Wednesday, June 6, 2007

Time and Timecode


The Sinister Symposium was this week. It was a sort of open house at my workplace for the Ballad of Sinister, and it was the first screening of my story reel before a live test audience. As I mentioned, I work at a major animation studio, so it’s the best place to troll for comments on the subject of storytelling in general and animated filmmaking specifically. The reviews were generally positive, a few negative. The negatives all had one thing in common: the perceived dark tone of the project, which boils down to a matter of taste. I like a happy movie as much as the next guy, but this one is not happy. When a story like this deals with timeless truths I see no need to candy-coat things, so I’m resigned to the fact that this project will never appeal to those who demand rainbows and smiley faces for their entertainment value. The sincere nature of this project is bound to garner mixed reactions; to me, a sign I’m doing it right. If my story ever gets watered down to the point where absolutely nobody has a problem with it, then it will be insignificant– a fate worse than just plain bad.

My story reel is now locked for the time being, That is, I will cease noodling with it long enough to use it as a blueprint to get some actual production work done. Now I’m buried under the staggering weight of bookkeeping. Already I have a shot list, and I’m stepping through the entire fifteen-plus-minute story reel and marking it up with the timecode of every cut between shots. This is important, because the next step will involve creating discreet timelines in my 3d application for each shot of the film. (I am defining the basic building block of my project as The Shot – a discrete camera cut. This neatly translates to one Maya file per shot.) I’ve decided to lend a twenty frame padding at both the beginning and the end of each shot to allow for crossfades; in editing the story reel, I discovered I favor this technique. Once I’ve done this, there remains to create a more granular breakdown in the form of an X-sheet per shot. An X-sheet breaks down the shot to its most fundamental landmarks in terms of audio-visual syncing: every major action, every sound effect, every phoneme of dialog needs to be marked with its exact frame numbers and timecode so that everything fits nicely together during animation. And, yes, I am using frames, not footage. Lots of folks at my workplace still count things according to footage, belying their analog upbringing. This is unfortunate, since a foot of film is 16 frames, which does not correlate to seconds (24 frames) or any other unit that would be familiar to anyone outside Hollywood. I’m glad the trend is to go with timecode rather than footage in this brave new digital world – but I still run into the old footage stalwarts at work – in fact, our production management infrastructure is still footage-based! King Henry I would be honored to know how we still canonize his 12-inch peds – even in a film-less world where the concept is increasingly irrelevant. Old habits die hard, I guess.

After I mark up the reel, I’ll create every shot as a blank Maya file and begin the 3D animatic – the first 3D pass using proxy geometry, rudimentary lighting and no textures. The skeleton provided by this pass will evolve into the finished film as I replace each element with finished product, one by painstaking one. Much to do, so I’ll quit procrasti-bloggin.

No comments: