The data given for the wiiremote is clearly for the gyroscopes. The IR tech will have considerably less lag.

Why are you so quick to say it's a Kinect bashing? All it is is comparing the 3 motion controllers. And it was apparently "uncommon knowledge" meaning it was something not that many people actually knew of. Just because Kinect *just so happens* to be the one with the greatest lag (although milliseconds are so miniscule), that doesn't mean it's a thread bashing Kinect. This is purely technological specifications. 

Or when the Gears of War 3 and Uncharted 3 come out, and the developers release the specs of their improved engines, and one engine is slightly less capable than the other, it's automatically bashing the "lesser" no matter all the great things it does, if you compare the two?

The Wiimote IR camera by itself is 100Hz in raw acquisition rate, but then the raw data has to be processed, laid out in a standard format and transmitted over Bluetooth by the embedded silicon, again at 100Hz. Indipendently from the acquisition and transmission rates, the time shift from acquisition to transmission builds the latency.

According to damnyouall's third source - it looks like a nice, independant measure set - the effective latency from IR detection to transmission is indeed about 50ms.

PS

@Squidz

Apples to oranges.

KZ2's so-called lag is from button pressure to on-screen animation going through a running game loop. Kinect's ~150ms is from input to output going through a libray call simple output loop. You'll have to probably add something like another 70-100ms on top when you measure the same in a real game with rendering, double buffering, running AIs and so on. Pretty much what the OP said, basically.

Also let me say that these raw lag numbers have very little meaning. We organically perceive times and delays very differently depending on our actions and focus. Since my arm movement or footing shifting will last a good fraction of a second, those extra 100ms Kinect takes for full movement detection might very well be not perceived at all, whereas they could be obnoxious when guiding an on-screeen pointer or in discrete sharp actions like a button pressure.

Still: apples, oranges ;)

What I really don't get is why Microsoft made Kinect so much weaker than it could have been. Instead of making the depth-finding resolution 640x480 as originally planned (which could easily have captured finger movement), they made it 320x240 pixels. PS Eye can do up to 120 frames per second. Kinect's lag is larger than it needed to be because it's running at only 30fps. Instead of giving Kinect a CPU of its own (like the prototypes had), Microsoft chose to "outsource" everything to 360's CPU, thereby increasing the lag and reducing the system resources. I suppose all these design choices were cost saving measures, but they really didn't help Kinect's performance at all:
They made Kinect less precise than Move, and gave it more lag than both Wii and Move.

Except the IR camera was always to run in 30 FPS, onboard chip or not. Thats the on PSeye doesn't have.

As far as the regular camera, there is no point for the regular camera running higher than 30 fps as thats not the camera that tracks movement. The IR camera works as fast as it could with the current tech and pricepoint.

As far as Sony and their camera being 120 fps, that is what they always do. As a hardware company they pack in the megapixels and GFLOPS in devices...

Anecdotal evidence. Must mean it's true for everyone!

Well, in that case: Funny, Gamestop execs are saying how well Kinect is doing in comparison to Move, but in my local Gamestop, I can't find any Moves and there are still Kinects on the shelves! That must absolutely be how it is for everyone!

This thread falls in line with my perception after using the devices.  The Move flat out feels more responsive/accurate than the Wii.  And the Move/Wii flat out feels more responsive/accurate than the Kinect.

Interesting, so that points to a 50ms lag with the IR portion of the controller, but doesn't necissarily represent the entire controller. Due to my past work experience with accelerometers I would expect the actual button presses and accelerometer readings to come through at a much faster rate, but I may be wrong. If it wasn't for the fact that none of this really matters as I don't notice any lag on the Wii anyways I would set up a test on my PC to see if my suspicions are correct.

Just out of curiosity, how would you setup such a test? Do you have access to a very fast camera? Because even a 60FPS one will give you a 17ms granularity, and that's quite bad to measure 49ms or less...

I thought about that when I posted. I dont' know if I even have good enough equipment at home anyways. I think the easiest would be to have the PC control a motor that moves (simply spins would work) the Wii remote. Since it would be a closed loop system I would know exactly when I sent the signal to move the controller and when the controller first responded to being moved. But I don't know how reponsive said motor would be anyways and how much lag that would introduce.