Project Natal: MS Full Body 3D Motion Detection

oh sure it's most likely next year .. maybe this Christmas though

 

If that was the case I am pretty sure they would have an on stage demonstration.

This is also is proof it´s still in a very early phase, final hardware spec may just have settled.

 

They did ...

 

That's up to their business guys to decide whether and when.

In the mean time, they have to recover cost or make some money from the device sales. It sounds like the exec confirmed that bundling is not available now. Anyway, this is a tech thread. Marketing and business model speculation goes elsewhere.

For a convincing and "realistic" Milo or Eyedentify, much tech work needs to be explored in many areas (e.g., Mo-cap facial expressions and movement like Heavenly Sword and Heavy Rain, integrated AI + animation like KZ2, ad hoc animated nuances like in Uncharted to minimize repetition, top notch natural language recognition like no other, etc).

 

Was jus thinking about possible hardcore gametypes for this and the only major problem i can see is navigating around the environment. I think offering a simple dual analogue nunchuck device could open up so may more choices. At the moment i cant imagine a zelda type game using this but add a analogue stick in one hand and a plastic sword from tesco in the other and you have possibly the greatest thing ever! What about a condemned sequel using this thing!

How could navigating you character be done with a gesture bassed system alone, any ideas?

 

I missed that part, I´ve just seen pre-recorded sample videos. Do you have a link?

 

Give me a well done boxing game with this tech please! And the minority report interface + character recognition is awesome. I'd like to see this out next christmas and marketed properly. Having kiosk's well done alone would just light the sales for this on fire. Ofcourse, they should be mindful of pricing it right and go for volume vs profit per unit.

Out of all the games and presentations, this is the best thing I've seen in this conf from MS and that's saying a lot.

 

IMO, best thing I've seen to hit our industry since 3d.

 

Sorry, no.

Any video of the conf should show it though.

They had the paint guy on stage and the blox game too.

 

The only thing that looks real time so far is this
The input seems sligthly laggy thus

 

EyeToy has a simpler model:

(That's a PS2)

MS's system should be able to detect Z-axis movement much better, but for a mini game like this, it's not a big difference.

Fine-grain gesture in a dark room is a better differentiator. That's why I think remote-free media playback will show its advantages the most. I don't think Sony opens up the EyeToy SDK ?

 

The lag is probably caused by the projector or however they capture the video feed, there was a shot from a different angle with her and the TV she was playing off of, looked like much less lag.

 

FWIW, I've had apps that've done this on my PC (bundled with Intel webcams) long before the PS2 EyeToy (or even the PS2).

So you're right in that 2D motion detection is not new, but that's far from what Natal is. Very different leagues.

 

MS's system can detect one more dimension implicitly, can see in the dark and should be much faster and accurate. Still, if the final app like the fighting mini-game doesn't show the advantages, it would not make a difference.

Both systems are tiring as a generic UI tool (You should know). So it's for short use. Doing it in XMB and the Blade UI may make sense since soon, you'll need to pick up a real controller for "deep" gaming.

The voice recognition part requires special attention too. e.g., The PS Eye has an array of mics to supposedly "catch" the speaker in a 3D space (so it's clearer). Not sure what happens if more than one talk. Overall, it depends on the programming and the application design. e.g., Echo may become a problem if a network is involved.

 

If it could detect individual finger flicks, then it would be pretty cool for controlling the Xbox Guide. I can't get to view any of the demos since this hotel's "high speed internet" is woefully slow.

 

I think the camera would work great . . . for T-bagging in Halo. I can see myself jumping up off the couch to T-bag BadTB25 during a multiplayer match.

 

It does seem like it can detect the difference between a fist and an open hand, as well as rotation of the hand. It'd be nice to hear more about the ricochet thing, though given how objective the media was about OnLive, I'm not sure I'd trust their impressions.

 

I feel Natal isn't ready for primetime yet. It's not coming out until 2010.

I feel that Microsoft should've held off on this, perfected it, developed several killer apps, as well as other decent games, saving it for 2011 and for the next-gen Xbox. I think that would've made a bigger impact. However, with Sony probably about to reveal their motion-control scheme, I can understand Microsoft not wanting to be left behind.

Also, I'm certain that Molyneux's Milo demo looked massively more impressive than what the reality is going to be. It's the motion equivalent of pre-rendered graphics.

 

@_@ What ?!

Is there a link or official source for this ?

 

first time I saw it, it reminded me of this

http://www.youtube.com/watch?v=aPENA1Bpm68