To give a quick glimpse of the last moments of Marcus Fenix and his squad before they go underground to fight what may be their last battle, using the same machinima technique as we did in the “Mad World” spot for the first Gears of War Game.

How did it differ than other motion control shoots?

It differed in that “Last Day” was created entirely inside the Unreal Engine 3 Game Engine. The movement of the characters and their faces was done utilizing the latest generation motion/performance-capture technology at Giant Studios; the same system used in James Cameron’s upcoming Avatar. This system includes a virtual camera that gave us a real time view of what we were shooting, making it feel more like a traditional live action shoot than a motion capture session.

Were the environments produced before the shoot?

Yes, Digital Domain prepped all the environments and characters with textures and simple lighting, so they could be imported into Motion Builder.

Talk about the advantages of the system/technique.

There are two main advantages. One advantage of the virtual eye-piece is that the director is able to control the movement of the camera by either moving it physically in the space or using joystick controls to simulate dolly, zoom and crane moves. The other advantage is that the integrated monitor provides you a real time view of exactly what you would be able to see as if you were in a live action environment.

Did you approach capturing the motion control footage differently in terms of framing/camera movement?

Yes, I had Claudio Miranda operate the virtual camera so that it felt like it was shot by a real DP rather than an animator sitting at a desk.

Can you talk about the other advantages and limitations of using the Unreal Game engine?

Because the whole spot had to be rendered in the Unreal Game engine, we couldn’t rely on compositing. Everything in the frame, had to be rendered together in near real time. A game engine isn’t really designed to be used in this way, so it required quite a bit more time to create than a spot created in a traditional CG pipe-line.

Could this process of using the Unreal Game engine be used on non-game related projects?

If you were creating either a feature or a series that uses recurring characters and environments then a game engine may make a lot of sense because once the elements are created, the time required to render is shorter. However, if you’re doing a project with a lot of different environments and lighting situations, then just the initial set up process could quickly erase the time-savings on the back end. Furthermore, although the graphics quality of game engines are improving all the time, they still do not match the level of photo-realism that you can achieve through a more traditional software-rendering pipeline.