“When we consider a new project, we really study it… Not just the surface idea, but everything about it.” – Walt Disney
(Thomas and Johnston, 1995, p.47)
Finding The Spark
At the beginning of this semester, this was the assignment I was looking forward to the most. I absolutely love animating, especially 3D animation in Maya and I was having a blast playing around with different rigs, trying out new techniques and workflows and trying to find inspiration and reference for this project.
Armed with a Disney Plus subscription and my four favourite animation books by my side, The Illusion of Life (Ollie Johnson and Frank Tomas), The Animators Survival Kit (Richard Williams), Timing for Animation (Harold Whitaker) and Cartoon Animation (Preston Blair), I felt ready for this assignment.
My first goal for this project was to tackle the fundamentals of animation. I wanted to ensure I had a solid understanding of what makes an animation work on the most basic level before I even opened up Maya or began planning out anything. The Animators Survival Kit and The Illusion of Life were amazing resources for this topic and I really enjoyed reading through them, but personally, I learn a lot better through video and visual demonstration and so I sought out lectures and videos online that covered the topic. Below are some of the best resources I found on the topic.
I was also lucky to have access to two amazing courses on the fundamentals of animation that greatly helped with my understanding of the medium, The Fundamentals of Animation by Aaron Blaise and Animation Foundations by Bloop Animation.
After a lot of research and study into the principles of animation, I had a look back on my own work and how I had previously been applying the principles of animation so I could learn from my own mistakes.
Beginning with this simple bouncing ball animation, when looking back through it I thought I could push the timing and spacing further. The ball is already pretty cartoony in this shot so I should have pushed the timing and spacing just a little more to really sell the cartoony look of the character. I also wasn’t very happy with some of the arcs that the ball follows through the action as at points the character seems to follow a sharp arc up and a more gradual arc down. Motion trails would have helped immensely with this animation as it would have allowed me to keep a closer eye on my arcs through 3D space, especially from the point of view of the camera.
The next animation I looked at was a bouncing ball exercise produced for semester one. I was a lot happier with the arcs and spacing in this animation but I feel I should have pushed the squash and stretch much further on this animation, particularly with the regular and light balls respectively.
Next, I looked back on an old pendulum animation I made as a study of follow-through and overlapping action. Once again I was far happier with the timing and spacing of this animation but there are some glaring issues within this animation such as the pendulum initially swinging to the right at the beginning of the movement without any physical force pushing it in that direction. I also would have liked to have pushed the animation further at the end with more ease back into its default position.
Finally, I looked back on my flour sack animation from semester one. My 2D skills aren’t great so I had a bit of trouble with this one but overall I was pretty happy with it at the time. I think this animation suffers most from a lack of frames, however, there are still some things to learn from this animation in terms of the principles. Particularly with the anticipation of the second jump. Personally, I feel like the down anticipation could have been pushed much further and maybe extended it a frame to really sell the wind up of his jump. I like the stretched out pose after leading into the jump but as before I feel like I could have pushed it even further to really sell the overlap in the flaps and the momentum of the jump. Another note I have with this jump is on the landing as it sticks in his stretched out pose just before landing a little too long which for me almost ruins the weight of the character in the animation.
Looking back on my previous work was really helpful at this point of the process as it helped me to think about my work a lot more critically and figure out what was working and what wasn’t which would help me greatly going into this assignment. I always try to take some time to reflect on my work both when I have finished it and further down the line to see how far my understanding of this medium has come and to analyse mistakes I may still be making in my current work that I need to improve on.
All Ideas Come From Inspiration
After reflecting on my previous work and revising my knowledge of the principles of animation, I began to look for inspiration for what I could do with my animations. I wanted to do this really early in the process as conveying a full personality and character with a ball with legs can be difficult and I wanted to make sure I had a good basis of inspiration for my animations so that I would have a slightly easier time with my ideas.
I was particularly interested in character and personality animation as I believed that would likely be the most aspect of this assignment and so I looked at different examples of both 2D and 3D animations and how they convey personality.
A particular goal of mine was to study different walk cycles of characters and how they are used to portray a character’s emotions and personality through motion. Below are some of my favourite examples that I found.
I particularly love that in this clip, we see several different walk cycles from different characters. My favourite part of this clip was the difference between the cat gang’s walk cycle and Jerry’s Cousin’s walk cycle. Both are used to convey a similar macho, gangster kind of walk but the small differences between the two really sell the difference between the personalities and attitudes of each which I absolutely loved. The idea of portraying two completely different characters (I’m counting the cat gang as one character to make it easier) who are expressing very similar emotions in a very similar walk cycle but still managing to sell the difference between their personalities was extremely intriguing.
I absolutely adore the animation from Walt Disney Animation Studios, their animators are masters at conveying personality through animation and Big Hero 6 is one of the best examples of this. Baymax’s movements are all very simple and almost floaty, really selling that he is almost like a balloon in this scene. Later in the movie, we see a really nice contrast from this when Baymax is suited up in armour but for now, I want to focus on this clip.
The animation of Baymax here to me really sells his unique personality throughout the movie, to aid at whatever cost. The contrast between Hiro’s extremely expressive animation and Baymax’s more reserved motion in this clip is absolutely brilliant. Studying this clip really helped improve my understanding of portraying character through the more subtle movements of Baymax, it’s a beautiful example of “less is more” in animation.
The HTTYD series is probably my favourite animated film franchise of all time, from when I saw the first film at the cinema to my near-monthly binge of the trilogy, my love for the series has only grown with time. It’s filled with little moments that really help sell the characters and their personalities and has some of the best acting animation shots in 3D animation in my opinion.
This particular clip was of great interest to me as similar to the Tom and Jerry clip, we have several different characters with similar walk cycles each portraying their individual personalities through their subtle movements during their walk cycles. When it comes to acting, I personally like to see more subtle touches that really sell the acting and personality choices of a character and my favourite scenes tend not to be the big grand gestures and speeches where the animators really get to push all the poses and really sell every little detail of a character’s personality, but rather the more subtle scenes, the ones where a characters pain can be read on their face, where the slight change in their movement says more than 1000 words of dialogue. Subtle acting choices and experimenting with expressing a character and personality through their motion to separate them from the crowd is something I will certainly want to be experimenting with more in the future.
I can’t talk about this movie series enough, it is filled with some of the best examples of subtle acting choices in animation. I chose this scene in particular as it has one of my favourite moments of acting in the entire franchise through Stoick’s change from an aggressive, strong and resolved walk to a very open, vulnerable and slow walk towards Valka. I especially love that the animators didn’t feel the need to focus on Stoick’s facial animation until the last moments of the scene as his body movements tell the viewer everything they need to know. It is a truly beautiful scene that really emphasises the small movements and how important they are to selling the emotion and personality of a character.
Moving on to body mechanics, I decided that rather than focusing on existing works from movies and shows like I was for studying the acting choices and how the animators portrayed a character’s personality through their movements, I wanted to look more at body mechanic exercises and studies as these tend to have a far bigger focus on selling the body mechanics with very little to no acting choices.
Animation Mentor progress reels were a huge help in this part as a lot of them contained very focused body mechanics shots. This clip, in particular, wasn’t overly helpful for what I was doing as it was focused on advanced body mechanics, however, studying how the animator achieved the motion of many of these shots by going through frame by frame was a great help for understanding how the movement was achieved in 3D.
I particularly loved the second part of this reel where we got to see some more shots using a ball with legs and even some more fundamental exercises, it was really interesting to see the different approaches taken by this animator in portraying body mechanics and mixing it in with acting.
I wish I could include more of Kevin Parry’s work that I looked at for this project but this blog is already starting to get a little long. I cannot emphasise enough how influential this man’s work has been for my understanding of not only body mechanics but acting choices as well. It is an animator’s dream to have videos like these, covering different body motions in full with nothing but a focus on said motions, it makes for the perfect reference on a rainy day where you don’t feel like recording your own, or even for just mixing it up and trying different things on the fly.
I adore his videos as they make studying the motions of the human body and how said motions can convey a feeling a breeze, I will certainly be studying his work for a very long time.
Setting The Scene
I mentioned above when critically evaluating my own work that I’m not at all confident in my 2D skills with both drawing and animation. It is something I am working on and slowly but surely improving, however, I’m still not at a level of comfort for it so my process of planning out a shot was a little different than most I would assume.
However, before I even began to think about planning my walker animations, we began learning about animating in Maya.
I’ve been using Maya for animation for a while, I began just a couple of months before starting the course so I was confident with using the software’s animation tools, but never tell me you cant learn something new as there’s always more to learn. For example, during our first animation classes using Maya, we looked at using the Dope Sheet. I’m kind of ashamed to admit that I had never used Maya’s dope sheet before this point, preferring instead to stick to both the timeline and graph editor for all of my timing. After using it during class for some exercises, I felt like I had been missing out on a major animation tool that I had never even considered using, which is a little strange as I used Blender’s dope sheet often when animating in Blender. It really helped me speed up my workflow and I was genuinely amazed at how much easier it was to time out my animations during the blocking stage using the dope sheet.
During class, we focused on simple exercises such as a bouncing ball and jump using the Blob rig. I found these exercises to be extremely fun and refreshing and it was nice to really get back into the swing of things with some fundamental exercises. I especially loved playing around with the Blob rig as it was an extremely fun rig to mess with and had some really nice features such as mouth controls for basic expression. I’ve included the animations made during class below.
I would like to note here that the ball bounce was using the Ultimate Ball rig by the same creator of the Ultimate Walker rig that I planned to use for my final animation exercises submissions. To be completely honest I wasn’t a huge fan of the rig, particularly the squash and stretch controls as I found them extremely limiting. In the animation, I decided to opt for the ball to rotate throughout its entire motion. This made it particularly hard to include squash and stretch as the squash and stretch controls were locked to the one axis and so would rotate with the ball. This meant that I would need to compromise my rotation to ensure that controls were in the right direction for the squash and stretch to work. I haven’t seen this addressed in many ball rigs I’ve found bar one by Bloop Animation that has a separate rotation control for the squash and stretch, allowing you to add squash and stretch from any point regardless of the ball’s rotation. Not particularly important but interesting to note.
For my two animation exercises, I had a fair idea of what I wanted to do. For the body mechanics, I wanted to try an exaggerated jump motion as a jump motion has a lot of really interesting motion going on throughout the whole body. As we were only working with a ball with legs, the motion would be focused on the ball leading the action with the legs providing the force needed to make the character jump. Following this, the force of the impact would have to be initially absorbed with the legs with the ball driving the force down into the ground. It’s a motion that requires a lot of small things to happen in a short period of time for the motion to be readable and believable. It also uses every aspect of the rig and so makes for an excellent body mechanics exercise.
For the emotion/character study, I wanted to do something a little different. I wanted to try conveying a character’s personality and emotions through a walk cycle. This would be a far more difficult task as I had to not only ensure that the character’s motion was believable throughout their walk cycle, but I also had to ensure that their personality was readable throughout the entire cycle without any arms, hands or facial features to help convey said emotion.
With my basic ideas in hand, I began looking at other animators work and more specific animation lectures for achieving weight and different workflows I could use within Maya. Below are some examples of my research.
This lecture from Animschool was extremely helpful for understanding how to animate weight in 3D space. This would help me out a great deal when it came to animating my body mechanics shot and I returned and watched this lecture over and over again to further my understanding of it.
I’m a huge fan of Sir Wade’s work and have been a part of his mentorship program for nearly a year now, his feedback and YouTube videos have helped me immensely with my animation work and this video, in particular, helped me a great deal when it came to figuring out what kind of workflow I should be using in Maya. I struggled a lot with finding a workflow that works best for me and I’m still trying out new things to see which one I really like to use so having a reference where I can look at the workflows of professionals has been extremely helpful.
Sir Wade as an animator was also a great inspiration for this project, especially his Instagram as while it is mainly focused around personal stuff, there are a few gems hidden there such as his idle link animation loop that can be seen here: https://www.instagram.com/p/B-XekyrjMt8/
I absolutely love how much life this character has in this simple shot, it’s amazing to see what people can do within 3D to make a character seem completely real.
Speaking of animators who I took inspiration from, I absolutely loved studying Adam Green’s work throughout this process. He’s a supervising animator at Walt Disney Animation Studios and has some amazing breakdowns of his animation work on his Instagram, one of my favourites is this shot from tangled that can be seen here: https://www.instagram.com/p/CFSfc5IDGIs/
I love studying the work of professional animators, particularly those working within 3D as you can see the difference in personality really come through in their shots. Another amazing example is ILM animation supervisor Jean-Denis Haas. You can really see how he approaches a shot and makes it his own with this breakdown of a short animation he made here https://www.instagram.com/p/CL6r2pwjdEh/
I love how each animator really puts their own swing in their work. With Adam Green’s work, a lot of it is more subtle and focused on character acting while with Jean-Denis Haas, his work is far more action-focused and he really likes to push his poses to really take his work to the next level.
Going back to workflows, I looked at a lot of different videos that really helped me break down exactly how I wanted to approach each shot, there’s a lot of them so I won’t be going into details but I will list them below.
Once I had some research done, I decided that for both of my animations I would be following a kind of pose to pose workflow. I had considered straight-ahead animation for the jump shot however when doing the jump test with the blob rig, I found that I preferred pose to pose blocking. This would mean I would have to manually offset keys to ensure that the movement was fluid but I still far preferred that workflow to the alternatives.
Production
Early on in the project, I decided that I would be using the Ultimate Walker rig. I really enjoyed the customisation that I could do with it by replacing the shaders on the geometry and I liked that the rig had set limits on controls such as the squash and stretch and the IK legs. This would mean that I would have to focus more on the subtle action of the shots without getting bogged down with too many free controls and would prevent me from taking shortcuts for things such as my acting animation.
I said before that my planning stage was going to be a bit different, and that’s because rather than planning everything out on paper or through 2D animation, I instead planned on jumping straight to the blocking stage of my animation while keeping my references close at hand. There were downsides to this workflow, I had a lot of controls to worry about and movement was slightly restricted but I got around this in my body mechanics shot by using a simple sphere to animate the initial motion of the up and down so that I could work out my timing and spacing before getting bogged down with all of the rig’s controls.
To avoid confusion, I’m going to talk about each shot one at a time, beginning with the body mechanics shot.
As I said before, I initially used a ball to get the general motion of the shot down, however, I also had a look at some reference, particularly these two clips from Kevin Parry’s 50 Ways To Jump:
Here’s what the ball’s motion looked like:
For animating the character, I decided to leave the keys in Spline mode as I prefer flipping through the animation using the “,” & “.” keys and I could quickly preview what the jump would look like in motion as well as see very quickly if there was anything weird going on in the graph editor.
I messed around with the timing and spacing a bit for the final animation as well as I found that when I included the legs, the initial ball motion didn’t work as well as I had hoped.
Here’s a clip of the final animation with the ball beside it:
And here’s the final animation:
https://syncsketch.com/sketch/b125e98b3436/
Overall I was pretty happy with this animation, but I’ll cover my thoughts in a little bit.
For the overall workflow of this animation, as mentioned above I animated the entire thing while in spline mode, using stepped preview when I wanted to double-check the timing before worrying about the general motion. I didn’t particularly do anything special with the rig aside from using the “Foot Break” and “Foot Roll” controls to further sell the motion of the jump.
I spent a lot of time focusing on my timing and spacing for this animation as it was important to me that I ensured that the character had a good sense of weight throughout the entire jump. I found that in my initial ball reference, the character seemed to lose the sense of weight at the end of the animation as the up motion was a bit too fast and the ball seemed to ease straight from its lowest point to its resting pose at the end.
I’m pretty happy with how this animation turned out but I’ll talk more about that in the reflection section of the blog.
Next was the character acting shot. This was particularly difficult to get right as I wasn’t sure how best to convey character and personality in a simple walk cycle with nothing but a ball with legs.
The first step I took was to gather reference, this time however rather than jumping straight to video reference, I went to Richard Williams’ book, The Animator’s Survival Kit, where I found a brilliant frame by frame breakdown of a default walk cycle as seen below.
This was extremely helpful as it didn’t use arms at all and focused instead on the basic movement of the character.
From there I began animating a basic neutral walk cycle, I wanted to add a little bit of personality but not too much as that was going to come later.
For my workflow, I stuck to animating in spline once again, using stepped preview to double-check my timing when I began messing around with it. However, this time I put a lot more emphasis on my reference, ensuring I had all of the key poses laid out down for my cycle before messing around with any timing or spacing. Once I had all of my key poses, I then used the “break connections” option in the attribute editor so that I would have better control over the different motions of the body, this also made it far easier to offset my keys for some nice overlapping action. From there I decided to let the spline do its work and I spent a good amount of time refining the movement in the graph editor.
Here was the finished result:
I got some excellent feedback from Alec on this shot and was ready to take it to the next step of giving it more of a unique personality. However I wanted to take it one step further, I wanted to see if I could take the walk cycle and rather than just give it one personality, I wanted to give it two distinct personalities.
My workflow for this was to save my initial walk three times, once as a backup, one for the first personality shot and one for the second personality shot. From there it was a matter of pushing poses, editing my keys and using a reference to really sell the different characters I wanted to portray.
Going back to my first example in my inspirations, I wanted to do something similar to the Tom and Jerry walk cycles where I’d take a similar walk cycle and apply it to two different characters.
For my reference, I once again looked at Kevin Parry’s work, specifically the 100 Ways To Walk (as seen below) and I decided that the emotion I wanted the two characters to share would be a mixture of confidence and happiness. Not the easiest thing to convey in a simple walk cycle but I was determined.
One of the best decisions I made during this project was to use the “break connections” feature as it allowed me to easily edit my walk cycle.
This was the final result, two characters with different personalities performing the same happy/confident walk cycle:
https://syncsketch.com/sketch/9ec5963b916a/
Reflection
This project was hard. A lot harder than I initially thought it would be and to be completely honest, I’m not 100% sure if I’m happy with my results.
For my body mechanics shot, I would have liked to have gone even further with it, really selling that landing and weight shift at the end of the animation but overall I was pretty happy with it.
More importantly, for my character shot, I’m honestly not sure how I feel about it in general. I think I would have had a much easier time tackling a reaction shot or perhaps a more exaggerated type of personality walk like a really depressed walk with a character dragging his feet. I did ask for feedback from the class and my family who have no real knowledge of animation on what they thought and if they could tell the characters apart and I received excellent feedback and reassurance that the characters do feel distinct from each other, but I’m not sure. Perhaps I’ve been looking at this project too long but while I can tell a difference, I wonder if I could have pushed it further, really played with the mechanics of the movement, perhaps given them two distinct walk cycles instead.
This project taught me a lot. I struggled a lot with getting a workflow that I was comfortable and could work fast and efficiently with as well as many problems I ran into with the rig itself with the geometry breaking in weird ways for no apparent reason (from my research it seems to be something to do with the way the constraints are set up). I also feel like I didn’t do enough research, perhaps I could have planned more, perhaps tried to plan out my animations in 2D to get the timing rather than doing it all in Maya, perhaps I could have found different styles or approaches from other animators that would have suited this project better. I have also learned that I will need to set up some form of reminder to record my progress while animating as I tend to get very much into the process and don’t think about recording what I was doing until after the fact, at which point showing how I went about it is difficult.
I will certainly be taking these lessons with me into my next project and I can’t wait to get started. This was a tough one, but one must overcome many obstacles to learn. Or perhaps just fail faster.