I've spent the last year avidly tracking down and scrutinizing as much Wii footage as I could get my hands on - especially any videos which showcased both TV and user at the same time. Having determined in advance some aspects of the nature of the wireless control scheme, and being aware of the limits of some of those aspects, I was keeping my eyes open for one particular phenomenon.
The phenomenon I had reason to anticipate seemed to be verified by several of the aforementioned videos showing people playing various games. But I had to be sure. So I picked up my Wii yesterday and have spent ample time utilizing it.
Understand that in my capacity as a video and audio hobbyist, timings, and particularly sync-related issues, are something I'm used to being meticulously aware of. Here is what seems to be happening, along with my suspicions as to why.
The Wiimote's traditional controls - the "plus" pad, the analog stick, the buttons - all seem to be utilizing the same, effectively latency-free wireless technology utilized by the Gamecube Wavebird controller. There is no detectable lag.
The functions related to determining the positioning and angles of the Wiimote and nunchuk are most likely accomplished exclusively by the infrared sensor bar. This would explain the considerable latency, which I generously estimate to be 150 milliseconds, or comparable to a speedy dial-up modem connection or an intolerably lagged broadband connection.
These phenomena are observable, and identical, in both Wii Sports and the Wii's built-in default system menus. Considering the well-established latency issues related to infrared technology, there is little wonder that such phenomena are observed.
The wonder, instead, is that such a patently poor communication technology was chosen for use in a control scheme on a videogame system - a platform on which latency is of paramount concern.
I am tempted to produce slow-motion videos, slowed down, to illustrate precisely how much latency is being exhibited, between flicks of the Wiimote and on-screen result. To such a video, I would add the result from a wireless mouse I've owned for years, which manages to be latency-free as well as sensor bar-free and yet works like a charm at three-dimensional hand motions. Others may feel free to develop their own video, proving the point, or, brave soul ye, attempting to disprove.
Now, the good news is that since proper wireless motion technology has existed for many years, the option is very much open for Nintendo to offer, say, a "high accuracy, low-latency, sensor bar-free" alternative to the Wiimote. There is no reason why such a product could not be fully compatible with preexisting games.