's time for us to start. Lining will open up the graph editor and I will want to have a look at the motion of the body to begin with, if you check the translate zed, which moves the character forward, Let's translate Zed is sort of okay, I think, except is maybe a bit too clunky in these are the screen you see there is a change in speed. So maybe we can just delete these keys and let the curve workout their magic alone. So you see, I'm just trying to understand the behavior of this curves and make sure that they look decent and smooth. Same thing down here. And for the most part I think we are. Okay, let's see as we go back out and maybe we can use this tangent to get an even nicer curve. And there you see that's kind of nice already. That sort of okay, now we want to check the Waitrose nations that translation on the way up and you see that the transition on the way up is sort of funky. So now I'm going to check the y translation of the graph editor. And on the way down, I feel we are sort of okay at the beginning as we go down into the anticipation, we did a decent enough job. Maybe they can keep the curve a little bit more harmonious in there. And then on the way up, however, that's when I feel that we keep changing speed and direction. So maybe I can find the top pose, which is this one. And I can just delete these key here to give myself a nicer arc on the way up. So maybe they go something like this will probably work. So we're good to stay up for a little bit longer before we start going down. I don't think I need this key and I think maybe go down a little bit faster there. Want to check this bows and see if perhaps it could go down a little bit faster. And in here, maybe I can push a little bit higher because I'm afraid of spoiling the contact pose which were supposed to be extended. Now I want to check that that works are right. And you see that now that jump works quite ok. There is a nice sense of push in there and the body is nice and clean. So you see your inks pipelining. You're not only tweaking the tangents, you're also deleting unnecessary keys that you don't really need down here on their rival. I feel like it's fair that we keep going down. I feel that maybe we could take a little bit longer to recover height and we keep going up in here. And then we could go down a little in the end. I'm going to check the forward motion as well. So you see how we keep going forward and year, which seems reasonable. Then as we start going back, we sort of hit a plateau in there. So maybe I could keep going back a bit more gracefully there we have a flat tangent at the end that you are, and that's a lot more believable. I think if I look at it from the camera point of view, a euro, please notice I'm only checking the body more shall struck the body. We did a decent job, so I won't go into create a sphere. If you have watched by Bruce Walker tutorial, you will know where I'm going with this one. After the sphere is there, I'm going to grab the body control thirst, this fear next in this order and go constraint, option of parent constraint. And I will make sure that maintain offset is set to off and heat. Now this sphere is going to follow the lady and you should think of it. Now you can tell if that sphere is going to work or right or not. I can actually grab the geometry of the lady and just hide it. And then I can just track the sphere and see if that makes some sense. And if that animation looks cool, if he looks clean, unbelievable, and it's easier to track it if you maybe use a red lumber there as a material because this way. Quite miss it. If that looks okay, then you're halfway through your animation because your body's moving nicely and clean and everything looks believable in there. I want to check the rotation of the body as well between 2631 or so, there is kind of a big rotation on the way back. So I would probably want to have a look into it. So we'll grow, grab the controller in there and I will go, and I would go check the curves and you see that there is quite a big deal of rotation going on in there. I want to use Gimbels and x seems to be the axis involved. And you see there is quite, indeed quite a lot of rotation going on in there. That's not necessarily an issue, but we want to make sure we know what goes on and if we want to know if that's what we really need. So we'll go and check that x rotation and see if that's something that maybe I can soft and a little bit before it becomes a bit of an issue. So I will go grab the rotate x. And this is the area where between 2930 toothless and massive amount of rotation there. So you see that the x rotation is sort of constant until between 2932. So maybe find, make the curve of bits move forward there. I will get as mu for rotation as well. And the body will progressively rotate as probably going to be a bit nicer to look at. Now in advanced Riggs when you are in production, you can hide legs and arms so that it becomes a lot easier to see what goes on with the body. I think this character has the ability to go into proxy, but I don't know if it works really well. So we've proximate, I can maybe think of hiding parts, but that means going and researching into rig and find out where the proxies. There you go. That's the proxy. I can go grab all the proxy groups apart from the torso and hide those groups. And let's see if I have the head as well that it can hide. Maybe I can also make a selection set and we'll also hide the rocket back geometry. I will also want to hide the gadget proxy and the head Geometry Group over make another selection set or those with a selection set, I can easily hide and show these Bart's. I don't want to see that leg belt that there isn't there. And this is this struck jail, so will temporarily hide it. In fact, maybe I can add it to a selection set again, we'll call it a strap onto the shelf and hide it. So now I can glean on in the body and I can focus on them getting the body really, really clean. There's something going on with the color in the middle of the colour goes up and then you see goes down a little bit again. So there's something going on with that rotation. So I'm going to grab the chest rotation in there. I'm going to check that box is moving it forward is the x-axis. I'm going to go into the graphite or in check rotate x. And sure enough, there is an inversion of tendency there. So the chest is indeed going down for the anticipation pose. And maybe it could even delay a little bit of that rotation just ever so slightly. And then as we are down into the anticipation, it keeps going down and maybe I could delay the chest ever so slightly on the way up. I could extend quickly. That's the kickoff pose probably into check. We just need to show the limbs and you see that's indeed the kickoff bows. And then you see the torsional starts going forward. Let's see how we can make this work. So we can maybe go back to straight a little bit slower than we do now. And then we can invert the motion a little bit slower as well. And then as we go back down and there, maybe we can delay a little bit in here. There's a lot of motion from the body control. So maybe it can delay a little bit more and then we go down. Probably you can keep going down little bit longer, a little bit softer, you see, and then we gently come back up. Now I want to see if this makes any sense. And you see that it does make a lot of sense. It's pretty clean in fact. And I wonder if I want to lift up these much, maybe can lift up a little bit less. If I press R, That stands for scale. I can Middle mouse drag from this pivot and you see I can decide how much I can come up in their strike and scale the curves ever so slightly. And that looks a little bit more realistic. I could even think of adding one more key and coming back down ever so slightly. There they go, Just to add a little bit of more motion towards the end. If I really wanted to, probably will have to animate something else, then you go and now that's a lot cleaner. I want to check maybe the other curves in there, the torso rotation y. I want to see what happens in there. And this is very difficult to track because you see we have really no indication of what's going on if you moves quite a lot. So maybe we could track it using another dummy object. So I'm going to go create a cube this time around is you think of it now, I can bet the troubleshoot that twist of the torso. You see that at the beginning is sort of a K, But at the end it feels very, very fast. So maybe I want to make it a bit less fast than there. I want to show the limbs in there. I don't want to break anything because it's very easy when you do this cleanup to forget that there are the leaves that need to work with the chest. So in here I'm going to check the rotation y and you see it's very fast on the way down. So maybe one thing that I can do is I can start twisting a bit earlier so that in the next pose, as the hand starts to go down, we are already twist in quite a lot, which probably makes some sense. And then by the time we are down there, maybe I can be already a lot more twisted so that I don't twist that fast words the floor. And after we touched the floor effect that we go down is going to counter twist the chest quite a lot and even more after we touch the Florida in, then I guess it will slowly counter to his again, which makes sense because if you think of it, the left arm is kinda pushing up. See that? So it makes sense that the Chess counter twist, eventually you probably want to avoid the plateau in there and go a bit twisty in this direction and then come back down on 55. That makes sense. Now let's have a look at the lending there. You see that now it makes sense. I think I have rotated on exhibit too much on 55, so maybe we'll just soft and that rotation there. Now you're that seems to make some sense. And now the body seems to be a bit cleaner. And even at the twist of the chest, however fast is still under control. I want to check the tweets for the chest here in the dissipation. And you see that on the way towards the anticipation at the beginning of their right arm is moving forward, which would probably call for some twist of the chest to help that let's see. The chest was going in this direction, so it was already helping. Maybe we can help a little bit more. Maybe we could start this motion a little bit earlier. So we start from the very beginning to twist in that direction to help them with twists even a little bit more as the arms moves forward. And then as the arm inverts its motion, we start counter twisting, but we probably want to be a bit delayed because that chest is going to be heavy to change direction for anyway. So we are going to delay that twist a little bit. See, we are fine tuning the motion now. And then as the body goes up, we keep twisting clockwise. After reaching the extreme of the twist, we counter twist as we go into the kickoff posts. So we'd expect maybe I could have twisted even a tiny bit more in there, a little bit less than the beginning, maybe a bit more. And then I think we can maybe increased a little bit more because of the arm pushing this much, I would expect a twist to keep going. Maybe it can even delay that twist. And then here we're already, this is a mistake probably you already countered twisting as the second arm comes up. But a thinker can maybe take my time to twist back bits lower and want to check that animation in there. And other Twist is a bit more organized. You want to have a look at it from various points. You see that fixing the chest as created some issues with the left arm in there, but that's not a big deal. We're going to take care of it anyway. They go. So now that chest is tracked and it's nice and clean, we should as well truck that translation or the chest. Once you understand that you have to start from the core of the body and then extend towards the rest of the torso. In practice, you know how to do it. The rest is just work really. So in here I want to see what the torso is doing. So the Cestius staying there and doesn't really translate, Maybe I could move it up ever so slightly at the beginning just to anticipate the way down. And then as we drop, we are dealing with the chest is to create a little bit of stretch in here. We already going down, but really I wanted to delay the moment when I will come back down. So after we are, we have reached the bottom more or less. I want to start going down and compressed not any earlier. So maybe axial compression will be at around here. There you go, keep going down. And as the arm starts to move up and that's when we start to move up a little bit as well. And until we are stretched on the way up for the, for the kickoff, and I think we are going to keep this a little bit stretched for a bit longer and then maybe you don't want to stretch that much, sorry. And then we start to recompress on the way down, except maybe that between 2932 as we start to go down, we don't want to compress again. We want to delay the moment of the compression to create a feeling of stretch. And then as we go down C, we already compressing back. That's maybe a bit too early. Maybe we can stay there longer and again, a bit longer. And I am mindful of the hand. I want to make sure that hand not stretching too much, that army is not stretching too much. And then after we go and hit the bottom point, that's when we are at our lowest, I think. So if we can delay that a little bit more, it's going to be cooler. I think they go and then we come back up in time, EM wave time. Now I want to check if that's something that makes sense. And I think that sort of makes sense already. That's nice, nice and clean. I will have to do the same for every single control, but in here we'll just have a quick look at things. So we'll check the pelvis. The pelvis is fairly important in year and for the Belle-V is similar finger duplicate this cube. I first select the pelvis than the cube and I go constraint parent that appeared and now I have a tracker for the bellies that I could easily use. And there were probably hide the tracker for the chest or make it thinner or we hide it for now. And I want to see what the pelvis is doing if doing anything rogue. And I have to say that the pelvis looks quite ok. There is a bit of jargon motion at the end. 27. Splining Neck and Head: Now that I fixed the body, I think it's time for me to do the head. So we'll go and call the head back and show it. And it will maybe get rid of the torso for now because I want to make sure I am tracking the head and the hand at the beginning is quite nice. It feels it's under control, but at the same time, it feels organic. During the jump at the top, there's a bit of back and forth in the head. But actually, one thing I could do, maybe at around 28, I could move their head a bit more to the back. So I'm going to grab the neck control and good to check at 28. The reason why I want to move it a bit more to the back, by the way, is because they had seems to be Amy forward all the time, which is fine. But then between 2932 it started moving back. Maybe I can anticipate them motion to the back and see how it works out. I'm going to use the extra rotation to do that. But you see that the extradition really isn't very active in here. And then all of a sudden that 32 other you go. So he starts to move back as I said earlier on. So maybe I want to start keeping it a bit more to the bank at 32 already. And I suspect that the position of the head is something a controlled through translation there and specifically on Translate Zed. So let's have a look and you see in fact that on translates that they had is doing stuff. It is doing what? It's moving backward. Okay, so I was already thinking about that, but you see that it moves backward quite lot more between 2932. Maybe it could anticipate that a little bit better, a little bit more, keep the curve a little bit cleaner there so that I have less of a jarring motion. They go now. I think it's a big cleaner now and then I don't want it to move forward that quickly. Okay. Now I get it. That's why I had to make it move forward because he was getting into the color there. So we tried to move it as close as I can to the color without having too much of an intersection so that a will delay a little bit the moment in which the head moves forward and not super happy about their hose, I need to check if the view port, this is looking weird and it's possible after resigned to the idea that I have much less translation going on there. And I will do that by scaling the curve. And then as we go down towards the down pose, the head will start moving forward once more and again, once more until eventually, it will slowly come back more or less. Towards the end, I want to check whether the head is doing in the landing. And I think in terms of translation, it seems okay, maybe in terms of rotation in the end is a bit strange. It seems that in terms of transition on the way in the air, the head is sort of stable, it looks okay. On the way down here on the lending, the arc is a bit bizarre and I'm not quite sure what's going on with their rotation. So let's have a look at the rotation. The rotation is on x, mainly, the rotation curve is quite interesting. Let's have a look at that. So the rotation curve, so far the animation was k. And then as we start to go down, the head is doing what it's coming down, because it's aiming, because it's aiming at the destination. I get it, I get it. Okay, so in here, however, the head is the late, WE defect that we go down. Maybe you can try and have their head look down a bit earlier. I think that works. And then as we start going down, the head will be delayed a little bit because of the fact that there's so much speed on the way down to the nose will go up a little bit versus likely will keep going up. And then after we hit the bottom, so not now, it will start going down, I suspect. So this is when I expect it will be going. Down. That's the sort of making it up. So we go down with the nose and then we come back up, which is going to take us a bit longer than it does now. And after we have come back up, we start to stay there. Well, it doesn't make much sense in my opinion is that as we come back up with the rotation of the head there, let me zooming. It feels like the translation is coming back down at around 47. So I want to check which axis is that, that's the y-axis. And you see I was correct that the translation is sort of going down. So let's have a look at what these axes is doing. Here is so you see it's kind of controlling the length or the neck. That's probably not something that we can't avoid checking. As I go back down, I expect the length of the neck to increase. So maybe I will just see that I added increase already earlier, which doesn't make much sense. I think if anything, you should increase as we go down. So we'll keep the neck a bit shorter. And then as we start going down, I will start increasing the length of the neck to simulate stretch. So I think this is probably the longest the neck will ever be. Which means that probably in here I can keep it even shorter. They go and as we start to go down, the neck will become a little bit longer. And a little bit longer is probably a bit too long, and it will stay there until we reached the bottom, at which stage it will go back to be a bit lower. So we'll delete this key because it will happen a bit later. So we'll go back to the a bit lower. It will probably keep going in that direction for a little bit longer, I expect. And then it will start coming back up again, I suppose. And we probably don't want it to come back down that fast, so i will keep becoming up and maybe in the end we can count down ever so slightly, but really not enough to be not editable. Now I am going to check the neck and you see that now the neck and the head move a lot better. They're a lot nicer effect as we start to fall, I think it started to make sense that the head starts to lag behind counterclockwise. And a little bit more maybe here than in the previous frame. They go And in here sort of make sense that we like a little bit more. And then as we fall down very quickly, we should be delayed probably until this is when we will start rotating in the opposite direction. And I think these already makes it look a bit more descent. Maybe in year, I can delay that rotation a little bit more so that once we reach the bottom, we start rotating. You see clockwise little bit from the camera, we come back up, we probably want to rotate clockwise a little bit more. They go, and whilst we're at the top, we start rotating back counterclockwise, little bit better. And let's have a look at an animation and that looks a lot softer now you see, I actually like it a lot more if not wonderful, but it's so much better than it was before. And I'm quite happy with the way it's working there. I want to check if there's anything else that I can do in there. I can maybe animate the exhibit better. I could probably do that. So as we go down, we can maybe pitched the nose lower a little bit more. So we hit the button with the head and then the nose pitches down a lot. And then as it comes up, it keeps putting down a little bit more. And then in the end, it will start coming back to its desired position there. And want to check the animation. There you go. And you see now that's probably big to lose. So maybe one thing that I could do is I will try and scale this curves down and that's a bit less loose, so it's still soft but it's less lose than it was earlier. They go I want to check the beginning of the motion. I think we're doing okay. So maybe the only thing, if anything, is that I could delay the moment the head squishes into between the shoulders. So let's see if I can do something with the translation there. So maybe the body goes down at 13, the head is still quite a bit higher. And then it squishes down in a couple of frames. There is a bit of a difference between the movement of the head and the chest. So I create a bit of an offset there. I feel like I could create a little bit more of anticipation of the head in there because it feels like she's just subject to the motion and not really controlling it with the head. So maybe I can try and give the head a little bit of a nudge on the way down as an anticipation there. Let's see. And I think that works a little bit better. Feels like she's a bit more part of the motion. There may be similarly, nudge it a little bit more into the body. So to compress, you see for the anticipation, I need to see the arms because of the know what they are doing. And then maybe this is when they had should come out again. So we nudge between the shoulders, we stay compressed between the shoulders. See how I'm tweaking the animation in camera now I didn't do it until now because until now I wasn't sure if he was working in 3D now that annoy works in 3D More or less and sort of more confident in tweaking a couple of things in camera. And in here I will stretch the head ever-so-slightly to give me a decent sense of stretch, no ego that looks decent moral less. Now we have the body working. We have the headboard king frame 25 at and know what that is. So let's delete this motion trail and see what's the motion trails for the main body control. Okay. So that's the actual motion trailer was a bit confused, really. Kept looking at that 25-year-olds thinking was that thing about now that we have cleaned the motion of the body and of the head, it's time for us to finally touch the rest. And I expect that we'll probably be doing the arms because they think there will be more difficult than the legs make. So we'll start with the arms first. So in the next video we are going to spline the arms. See you later. 28. Splining the Arms: For this video, I don't really need the pelvis anymore, so we'll hide the pelvis tracker and it will bring back the arms proxies. So I will start with the arm which is closer to the camera because that's the one that will be the most visible. So first of all, I want to see that at the beginning the arm does its job as the first pass. I want to tweak the poses again just to make sure that everything works. And if I see some imperfections, I will try to convert the parents space of the arms from body to world. So as the body moves forward, the arm was they a little bit more behind there, which it is doing, and then it will start moving forward. We have seen that. And in here you see that we move forward very quickly and then we invert the motion. So one way to make this thing look a little bit less harsh would be to add overlap. Because you see in here the change of direction is so quick that it's very hard to read and you see it's super quick. So maybe one thing that I could do in here is move forward a little bit less to begin with and overlap backward a little bit more. So that will give us a lot less of a sense of movement forward, you see? And then as we invert the motion, I can move back a little bit less and overlap to the right a lot more. You see that the overlap is nearly invisible because we changed the direction really, really quickly in here between 78. So in there I will have to go and somehow either manually set a key there so that you see that the overlap becomes lot more visible now or used curves in here, it's up to you. I think I need to stay in this pose a little bit longer, so what I can try and do, and I'm not sure this will work, but I would try it anyway. First of all, I will save, of course. And then we'll set of helium three because I know that animation works. And then on frame five hour set another key and unframed five. I will use the twin machine to try and create a pose which is a lot more similar to that of frame seven. I will delete frame four, and now I should get a bit more motion a bit earlier for the arm, I should stay there. You see a bit longer before inverting the motion of the arm. And in fact, I can invert a bit earlier and overlap the opposite way. And that should make the action a bit more readable. It's still not excellent and the transition of the arm from the back to forward is very linear. So maybe I could think of it in terms of arcs. So as we go down, we already are going down with an arc. And maybe in here it's too early to overlap forward the reddest, so we stay overlapping backward and as we invert the direction of the motion, we overlap. And then to be honest, I feel that maybe I could have delayed this. There would have been a lot easier to manage probably. Let's have a look at it from the camera point of view. It's too far forward. It's too far forward. I don't like it. Sorry. So we'll just keep it a bit more to the back so that we move it a lot less than the beginning. They go think, let's have a look at the camera. Yay, it's lot better. There's no comparison. There you go. So now we go to the back and he heared The problem is that we get to that extreme pose on the way up. And then all of a sudden we start with no warning whatsoever. We just start and call it a day. So maybe we want to do something about that. I will go to Frankfort dean. I'm going to add a little bit of overlap to the there. And then as we go back down in here, I will try to add a keyframe Tibet, the truck, the arc in here, something like these. Let's see. We have the whole arm drop-down and I think is going to give us a better sense of what's going on. I can overlap, buffed the W3C and the forearm on the way down. And similarly here I can overlap the rest a bit more in here between 2018, that's the probably of IK you still need to track the Arkin there on 19 and that's not going to happen on its own, so we will need to do it manually. And that's nice and it sort of works. I think that maybe I don't need that many rotation keys in there. Normally I don't really like to animate frame by frame. So in here, make sure I apply any other filter just in case I'm messing up with the rotations. And I want to see what happens if I keep only one of two keys of rotation there. Let's see if I delete this key, you see that I am losing the overlap, so we'll need to keep those keys. So we go back down in there. Maybe we do not need to rotate down the clavicle that much at the beginning. So we'll maybe rotate it up a little bit. That will mean fixing the position of the arm. There you go in maybe 190 will accelerate a little bit. These actions let, should create enough momentum to keep going up your thinking here you see stretching because we changed the chest. Want to check the animation there and then emission there is quite clean. I think maybe a bit too fast. Kind of feel that maybe we are RSpec a bit too much when disposed. And then there's a moment at around 23 to 25 in which the arm is moving forward again, which I probably don't want to have, I will keep moving backwards, is that this is the effect of having too many pauses during your blocking. It's very hard to get them all to work the way you want and to get every position and everytime incorrect. So when you spline, you often find yourself having to get rid of stuff now ego that's sort of clean now. And as we go down is the tough part. We go down, we want the hand to stay behind. So you see that if he's like the body goes down alone at the beginning and then disposing here we want to check if it's a problem we alert, so we'll run an elite few third there. And it's not a problem either. So it's a problem with the interpolation of the arc. So I will have to go in the middle there, in a, will have to create a pose there to better describe the arc in there. The pole vector is somewhere I'm plays and for sure it's moving back quite a lot and quite fast. So we'll have to keep the pole vector down is a check that both on the way down, I want to maybe the late at ARM, all I'm doing here is struck in the RX. And if you're not sure what you're doing, you can always create an editable motion trail there just to see if the arc is doing something that makes sense. You see that for the most part the arc or making some sense though, however, maybe on 32 accurate. Push the pose a little bit more on the way up the ark seems to suggest that in here between 3233, that's quite a big gap. That is very difficult to fail anyway, because it's a single frame, the pose in which we are about to go down and contact the floor. And then at 36, that's when we translate into world space there. So in here you have two ways about this problem. One is to create the arc again manually here, which I don't necessarily suggest, but it's just wireframe, you should think of it. The other way would be to convert the space with the board bake tool. Right now, however, I feel that I would be maybe better off if I just duplicated 34, 35. And then in here you see we lose as nice. And here you see we lose REG. But don't worry too much, we just need to move the master and undo. So by duplicating the pose, we are duplicating position and also the pint space. So we're going to change by an space manually. And maybe we can just rotate the hand down and tried to reach for the contact bows. And we see from the 3D view where these hands should be because they feel that the contact was lot more centered that you go and see was. So that means that this guy has to be a lot more towards the center. In here, we have quite a few sections in which we animate frame by frame. Normally I tried to avoid it, but if you have to track arcsin IK, sometimes that's quite fast as away if you're just filling up a couple of gaps and that poses are working about it. I'll suggest you on your mate frame-by-frame In any case, unless you can avoid it. So we'd go down there and we touched the floor. Once we touched the floor, we spread the fingers down and we stay there. Nothing changes anymore. They go and that seems to be okay. And I don't quite like these bows that I stumbled into right now. I don't like it because it's hiding the face of the character. I don't like it because it looks static in general. And I think I should give her a little bit more time to land. Now let's do the other arm. So the left arm again and going to check what goes on in here. So between 14, the body's moving forward, the army's not doing much, to be honest, the arm is just staying there and moving forward with the body, maybe we can delay the arm ever so slightly so that the body will move forward. You see that the arm will stay there. She's going to look a lot cuter. And then they arm will start its arc. I will make it start. They aren't a bit earlier. They're just want to make sure and not moving both arms at the same time. And in here I will just need to track the arc, so maybe it will not break down. I want to see in 3D where the arc is moving and you see it in 3D, started staying there at the center. If you can open it up a little bit more and then have it closed a little bit later. Here. There you go. So you see keeps working. Evening was static pose, like it better that way. And then it gives us a little bit of an anticipation of the way down. And then it starts, I guess pushing up as well is way. And then it will probably speed up quite a lot and follow an arc that you see opens up clockwise. Maybe let's not stretch it and keep the port vector a bit more open there so that after we reach these extreme pose, we can keep following the arc outside a bit longer, maybe until 29. So keep moving outside and the elbow will come back in and need to check from this view as well. Let's see. So we go back since ok. And here we opened up, we follow an arc, we open up, that's cool. And then it sort of counts back keen in here, which I don't like too much. So maybe I will keep it outside again, which I think looks more appealing from this angle anyway. And then I will start moving towards the next pose, which is down here at the back in here. Remember what happened? We accidentally stretched it so we'll keep it going up, then it will start coming down, coming down, coming down. I don't think I need frame 60 effect. It's left over from the world bake tool they bought. The thing is that you don't lock it at the end. She rows going down at the end. Keep going down probably even at 55. And I think that's sort of okay already. Let's have a look at it in 3D to understand what's going on. So we go back up, maybe we keep the overlap down ever so slightly. And then as we hit the top, the overlap starts to work on the way up. And I think in here the problem is that we should be overlapping for a tiny bit longer. Yes, I think so. And then we should go back down with a bit more time. So not at 47 AU. And I want to see if that makes some sense for what we need is more than okay, now it's under control. We probably want to have a look at the shoulder blades just to know what's going on with those. The most important thing I think is when we touch the floor with the hand, we want to make sure that the shoulder blade gets pushed up and you see that's exactly what's going on. And wonder if I could push it up even more that see what happens to the hand if it changes orientation, think I can't push it up even more and even more and then stay there. And then as we lift, that's when the shoulder blade comes back down and keep coming down a little bit. So we want that shoulder blade to be fairly flexible there. You see, we want it to be clearly connected to the chess, but at the same time not to move one-to-one with the chest. We want to feel like it's the arm that is pushing up their shoulder blade. And this is exactly the case in their nice that's about it. I think for the arms, I didn't really quite check the rotation of the wrist, but I could have converted the breeze to world space and just clean the rotation a lot better. But I think for the most part, this is okay. And now that we've cleaned the body and the arms in the next video we're going to display in the feet. 29. Splining the Legs: And now that we have cleaned the body, the head, the arms, it's time for us to clean the legs. So blacks are more or less working already, but we want to check the motion anyway. One thing I noticed as I watched an emission or the knees is that they tend to move quite quickly at times and sort of lock like here in the end. I don't mind the movement on the way out, to be honest, but I think maybe we could have better control over it. And one problem we've had throughout the animation, if you remember, is that the nipple vectors were moving quite quickly in an erratic fashion, remember? And that was because of the fact that they were following the feed. I think in order to control these nipple vectors, we should probably want them to be in parents space. But remember that we did convert the parents space to world for the knees pole vectors. That means that now we can clean that motion quite easily hopefully. Are you the two nepa vector? So I will grab both knees, pole vectors, and especially during the jump, I don't want this kind of crazy motion to happen. So I will find a pose at 29 where the pole vectors seems to be in a decent position. And I will maybe regularized that position and even more so that they are more or less in front of the character at a certain distance. Then it will delete all these keys between 2327, which are just making the pole vectors go all over the place. Now, for the lending the pole vectors where more or less in the right position, maybe I can delete 30 to see what happens there. You see not much of a difference and I think has the girl lens on frame 40, I would like to pull vector or the right leg to be a bit more in front of the leg itself down there. That seems to be more reasonable as a position, we want to avoid the pole vectors being above the leg in a position like this one because that will make any change in the co-vector create a very big change in the post. So we want them to be at the position where by moving them we can control the pose with fine detail like this you see? So I will delete all the keys between 3440 and now if you check the right leg isn't, the right leg is still moving out and that we want to maybe replicate this pose at 47. We delete anything in the middle between 4047 and maybe I want to recreate them. Motion of the leg, moving back inside. And then in the end, I don't really need the pole vectors to do much. In fact, I can just delete those keys, go back to 50 when I know that it's the last bows of my character and see if the pole vector needs some loving their Just keep it going a little bit. And now let's have a look at the motion. And now you see that the motion is lot cleaner. Now let's have a look at the beginning of the motion because that's going to require some love probably I figure could maybe keep 16 and get rid of 18 just to keep the motion had been more fluid between 1618. You see that now the pole vectors or doing a nice arc for, I don't know if I need four, let's see, not really. And then I will go to ten, the lead then to see what's going on in there. I think maybe I can try and see what happens if I delete 13 instead. What I'm trying to track in here is the right path vector more than anything else because they feel that the right knee is pushing out. It's sorta stops wishing out, you see as the girl goes down. So maybe I can keep pushing the NEA out just so that the knee as well will follow some sort of arc. So we'll maybe delay the opening at the beginning and then there will be more opening you see as we go down towards the contact. And then maybe I want to have even more opening between 1316. There's a bit of a glitch in there in which the knee goes forward and then sort of comes backwards. So I want to have a look into it as well. Let's see from the side view, it might be an effect of the proxy being opened on the need to be honest, let's have a look. You see that the motion of the knee on the way down now is lot more believable. They're right near spatially. Left knee, I didn't quite check. Let's have a look at it. So the right nice seems to have a little bit of a glitch between seven and then where it sort of locks you see? So maybe between 710, I could move it out a little bit so that it doesn't lock as much. And then between 1013, we can move it out a little bit. And x 16 Keep it out a little bit more and then it will just fly out. Let's have a look. There you go. You see now the kidneys are a tiny bit wobbly, but in a good way they're not locking anymore. Now I want to have a look at the motion of the body. So I will grab the body and the pole vectors or the arms. Remember, as we go down with the body control, we're doing some curve like this one. And I want to see if that's the case because maybe we could make the arc a little bit better if I track the sphere. If I track the big sphere in here, will make it smaller. What to see? What does that do? You see that that's true, that we go left and we want it to go left because we needed that to be able to lift this food, if you remember. And then after going left or however we start to go right all of a sudden and then we go down vertically again. Similarly, we can better care for that curve and maybe improve the arc. And we could get an arc like this, for instance, something a bit more continuous or maybe something like this, but maybe not something like we had now, which is going down vertical in practice. So let's have a look and see if that's a thing. So we'll go grab the root. First of all, identify the axis and its translate X. And you see, sure enough, translate X was going probably to the left in here, you see? And then it starts moving quite a lot to the right, but then doesn't move anymore to the right in there. So maybe one thing that I could do is to delay the movement to the right as we go down. So we go down a bit more diagonally. And as we reach the button, we delay that plateau. We sort of keep going a little bit more to the right. If we got this fear by going under the animation toolkit, visualize, and let's go into the options. And I've got selected custom frame steps in here. And I want to see something like five or six. Let's do six frames before and after the current frame at a step size of one. And I should give us a decent idea of the arc there. You see that the arc there is now a lot better, nice and soft. Yes, and I think that's a lot better. Let's hide the sphere for now. And the way the goal now goes down to the anticipation looks a lot better. Yes. Now let's go back to the camera and let's not forget we were here to fix the feet. I was about to forget that I think so. The right food is lifting up ever so slightly in here medic and reduce that even more so its foot rolling and then as the weight has moved to the left foods see there, then the right food can start moving. So maybe I can delay by one frame the whole right food motion feel like it's moving a bit too early, so I will just move everything one frame to the right and I'm not quite sure what's going on in there. You see that the food is started trembling on the floor. So I will go and check the curves and we'll set them to auto. I don't think that's going to change. Actually that changed or improved it. So now it's not trembling anymore. So it's something to do with the curves in there. Maybe I could try keep the food sliding to the back for a little bit longer because it feels like it's just lining back and stopping there. Somebody could grab the translation, their use in there, find out where the Backsliding is happening here. And maybe instead of having it stopped immediately, I can keep it a little bit more alive. There's the tip going down there that I don't want to have proxies not showing us the wiggle of the toe. So so I think we will have to go back to the full jail. So we'll go medium resolution there and we'll go back and evaluate. So you can't quite trust proxy all the time. Of course, we've seen here till we go and we'll delay that. And then in here, I will keep it higher. So now the fruit is sliding to the back a bit softer. Maybe it could even keep it softer. So by moving it bit more forward, so doesn't quite locking position as if you were glued. And then by the time we are at 1380 is probably glued. And I could maybe use the boat pivots to give the impression that the food is still working there and started anticipation in here. I feel like it's lighting to the bank a little bit between 182220, but it's actually staying there. The food Roland is rig is creating a little bit of an issue with the TPP racing without us wanting to. So I will have to counter animate the tip which is not ideal. It's sorta works. When do we stop doing this fine tuning? It depends on the camera is a camera is this far away from the character. That fine tuning we are doing is well enough. You see that you don't even see all those details not working. The camera was the one we have right now. I would probably need to make sure the food moves perfectly whale because on a big screen these will become an issue. Now, that's the push. And in here, this doesn't quite work in my opinion. This section doesn't quite work in my opinion because the food is leaving the floor before the girl S even pushed. Usually your foot leaves the floor after pushing, not before. That's why you can jump right? Because the food is pushing off the floor. So I will need to duplicate frame 22, say about 22. And you see that gives me a much better push. In fact, now that I think of it, we go down, I could maybe roll down the food and stay a bit lower for a bit longer just to have a more even push on the way up with the heels. So now that we had the bush, you see that the pose look a lot more interesting. It would look nicer if the leg were extended after you lift off the Post usually looks better if you reverse the curvature of the food. If the food was rotated this way, the whole action would feel a lot more proof sheets that maybe I can push even more. And now it's time for us to track the arcs that those feeds. So we'll use the editable motion trail there. Maybe I want to delay the food on the way forward. It's nice to be able to rotate the food without the pole vectors getting in the way. If I delay the foods, the effect will be that during the translation, I will be able to see the food moving independently from the rest of the body is going to look a bit nicer in here. I have the feeling that maybe you don't need to raise the food that much. After all, I want to go down with the food, right? So it doesn't make sense to lift it too much. Now that I have looked at the motion, the motion trail is destructive me maybe I can go back to proxy just to try and see only the food. And I will go find the dorsal controllers, these guys I think. And I will hide them. Second that you're now we only have the legs and we want to check the result there. Everything is so fast that it's very hard to track in the end, the animation looks a bit strange because we touched the floor at 34 and that's alright. Then we roll back up. Maybe we don't need to roll back up this much. And then again, we don't need probably to roll that much on the way up. And you see that the fruit is translating around, that's not really good. So we'll need to have my graph eight or the radian there. There is plenty of x and y translation going on there. So maybe we will copy those value Control C or my keyboard actually will copy them 36 and at 38, and we'll click on the graph editor and paste. Whoo. I will have to pay them manually. I will just copy the values instead. So between 3638, we have no sliding here. There's sliding occurring again. So I will try and copy and paste it again. Oh, that worked. I wonder why sometimes it works, sometimes it doesn't. I very rarely use copy and paste and Maya because in the viewport it doesn't work well in the graph H or he works. Okay. The copy and pasting blender is actually better in my opinion. I don't think I even need those keys for translation. And to be honest, because the food is no longer moving there. So once it starts the ground, it will stay there. So that's sorted. Now let's have a look at it from the camera. I may even want to hide the left leg and just check the right leg. And that sort of works. You'll see that leg is pushing quite a lot. Now, I wonder if for these frame and I think a 22 hour, we'll roll back the food a little bit just to keep the leg straighter because I feel that the push wasn't being felt enough. So if the leg is straight, I will feel the pose a lot better and further contact, I feel that we only have one frame in which the leg is straight in there. So maybe I wanted to think it would be nicer if I think it would be nicer if even the contact we're a bit straighter there. So maybe I will first of all fix this pose little bit so that it's not stretched, just ever so slightly. And then for this I compose, I will roll down the food a little bit so that we land on the bowl of the food a bit better in the leg is still a bit more straight. And I wonder if maybe a 29, I couldn't just keep the leg a lot lower so that at 30 we will get a straight leg. The idea here is that I want to read that straight leg for a bit longer, otherwise it won't feel like a contact. I think now it's working a bit better. At least I see the straight leg on the lending and I also see that on the take-off, you now will never look at the left leg and see what the left leg is doing. So we have that dissipation here and the leg is doing or write for the kickoff. I have a little bit of a problem because you see that the tip is coming up. I don't want that. That again is the effect of the foot role on this particular rake. Have a nice stretchy posing there and maybe I can reduce the food role to make it even straighter. And then in here you see we have a stretched leg on 2425. So we have a stretch leg for quite a long time there, so that's going to be readable. Maybe we'll have to reduce that stretch and then we recover. And the problem is that we go down to the contact and we are not even straight. So we'll end with the leg with quite a big band in there. So I think I want to anticipate the contact by one frame if possible. If the contact was on 35, I want to try and have the contact at 34 instead. First of all, I will copy 35 or 44 to make sure that 34 I have the contact there. In fact, I think I can delete 35 altogether there. And now we can definitely have the context right there. But I feel that I'm not seeing the whole picture in here, to be honest. So I will go back to the full geometry and you see that I wasn't quite seeing the whole picture. It's very tricky to work with proxy when the proxies are not updating everything. So in here, I would that's when I have this stretch. You see, so a will try and you see that's reducing the stretch. So you see I'm trying to get to a straighter pose a bit earlier there, just to feign that contact really, there's just so much I can do if they're stretched going on maybe at 35, I'm going to move forward a little bit less. Now let's have a look at the motion and I like it more. I mean, at this stage while the provinces you're going to encounter is that staff that I see because I've been working for many years, is not necessarily evident to everybody at the beginning in a class. It's something you notice with dime anyhow, the key concept in here is that we want to show straighter line of auctions for the contacts. Otherwise, it will feel a bit like the characters are lending with bent legs, which in reality is a fairly efficient way of lending. You don't want to lend with the leg fully straight. You want to land with the leg partially paint just a tiny bit because if the leg were straight, you would land on the joints of your leg not under muscles and that would risk breaking the joints maybe at 33, I can try and push for oppose, Which is a lot more similar to the next one. That means that 32 will probably have to move a little bit faster. There they go. Now let's have a look at this. And I think now it's a bit cooler as an action. What I also like is the fact that the food reaches our heart and then the rest of the body pushes forward, but you see finds the food stopping the motion. And so the leg has to bend and then come back. I think that's quite cool and effect. Now let's have a look at the stretch. Remember there was a stretch on the kick-off. So I want to see what goes on in there between 1820, you see that the tip is raising there, so I will have to keep it low. That's the effect of the food drawl again. And here there's plenty of withdrawal. The only thing I can do on 21 is to counter animate the tip so that it feels like the tip is going down onto the floor. And at 22, same thing when a rig doesn't give you predictable results, that's when you end up counter animated this way, which is not ideal, but if it's just a few frames, you see still works. And in here there is quite a bit of stretch if you think of it, but I don't mind it too much. The only problem I have is that we go from very bent to very extended in just one frame. So that maybe I could think of foot rolling back a little bit, the beginning on 21. That means wiggling the toe forward that they go. And then at 20 to keep diff withdrawal lowers still, maybe a 23. I can have actually a bit more of withdrawal. So this way we go from bent to slightly bent to fully extend it in free phrases that are one, there is stretch going on for a long time on that leg, but I don't mind the line of action that I created. Maybe one thing that I could do is move the leg a little bit forward. There. You see, without changing too much the pose we removed most of the stretch in there. Problems that if between 2425 we accelerate them much with the left food, you see how much we travel, and it's strange to see 26 not traveling that much. So maybe I can try and travel a lot more. That means food will have to go higher up later on. Maybe you can try a little bit less than 27. I want to have an arc that keeps going up. If you see what I mean there you see, that's the arc or want to have, which is exactly what in getting there, I want to have a nice arc and on the way down, that seems to work and not quite happy with this frame in which the food is completely hidden by the right leg. We should always try to keep the silhouette visible, even if it's just one frame. There you go. I think I like it better that way. Now let's have a look at the whole motion as it is now. I think I don't need the motion trail anymore. I think that all in all, the innovation is doing his job already. I think the character could probably jump higher, but overall it feels like the body is under control. We have nice clean motion all over the place so I can't complain too much or bring back the head and that's our emission. They're not super happy about the head at the end. I think it's a bit too soft. Now that's our animation there. And now that we have the animation in place and we cleaned what we needed to clean. It's time for us for the last two weeks. In the next video, we're going to tweak the face a little bit, and then we're going to tweak the fingers a little bit. Just a couple of things. 30. Tweaking Face and Fingers Animation: Now that we have the body and the limbs blind and the arcs trucked, I think it's time for us to have a look at the phase very quickly. This is not a performance shot, so we're not going to spend too much time on it. But I think it's a good idea to understand a couple of principles or face animation while we are at it. So I will select all the facial controls and I will make a selection set of those because I will be employing them quite a lot. So first of all, one of the issues of facial animation is that if a character is moving a lot through space and they had these traveling a lot for space. If you put the transition between one facial posts and another in that time where the character is moving a lot, we're not going to read the transition. If you focus on the face, you can start to understand that the phase changes direction, but that becomes super difficult because the face is traveling all over the place. In my opinion, you need to place facial transitions in places where the face isn't traveling too much so that the audience can see that. So what are we doing here? Instead of having a smooth transition from the starting pose to the anticipation that 16, I'm going to delete all those keys. I'm going to keep the initial pose for a lot longer. And then once the lady is down there and the head doesn't move that much anymore for a space that's when I will crunch the expression. So maybe I can even delayed his finger. I don't need 24 now, something like this, let say so the add goes down, you see, it stops moving. That's when things become serious. Now let's have a look and see if that works. These I glitched and we love it. Let's move the master, let's undo that. And now the transition is a lot stronger because it happens in a place where there isn't much motion. However, it also happens when there is a lot of motion from the arm in there. We want that transition to happen before the orange starts because otherwise people will be looking at the arms. So I will move the transition a little bit to the left so that the transition was start when the arm is not moving that much and that's when we will be noticed in the transition the most. Now I press Play, still have that glitch. At any rate, everything is so fast in here that you will probably not notice it too much. One thing you can do to make the transition look a little bit more appealing is as you go down, since we're not going to see what happens on the face much, we can create another pose just before the transition starts with the purpose of creating more contrast between the starting pose and they focused post. So I'm going to open up the phase a little bit more on the way down. So that is we go down, the face will open up and you won't even notice it because the head is moving so fast. And then this way, when we crunch, you see it will become a little bit more visible that we are crunching and maybe we can grab the eyebrows and we can delete them by wireframe or so. Look at this, see fear visible or that glitch, moved master and do. And I feel that now I can feel it transition a lot better because the beginning of the eyes are open but then they become closed would be nicer to stay there for a little bit longer to appreciate the facial expression doesn't really make sense to animate the face more than we already did. I still don't like that neck there. I think it's too stretched. Now let's have a look at the fingers. And the fingers have a similar issue to that at the phase, as in they tend to linearly move to the next pose. So we don't quite see that. So I will try and play a little bit of a trick and see if I can make that transition a bit more interesting with the fingers. So maybe I could keep the fingers open and then have them close before we start the downward motion of the right food, Medea can keep the fingers open for a bit longer and then maybe I could try and close them before we start the arc on the way down with the right arm there. So maybe until 13 I can keep the fingers as they are that is open and then they would close and that's when the arc would start. Maybe you could make them closed in a longer time. And for the rest of the time these fingers will be in a fist probably until this point. Where was it until 34, there will be in a fist. So I will make sure that Feast pose stays there until 34 and we delete anything else in the middle. At this stage, I'm even delete in-degrees Pencil keys. I don't need them anymore. So let's have a look at the fingers. And I feel that now the fact that the fingers are closing there becomes some sort of anticipation to the arc forward and not too happy about the art forward, to be honest, I feel like it's starting a bit too fast, so maybe it will delay little bit then, you know, I feel like the Feast, you see the fact that we are going to a feast. It's what makes the arm start moving forward. Obviously now we are very far away from the cameras are not fine-tuning the fingers quite a lot. That maybe I could delay the thump closing so I could give it another pose in which the thumb is still open, so it opens up a little bit so that the other fingers can close, can curly, the thumb can cover them like in a proper face that will make it look a bit nicer. And I want to see what the left hand is doing. And I think similarly, I will probably will want the left-hand to close a bit later. So we'll delete frame four, and I will just keep the pose c for a bit longer. And then maybe I can between 912. Let's see. There you go. I can close the feast and there I think I've kept the feast until 34. So maybe I will grab the fist pose and at 30 to set a key and delete anything else in the middle, I don't really need those keys anymore. Maybe I could transition a bit earlier to that open pose. Yeah, that could work. Maybe I want to keep it for a bit longer as we translate into next boards. Now let's have a look. 31. Tweaking the Arcs and Preparing for Playblast: The arc of the Feast is not great there. I feel like it's a bit confusion area in this area of the animation also wants to play the same trick I played with a thumb on the right side. So let's close the thumb last year, that makes a lot more sense. The arc of the feast and there is a bit all over the place, let's face it. So if I go in 3D, I want to see what goes on with that arc. See that is strange. It goes camera left and down, then again up. So maybe I want to convert these arm for a second to world space to clean the arc in there. Let's save, and I will grab the risk control, open up. We'll bake, bake this lecture to locators. Then I will grab the wrist again, change this space to world so that the rest will no longer answered to the body controls, seats, Freedom, space. Make sure world stays world by breaking the connections and then select the locator there and bake from locators. Now, I expect the transition will be bits move for to begin with there, because it's no longer affected by the body controlled moving. And if it weren't smoothed her, it wouldn't be a big deal because we can make it. I think it's already working a bit better though. Still not great. Still bit of indecision. There is still goes up down what goes on there? I don't know. So let's have a look at the curves and maybe you can start by deleting the translation curves there to have the motion a lot more continuous. You see that now that already fixed it. So it was a rogue translation key coming from my blocking. The denser you're blocking, the more you will encounter these issues. Let's see the rest of the motion. You see that the motion of the wrist is a lot cleaner, Not really good. The fact that they will lose the hand feels like it disappears. You see feels like it's glitchy there. I feel like I should be doing something about that as well. Some of you want to have their hand translate that much to the back IN thereafter row just because I don't want it to disappear. It's ever look. Yes. And I think it works much better if he stays in the picture all the time. It gets a lot cooler if it does it, they, and now it's time for us for blablabla. Things seems to work for the sake of brevity. I will keep the animation this way. After all, it does its job, it jumps. It's probably tough for us to go and find the backpack, which is your rocket buck. So it's time for us for our platelets. You see, we didn't have to do many platelets so far. We got the scene to work from the planning through blocking. And then when we were doing explaining, we didn't quite do that many play blasts. So now let's right-click played last. So I will right-click on the timeline go play must option. And in the playlist options, I want to select as a format QT. If you're on a Mac, you will read AV foundation there, which will work just fine. And if you don't see cutie in there and you are on Windows, make sure you have QuickTime installed. You will find it by Googling it. Or I will put the link in the description of the video anyway, as an encoding, I will select H.264 down here. Quality a 100% will do for me display size, render settings. It would probably render at off HDR, a quarter HD scale. One. I want to have the same scale of the Render Settings, and I want to remember to save the file in there. Now let's hit Play blast and wait. And now we have done our animation. There. They go, start that works. I think I could push a few things a little bit more, but more or less it works. And for the time and considering the time it took us to make it, I think the result is more than decent. If you want the platelets to have higher quality, you want to go into the Render Settings in Maya, scroll down until you find the resolution and go HD 1080 so that you will get much higher resolution out of the playlist or so you see that in the platelets windows are not really refined. If you click on this button at the top of the viewport, you see that that refines the edges quite a lot. And if you click on this button instead, that generates some nice contact shadows. You see I'm with inclusion in the contact areas of your character. So in the next video, we're going to see how to light up the scene for play blast purposes. 32. Embellishing the Playblast: Now that we have our animation and we made the playlist window look a little bit better. I think we can make the whole thing look a little bit better. And my a view port is not really good for this kinda stuff. I think it's never been developed with that thing in mind. So maybe one thing I could do, first of all, is create a nice gradient in the background. If you hit out, be like Bravo, you can swap between different background in Maya. I've created a directional light in here. And if you press seven on your keyboard, you will see the effect of a directional light. So I will maybe point directional light towards the character coming from the top. Maybe something like these are suppose will do. I can enable shadows in the view port by clicking on the shutters button, the viewport. And you see that gives me shadows which are, however, if you think of it, a tiny bit jagged there. So you have to open up the attribute editor for the light, go under shadows and enable US depth much shadows. And in there, you want to up the resolution to 2048, for instance, and then increase the filter size to five. That should give you better shadows. You can maybe lower the resolution a bit to have them blurrier. Something like this will be less noticeable as a problem. Or you can try 4096 and you see that gives us nicer shadows there. So now the character also as shadows, which is something that makes the scene look a little bit more interesting. If you think of it, you can't change the shadow color as is c. So that will give you, again another advantage. I can make the shadows bluish. So he will feel like there's a sky dome lighting the thing up. And I can change the color of the light to something yellowish, just ever so slightly. You don't really want to push it too much. That already makes a difference between the previous label as we had and the new one. Sometimes you need to produce some animation for this CH4O and you don't have time to render maybe or you don't know how to use Unreal Engine or blenders IV. So the only thing you are left with is the playlist, so you can still get something decent out of it. I can maybe create a rim light so I can delete the light in here. I can rotate it towards the camera. You see that with that light, we get a line of light painted on the edges of the character. That's quite cool because it helps detach the character from the background. So maybe I will put it here this way. I don't want these light to cast a shadow however, so I can disable it at the depth map shadows so that I won't get double shadow's going on. And now consider what you seen here and compare it to what you had earlier on. So if I press F5 in the viewport, I get to see the shaded version six, I get to see the texture. If I hit seven, I get to see the elite version and you see the leak version. It's actually a lot cooler than the previous version. I mean, it's not as beautiful as a render, but it's functional. Now yo, so now we should get a much better playlist in HD when we're in there out, be careful because all those boxes up here in the viewport are going to slow down your computer quite a lot. If you have an old computer and a slow computer, you want to switch them on, on the when you need them for the playlist. And now we have our animation or in their doubt. And now that we are done, I would like to thank you very much for following the tutorial so far. I hope you found it useful. And if you found it useful and you enjoyed it, please consider leaving a review that will help me a lot in my journey to become an independent creator. And at this time, I'm actually recording usually at night and on weekends while I have another job on the side. So I would really like to focus on the doors instead. Have fun and thank you very much again.