In-depth comparison of the big three's motion controllers.

[NJ5]

WereKitten said:

Where are you pulling this stuff off?

The Wiimote has an IR camera for pointing and 3d accelerometers, thus it can't detect its absolute positioning if not using the IR. When not pointing towards the led bar, all you have is 3 angles with respect to gravity direction or 3 accelerations. The Motion plus gyroscopes only make those 3 angles more precise. The software can try to integrate the accelerations - if WM+ is present to subtract the gravity- , thus "sort of" computing a 3d position, but it will need to be corrected every now and then by using the IR camera or errors will accumulate.

The EyeToy was not just used "to see yourself": the XYZ absolute position tracking of the wand from the camera was clearly used. See here for an analysis of the patent application Sony submitted, regarding the tech that is probably being used.

Take for example the two-wand manipulation of virtual building blocks demo-ed on stage. You can't do that continuously and precisely with Wiimotes or WM+ unless you restrict yourself to keep them pointing at the screen all the time, in which case from each you still only have 3 positiong degrees from the IR camera and a roll angle (around Z axis). Not the whole 6 degrees of freedom.

Edit: thanks to av36 for really useful and in-depth details. This confirmed what I knew about Wiimote and EyeToy+Wand setup, and I learned more about Natal.

 

Let's say you're right, and let's imagine IR is not even available on the Wii to perform corrections. Would it really be a problem not to have absolute 3D positioning? I think absolute positioning is not very important for games, since people's living rooms are too small to make people walk around in a way which really impacts what happens in the game.