This whole idea of playing a film like a musical instrument is going to get one step closer to reality with the release of Rock Band, a new game from the brains behind Frequency, Amplitude and Guitar Hero.
The process by which Harmonix, the developer, secures original session tapes and converts them to a real-time playback format is very similar to my idea of how the film RIG.
http://www.wired.com/gaming/gamingreviews/magazine/15-10/mf_harmonix_sb
Still, we're not playing with Bowie classics, we're trying to make a new work of our own. That's the real hard part.
http://en.wikipedia.org/wiki/Rock_Band_(video_game)
I posted this bit of a rant today which has some vague relevence to the film soundtrack.
http://thequality.com/people/michela/weblog/archives/002613.html
Distinctive Records and Hybrid have agreed to let us use that track from the Morning Sci-Album which I've been using for the previz-comic and open it up for re-mixing. We may not get anything in the way of multi-track recordings as the group are deep in the midst of recording their new album but it's a good step anyhow.
.
http://trac.modfilms.com:7861/realmedia/2005/mp3/sbs.mp3
Inaugral post on sound. Thanks JohnB for the opportunity to hang out with the game industry's audio folk at Excel for a day yesterday. Good food for the thought.
SUMMARY
Audio was the starting point for this idea (approaching film as musical instrument) and continues to be a large part of what drives me (spot the frustrated muso). Interactive audio design is a large component of what will make it work.
Christian and I have been looking at using real-time techniques as a route to getting the film CG elements in line with concept and budget. After seeing what game audio designers are using, I think we need to start down the same exploration route with audio as well.
Time to get sound people in the mix and thinking about this.
NOTES
Audio remains subordinate to game-play in design terms. No surprise there but it means that speech in games is still problematic even at places like EA (Harry Potting away) because of the way projects are managed. Too many cooks.
One anecdote I heard - when sound designers plump for there being audio-only cues in a game, they normally get knocked back by producers saying "But what is the person is deaf?".
An audio-only virtual set for each location may be the best starting point to designing Groover. Establishing a real-time soundscape for each (using variant samples and semi-random triggering) could be possible as part of the linear film's design process. The impro/pre-viz capabilities of the audio tools in use on games are fab. The sense of having access to the surveillance networks could be seeded in this way.
XBox is _the_ console for sound design. Lots of real-time support, already marketing 'custom soundtrack' capability to the audience. Opportunity for free eval tools (e.g. Sensaura Cage Producer, Microsoft XACT)
Production pipeline for film needs to take real-time audio into consideration. Need more research and consultation with film audio people. A lot of game people doing short films as well so no clear obstacle to converging an approach _but_ need to understand how feature film audio bods think about all this.
What should happen research/planning-wise with sound design in parallel with film pre-production? Need help figuring this out in more detail.
Stress the importance of improvising with actors and early workshops while there is time to get sound design framework in line with what the performances can be. Development on CD as a game character needs to start in film pre-production.
Only quibble with industry advice at forum was on the relevence of impro towards the end of production. The view was that everything should be totally locked down by production and pick-up scripts should be minimal. I don't think this is right for our model. If there's an opportunity to reshape additional audio (i.e. for Groover) at the end of the prototype build stage, I think we do it. Not desirable but I think we need to be open to it needing to happen. They redid 90% of the Fellowship of the Ring dialogue as ADR.
Problems with game voice-overs is partly historical. Voice agents think Commercial Voice Over as the model when Radio Drama is more appropriate.
Casting people need a clear brief as to what we're looking for (re-visit character bios to ensure they work for this). Need to pre-screen wherever possible.
Give kids longer to audition (EA experience). Sight-reading isn't that common.
Test across localisation at an early (Prototype build?) stage. Take this into account in designing flic assets. Make sure the IA supports localisation.
Interesting lead to a software synthesis instrument technology which could be fun for CD and the like. The idea sounds like the old MOD format except that the film embeds an algorithm and sequence data. Need to research that one.