I’ve always been interested in methods of learning, whether it be animation, language or any other topic. In the creative fields a musician will start by playing popular music, a writer will read profusely and a painter will spend some time copying masters. With modern CG animation schools and their relatively short course lengths it seems that idea of copying masters as a way to learn the craft is often sacrificed.
Long before going through Animation Mentor I stumbled across John Kricfaluci’s blog and his thoughts on learning animation through his self made curriculum. For about a year I followed his process of copying from Preston Blair’s book as a way to study appeal, poses and construction of characters. It was essentially the way I learned to draw.
Fast forward to 2014, where I attended a talk by Mark Oftedal on cartoony animation. He outlined ways of studying and copying the masters of animation then applying those observations into your own shots. It reminded me a lot of John Kricfaluci’s methods and was also a bit of a kick in the butt to get back in the habit.
Below are some examples of the studies I’ve done recently.
Some thoughts on using this method of studying animation:
– As a CG animator I find drawing quite a refreshing way of skill building and observation, especially after a day of sitting in front of Maya.
– Mark recommends doing quick gestural type sketches, John recommends taking time and aiming for accuracy. I tried both and felt I was getting more out John’s method of paying attention to details, I thought more experienced illustrators may prefer Mark’s method.
– In both the poses and animation studies, it’s often quite clear where your understanding or analysis falls short. John recommends overlaying your sketch on top of the original and noting the differences. I found when doing the animations, I’d often miss getting the feel of the original. Going through and finding the points where the 2 differ were often eye opening.
– I also found this method a great way to ramp up for a project. My current project at work is quite cute, so I spent some time going back to Preston Blair’s book and also copying Mickey Mouse poses as a way of getting used to the style.
One of my regular struggles is breaking up movements so that my animation doesn’t feel pose to posey or that different body parts have the same motion. A few months ago Jason Martinsen gave a talk at Animation Mentor about breakdowns and inbetweens, highlighting the work of Charlie Bonifacio and also the breakdowns used on Mushu in Mulan (supervised by Tom Bancroft).
I went through a couple movements from random points in the film and tracked how the head an hands have been broken up. Note the mix up of spacing, arcs, easing, anticipations and overshoots. I especially love in this first example how the arc of the mallet brings our attention back to Mushu’s face.
*Edit* Thanks also to Charlie and Tom for commenting and crediting the animators. The first was example was by Rob Corley and the following 3 were by Tom Bancroft.
If I had one shot that had the character’s hands in the same pose 8 times, is that a bad thing?
I tend to find students are quick to point out twinning when giving critiques, as if to say it shouldn’t appear in an animated performance. But what is it about twinning that makes it something to be avoided? It happens in everyday life and is something we sometimes naturally do. Give “Barack Obama hands” a google image search for a quick example.
If it’s something that naturally happens in everyday life, why try to avoid it when animating? I feel there’s an assumption that twinning = unnatural posing/movements. Yes very symmetrical twinning does appear unnatural, stiff, mechanical, etc, but it doesn’t always have to be that way and there are ways animators can help reduce that feeling.
Coming back to my opening question, I noticed this shot in Ratatouille. It’s a 12 second shot and by my count twinning occurs 8 times.
I thought this is a great clip for studying how to pull off twinned posing without seeming unnatural.
Check out the clip for yourself and look out for:
– Small variations in the posing of the hands
– Different timings. The hands come in and out of poses at different moments.
– Different movements. Watch how the hands will move within a pose, one might go up while the other goes down for example.
– Moments where the hands break up into completely different positions. The hands aren’t twinned through the entire shot.
The other day I was watching some of the first shots I did on a production and could see one of the main problems I had early on was not fully understanding the style. At A.M shots tend to fall into a “Disney style” or very cartoony, but I think CG animation has become much wider than that and the industry more demanding.
I’m now working on my third production, each time a different style of animation has been required and each with little to no ramp up time. I think it’s a valuable exercise to not just study but try different styles of animation, I’ve found understanding and being able apply a style are two different things.
I’ve made a list of what I think are different styles with examples below, the categories are purely opinion and meant to be just food for thought, but feel free to comment if you think I’ve missed anything glaring.
With motion capture
“Disney/Pixar style” A very loose way of describing it. Both studios use slightly different styles in different productions.
1. 2D Pan/Zoom
Often when animating things like lip sync, eyes or fingers I’d like to get a closer view of those body parts from the angle of my shot cam. Maya has a simple function that will allow you to zoom in, without actually changing your camera set up.
First, enable the 2D Pan/Zoom in your Viewport (the icon with the magnifying glass).
Second, select your camera, open the Attribute Editor and adjust the settings under Display Options > 2D Pan/Zoom.
Simply clicking the 2D Pan/Zoom icon in your Viewport will now let you flick between your normal camera set up and a close up view.
Update: Thanks to Iestyn in the comments for this tip! If you really want to speed up your workflow, the ‘ \ ‘ key is the hotkey for 2d pan, if you hold it down (instead of the ‘alt’ key) you can use it to navigate, zoom in & out, pan etc with the normal mouse buttons.
2. Isolate Select
At work I almost always animate with just my character Isolate Selected in the viewport, just for getting better playback speeds and for clarity without having to set up layers in every scene file/unload references. I’ve combined a few different scripts found online into one to make the process isolating a characters much quicker. Basically all I have to do is select one of the main controls (COG, Root, Head, Spine, IK hands), run the script and that character will now be Isolate Selected in the current viewport. The script includes a switch to turn the Isolate Select on/off and also works for all the Animation Mentor rigs I’ve tried.
I was recently invited by Animation Mentor to do a shot with their new Jules rig as a way of both testing and showcasing it. It was a tight schedule of 3 weeks in just my spare time, but am happy with how it turned out.
Thanks to Albert Morrissey for his rendering, lighting and modelling skills. Always a pleasure to work with.
This is a simple script/keyboard shortcut I use daily, when setting up my work computer I couldn’t find it anywhere else online so am sharing it here.
All it does is change the playback range of your timeline. In a similar fashion to QuicktimePro, pressing “i” will change the playback range to start on the current frame, pressing “o” will make it end on the current frame. It doesn’t affect the length of your timeline, just what section will be played. It’s much quicker than dragging Time Slider or entering frame numbers.
Set the start of the playback range (assigned to the “i” key): playbackOptions -min `currentTime -q`;
Set the end of the playback range (assigned to the “o” key): playbackOptions -max `currentTime -q`;
I’ve heard so many differing opinions on the importance of drawing by great CG animators, writing my own opinion seems almost pointless. It is however the number 1 skill I’m working to improve on for this year. I have a few books on drawing (mentioned below) but I find there’s enough online to work with, it’s more a matter of sitting down and doing it regularly.
New Masters Academy has some great over the shoulder videos of Glen Vilppu and Steve Huston drawing poses (both artists have taught at Disney and Dreamworks). Even if you have no intention of picking up a pencil they’re great to watch as they both talk about the essence and movement within a pose, a great way of thinking when analysing your video reference.
For those who are complete beginners, this thread might be worth a look. It documents the progress of Jonathan Hardesty who put himself through a fairly regimented program of teaching himself to draw over a few years. As his site shows he became quite successful, but I was interested to see how far he got in just one year:
*Just updated this post from last year so decided to bring it back to the top of the blog.
The A.M G.U.I picker can be a bit of a hassle to work with as it’s quite large and doesn’t allow for shift selecting multiple controls. 2 alternatives I used in A.M are Awe Control Picker and animSelector. Both are easy to set up, take less screen space, allow for controls to be grouped together and also shift selecting multiple controls. While they’re not visual pickers, I have found they are fast to work with.
Of the two I now prefer Awe Control Picker, it’s quicker to set up and found I had less hassles with it. It’s also developed by an A.M alumn who continues to improve on it.
Awe Control Picker:
In addition to those two, AnimSchool offers a G.U.I Picker. It’s resizable, has the ability to zoom, is easy to set up and allows for selection sets and shift selecting. The only downside is that the Terms & Conditions prohibit the use of the picker with rigs from other animation schools.
One other tool worth mentioning while on the topic is the Studio Library. It’s a pose library that is easy to set up and great if you use the same rig over many classes/shots (thus building a useful library over time). It’s also useful if you have many shots and need to hook up poses between them, if you work in a team and want share poses/animation cycles or like using pre-made phonemes when doing lip sync.