Topic: Wiimote has limited angle of "pointer" detection?

[Ceric]

Look at what Nintendo said in there stats of the controller range.  We know that a Wavebird is actually goof for 60 feet.  Nintendo stated that the Full feature of the controller was good for about 5 meters.  That's about 16 feet.  Now RF is a radio frequency.  If you take a PS2 Pelican wireless controller and rest it on you leg suddenly your signal will go away.  Why?  RF has a hard time going through soft matter like flesh.

So since we know that the Full feature Wiimote has about a 4th of the range of a Wavebird (or half, I think Nintendo's Official Wavebird spec is 30 feet).  We can surmise that if Someone fat walks inbetween you and the Wii or if Someone wearing a dense sheet or you try to play from under a dense sheet that it will probably get muddled.  If I knew more about the Bluetooth Spec we could have more fun with stats.

Not to surprising.  Nintendo probably had to lower the power because of all the stuff they are putting in the controller.

[31 Flavas]

Quote
Originally posted by: vudu

Ian, I hate you so much.

Please, don't hate the village idiot... That's like picking on a cripple...

[31 Flavas]

Quote
Originally posted by: RickPowers
Actually, what I meant is that if the controller loses calibration, you simply hold it still, and it syncs up again.

Yet, Ian figures that takes 5 minutes or something...

Edit: It's really something to be defending Ian in post and bashing him in the next, but Ian's leaps of logic always amuse and amaze me.

 

[Sir_Stabbalot]

Quote
Originally posted by: 31 Flavas

Quote
Originally posted by: vudu

Ian, I hate you so much.

Please, don't hate the village idiot... That's like picking on a cripple...

Ian's not really a village idiot, he's way to smart. He's more of a village super-pessimist.

[UltimatePartyBear]

I finally paid attention to the few specs Nintendo mentions on the official Wii site:

Quote
Up to four Wii Remotes can be connected at once using wireless Bluetooth technology. The wireless signal can be detected within 10 meters of the console. Both the Wii Remote and Nunchuk controllers include a three-axis motion sensor. The Wii Remote also includes a speaker, rumble feature and expansion port, and can be used as a pointer within 5 meters of the screen.

If it worked the way I originally thought, there would only be one conceivable reason for the pointer ability to be range limited, which is that the triangulation of the controller's position would be less accurate the further away from the sensor bar it moved.  It seems unlikely that it would be imprecise enough for such a short range limit, however.  I see the writing on the wall now.  The good news is that only one of my fantasies is actually crushed by this revelation.  I had previously thought that if the Wii knew where the controller was in relation to the TV, then it also knew where it was in relation to the game world, meaning that you'd actually have to shoot around things.  I would have enjoyed that added level of immersion, but it's not a major loss.  It does mean that Sony's ripoff is closer to the real thing than I thought it was, though.

Eesh.  I'm trying not to let this shattering of my illusions dull my excitement.  Time to go read more Super Mario Galaxy impressions!

[zakkiel]

Quote
If it worked the way I originally thought, there would only be one conceivable reason for the pointer ability to be range limited, which is that the triangulation of the controller's position would be less accurate the further away from the sensor bar it moved.

Or, uh, the standard limit for a good signal from remote to receiver? This information actually means nothing at all.

Quote
I had previously thought that if the Wii knew where the controller was in relation to the TV, then it also knew where it was in relation to the game world, meaning that you'd actually have to shoot around things.

It does. Did you notice the ping-pong demonstration? With sufficiently precise accelerometers you need only the occasional input from some fixed reference (like a pair of IR sensors) to keep track of space. The fact that there are only two sensors - which doesn't allow any vertical location at all - suggests that the accelerometers are the more important feature.  

[Jensen]

Here is some info on the accelerometer used in both controllers.  They probably cost nintendo a little over a buck a piece.  It works in conjuction with the sensor bar, which I am assuming is IR.  The PS3 won't have the IR bar, of course.

Analog Devices Wii press release

ADXL330

Tilt functionality is just a detection of the acceleration of gravity.  To the contoller, gravity is sensed the same way as any other acceleration.  The IR gives a frame of reference.

Only being $1 to $2, I wouldn't be suprised to see the traditional controller have the 3 axis accelerometer.

[31 Flavas]

Quote
Originally posted by: PartyBear
I finally paid attention to the few specs Nintendo mentions on the official Wii site:

Quote
Up to four Wii Remotes can be connected at once using wireless Bluetooth technology. The wireless signal can be detected within 10 meters of the console. Both the Wii Remote and Nunchuk controllers include a three-axis motion sensor. The Wii Remote also includes a speaker, rumble feature and expansion port, and can be used as a pointer within 5 meters of the screen.

Bluetooth! Wai! I wonder if we can use it and the sensor bar with our PC computer's too then? Maybe even future PC games could use the Wii-mote. Oooooo.

[UltimatePartyBear]

Quote
Originally posted by: zakkiel
Or, uh, the standard limit for a good signal from remote to receiver? This information actually means nothing at all.

The range of the controller and the range at which it can be used as a pointer would be the same if that were the deciding factor.

Quote

Quote
I had previously thought that if the Wii knew where the controller was in relation to the TV, then it also knew where it was in relation to the game world, meaning that you'd actually have to shoot around things.

It does. Did you notice the ping-pong demonstration? With sufficiently precise accelerometers you need only the occasional input from some fixed reference (like a pair of IR sensors) to keep track of space.

That's not quite the same thing I was thinking of.  The paddle in the table tennis game has to start somewhere.  From that point, the accurate accelerometer is good enough to map real world motion to virtual world motion.  However, there's no way for that starting position to be related to any real world position.  If it were possible to know exactly where in the room the controller was, you could take it one more step.  Imagine a room with a column in it.  Now imagine that the player's viewpoint is directly behind the column, so that it's in the middle of the screen.  If the player is sitting directly in front of the TV, he can shoot straight ahead and hit the column.  If he reaches around to the right, he can shoot around the column, hitting the wall behind it even though he can't see it.  With the controller's absolute position known, the bullet's trajectory can be traced from the controller all the way into the virtual environment instead of from a virtual gun in that environment.  That can't be done on relative position alone.  It can be approximated using the pointing function combined with the motion sensing, but that's beyond the point I'm trying to make.  Knowing the absolute position of the controller could have allowed for some gameplay ideas that aren't feasible without it.

Quote
The fact that there are only two sensors - which doesn't allow any vertical location at all - suggests that the accelerometers are the more important feature.

I was operating on the assumption that it was the controller that had three transmitters in it.

[Shecky]

Look it can't be using IR... how the heck would Excite Truck work if it was?

[UltimatePartyBear]

The IR is for the pointing only.  There's still an accelerometer in the controller for sensing both movement and tilt.

[stevey]

Accelerometer only sense rapid change in speed, but the wiimote has gyro's that are the tilt / twist senser that can make up for the limited pointing detection and the wiimote not even finished yet, give nintendo time fix stuff before bitching.

[UltimatePartyBear]

Quote
Originally posted by: stevey
Accelerometer only sense rapid change in speed, but the wiimote has gyro's that are the tilt / twist senser that can make up for the limited pointing detection and the wiimote not even finished yet, give nintendo time fix stuff before bitching.

Here's the most relevant text from the page Jensen found (emphasis added):

Quote
About the ADXL330
The ADXL330 is a small, low profile, low power, 3-axis accelerometer with signal conditioned analog voltage outputs. With a ±2 g range, it is capable of measuring the static acceleration of gravity for tilt-sensing applications, as well as dynamic acceleration resulting from motion, shock, or vibration. Power consumption is extremely low, less than 200 microamps at 2.0 V (typical). Supply voltage range is 2.0 V to 3.6 V. The ADXL330 is compatible with applications requiring bandwidths ranging from 0.5Hz to 1.6 kHz. The ADXL330 is offered in a 4 mm × 4 mm × 1.45 mm, 16-lead, plastic lead frame chip scale package, with a shock rating of 10,000 g.

Judging by this, the accelerometer is way better than I expected.  There are no gyroscopes in the controller.  It doesn't need them.  Also, I'm not complaining.  I guess that may be assumed since that's been the norm around here lately, but I'm just talking about something I find interesting.  I hope that doesn't offend you too much.

[Jensen]

This is very simple guys.  There are 3 acceleration sensors in the controller.  The only feedback is 3 numbers: the amount of acceleration in each of the three directions.  

Holding the controller upside down will give the same 3 numbers as pushing the controller downwards at 2g.  But turning the controller upside down requires it to first be on its side, so you can infer that the controller was turned.

Without harsh lateral movement, Pitch and Roll detection is absolute, because it is sensing the constant force of gravity. Yaw detection is relative, so it can drift.  Lateral vs. Rotational motion is just infered.  You could use the remote as a pointer without the sensor bar, but the horizontal orientation would quickly drift.  The IR sensor bar gives absolute orientation along the horizontal/yaw axis.  The IR can also detect distance by comparing how far apart the sensors are.  

[Ceric]

Ok Jensen to much technical in one setting.  What's that expanded into laymen?  I think I know but its to late.  Maybe I'll read it again tomorrow.

[jasonditz]

Have they said yet where the sensor bar has to be positioned? Is it restricted to being below the TV like was shown in the E3 demos, or can it, for instance, rest on top as well?

[UltimatePartyBear]

Quote
Originally posted by: Jensen
This is very simple guys.  There are 3 acceleration sensors in the controller.  The only feedback is 3 numbers: the amount of acceleration in each of the three directions.

Your own link makes a good case for there only being one sensor in each controller.  One chip can detect acceleration on 3 axes, and it can detect the pull of gravity.  Color me impressed.

Quote
Holding the controller upside down will give the same 3 numbers as pushing the controller downwards at 2g.  But turning the controller upside down requires it to first be on its side, so you can infer that the controller was turned.

Not to mention that as soon as it came to rest, it would detect a sudden, constant 1g pull in the opposite direction that could be assumed to be gravity.  There's no way anyone could keep it moving in the same direction at 1g for long.

Quote
Without harsh lateral movement, Pitch and Roll detection is absolute, because it is sensing the constant force of gravity. Yaw detection is relative, so it can drift.  Lateral vs. Rotational motion is just infered.  You could use the remote as a pointer without the sensor bar, but the horizontal orientation would quickly drift.  The IR sensor bar gives absolute orientation along the horizontal/yaw axis.

That's an interesting point.  I'd wager the sensor is actually in the controller, and the bar has IR emitters on it.  It wouldn't be unreasonable to assume the bar was horizontal.  The sensor would see two bright spots in the IR spectrum, and the controller's orientation could be determined based on the line between them.  However, twist the controller 180 degrees and it will see the exact same thing.  I don't think that's what the IR is for, regardless.

Quote
The IR can also detect distance by comparing how far apart the sensors are.

True, as long as the sensors are in view.  Acceleration detection can make up for when they aren't, and it may explain the need to continuously recalibrate under some circumstances.

All of that still only gets the position of the controller within its own reference frame, though.  The virtual world is a different reference frame, and the only way to know the controller's absolute position in the virtual world's reference frame is to start with a known point in both.  That would be the TV screen.  The game knows which point in the virtual world is being used as a POV.  It could paint a dot on the screen and ask you to touch the controller to it to calibrate, and then the two reference frames would be in sync right up until the moment you exceeded 2 g with the controller, which is probably pretty easy to do based on the examples at the end of this article.

Let me reiterate that I am not whining.  The controller does not suck.  The controller is awesome.  I just want to understand it.

[Jensen]

Quote
Originally posted by: Ceric
Ok Jensen to much technical in one setting.  What's that expanded into laymen?  I think I know but its to late.  Maybe I'll read it again tomorrow.

It's like a bubble level, with one bubble thing along each axis.

I'm not the best at explaining technical things, so I probably understand it better than I am explaining it.

[Jensen]

Quote
Originally posted by: PartyBear

Quote
Originally posted by: Jensen
This is very simple guys.  There are 3 acceleration sensors in the controller.  The only feedback is 3 numbers: the amount of acceleration in each of the three directions.

Your own link makes a good case for there only being one sensor in each controller.  One chip can detect acceleration on 3 axes, and it can detect the pull of gravity.  Color me impressed.

I don't know how the sensor works,  but the output is a 3d vector (magnitude of force and direction)

Quote

Quote
Holding the controller upside down will give the same 3 numbers as pushing the controller downwards at 2g.  But turning the controller upside down requires it to first be on its side, so you can infer that the controller was turned.

Not to mention that as soon as it came to rest, it would detect a sudden, constant 1g pull in the opposite direction that could be assumed to be gravity.  There's no way anyone could keep it moving in the same direction at 1g for long.

My example was a bit extreme. Moving laterally up and to the left would make the same force as rolling to the right and then back to neutral. (Without the IR and IR sensor)

Quote
I'd wager the sensor is actually in the controller, and the bar has IR emitters on it. It wouldn't be unreasonable to assume the bar was horizontal.

Yeah, you're right, I was just used to it being called a sensor bar. And based on photos, it is horizontal.

Quote

Quote
The IR can also detect distance by comparing how far apart the sensors are.

True, as long as the sensors are in view.  Acceleration detection can make up for when they aren't, and it may explain the need to continuously recalibrate under some circumstances.

 The gyroscopic mice work because the only sense input while holding down a button, they are relative, like a mouse.  The Wiimote has to work differently.

Quote
Let me reiterate that I am not whining.  The controller does not suck.  The controller is awesome.  I just want to understand it.

The controller isn't magic :-) It can't always detect 6 discreet degrees of absolute control.

[UltimatePartyBear]

Quote
Originally posted by: JensenThe gyroscopic mice work because the only sense input while holding down a button, they are relative, like a mouse.  The Wiimote has to work differently.

Who said anything about mice?  Have you read the MP3 impressions?  There's mention of recalibration.

Quote
The controller isn't magic :-)

Aw, crap.  Why don't you just crush all my dreams while you're at it?

[Jensen]

Quote
Originally posted by: PartyBear

Quote
Originally posted by: JensenThe gyroscopic mice work because the only sense input while holding down a button, they are relative, like a mouse.  The Wiimote has to work differently.

Who said anything about mice?  Have you read the MP3 impressions?  There's mention of recalibration.

That was me thinking to myself... On why gyroscopic mice don't need IR or calibration.

I'd be more excited about MP3 with a traditional mouse and the nunchuk.  The edge sensing stuff makes it harder to watch videos.... it just doesn't look as slick.  PCs are still the best platform for FPS games.

[Athrun Zala]

Quote
Originally posted by: Ian Sane
"According to Jonny, calibration takes place automatically if you simply hold the controller still."

I'm sure we'll all enjoy holding still while an enemy is killing the sh!t out of us.  I think they need a method we're more in control of.

how about pausing the game?

[Jensen]

The nunchuck uses a different company's sensor, but the functionality is equivalent.
Link