plus, blue is nice.
For those interested, you can do some quite neat machinima using older game engines like Quake, using engine mods like Tenebrae or Darkplaces (which can significantly increase the prettiness of the engine). Being older, they're easier to use and map/model for, but perhaps this is a G+VG topic in and of itself.
I also hear that the Sims 2 is also an excellent machinima tool, what with the facial animation systems, and highly customisable sets and characters. The Strangerhood (linked above) uses this.
I feel that the really cool stuff is waiting to be made in engines like Doom 3 - a horror movie just waiting to happen, or Half Life 2, which also features excellent character and face models, with the bonus of physics and the ability to build whatever takes your fancy.
Filming the buggers though, that's the problem. You could go down the CGI-lite route, and animate the models yourself, scripting the action with the engine, as well a plotting some fairly impressive camera moves too. However, this requires the filmakers (for it would have to be a team, at this level) to have some serious skill ie. the kind of skill that would be landing them jobs in the industry, meaning they'd hardly ever make the things in the first place (an exception to this is the film Bot, linked above, that was produced and directed by a hollywood director, digitally slumming it. It also won the first prize in the official Unreal engine contests, it's well worth a dl).
On the other hand, machinima can be produced like a play in the multiplayer environment, with online players becoming the actors, and other players being the camerapeeps, recording the action. These can be dubbed over later. the problem with this, that while being a lot easier, there's not any of the fidelity of lip-synch, custom animations and finely programmed camera movements that you'd get from programming. Red vs. Blue was produced in this way (using an X-Box no less!), as were lots of films in the older, Quake 1+2 generations.
what is interesting though, is what this approach to film-making holds for the furure. id head honcho and engine guru John Carmack has said that within two generations, we'll be seeing CG quality graphics rendered in real-time. Add advances in procedural animation, where the models learn to walk and fall and jump according to their physical properties, reducing the need for hand-animation, and machinima could be a quick and slick alternative to cheap-cgi. Could we be seeing this technology becoming another accepted form of animation, or even in a feature-film context? Or, why not? |
