CrazyHorse said:
So is this just a case of different technologies creating the same outcome or does the skeletal system Natal creates make for more accurate body motion detection (when the PSWand isn't being used)? It does seem that for all the media hype surrounding Natal it doesn't really offer much more than the PSEye. |
Creating a skeletal system using pixel colors and edges can be quite difficult. Add the need for detecting multiple people and it becomes even more difficult. Not saying it cannot be done, but detecting distances using only those methods can be quite inaccurate. There will not be a 1:1 relationship between the PSEye and the player. That is where their wands come from. It makes it much easier to detect position, motion, etc. But this is only for two points. It doesn't detect the legs, waist, etc.
So the difference here is that Natal offers full body motion detection. Legs, arms, head, etc, are all accurately detected. The problems seen at E3 were with animating the skeletal system during a rotation. Natal only sees a 2D set of pixels, but each of these pixels has a depth associated with it. Much like that toy with all the metal rods and you push your hand in it and it creates a mold. That is similar to what Natal does. So Natal must have algorithms to predict how the skeletal system should be during a rotation. This isn't too diffuclt as the skeletal system is quite standard among humans:) They obviously had some flaws in their code. Rest assured this will be fixed upon release. This skeletal system can be directly associated with an in game avatar. This isn't the only use, but it is definitely one of the big ones.
This allows for complete 1:1 interaction with a surrounding of around 20ftX20ft(I cannot remember the exact dimensions). So there are a number of game designs where you and your character inhabit the exact same 'space'. Dodgeball would be a good example. You can jump, dodge, throw, catch, and interact 1:1 with the game environment. Doing this with only the PSEye would be extremely difficult. Even the wands would be hard as you jumping and moving around would be hard to do as well. Especially if there were environmental things on the ground where foot placement was key. Such as holes or bars that needed to be jumped over. The wands could only tell that you moved, but the video camera would have to handle the feet detection and distance detection. So it becomes extremely complex to handle this with the PSEye/Wand combination. Whereas Natal lends itself quite easily to this situation.
What does make Natal stand out is the standard features that it has. Voice recognition, facial recognition, gesture recognition, and 1:1 motion recognition. Developers can rest assured that all Natal hardware will be capable of these things. Customers are not required to be connected to the internet and download patches to aquire these features. The computation of these things also occurs completely on the Natal hardware. Developers are not required to sacrifice any of the 360's time computing output from the Natal hardware. It is basically like handling input from another controller.
