Sunday, June 29, 2008

animating fast: inspiration

The other night I saw Michel Gondry's Be Kind, Rewind, and I think it serves as a good inspiration for the upcoming spring Animating Fast class. Their Lion King rendition was particularly funny. Maybe this film will be the first assigned screening? Their technique reminded me a little of the production methods used in Dracenstein by Tom Priestley (a fellow submitter).

Thursday, June 19, 2008


I've been having a lot of fun watching the Big River Show lately. It's Bill Bowles's latest adventure, this time a trek down the Mississippi river in an old pontoon boat. Check it out and enjoy.

I am such a fan of fresh media like this.

Sunday, June 15, 2008

calling anyone and everyone

In the last few days I've had many thoughts related to distributed production. The biggest most productive phases in Tower 37's production were of course when a class's worth of students (and, when available, alums and others) were cranking away on the project. Anyone following the history of this film should know by now that my advanced animation courses at Hampshire are built around the idea of collaborative production, with students in the course becoming contributing creative crew members on a substantial film project.

This is a great model for bringing to life independent animation work that lies beyond the capabilities of an individual, and I have argued elsewhere for its merits as a pedagogical strategy as well. But this began as a single-class endeavor with my first film in 2002, and has obviously grown way beyond that.

Over the course of this production there have been four periods during which the crew was greater than, say, 5 or 6 people: the fall of 2005, spring of 2006, January of 2007, and fall of 2007. In other words, we've had to adapt frequently to crew changes. This means bringing folks up to speed quickly and ramping them off the production quickly as well. There's been much more of that than I hoped for, and it's taught me a few big lessons.

I can't get into them all here, but a clear vision for the production (as in theme, story, etc) coupled with easy-to-follow standards and protocols equals a fairly low "crew change" cost. These of course are coupled with a set of tools that permits remote work at odd times of day on multiple platforms, which HELGA already offers us.

Where this is all going comes from answering this question: what if the crew was ever-changing, with perhaps people coming on for single production tasks instead of long stretches of time?

Check out the Open Source Cinema manifesto. I wouldn't say that I'm adhering to every point in the manifesto, but I've enjoyed watching The Basement Tapes grow and evolve as a project and I think there's a lot to learn from this approach.

Anyway, I'm thinking that my next production should be structured so that anyone can jump onto the film's site, grab one of the many production tasks that are available, and contribute. This would require some new HELGA infrastructure to better encapsulate bits of the film into easily download-/upload-able archives. It would also help to use the most ubiquitous production tools possible, with Open Source being perhaps the most desirable.

What's nice about this is that it doesn't preclude the same semester-based approach to production, it simply opens up the project to a larger, more distributed crew. Details would have to be worked out, obviously, such as how people can find assets that are available, get them, work on them, get feedback on their work, and ultimately get their finished assets back into the project.

Perhaps I will be able to explore this approach in more detail during my upcoming Animating Fast class in the spring.

down to 35

We are down to 35 shots left to final. Will is on his own for the next two weeks as I travel through California, but before I left we lined up a whole slate of shots for him to attack in my absence. Those were all shots that, pretty much, could be solved with our existing lighting infrastructure and lighting/compositing strategies. Of course, what that means is that when I return we'll be left with a pile of what we casually refer to as the "hard" shots.

What makes a shot hard in this film? The biggest difficulties probably relate to water, and we have our fair share of splashing, pouring, dumping water shots to contend with. But we have other issues, such as the shot with a few thousand versions of Leed swimming around (hard because we have to keep them from looking identical AND because we have to get them through the actual RAM-limited render process). There's a file which doesn't even open in Maya any more, and we have to try and repair that one. There's the film's first shot which is hard because it's 1000 frames long and features a huge scale change that has to happen seamlessly. Another way in which water is challenging is that we have three shots that occur under water, and thus we have to figure out a nice way to simulate light coming down through a reflective/refractive surface from below.

But I like our chances. I am almost finished with my responsibilities from the spring term, and thus I will soon be able to focus more on these challenges.

Monday, June 2, 2008

a long-awaited Tower 37 (nee Uprising) update

So the 10-minute CG animated short I've written about here before (Tower 37, formerly called Uprising) is actually nearing completion. Which is a good thing, 'cause it's basically its third anniversary this summer. Three years ago I was working on the story and the reel, now we're down to 44 shots left to finalize.

You might get some kind of perverse enjoyment out of reading prior posts about the film, especially those from late summer 2005 when I thought we could actually finish it in just one term. Ha. Well, over the three years it ballooned to 11 minutes, and now has been trimmed back down to below 10. There are 114 total shots as of this writing. But I do see the light at the end of the tunnel.

Bit Films Intern Will is currently leading the lighting/compositing charge, and I'm doing what I can as my other responsibilities allow. I think we are much more efficient with a small crew than with a large one, and I hope to update with some regularity as our number of remaining shots drops.

If you're interested in reading about how I structured a collaborative animation production class around this project, or if you want to learn more about the film itself, you can download a PDF of my MFA thesis and knock yourself out. Or just watch the movie as it stood a year ago when I got the degree. The site hosting the thesis keeps telling me people are downloading it; I'd be curious to hear from those downloaders what they think about it all.

Here's an image from shot u2_01. It is 1/3 of the final full resolution. I'll start adding more images to my posts if this works and if people seem receptive.

For those interested in the technical specs, this film is being made using Maya 8.5, RenderMan Studio, and Shake. To a lesser extent we also use After Effects, the Maya Software renderer, and hundreds of custom scripts (mainly written in TCL).

Tower 37 is using the HELGA system (developed at Hampshire College) for production management and web-based render/comp spooling, asset management, and job tracking. I will update this post or blog with a HELGA link as soon as one is available.