I previously wrote about a new model for game development that a few people have thought of independently and are starting to experiment with. The basic idea is to formalize the game development process and make all high-change/high-dependency changes early in the process, when the cost is the lowest. The game’s core design is done by a tight experienced team from one location and they create a blueprint that describes how the rest of the game is to be implemented. This seems to share a lot with the model that Hollywood uses to make large-budget pictures while reducing risk.
There is so much more to learn from Hollywood though! This past month at the San Francisco IGDA, Industrial Light and Magic and LucasArts came in to discuss their new cross-disciplinary focus. They are bringing their film and game efforts together under one roof, in the old WWII army district of San Francisco, the Presidio. It’s wonderful to see the city provide tax incentives to bring exciting business and job prospects to a beautiful (but unused) part of the city.
They hope LucasArts learns how to build and work with film quality art, and they hope that ILM learns how to use real-time game techniques to speed and improve film production. The concept may seem abstract, but I think they are going to see some huge benefits. Some techniques and benefits they could see include:
1. Creating artwork and shared assets once, and using them for both their films and their corresponding games. If they acheive this scalability and versatility, they should be able to target multiple game platforms more easily as well.
2. Motion-capturing cameras during filming to use within games. This could be use to re-enact or reference key scenes in the film. As motion-capture technology gets better, this may even allow actors to be motion-captured at all times, and for this data to be used by characters in the game.
3. I don’t know enough about how the “dailies” are created or used, but it seems to me that “instant dailies” could be created with this real-time data and technology. Instead of waiting for the end of the day to see how shots turned out, a rudimentary version of it could be seen instantly. Conceivably, the film crew could have level designers on set to quickly change backgrounds, architecture and lighting to see if the footage will be good enough before finishing at a location.
4. Allowing directors to pre-visualize scenes by using placeholder artwork and real-time game technology. This would be an intermediate step between storyboards and set design, and could provide more insight into a scene before committing resources (set-building, location scouting, etc.) to it.
5. Non-realtime lighting technologies and techniques could be brought to games. Some level of Avid-type technology could definitely be brought to a real-time game platform, for the game director to use for cutscenes and even in interactive scenes. This is illustrated to amazing effect in Lord of the Rings, with palette manipulation, image highlighting and glowing. This is one of the techniques we want to use for Natural Selection II and it seems very doable.
The list goes on and on. I don’t know that much about how films are made, but I know that film production methodologies could remove a lot of risk from game development and improve our work dramatically. As games are compared more and more to film, game developers will increasingly need to look toward that industry to manage budgets, mitigate risk, define creative responsibilities and create sophisticated visuals and sets so we can create works as important and well-received as the best films.
Charlie Cleveland 