Originally posted: Saturday, 24 November 2007
A lot has been posted over the past few weeks about Beowulf and whether or not it should be considered “animation”. I do know a few incredibly talented individuals who worked on it, and am sure they poured their hearts and souls into the film, delivering whatever was asked of them in a timely and professional manner. They don’t deserve the lambasting and flaming that they’re receiving from the animation community.
You can read various arguments for and against Beowulf’s animation credibility at all sorts of sites, including Cartoon Brew, Mark Mayersen’s site, etc. I won’t rehash those arguments here.. I haven’t seen the movie yet, so I can’t comment on it as a film. Nor, do I want to argue over semantics about wether or not it’s animated, as I’m sure MUCH more keyframe animation was done on the movie than is being publicized. Most likely it’s the same sort of PR that was done for Gollum.. you know, the blitz that says it’s all mocapped instead of being keyframed. For some reason the media seems to get a hard-on over the idea that our beloved actors can simply show up somewhere, act for a few hours, and then a bank of computers can magically turn that into the performance we see.
I don’t get why the idea of animators working hard to create a performance isn’t sexy to them.. but, whatever. It is what it is.
Anyway, I DID want to post my thoughts about what films like Beowulf mean to the animation industry as a whole.. so I’ve included my comment from Cartoon Brew here, so you can see what I think is likely to come in the future.
Quoted from comments of Cartoon Brew:
You know.. I haven’t seen B-wolf yet (but I plan on seeing it sometime this week), so I can’t really say anything about the film.
However, as an animator.. I DO have to say that I’m not really all that threatened by b-wolf. I worked on LOTR & on Gollum and was able to directly see what worked & what didn’t, and I know there were incredibly talented artists both tweaking the mocap data & creating animation from scratch.. all of it working towards Peter’s vision. That’s what made gollum so good.. peter, fran, andy, and randy cook pushing us towards creating the best performance possible.
As for whether or not performance capture will replace animation.. I think the live action people have more to “fear” than those of us creating stylized animated films. See, in live action you’re completely constrained by location, time of day, the physicalness of what was shot WHEN it was shot. What props are available, what costumes can get created by the time it’s shot, etc etc etc. Then, once the scene is shot it’s SHOT. You’ve got the coverage you have, from the angles you’ve got. Any kind of “tweaking” or anything you have to rely on 2d techniques, or clever editing. You can’t easily change the camera angles.. lighting.. costumes.. anything. I say “easily” because you CAN do quite a bit with compositing & other techniques.. but currently that’s expensive & time consuming.
Let’s pretend that in 10 years the “performance capture” technique has improved to the point where it actually DOES capture everything we’re saying it’s missing today.. meaning, it’s so damn good that you really can’t tell the difference between a live action actor and their digital counterpart. Well, you CAN tell the difference.. but only if you suddenly change the angle, adjust the lighting, give them different clothing, etc etc. See, the director/art director/dp/etc will all have the ability to tweak what was shot until it meets their “ultimate vision”.
But what if they want acting changes? Well, not only will they capture the physical acting of the actors, but they’ll be analyzing that data.. capture the actor long enough & they’ll be able to build a fuzzy logic brain to go along with the actor. Soon enough we’ll have enough data about keanu in order to recreate almost any reaction to any given situation. So while you’d still have to get the actual performance for any of the important beats.. much of what happens can be figured out by the fuzzy logic system.. think “massive” (the crowd simulation program written for LOTR, now used all over the place) but a bit more advanced with a LOT more data running through it. Of course, we’ll still have frame-by-frame control like we do now, so even if the performance isn’t 100%, we can still go in and tweak things to make it perfect.
Directors will have ultimate freedom to create “their movie”. THIS is where the technology is going.. it’s going to affect live action films way more than our cartoony stylized ones.
I’m not saying this is a “good” or “bad” outcome.. I’m just saying that this is where it’s headed. And for some directors, it’s already there.
So for me, I’m not concerned about mocap. As long as there’s still an audience for funny, stylized characters that do things that people can’t do.. animators jobs are safe. No matter how talented any performer is, he’s never going to be able to have the ability to get from point A to point B in as many ways that I can dream up in my head.