Actually I think Bakshi is probably thinking in terms of 2D animation, and using the computer for ink and paint, compositing, some background plate generation, photography, tracking, lip sync programs to generate exposure sheets, sound mixing, etc. Then you can get 8 guys together to make a hand-drawn movie without requiring a paint department, multi-plane camera, camera operator, sound studio, etc.
Sure for a small group (or individual) to pull off a movie or series they need to be more generalist than specialist. But it isn’t impossible. Film school students learn to wear many hats. Same basic principal here. And you hire out for areas that you can’t do yourself. For the online comic The Dreamland Chronicles, Scott Sava gets help with the models because he said he isn’t the greatest of modelers. Play to your strengths, get help for your weaknesses.
And to keep on track of the topic of “how do we actually do it”, here’s some of my own thoughts/plans for the first part of production. My focus is doing a production on Linux, with open source software when I can. This saves on software cost, and allows better tweakability of software. Many of the software programs I will mention in my posts are available for Windows and Mac as well, however.
Script: Use Celtx for the script writing phase. http://www.celtx.com . This also has storyboarding capabilities, but I plan to do more pre-vis than storyboarding for speed.
Sound: Once you have a script, you create a radio play from it. This is your scratch soundtrack. You can initially record all the parts with you and a few friends, and then replace characters with more professional recording once the script is finalized and you can arrange for the voice talent.
There are many podcasted audio dramas out there, so pull voice talent from that community. They have experience with voice acting, have experience working on indie projects, and often have their own sound recording equipment so they can work remotely with you over the internet. For local recording and scratch tracks, I plan to use a Zoom H2 recorder (http://www.samsontech.com/products/productpage.cfm?prodID=1916) I’ll probably have to work out some sort of setup to control ambient and bounce sound, since I likely won’t be working with a sound booth. But if needed, you can always rent out time at a local sound studio and record your actors. Just be prepared, and rehearse your actors before time so as to keep your in-studio time and costs down.
Voice actors can be found in forums like:
I plan to cut audio on Ardour (http://www.ardour.org/), and use xjaedo to view any needed video clips that are synched to the sound. An article on scoring with Ardour can be found here: http://www.out-of-order.ca/wordpress/tutorial/composing-soundtracks-with-ardour-on-mac-os-x/
I will mix sound in a setion per sequence, and then export a sequence-wide audio clip for animating and performing previs against. The soundtrack will be a back-and-forth thing, with some audio/voices laid out in Ardour, exported to a wav file, read into Maya for pre-vis, image sequences rendered out of Maya, brought back into Ardour with xjaedo, scratch foley added, etc. Retiming of sequences will also need to be a back-and-forth process, since I plan to do all my shot length editing in Maya.
Lip synching the animation to the voices will be done with Papagayo (http://www.lostmarble.com/papagayo/index.shtml), probably the 1.3 mod version. I’ll have to work out a script to apply the papagayo files to my character rigs, but that’s not a problem.
After animation and lighting are complete, and I get to the foley and scoring stage, I plan to record as little as my own foley as possible. Ideally I’ll be able to work with sound clip libraries like:
I’ll likely comission a composer to write and arrange the theme song, and perhaps one or two thematic pieces. The rest of the background music I hope to fill in with music library samples. I’ll probably have to learn to arrange music myself so that I can adapt parts of the commissioned pieces to different scenes. Most of the music created specifically for the show will be midi, which I will arrange and time to the picture through Rosegarden (http://www.rosegardenmusic.com/) and then rendered with TiMidity (http://timidity.sourceforge.net/) and some good sample libraries. Not as good as a live orchestra, but passable. I’m not trying to win an Academy Award here, I’m just trying to get something that is workable.
I’ll stop my rambling here, and continue thoughts on other parts of the topic in other posts later.