After the official launch timeframe recently hit the news, Natal’s lead developer, Alex Kipman has shed more light on Microsoft’s upcoming controller.

Apparently, Kipman has demoed Natal at this week’s CES and revealed how this controller came to life by Microsoft collecting “terabytes” of data on people in typical gaming poses. They’ve done this in homes and motion capture studios and analyzed the data which resulted in a 50MB software package.

Mr. Kipman said how it was crucial for Natal to work on the existing hardware without sacrificing much of the processing power and he says Natal eats up between 10 and 15 percent of Xbox’s computing power. We’re not sure whether that can be classified as “not sacrificing processing power” but given the nature of the controller and the fact that motion capture is in no way a simple task, it’s understandable.

He says that developing Natal has been similar to training the actual device and teaching it human anatomy, and it analyses info over millions of frames. In fact, Natal recognizes up to 31 different body parts in up to 30fps resolution. It will recognise any pose in 10 milliseconds and it only takes 160 milliseconds to detect a new user that steps in front of the camera.

Furthermore, it "correctly positions your hand even if it's held behind your back… It knows the hand can only be in one place,", Kipman added. This is performed via a software with understanding of human anatomy  and will probably be crucial in scenarios where there are for instance two players, and not everything is clearly visible at all times.

More here.