That's just tracking the hands and one foot, not full-body tracking.

Not downplaying some achievements, just saying the quotes in the OP look way inflated.

LOL, so you think it was only programmed to follow the hand and foot? :P

I'll see if I can find the article I read which discusses this but next to the demo setup they had in the room for press was a laptop that was running the software for Natal, on it you could see your entire 3D structure as you moved.  It was tracking the whole body, it's just that the game was set up to be controlled by the hands and feet.  It not like they couldn't have used your elbows or shoulders to control the car but that would be silly.

@daroamer: No that's not what I'm saying, geez... I'm just saying the Burnout case doesn't require correct tracking of the whole body, unlike the Avatar demo.

It's obvious that they're tracking the whole body in general.

But you said that that tracking isn't resulting in anything tangible. So I'm not sure what you're point is? The point I'm making is you're blaming the failure of Natal in that avatar demo on the tracking software, when the fault was almost certainly with the avatar rigging, not the tracking.

Here is the article, by the way:
http://gizmodo.com/5277954/testing-project-natal-we-touched-the-intangible

@daroamer: I meant tangible in the full body tracking sense, which is what Lee's declarations are about.

Do I have to spell out everything in full?

Apparently because I explained where the failure was and you're still claiming that Lee's assertion is wrong or at least overly optimistic? He obviously works with this device in the lab and would know much more about how well it operates and what the internals are doing whereas you're just basing your opinion on one live demo where the failure was probably with something not related to the device's tracking at all.

@daroamer:

A simple question: what's your evidence to say the failure of the demo was probably not related to the tracking?

In that demo, they show not only weird movements but also limbs going completely out of place compared to what the demo guy was doing. If I'm seeing it wrong, please explain how.

In other words... if the tracking device was tracking and identifying all the limbs correctly, how did the end result fail like that?

I can't believe I read so much BS in this report.
Gamers are very easy to fool, too bad for them.
The things written there are just plain impossible and pure nonsense.
Any scientist would notice all the impossible things they say in this.
I'm mostly impressed at how MS can get away with saying so many lies, and can get so many people to believe such nonsense.
The Johnny Lee guy was also doing useless things before, but now he's even called an expert (LOL) and I sure hope what he says is taken out of context.
Because if not, it just shows how low some people can go.

All of this looks like blatant vapourware after I read all this nonsense, on the scale of a huge scam.
And I'm sure MS will yet again get away with this once everyone cheering for them realize the fools they were (like always).
Seriously, this is embarassing, including for those that supposedly tested this, who I am sure have their excuse ready.

That glitch was almost surely in the software running inside the camera. Firmware if you will. I don't think MS wants every developer to have to recode basically the same thing.

@JaggedSac: Yeah that was what I though too.

I'm sure there are a lot of improvements they can do.