famousringo said:
There isn't much info on the IDG-600 gyroscope out there yet. The best I could find is this Ars Technica article: http://arstechnica.com/gaming/news/2008/08/wii-motion-sensor.ars "When you think of gyroscopes today, though, especially with regard to heading information, they're around $300,000 and are capable of accuracy of ten-thousandths of a degree per hour so that a 747 after an eleven hour flight can land where it's supposed to land," explained Virginia. "Game controllers, such as what Nintendo has selected, don't need that kind of accuracy. What they were looking for something in the area of one-tenth of a degree per second. [The IDG-600 gryo in the MotionPlus] measure up to 1500 degrees per second; it offers accuracy and full-range motion. We worked towards [a cost of] $1 per axis." One-tenth of a degree per second would drift up to 6 degrees in a minute of play, or a full 90 degrees in 15 minutes. Invensense's PR guy seems to be talking about Nintendo's minimum spec with that sentence, though, so it may not describe the IDG-600's actual capabilities. |
^Thanks, I looked through the documentation at Invensense but I couldn't find the accuracy numbers.
Please also note that the gyro won't drift that much, even in the worst case, unless it always drifts in the same direction all the time. These kinds of behaviour are described by the "drunkman's walk model" :)
@sethnintendo
Yep, but the question was not about the time it takes to make the calibration (it will be istantaneous) but the absence of any player input. As in: what happens if the player doesn't point the wiimote at the screen during all the loading time? You can't correct the horizontal drift.
My suspicion is that a "minor" input will be required at least once per loading, in the form of one on-screen button click or something like that.
edit: I see our monkey friend said the same yet. Oh, well, repetita iuvant :)
