Project Natal drops hardware Motion detection to save costs!

[NJ5]

WereKitten said:
^what drkohler said still stands, though: the "elaborates trillions of body configurations every frame" is clearly some sloppy PR speech. There's not remotely the power to elaborate anything trillions of time per frame in a cheap device.

The guy probably meant that their image->skeleton algorithm can potentially distinguish trillions of different poses, and does that 30 times per second.

That said I'm slightly concerned about this:
"The system locates body parts to within a 4-centimetre cube, says Kipman. That's far less precise than lab-based systems or the millimetre precision of Hollywood motion capture. But Douglas Lanman, who works on markerless 3D interaction at Brown University in Providence, Rhode Island, and is not involved with Natal, says that this will likely be accurate enough for gamers."

...because 4cm is very rough and could give troubles if we want to, say, use a finger as a precise pointer for on-screen interaction a-la Wiimote/Sony wand.

I've suspected since almost the beginning that Natal wouldn't be able to do IR aiming or anything similar... that suspicion has only grown to almost a certainty.