The Spirit Within
A long time ago, I heard someone say that every new technological advance involves both a gain and a loss. It's easy to cite examples in animation. Take a look at the acetate cel: suddenly you didn't have to redraw every single frame, but the static line removes a certain amount of vibrancy from the final image. The cel also helped pave the way to an animation studio system, but at the same time it allowed for mechanical repetition of movement. How about Xerography? It sped up the cleanup process, but resulted in a rough, slightly grainy look. It also precluded the possibility of painting coloured outlines around characters and objects for a softer look.
In general, though, the benefits of a new technology vastly outweigh the drawbacks, to the point where we don't even see the drawbacks anymore; most people don't miss the shimmering line that was the hallmark of animation before the advent of the cel. But there are times when you can't help but wonder if we've taken a certain technology too far.
I haven't seen Spirit: Stallion of the Cimarron, but I observed with keen interest how Jeffery Katzenberg played up what he calls "tradigital" animation processes, those that combine traditional and digital techniques. I also chaired two sessions at last year's SIGGRAPH conference on Spirit: one relating to the challenges of combining hand-drawn and computer-generated (CG) animation, and the other on creating painterly fire effects. I found both topics fascinating as they dealt with the struggle to combine computer animation, something often a hair's breadth away from being accused of being cold or inhuman, with an organic aesthetic. In both cases I think the animators and the programmers applied their considerable talents well and pulled off what they set out to achieve; I have to applaud them for that.
This is a fulfillment of the promise of computer animation. It was used in this case to solve a complex problem, and what mattered more than anything was to look convincing—the end result didn't call attention to the tools used.
In contrast, Spirit's forest fire is an indicator of the wrong path chosen. The decision to make it "painterly"—that is, to approximate natural media—points to a desire to make the fire look convincing, but not necessarily realistic. But if realism is secondary to believability, why use a computer at all? Why go through the process of developing a simulation, implementing it, and then working through colouring, lighting, and rendering issues?
Here's why: because as computers get better at simulating reality, the trend is toward relegating animators to character animation, and leaving as much as possible of everything else to computer tools. I can understand this sentiment, but it ignores one of the tricks of the storytelling trade: that inanimate objects can be characters as well.
Try the Chuck Jones test on the forest fires in Spirit and Bambi by watching both of them with the sound off. Bambi's forest fire gets across the idea of menace, of the claustrophobic terror of being trapped in a living tinderbox. By animating the fire using the same medium as Bambi, they seem to coexist in the same sphere, making Bambi's fear more believable. Spirit, on the other hand, has a forest fire that looks like a special effect. Partly that's because of the aesthetic difference between the painterly look and the largely flat colouring of the scene; mostly, it's because the CG animation moves differently from hand-drawn animation. It ends up calling attention to itself, and for all the emphasis on being painterly, our unconscious minds still place the fire outside of the scene.
We need to pull back from this notion that anything that isn't an individual living creature can be reduced to a simulation. While these technical achievements are formidable, they're robbing animators of one of their essential tools—the ability to infuse any object with life, and to use that life to enrich a story.