After a bunch of (mostly) expected games, Microsoft's E3 press conference finally revealed something of greater interest and scope: Project Natal, the code-name for their well-rumored motion-sensing bar.
The name says it all. Crave is our blog about gorgeous gadgets and other crushworthy stuff. If you would like to contact Crave with a tip or comment, please write to: crave@cnet.com
Add this feed to your online news reader
Web sites launch all the time, but they also shut their doors. We highlight 15 that bit the dust this year.
Let the debate begin: Was the iPhone more important than iTunes? Was anything bigger than Google finding a great business model? CNET offers its list of the 10 most important stories of the '00s.
Among the demos, Milo is the only one that is amazing; however let's not forget that Milo has been taking 8 years of development since 2001. It will not be possible for games that are made in just 2-3 years to compare with it and by the time developers have remake Milo Xbox will be Amiga. The world of AI is not yet ready for full-scale, widespread computer-based emotion, and no amount of business will power will change that for at least another 5-10 years.
With the possibility of having a Milo-type game ruled out, Natal become merely a motion-sensing camera, something which is already possible for quite some time. Just like another commenter says its fate would just be a beefed-up Playstation Eye Toy.
Another thing with Milo though; their emotion engine is quite impressive from the look of it. Lionhead has had quite a good reputation to me since their games are generally quite original and unusual unlike many of today's games. With or without Natal, when they decided to release Project Dimitri (Milo) it could be a real breeze if it really is as good as it seems to promise.
I prefer a controller over a camera ANY day.
The Canvas game shows the current depth of field detection requires all participants to be relatively in line with each other or they will become the background or foreground. In addition their scene is not in a living room which may have objects in the sensors field of view which could cause false detection.
The voice recognition will have a hard time with multiple participants as it will likely ask everyone to talk in turns in a slow and recognizable tone to it. After which its recognition will be 95% at best but if its not 99.9999% people will get frustrated with the system. In order to correct you likely need the controller and that would defeat the whole purpose of the system.
The face recognition which isnt likely to be working right now will be very inaccurate as current commercial systems with HD cameras and servers running analysis software can only give a chance that it is the correct person. Have a couple people in the scene and it might just mix up the users easily and also add frustration. Its easy for a camera to detect a face is present and log into the few accounts it has (probably one). But if you have lots of friends over all in the view at once its going to be very hard to ID them with enough accuracy to work.
The 3d motion capture system is also not likely to work well as it has already been shown it only has two cameras to detect motion so if you are moving out of its plane of view it cant see that. Movie motion capture systems have to use countless IR cameras around the user to capture the full motion. That is why their demos are all 2d with limited depth.
This will make playing games even more frustrating as your actions may not be accurately captured by the system. As shown in the avatar posing demo it was very noisy and jittery which in a HD game my be disturbing and distracting if the character your controlling starts to flay around randomly.
In addition the technology for multi-user 3d capture doesn't have enough maturity to be used in a consumer product. The processing and hardware just isn't there and for consumer pricing it really isnt going to be there. Computers have a very hard time just identifying objects from the background and adding multiple targets to track and generate an accurate bones model for each isnt possible even in research environments.
Project Natal is just a fancy concept but in reality its vapor ware at best. If they do attempt to release it as a product it is doomed to failure. In addition game developer support and the fact that it is not included with every console will complicate making games for it. Using it and a controller will make for an even more awkward and frustrating experience.
If they do get it to work it will be revolutionary and should be used in a lot of other applications.
I would hate to be playing Forza 4 or a great racing game and have to wave my have to steer you cant beat A good solid racing wheel with force feed back "you just cant" And about COD Modern Warfare imagine having to dance your way through the levels sweating to death "NO THANK YOU".
However this is still a great way to play some games with the addition of having this feature in the game.
For example playing a sci fi game like Mass Effect and in a part of the game imagine controlling the computer screen moving the planets exploring deep space and communicating in the holographic computer. now having that kind of control in the game would definitely work"
I'm not a person that reads allot, but at least I can see this race was started by them.
- by sai-tumu September 28, 2009 12:17 PM PDT
- Did anyone notice the thing where the hard drive is supposed to be??
- Like this Reply to this comment
-
-
- by xXSunaBouzuXx December 23, 2009 11:12 PM PST
- Yup there is a gray thing instead of the HDD, maybe that attachment is going to make the difference. . ^^ who knows ^^
- Like this
-
Showing 2 of 2 pages (66 Comments)