Talk:Wii Remote/Archive 1

Archive 1

Archive 2

Archive 3

accelerometer

I had to remove this:
" An accelerometer sensor from Analog Devices in the Wii Remote itself allow it to sense rotational and translational movement along all three axes, providing a full six degrees of freedom with just the Remote."

There is no doubt (now) that there is this device in the remote - but the text is still wrong - an accelerometer (probably) does not sense rotation etc - the accelerometer may be only one of a number of devices in the remote giving positional information. It's more likely that the accelerometer is the only active device in the analog stick, but there is more to the pointer/controller. Hope I'm not wrong 'cos doing this will make me look stupid...HappyVR 21:57, 10 May 2006 (UTC) Note also the 'sensor bar' - probably not necessary for just a accelerometer controlled controller.HappyVR 21:59, 10 May 2006 (UTC)

Actually, accelerometers sense rotation better than they do translation. If the vector magnitude that the three-axis accelerometer senses is 9.8m/s², then the controller knows the absolute direction of gravity, and therefore knows exactly how it's tilted. On the other hand, if it senses more or less than 9.8m/s², then it knows the player is accelerating it in some direction, and it may try to sum up the distance that it thinks it's travelled, but still, for translation movements, it can only sense relative (not absolute) position. --Interiot 01:02, 11 May 2006 (UTC)

That sounds like a direct quote from Sony's press conference, it describes PS3's controller.Qwerasdfzxcvvcxz 02:01, 11 May 2006 (UTC)

C'mon, I'm not that good, especially when it comes to technical writing. I was just trying to keep it from sounding like mumbo-jumbo by relating the technical aspects to what the reader was already familiar with (eg. holding the controller and tilting it). --Interiot 02:24, 11 May 2006 (UTC)

I don't know what you two are on about - but I don't recommend 'cut and paste' editing if you don't understand what you're doing.HappyVR 06:36, 11 May 2006 (UTC)

This makes a lot of sense, Interiot, but I'm still a little confused. How would a three-axis linear accelerometer distinguish yaw? I don't think you said that it did, but is it wrong to assume there is another sensor in the remote? Perhaps another accelerometer off-center? I thought that perhaps the sensor bar could account for that last part, but when you're holding the Remote gamepad-style, your left hand is covering the IR(?) window. And the videos on Nintendo's Wii page seem to indicate that yaw can still be detected. Any ideas? Dancter 16:57, 11 May 2006 (UTC)

I'd assume that the sensor bar has something to do with detecting yaw - can't answer your question about ir windows - perhaps it uses radio/microwaves? - I definately assume that there is more than just an accelerometer in the remote - otherwise there would be no need for the sensor bar - one previous suggestion has been ultrasound (see pre e3 Wii page in rumour section for an arstechnica.com article) - a simple pendulum would detect 2 rotation degrees of freedom - and I guess the sensor bar gets the third.HappyVR 17:06, 11 May 2006 (UTC)

I was assuming that the sensor bar is IR-based, which is why there'd be that opaque window at the top of the remote. (Unless it's just there so you can use it as universal remote. You know, change the channel, program your VCR...that kind of stuff.) If it is IR-based, then there would need to be a clear line of sight between that window and the both of the sensors at the end of the sensor bar, which would need to be long so the sensors could be spread apart, making it easier to accurately triangulate a position. (There's no real confirmation of any of this, but it's the best explanation I can come up with.) But in those videos I mentioned[1], there is footage in there of a guy still utilizing the yaw function (arguably) even with the window covered. Dancter 18:10, 11 May 2006 (UTC)

Actually, the + and - buttons on the Remote would seem to corroborate the idea that Wii Remote does have an IR transmitter. This, the power button on the Remote, the opaque window at the top…this would seem to support just the universal remote idea, except for the fact that there is the same opaque plastic on the front of the sensor bar[2]. It's all speculation right now, but still… Dancter 20:33, 11 May 2006 (UTC)

Interiot is right, a three-axis linear accelerometer can determine tilt. To my best understanding, this gives you up to five degrees of freedom (I could be wrong, though). But I still can't account for yaw. So, yeah. Still some unanswered questions. But I'm betting there's at least one other sensor in the Remote. And though it doesn't need be, I'm betting it's based on that Gyration technology Nintendo invested in all those years ago. I'd imagine that a gyro-based rotation sensor would be more responsive, more accurate, and more resistant to vibrational disruption than an accelerometer-based one. (Though I have absolutely nothing to back that up. Just my thoughts.) The postition-sensing sensor bar would be the final piece to give the Remote pixel-point accuracy. Dancter 18:27, 11 May 2006 (UTC)

Without going into details - with enough 3 axis accelerometers (not more than 4) it will be possiblr to detect 6 degrees of freedom - 3 translational (movement) and 3 rotational. However what an accelerometer can never do it detect its absolute position in space.HappyVR 17:57, 12 May 2006 (UTC)

Not more than 4? Actually, I would like some details about how you came up with that number. As I mentioned, to my understanding, with a three-axis linear accelerometer, you would only need one more sensor for that sixth degree, and all it needs to be is a simple one-axis accelerometer, placed off-center. And I'm not arguing against the usefulness of the sensor bar. Like I said, with the sensor bar, the Wii Remote can achieve pixel-point accuracy. I'm just trying to reconcile the known facts. I mean, why hasn't anyone else asked the question of how the sensor bar works? What the opqaue window at the top of the remote is for? Is it to use the controller as a universal remote? Is it for the sensor bar? Is the sensor bar utilized for that sixth degree of freedom? If not, then how does the Remote achieve it with just a 3-axis linear accelerometer? Dancter 18:52, 12 May 2006 (UTC)

One three axis accelerometer (giving relative position,speed etc) and 3 one axis accelerometers located along x, y and z (orthogonal axises) axis with the three axis accelerometer at the origin, the detection axis of the off axis accelerometer would of course be at right angles to the line connecting the 'center' of the 1 axis acc. and the 3 axis acc. Also the detection axises of the 3 off center acc. would not coincide. That's how I got 4. There might be a simpler way with less but this setup would work to detect 6 types of motion - x,y and z linear motion and rotation about x,y and z axis.HappyVR 19:05, 12 May 2006 (UTC)

(Actually the 4 acc. i mentioned would contain in total 6 linear acc. - I should have been more clear)HappyVR 19:32, 12 May 2006 (UTC)

No worries. I was starting to guess at what you meant before you clarified it. But as Interiot indicated, with the presence of acceleration due to gravity, a three-axis can determine upward or downward tilting by factoring in the changes to the gravitational force vector relative to the controller. That would eliminate the need for two of the one-axis accelerometers, leaving one whose detection axis is off-center and orthogonal to the z-axis. Dancter 20:21, 12 May 2006 (UTC)

Yes - maybe they should be properly described as 'force sensors' then - could be a strain gauge attatched to relativly high mass. My guess is still that any accelerometer is used for just that - detecting acceleration eg flicks of the wrist etc and the positional sensing done by something else. HappyVR 13:53, 12 May 2006 (UTC)

You didn't sign your comment, so I added it. Anyway, I think we are in agreement. Dancter 21:17, 12 May 2006 (UTC)

As for the sensor bar - you could be right about IR - or it could be microwaves etc - who knows? The controller could just transmit millimeter radiation (point source) and the actual position detection be done by the sensor bar (like some sort of radar!)HappyVR 19:13, 12 May 2006 (UTC)

I posted this link [3] below, but it's probably more relevant here. It seems the controller uses a combination of a gyroscope, accelerometer, and an IR signal generated by the sensor bar for its positioning and motion detection. 129.186.140.209 09:56, 16 May 2006 (UTC)

Well, actually, the technical discussion about the remote has actually been all over the place. (Much of it is my fault. I've been thinking about refactoring it all.) There was even more further down the page, where we were discussing the tracking technology that was revealed, and how it fits in all this. I was thinking that this info would be the last piece of the puzzle before we could start including stuff in the article, but I'm not sure anymore. I included in the article a bit about the tracking technology that I thought was minimally speculative, but it was still removed. Dancter 15:26, 16 May 2006 (UTC)

Technical specifications?

Just to say, I think that the link to ADI's press release is too hidden away. I believe the information contained within it would be of great interest to anyone trying to understand the technology behind the remote. 87.81.125.8 11:52, 12 May 2006 (UTC)

As in the example above, mention of the Analog Devices' connection to the controller has been added a few times, but usually have given it too much weight. Perhaps a technical specifications section? Then information can be added about the Broadcom providing the Bluetooth chip[4], and any other technical stuff that might be of interest. Dancter 16:11, 12 May 2006 (UTC)

Like you say it is important not to mislead people, but I feel it is as important to display all the facts we know of in order for the puzzle to be pieced together more quickly. A short summary with clear links these press releases might suffice? 87.81.125.8 09:13, 19 May 2006 (UTC)

Well, whatever you have in mind, I say go for it. Be bold, as they say. I made a list further down the page listing some links that may be helpful. Dancter 14:13, 19 May 2006 (UTC)

Pictures

Someone needs to come up with a decent way of arranging the pictures, as the current one doesn't work. Rather, I should explain myself: I'm of the opinion that the layout should be optimised so that it looks good on as many resolutions as possible. Unfortunately, the current picture layout, with the wide controller image, doesn't work well at 800x600. The picture itself is 700 pixels wide, causing the horizontal scroll to kick in when combined with the menu to the left (and the H-scroll is generally considered a major negative point when designing webpages).

However, the previous layout (and the current one, for that matter), don't work well on higher resolutions. The pictures have been placed in such a way that, at higher resolutions (anything above 1152x864, to be honest, meaning that the current layout really only works at 1024x768 and 1152x864), the pictures sort of stagger themselves, resulting in bits of text actually being covered up by the pictures. I'm thinking a gallery along the bottom might be the way to go, with only one or two images in the article itself.

Opinions? g026r 01:26, 11 May 2006 (UTC)

I've done up a sample layout with a gallery. Obviously the exact locations of things in the gallery can be fiddled with, as can the locations of the other items in the text. However, I felt that 1) there should be a picture of the controller itself first thing, rather than the method of using it, and 2) that the wide functions image, while it works fine on the original Nintendo page, doesn't work in a Wikipedia article, as it's both a bit too large when not thumbnailed, and when thumbnailed to a size that's not too big, but still big enough to understand, that it essentially divides the article, where ever it is placed. g026r

...and I've applied my layout design, since I got no feedback on it. Let me know what you think. g026r 04:33, 11 May 2006 (UTC)

It liked it, but then I added sections to the article, which messed up the aesthetic quite a bit. I don't really have time to fix things, but I liked your style. Perhaps you could fix the layout again, while keeping the sections? Things seem a little bare with them now, but I think the sections are necessary for organization as the article expands. Dancter 12:27, 11 May 2006 (UTC)

I'm not happy with the way that the top of the page looks. Suggestions? Ladlergo 13:41, 11 May 2006 (UTC)

I thought it looked fine before I added the sections. Off the top of my head, I'm thinking that either the introductory text could be expanded (can't think how right now), or a wider image can be used in place of that top image. Or both. Also, would it work to move some of the pictures from the gallery back into main article? Dancter 15:03, 11 May 2006 (UTC)

I like the Wii remote image, but it stretches the section too much. Can we get an appropriate image that's wide instead of tall? I think that it would be fine to move the nunchuck, classic controller and gun shell images into their own sections now. Ladlergo 15:25, 11 May 2006 (UTC)

Actually, there are some more illustrative images out on the Internet now, including some showing the sensor bar's placement around a television[5], the Classic Controller connected to a Wii Remote[6], and even the Virtual Console[7]. I doubt any of the ones I linked to are usable on the article, but hopefully some will turn up soon. Dancter 16:11, 11 May 2006 (UTC)

It's simple, really

There's too many pictures, and not enough information.

There was plenty of information that was not really appropriate to elaborate on in the original Wii article, though I tried to make sure that at least the references survived through citations, but I don't see why it can't be done on the Wii Remote page. For example, if someone wants to re-introduce that blurb about Analog Devices providing the linear accelerometer, I think it's possible now, as long as it's put in the proper context. Keep in mind that just one of those things cannot provide the full six degrees of motion the remote is capable of. There is at least another sensor in there, and likely a different kind. Although I haven't found any evidence yet, there is probably a gyro-based angular accelerometer in the Remote like the type Nintendo invested in with Gyration Inc. Dancter 11:39, 11 May 2006 (UTC)

Actually, the information about the movement sensors is gone entirely! All that's mentioned is the sensor bar. While it doesn't need to be as verbose as I tend to make it (which I do keep things factual), I think this should be put back in. Dancter 11:56, 11 May 2006 (UTC)

Fixed. Dancter 18:33, 11 May 2006 (UTC)

Organization

I suggest separating them by section:

• Remote and nunchuck
• Virtual Console controller
• Gun shell

And perhaps another section (at the beginning?) for discussing what's known about the technology. Ladlergo 02:22, 11 May 2006 (UTC)

contradiction?

The article says the remote can be used 10m away yet later states the sensor bar works up to 5m..Could anyone supply references for either of these two facts and maybe an explanation?HappyVR 06:31, 11 May 2006 (UTC)

Actually, I was the one who added both of these facts to the original Wii article, and later added the citations. They both seem to have survived to this article. I thought it was clear enough: the Remote can communicate as far as 10 meters from the console, and can be sensed as far as 5 meters away from the sensor bar. (Actually, I wrote "5 meters away from the screen", but I thought it was appropriate for clarity, as the sensor bar is supposed to be placed proximal and parallel to the display, anyway.) You can try and clarify this more, if you like. Dancter 11:39, 11 May 2006 (UTC)

Thanks, I understand that now.HappyVR 16:55, 11 May 2006 (UTC)

Design

Some information should be added on the other design elements of the controller, such as the buttons, their layout, and what they do. Perhaps rename the Remote control design section and expand it? Also, some mention should probably be made concerning the battery. I believe the current design uses two AA batteries, rather than an internal rechargable cell, but I haven't found any confirmation that this will be the case for the final implementation. Dancter 13:45, 11 May 2006 (UTC) I guess I was right about the battery[8]. Dancter 23:19, 11 May 2006 (UTC)

Well, the functions of all of the buttons are currently unknown, but if you'd like to start adding what is known, go ahead. I think the retro controller would benefit from more details (Z, L&R, etc). Ladlergo 14:05, 11 May 2006 (UTC)

How about size of wiimote? The article should say the controller's dimensions. —The preceding unsigned comment was added by 68.66.203.35 (talk • contribs) 02:39, 20 May 2006 (UTC).

Not forgotten. I just don't think anyone's found that information yet. If you have, though, feel free to add it yourself! Dancter 03:02, 20 May 2006 (UTC)

Would like to just state that this page is really coming along. It looks better and better every time I check it out. Of course when more info is release, someone will put it in.--DivineShadow218 04:26, 12 May 2006 (UTC)

I agree; it's already in a much better state than DualShock. smurrayinchester^{(User), (Talk)} 06:23, 12 May 2006 (UTC)

how it works

Quite a few people have been wondering how this thing actually works - well maybe just me. This link[9] doesn't have the answer but does give one method for making such a thing to work - for any one who is interested.HappyVR 19:58, 12 May 2006 (UTC)

Thanks for the link. I didn't read the article when you first mentioned it. Now after reading it, I probably should have. Would've saved me some of the trouble. ;) Dancter 20:21, 12 May 2006 (UTC)

of course Nintendo will update their site so once that comes up, lets update it per Nintendos site. --DivineShadow218 22:11, 12 May 2006 (UTC)

None of this has shown up on the article, has it? Also, Nintendo is not known for being either timely or thorough on their website. I'm as much for verification and fact-checking as the next guy, but if we went just by Nintendo's website, we wouldn't know much at all. Dancter 22:24, 12 May 2006 (UTC)

not for a while atleast--DivineShadow218 03:36, 13 May 2006 (UTC)

According to this article [10] ArsTechnica seems to have it backwards. The sensor bar generates an infrared field, and the remote detects the movement through that field. I'm not quite sure how the controller is able to detect full position and orientation, but this sounds like a start. --129.186.140.209 09:45, 16 May 2006 (UTC)

classic controller

"The purpose for these notches remains undisclosed" Just wild speculation this but could it be that the notches are actually for plugging somthing in eg a position sensor. Also it's possible that the controller overall design is not totally finalised and that the sensor part of the controller might be detachable - therefore it could be 'unplugged' and re-attatched to the classic controller on the underneath - giving a fully position sensing classic style controller (at no extra cost) - I just wanted to get that off my chest.HappyVR 20:53, 12 May 2006 (UTC)

As a far simpler solution could it be that the notches are simply there for demonstration purposes so that the controller can be solidly attatched to a support. (preventing over enthusiastic nintendo fans walking off with it?)HappyVR 21:04, 12 May 2006 (UTC)

In the page I cited, the rep said that it was a secret. Dancter 21:10, 12 May 2006 (UTC)

Gyration

As far as I can tell there is no mention of 'gyration' in the article? It has been inferred previously that nintendo's investment in the gyration company could be significant. Anyway here is a good link giving some background...http://www.gamecubicle.com/news-nintendo_gyration.htm HappyVR 20:59, 12 May 2006 (UTC) I can't guarantee that the info in the link is actually correct (being cautious)HappyVR 21:04, 12 May 2006 (UTC)

Well, it's been pretty quiet on the Gyration front as of late, but I managed to find this[11] as recently as February. Still far from a confirmation, though. Dancter 21:08, 12 May 2006 (UTC)

Looked at it - can safely say I couldn't make head or tail of what it was saying or the picture. quote "sensing element is a non contact metal beam vibrated electromagnetically"(abridged) - I honestly just don't get it. But nevermind.HappyVR 21:20, 12 May 2006 (UTC)

read my replay in How it works...--DivineShadow218 22:13, 12 May 2006 (UTC)

6 degrees of freedom possibly wrong

I've noticed nowhere does Nintendo claim the controller detects all 6. The Japanese nintendo site mentioned in the article (via translation tool) says the controller detects 3 linear axises and 'change of inclination' plus pointer function (as per a standard mouse for instance). For this only 5 degrees of freedom need to be detected - the missing one would be rotation about the long axis of the controller - furthermore I've seen nothing to suggest that the controller detects this 'rolling or spinning motion'. Hope thats clear. If I'm wrong, again apologies - but it seems to me that the article at present has assumed too much. (as a side note re previous discussion 'accelerometer' it would seem that implementing the 6th degree of freedom is the 'hardest' bit to do..) I can see that implementing roll about the long axis of the controller could be useful (eg turning a key in a lock etc) I'm not sure that it has been done. Anyone got evidence that it has? If not suggest rewording the article just to say that the controller detects 3 linear degrees of freedom and a further 2 rotational ones...HappyVR 22:26, 12 May 2006 (UTC)

See, I like it when these things are talked out. I re-read it, and I think you're right. Well, actually, it's yaw that would be the issue. The side-to-side turning. (Sensor bar?) Roll and pitch are fine. But it doesn't actually say that the angle detection is also in three axes. Good work. See? Would you have looked for this had I not started asking? Dancter 22:39, 12 May 2006 (UTC)

I'm using Flight dynamics as a reference since it defines three angles in an easily understandable picture. (I'm assuming the longest length of the remote is considered to be the body of the plane in the picture - hope that's clear) - I hope we are in agreement that in terms of the picture [Image:Flight dynamics.jpg] it's roll that probably isn't implemented.
I'll leave this for a bit to see if anyone else knows more, but it seems best to ere on the side of caution and change the wording of the article slightly (removing the 6 degrees thing obviously)HappyVR 22:47, 12 May 2006 (UTC)

I still think it's yaw. Note that in some of the WarioWare: Smooth Moves footage (you can see it on the promo videos on NOA's Wii page[12]), someone actually does turn a key in a lock using the remote. Either way, it's probably best not to be too specific about the rotation part. Just that it senses tilt. It's a safe enough and informative enough statement. And also, I don't think it's been confirmed that the sensor bar detects position. Dancter 22:57, 12 May 2006 (UTC)

Also http://wii.nintendo.com/controller.html shows what is definately implemented - pictures (set of 4) entitled wave,point,rumble, listen. show two rotations mapped in 'wave'HappyVR 22:52, 12 May 2006 (UTC)

Even the picture's ambiguous. It doesn't necessarily indicate rotation. The controller is moved. Regardless, the Japanese website does say that it detects tilt. Dancter 22:57, 12 May 2006 (UTC)

I made the changes. What do you think? Dancter 23:29, 12 May 2006 (UTC)

Yes, I think it's best to ere on the side of caution as to what the controller can actually do - it's clear? that linear acceleration is detected and hence position, my interpretation was that yaw was detected as well but for now I'll just wait and see.HappyVR 06:39, 13 May 2006 (UTC)

Actually, with what we know about Wii, yaw detection is more than possible. My question was just if the sensor bar was needed for that. It would only make a difference in limited cases; say, if you were simulating spin-the-bottle. If you needed a line of sight between the sensor bar and the nose of the remote, instead of the bottle spinning, it would be interpreted to be rolling away. Dancter 09:31, 13 May 2006 (UTC)

Good Article

With the progression of this article as days go by and Nintendo releases more and more info, I suggest se nominate this article as a Good Article or Featured Article, depending on how much info we get. but since this article will change as the weeks go by, we cannot nominate just yet. --DivineShadow218 03:39, 13 May 2006 (UTC)

The article is good - but there's still some confusion as to what motions are actually detected - in my head at least.HappyVR 06:39, 13 May 2006 (UTC)

since there is a sencer, the the contoller allong side the Wii can ditect every motion including speed and depth. --DivineShadow218 19:03, 14 May 2006 (UTC)

I would suggest waiting until the console is actually released. Most information here is about a device that has not yet hitted the market and may still be greatly modified (who would have thought about the speaker?). Also, we need free images. The article can be a good article, but is too small for a featured one (in fact, I would not be against merging this article back to Wii). -- ReyBrujo 16:10, 16 May 2006 (UTC)