Wednesday, April 25, 2007
Work-wise, most of my time was spent supervising a BOZ Christmas special which is slated for release later this year. In the middle of that, I had an opportunity to help with our portion of animation on TMNT. In the end I only animated one shot, which is probably why my name was omitted from the credits, but it was fun all the same. I was also able to contribute a little bit to the cinematics for the upcoming Transformers video game. Animation-wise, it was only one shot (again), but I actually appear -- in silhouette form, anyway -- in one scene of the game.
There's been a bit of a break recently in BOZ video production, so I've had the opportunity to animate on some other projects. Recently we've been doing a number of promotional bits for Shrek the Third -- TV commercials for product tie-ins, character interviews, etc. -- and earlier today I put the finishing touches on a rather long (25-second) shot of Puss in Boots. The model is nicely rigged and was fun to animate, although it was strange looking at the fur-less version of Puss and wondering how the poses I made would appear in the final render. Anyway, I saved out a number of videos showing the progress of the clip from start to finish, and once I have my hands on a rendered version, I'll post the collection to the Animation page of my site.
An interesting non-character animation opportunity arose while working on a shot for one of the commercials. I probably shouldn't give the details away until the commercials airs, so for now I'll just say that it involved animating a number of cylindrical objects bouncing and spinning on a tabletop. Hand keying all that spinning action in a believable way would have been a huge pain, but luckily I was able to draw upon the work I'd done roughly six years ago for an Animation:Master rig experiment and build my own rig for the objects that made the spinning animation SO much easier. I was even able to add an element missing from my original A:M rig that automatically offsets the contact point as the object tilts, which makes the spinning action more realistic. There have been times when I've questioned the value of those old non-character animation experiments, so being able to go back and draw upon one of them to get a job done was a fun sort of validation. Unfortunately there's still a hitch in the Maya version of the rig that I wasn't able to overcome due to my limited understanding of Maya's constraints, but working around it for the purpose of getting the shot done was fairly easy. Once I figure it out, I'm thinking of constructing an auto-rigger and sharing it on the MEL scripts page of my main site. No guarantees, but it'll probably happen...eventually...
That's about all the time I have for updates at this point. I'll try to be more active with this crazy blog thingamabob as time permits. As with the auto-rigger mentioned above, though, there are no guarantees, but it'll probably happen...eventually. :)
I was contacted not long ago by a former AM student who was taking some time off, and was planning on studying a lot of classic animation (Disney, Warner Bros., etc.) during his break from school. In part of his message, he asked if I had any "tips for studying these clips frame by frame." Initially I couldn't think of anything specific that I do that's terribly different from what most people do...study poses, arcs, overlap, breakdowns, etc. But then a new method of studying animation occurred to me, and my reply went as follows:
I rarely have/take the time to break down clips frame by frame any more. I know I should, but I'm either busy with other things, or when the opportunity arises, I kinda breeze through a single shot, see some interesting stuff, log it away, and move on to something else.
Anyway, I think it's cool to pick out things like arcs, line of action, poses, etc. when studying a clip, and I'll often focus on those myself. However, one thing that I haven't yet done -- and that I would really like to try some time -- is to fully dissect a clip I want to study. By "fully dissect," I mean start clear back at the beginning, treating it like it were an assignment that I were receiving to animate myself.
For starters, take a step back and look at what's happening in the story. Look at the surrounding shots for context. Where is the character coming from, and where is he going to? Track the path of his emotional arc and figure out where he is emotionally at that given moment. Also take some time to figure out the character's personality (if you don't already know it). Is the character easy-going? Uptight? Aggressive? Passive? Whatever it is, work it out. Then ignoring the finished animation for a moment, look at the shot and identify the vital action bits...what does the character need to accomplish in the shot? Perhaps Wile E Coyote is pushing a cannon to the edge of a cliff, so the pushing of the cannon to the edge is the needed action.
With that, you have pretty much everything the original animator had when they animated the shot initially. You know the context of the shot within the larger story, you know the character's emotional state, the character's personality, and the needed bits of action. Let those settle and gel for a bit, and then start examining what the animator did with that information to turn those raw facts into a piece of entertainment. Let's say it's Wile E pushing the cannon. What did the animator do with Wile E as he pushed the cannon? Does he push the cannon directly to the edge? Does he stop and catch his breath along the way? Do his feet have perfect traction, or do they slip? Is this the first time he's pushing it, or the tenth time, and how does that affect his performance? How does his personality come through in the action of pushing the cannon? How is he affected by earlier bits of the story? Are there any obstacles in his way that he must overcome to get the cannon in position? If so, how does he overcome them, and how is THAT action tainted by his personality, emotional state, etc? Is this a key shot in the scene, or is it secondary, and based on that, how did the animator massage the performance based on that understanding? In other words, ask yourself all the same questions that you would if the shot were yours to animate. Then observe what the animator did to answer those questions, and notice how those answers led to specific poses, and timing, and spacing, etc.
It's easy to approach the process of studying classic animation in terms of breaking it down into the mechanical components...arcs, overlap, poses, etc. And while that stuff is important, the decisions made by the animator in creating the performance of the character are equally important (if not more so) because of the influence they had upon those mechanical bits. However, they're not always as carefully considered as the mechanics and technical princples, largely (I feel) because we can't directly see the animator's decisions. We only see the results of their decisions. It takes much more effort to reverse-engineer the thought process that led to the performance than it does to reverse-engineer the mechanics of that performance, but both are equally valuable when studying animation.