When developers take the time to minimise Kinect input lag it's barely noticeable, but the sensor's response time could get even faster in future, according to a newly published Microsoft research paper.
Essentially, the current method works out which of your limbs is which and then tracks them, frame-to-frame, says researcher Harry Fairhead:
The basic idea is that if you have an area identified as an arm in the first frame then in the next frame the arm can't have moved very far and so you simply attempt to match areas close to the first position.
This partly explains why Kinect struggles to read body parts if they become partially obscured or lost from view. The new method seems a smarter way of doing things: rather than track from frame to frame, it compares your body shape to existing maps of other players, working out which body part is most likely to be moving.
According to the research paper, the Xbox 360 GPU can perform this process in under 5 milliseconds, equivalent to 200 frames per second, a significant increase in speed on the current method
Could this be the new Kinect technology hinted-at for E3?