40 "Rayman® Raving Rabbids" listed at Redirects for discussion.4 Rabbit saying 'Wiiiiiii!!' Screenshot.The remote functioned normally, even with the sensor bar set roughly 85 degrees off of the horizonal. Everything else is irrelevant.ĮDIT: To confirm, I just did a test where I placed the sensor bar in a near vertical orientation by standing it on one end and leaning it against the side of my TV. Start from 6 feet away and 20 degrees right and it can still determine position and distance just as accurately, provided it can see at least 2 IR sources in each case.Īll it needs to determine the remote's position in 3D space is clear line of sight on 2 distinct IR sources (whatever those may be). Start from 3 feet away and 40 degrees left and it can determine the appropriate movement. It's all relative from the starting condition. However, when determining distance or the remote's position, neither the starting angle of the remote nor the starting distance of the remote from the IR source matter. I may not have been as clear as I'd intended though. It can determine distance, as can be evidenced from the photo channel. The Wii method is just different enough that it can manage that with only 2 targets. Give it 3 targets to track instead of just 1 and it is able to determine 3D movement based only on the 2D respresentation of the relative sizes and orientations of those 3 points. ![]() It's a similar system to what a TrackIR uses with the "Vector expansion". ![]() That's all it needs to do to work, so why complicate things? It looks for emitters and tracks their relative movement. Since it clearly does work just as well under these conditions, that suggests that there's not a lot of complex math at work here. Otherwise you wouldn't be able to use 3rd party or homebrew sensor bars with different numbers of emitters, or a couple of candles spaced farther apart than the sensor bar and still have it work. Absolute angle and distance don't matter. Deviate from that and it moves the pointer accordingly.Īs long as it can clearly see at least 2 IR sources it can determine the remote's position. Get both sources in the middle of the remote's view and it centers the pointer on the screen. As long as the IR sources are visible to the remote, it can track movement because it's only sensing relative movement. Angle doesn't matter, and doesn't come into play at all. But I haven't seen any evidence the system does that beyond simple triangulation (ie, it sees 2 sources, and based on their apparent movements determines the motion of the remote, which is point 3 of the triangle). I'm not willing to work through the math for a forum post, but it seems reasonable enough to me. Based on the apparent distance of those LED's from each other (and how many distinct sources you can see), it's probably possible to get a pretty good estimate of the angle (and, thus, the distance). ![]() I think this is why there are five IR LED's in each corner of the bar. The tricky bit is when you are sitting at an angle, which would make the two emitters look closer together. Geometrically, there's no reason why they wouldn't be able to tell exactly where you are. ![]() The Zelda pointer calibration also seems to know exactly how far away you are.Įdit: Geometrically, there's no reason why they wouldn't be able to tell exactly where you are. Well, Rayman sure gets grumpy if you sit too far away from the screen. If it sees the IR sources getting closer together and dimmer, it assumes the remote is moving away. If it sees the IR sources moving farther apart and brighter, it assumes the remote is getting closer. I do not believe that would be the case, as the Wii seems to sense relative distance and not absolute.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |