Official Nintendo Fanboy Smugness & Gloating Thread

[xav]

Mind you I've just though of one problem, if the lights are in essence
resolved on the sensor as points, how do you distinguish between
the points appearing closer together due to distance as opposed to
the viewing angle.

--*--------*-- <- normal distance, straight on
---*------*--- <- further away, straight on
---*------*--- <- viewing from the side, first * is distant, second * is closer

You don't need to worry about distinguishing between them unless you're trying to work out the absolute position of the controller. For relative position (up/down/left/right/forward/backward) you only need to compare the position of the points with how they were in the previous frame: have they moved, and/or has the spacing between them changed?

This won't take account of someone moving the controller from one side of the screen to another - i.e. moving from one seat to another. But if the controller-as-laser-pointer works best for small wrist movements, then this probably isn't considered a likely scenario. The "large movement" games - golf, sword fighting and so on - would be using the MEMS sensors more than the screen tracking.

Consider the SMG video showing someone playing the game. The first thing they had to do was to aim at Mario and press a button. This could be to set the original reference image, against which positional changes will be tracked. I would guess that the "Pause" screen in most games will offer a similar means to reset the controller's central position, to take account of the "switching seats" scenario.

I don't think a camera tracking two points can be the answer.

I suspect it's not the answer, just because of the lack of IR transmitter glow on the sensor bar. But I don't see any technical reason why a camera in the remote, together with the MEMS devices, couldn't achieve everything that we've seen the controller do.

[Kemper_Boyd]

VGA is relatively high res, compared with an optical mouse .

Well the $5 VGA camera was just an example of how
much a far more capable sensor costs. I'm sort of
assuming they've got a specialised sensor and that
with Microsoft practically giving away excellent Mice I
can't see that such a part can cost too much. Anyway
I'm no longer convinced that the camera is good
explanation of how it works.

[Kemper_Boyd]

I suspect it's not the answer, just because of the lack of IR transmitter glow on the sensor bar. But I don't see any technical reason why a camera in the remote, together with the MEMS devices, couldn't achieve everything that we've seen the controller do.

It's certainly not an open and shut case. There are all sorts of possibilities,
with corresponding problems and solutions. It seems to me that Nintendo
would want something effective, cheap and simple (no fiddly calibration,
or annoying limits). It'll be interesting to find out what they've actually
done. I bet it'll be one of those 'oh that's so obvious now you've told me'
moments.

[Roobarb]

You can see the frequency response of the Sony Nightshot camera here (curve to the right):

The main difference between the Nightshot and a normal videocam is the normal cam has a near-IR filter in front of the sensor.
IR remote controls generally operate at ~850-950nm.
You can get off the shelf IR LEDS and detectors that operate up to 1000nm (Silicon) and 2000nm (InGaAs).

If there is a camera in the controller, it is very likely to be a silicon based part, so it will have a similar IR frequency range to a standard video/stills cam. That, combined with materials cost ($2-5 is pretty significant, when you consider the wavebird probably cost ~$10 to make originally, based on its RRP), leads me to think that whilst technically viable, it doesn't seem likely to me to be what they've actually done.

[zbandito]

Many IR remotes will show on non-SonyNightShotâ„¢ video cameras. Dunno why, haven't tried recently though.

I don't think there is a camera. Too complex and expensive, too much electrical power and too much CPU power required. I would guess ultrasonic, but I'm not sure how they would work around reflections.

Is there some reason we don't think they are plain old radio waves?

[Roobarb]

Many IR remotes will show on non-SonyNightShotâ„¢ video cameras. Dunno why, haven't tried recently though.

They use the same CCD as standard videocams, the difference is that standard videocams have an IR cut-off filter in front of the CCD. With the Nightshot it can be moved out of the way. Those curves show the frequency response with the filter in and out.
Even with the filter in, a normal videocam CCD would have the same response to a common 900nm remote IR LED as it would to a 550nm (green) light, which is 50% of max response (centered at 700nm - red)

Is there some reason we don't think they are plain old radio waves?

How are you thinking this would work?

[Atty]

Very, very positive impressions from IGN on one of the controller demos called 'Obstacle Course'

http://uk.revolution.ign.com/articles/707/707953p1.html

The controls of this game are where the demo really shines. The aiming is 100% spot-on and sensitive enough that you hardly have to move the controller at all. It demonstrates perfectly that the remote will be able to be used with no effort or strain on the wrist if developers decide to program games to perform at that level of this sensitivity. With the smallest of movements the icon and your character went exactly where you wanted and it seemed nearly instantly second nature. This demo proved what Nintendo has been saying all along: Anyone can use Wii, and the remote responds with perfect control when done right.

After a few rounds of the game demo, hopes are very high that more developers somehow unlock the magic that is contained inside that tiny little remote, because the games that could be made would blow anything up to this point clear out of the water.

It sounds like it's the implementation of the controller technology that's at fault in games like Red Steel, not the technology itself. If this hands-on is anything to go by, the aiming on many games should improve dramatically before launch.

[Samus]

Snakey (and everyone else of course), here's some different and also truly amazing footage of Mario Galaxy and Zelda (including the much discussed shooting with a bow, only this girl can actually do it properly). The game design in MG gave me goosebumps as did the Zelda fishing and bow shooting. It also shows how people are swinging the Wiimote while playing Tennis, so it can definitely work like that as well. Better yet, the guy playing Mario Galaxy didn't seem to be a big fan of the Mario series and you can hear how his opinion evolves while he plays this one. He didn't even do the basic jump combos or backflips or anything (guess he really hasn't played a 3d Mario game properly before, or he's just too excited?), but was still having fun!

Now this is why I'm excited abt it

Dowload here (QuickTime format): http://zdmedia.vo.llnwd.net/o1/Podcasts/051106.m4v

[Darkflame]

cheers for the link.

And thats a fantastic tshirt.

I bet it'll be one of those 'oh that's so obvious now you've told me'
moments.

Ninteligence you mean?

[Anders]

Does that t-shirt come in black?

[marf]

I didn't notice at first that he's spelling Wii with his fingers. Love the T-shirt as well.

[Alarm]

The latest Mario Galaxy vid, shows the robot boss scene, high quality clip!

Blimey, that looks ace!

[grue]

I didn't notice at first that he's spelling Wii with his fingers. Love the T-shirt as well.

Is that a Katamari made of all the controllers ever? Is that a little prince there?

[Sir_LANs-a-lot]

I didn't notice at first that he's spelling Wii with his fingers. Love the T-shirt as well.

Is that a Katamari made of all the controllers ever? Is that a little prince there?

Yes.

[grue]

Then? It Wiill??