“Natal is challenging because it’s so different,” says Molyneux

[JaggedSac]

greenmedic88 said:

The single lens camera only senses outlines of moving objects. It's enough to allow for input on the X and Y axis, but limits player movement translation to a single plane. That being said, it's still enough to do something like Rebound or kayaking, or running in place, slapping hot spots on screen, etc. which is more or less what I expect to see with the first batch of Natal games.

Stereoscopic cameras should be able to provide depth perception input from player actions to be translated directly into the game playing space.

If a game doesn't make use of the Z axis, then it really is something that could be done with the EyeToy or PSEye, presumably with the ability to sense more and smaller moving objects (if it can actually sense individual independent finger motions as hyped).

We'll see what happens. I'll consider it worth my $50-60 if it actually works well with Windows 7 as a general user interface for navigation.

Richochet makes use of the z axis.  You can walk forwards and backwards.  You have to swing at a ball that is in front of you, in relative 3D space.  Hands on have described being able to put spin on the ball.  Holding your hand directly behind the ball(on the Z-axis) and only moving foward to hit it would be more difficult on an RGB only output.

However, as far as game design goes, it doesn't seem compelling to me.