Support NeoGAF

[Jokeropia]

How does it do that anyway? Lumens?

Here's the relevant text I believe:

[0149]Next, by using the width w calculated at step 102 and a viewing angle .theta. of the image pickup element 743, the CPU 30 calculates a current distance realDL (see FIG. 22) which is a current distance between the marker 8L and the image pickup element 743 (controller 7) (step 103). The current distance realDL is obtained from the following equation.

realDL=(w/2)/{tan(.theta./2)}

Since the viewing angle .theta. is a fixed angle, the viewing angle .theta. is prestored in the storage means (not shown) within the game apparatus 3.

[0150]Next, based on the second size data (diameter diamR), the CPU 30 calculates another width w which indicates, with respect to a setting position of the marker 8R, a width for which the image pickup element 743 is able to take an image (step 104). Said another width w is obtained from the following equation.

w=wi.times.diamM/diamR

[0151]Then, by using said another width w calculated at step 104 and the viewing angle .theta. of the image pickup element 743, the CPU 30 calculates a current distance realDR between the marker 8R and the image pickup element 743 (controller 7) (step 105). Here, the current distance realDR is obtained from the following equation.

realDR=(w/2)/{tan(.theta./2)}

[0152]Thereafter, the CPU 30 calculates an angle .delta.1 by using the current distance realDL, current distance realDR and setting distance m between the markers 8L and 8R (step 106). As shown in FIG. 23, the angle .delta.1 is between a line connecting the marker 8L and controller 7 and a line connecting the markers 8L and 8R. The angle .delta.1 is obtained from the following equation based on the cosine theorem.

cos .delta.1=(realDL.sup.2.times.m.sup.2-realDR.sup.2)/(2.times.realDL.tim- es.m)

[0153]Subsequently, the CPU 30 calculates the current distance realD (see FIG. 23) between the controller 7 and the middle point between the markers 8L and 8R (step 107), and then the distance calculation process in the subroutine ends. Here, the current distance realD is obtained from the following equation based on the cosine theorem.

realD= {square root over (realDL.sup.2+(m/2).sup.2-2*realDL*(m/2)*cos .delta.1)}{square root over (realDL.sup.2+(m/2).sup.2-2*realDL*(m/2)*cos .delta.1)} [equation 2]

[0154]It is also possible to obtain an angle .delta.2 between a line connecting the controller 7 and said middle point and the line connecting the markers 8L and 8R. The angle .delta.2 is obtained from the following equation based on the cosine theorem by using the current distance realD, current distance realDL and setting distance m.

cos .delta.2={realD.sup.2.times.(m/2).sup.2-realDL.sup.2}/{2.times.realD.t- imes.(m/2)}

Obtaining the angle .delta.2 allows the CPU 30 to calculate an angular position of the controller 7 with respect to the markers 8L and 8R, and various processes can be performed in accordance with the angular position of the controller 7.

That means that in this scenario the X positioning will have a margin of error more than 3x greater than that of the distance measurement. So we would need to know the finest the Wiimote can gauge the IR light distance down to, so we can therefore roughly figure out X position accuracy.

Isn't that what poppabk calculated earlier?

But anything which is moving will create aliasing which allows its position to be calculated more accurately based on interpolation. Sure there is a distance where you will see nothing but a single pixel point of light and you are no longer able to interpolate anything, but you are talking beyond living room distances for this.

The actual resolution of the wiimote camera is 128x96 and it has a field of view of approximately 30 degrees by 20 degrees. For a normal viewing distance (I used the THX recommended viewing distance for a 42" TV) this results in the sensor bar being in view over an x distance of about 75cm (which is the maximum you can move laterally before you lose sight of the sensor bar). So a pixel equates to approximately 6mm. Each of the two LED light patterns on the sensor bar is probably 3cm wide, therefore without interpolation you can still resolve individual points, allowing for sub-pixel interpolation to occur. Using the 1024 sub pixel accuracy - each interpolated pixel will equal 7/10ths of a mm.

 

[Shaheed79]

It's kind of funny how I never realized just how small the latency in the wiimote and m+ really are since I never had anything similar enough to compare it too. Now with the Move out on the market you really do get a feel for how impressive the response time of the wiimote pointer and m+ really are when used and calibrated correctly.

I think that now with Move and soon to be Kinect on the market the issue of "How much latency is acceptable latency in a motion based game?" will become an important and common topic among gamers and developers.

In our traditional controller games we have sort of a zero tolerance policy on latency between input and action. I see that becoming a similar cry from gamers once motion games become more developed, involved and competitive.

Online support of 1:1 motion games will become the biggest obstacle for developers who continue to evolve these kinds of games.

 

[distrbnce]

It's kind of funny how I never realized just how small the latency in the wiimote and m+ really are since I never had anything similar enough to compare it too. Now with the Move out on the market you really do get a feel for how impressive the response time of the wiimote pointer and m+ really are when used and calibrated correctly.

I think that now with Move and soon to be Kinect on the market the issue of "How much latency is acceptable latency in a motion based game?" will become an important and common topic among gamers and developers.

In our traditional controller games we have sort of a zero tolerance policy on latency between input and action. I see that becoming a similar cry from gamers once motion games become more developed, involved and competitive.

Online support of 1:1 motion games will become the biggest obstacle for developers who continue to evolve these kinds of games.

Wow, sounds pretty troublesome and dire.

Glad I'm having fun and haven't noticed any of that lag, or heard it mentioned by anyone I let try it out. They all think it's fucking magic. Hopefully the devs will get even better from this point.

 

[redbarchetta]

And Sixaxis doesn't have a gyroscope. It's just accelerometers like the bare Wii Remote.

I'm a bit late to this, but it's been well documented that the SixAxis does contain a single-axis (IIRC) gyroscope, used to detect yaw.

Edit: Well, perhaps not "well document" so much as "documented" :lol

 

[cakefoo]

Here's the relevant text I believe:

Where are the images? It's going to take me a while to understand that without pictures. And can't you just explain it, since you seem to be already well-versed.

Isn't that what poppabk calculated earlier?

No that's not. That's about XY camera resolution to determine pointing accuracy. I'm talking about determining lateral XY movement of the Wii remote itself in space and how dependent it is on knowing the distance from individual IR lights, so I need to know what's the minimum distance on the Z-axis you can move the pointer and have it detect said change in Z-axis position based not on the traditional 2-LED distance measurement, but on a single LED.

 

[MYE]

Sounds flattering too, read mine! Perfect opportunity for an example missed though.

I did...

:lol

Jesus man


And flyinpiranha, you missunderstood me. I'm not the one trying to prove anything or pull teh ownage here. If you're having that much fun, good for you.

 

[mrklaw]

how does the wii detect that you are moving a cursor on screen upwards? I'd assume a combination of gyro/accelerometer inputs, as well as the camera in the wiimote seeing the sensor bar move down in its view.

I'd assume the move basically does the same thing, only the camera is on the TV, so it'd note the ball moving slightly upwards.

I can imagine though that the wii has an advantage here, as a slight movement in the wiimote (eg moving a pointer around) relative to the TV screen, would register relatively large movements from the sensor bar, as the camera is kind of on the pivot point. Whereas with the move, the PSeye would only see very small movements of the ball for the same cursor movement?

 

[cakefoo]

how does the wii detect that you are moving a cursor on screen upwards? I'd assume a combination of gyro/accelerometer inputs, as well as the camera in the wiimote seeing the sensor bar move down in its view.

I'd assume the move basically does the same thing, only the camera is on the TV, so it'd note the ball moving slightly upwards

Move has a gyro that can detect over 2400 or 2500 degrees of rotation per second, and an earth magnetic field sensor for maintaining orientational calibration. Wii is purely visual, based on where in the camera's fov the 2 IR lights are.

In terms of how they feel in the end, my Wii pointer feels sub-SD, whereas Move's is very fine and feels like it could pinpoint any pixel on an HDTV. Conversely, it seems that some games don't yet know how to keep Move's orientation calibrated. I find this hard to understand why it affects the pointer but not the orientation of your sword or table tennis paddle in Sports Champions or Start the Party's augmented reality objects, but to me it seems like something that can be corrected in future games/patches.

 

[poppabk]

The camera allows the Move position to be tracked in XY within a millimeter, and 1-2 centimeters in Z. Remember that the Wiimote camera is a much lower resolution than the PS Eye, and that the Eye performs subpixel tracking to a degree of about 10:1. For orientation, it's reliant on the internals. Don't know if you knew that or not. The camera is just there for the optional task of providing the positional data needed to project a virtual beam from the remote to the tv so that it's consistent even if you change seats/stances.

The practical resolution is very similar. The eye has a max resolution of 640x480 at 120hz and 320x240 at 240hz with a viewing angle of 56 - 75 degrees. The wiimote has a resolution of 128x96 and a viewing angle of 30 degrees. What this amounts to, through basic trigonometry is that at the same distance the number of pixels per meter of movement is approximately the same as both see a similar improvement in resolution through sub-pixel calculations. The difference is all in where the camera is located. Once you take into account rotation, the wiiremote detection of translational movement is greatly reduced, because a slight rotation produces a large shift in where the remote is looking at, whereas a large translational movement still produces a small change. Great for pointing, lousy for determining position. The move camera sees no translational movement when the remote is rotated round the center of the ball, that is all left to the gyros (and possibly some occlusion and eccentricity interpolation).

ie if you could keep the wiiremote perfectly horizontal it could track position as well as the eye - but this is physically impossible and would be useless for a game anyway. The move set-up is the way forward for motion controls, as it allows positional tracking even when you are swinging like crazy. The wiimote can track position through the camera only under idealized settings which are uselsss for 99% of any game application.

 

[redbarchetta]

In terms of how they feel in the end, my Wii pointer feels sub-SD, whereas Move's is very fine and feels like it could pinpoint any pixel on an HDTV.

That's not a fault of the Wii Remote's pointers as it is the hardware's output resolution itself though.

 

[Maxrunner]

So the wiimote lags less?i assume that the next step for nintendo is to put a camera integrated in the sensor bar like the one in the controller itself, and perhaps put some leds in the specific positions so that theres always a way of the controler to know where it is.....is this correct? Also is the infrared a better solution for the type of camera being used?

 

[onipex]

Not really a discussion I was in, I was just commenting on the failed support of M+

30 games in a year is failed support? Its not like it can be patched into games like Move. I will be surpised if Move would reach the penatration rate of motion + even with more games supporting it.


No one has mentioned that with motion + the Wii can point without looking at the sensor bar. Maybe it doesn't matter in the discussion but I think it is interesting.

Edit: actually I do see that it was discussed.

 

[Lonely1]

But that's a game that has you pointing at the screen at all times anyway, so it doesn't have to risk losing itself from drift like if the accelerometers were made to have to fend for themselves in table tennis where your paddle isn't pointing at the sensor bar as often.

Hey, I'm not disputing Move clear superiority in Absolute position tracking. .

And many people are reporting the pointer to be as good or a bit worse than the Wii's.

 

[defferoo]

Oh I see, so maintaining positional accuracy down to the millimeter isn't enough for true 1:1. Semantics much?

If someone could detect sensor drift occurring within a 60th of a second, then I congratulate them.

That's just ridiculous how you're trying to blur the gulf of difference between how much drift is accumulated between recalibrations in the Wiimote compared to the Move by classifying them both as "pseudo 1:1" with no distinction. Going a 60th of a second between calibrations is nothing. Going 15-30 seconds on the Wiimote with several large, high-speed swings contained within, that position is going to float into the stratosphere, rendering its positional data completely useless for gauging where on XYZ you are, and has to be relegated to a more general kind of 3D gesture recognition scheme-- or they can go the WSR route and use orientation-based trickery to make it look like it's following your 3D movements when in reality it's just following the angle of the remote.

Honestly, my experience with WiiMotion+ has led me to believe it's not as drift prone as you think it is. Maybe you can't play games with the same amount of accuracy as you can on Move, but you can still play games with a reasonable amount of pseudo 1:1 tracking. Enough to make you feel like you're affecting the outcome. As long as you aren't swinging your arm around like a madman. And yes, while I agree I'm generalizing here, I don't think the difference is large enough for me to say that the Wii Remote w/ motion+ doesn't do 1:1 while the Move does.

As for the WSR implementation of tracking your movements, I was disappointed that it wouldn't use the accelerometer at all and incorporate linear motion, but that doesn't mean it couldn't be done. It's obviously possible as seen in the LiveMove2 demo.

If the Move doesn't use the compass to calibrate the gyro, then I don't really see how it's doing it. The ball is only one point, which is not enough to determine the orientation of the controller. If that's the case, it may just be that the gyro sensor in the Move is less prone to drift than the WiiMotion+ sensor...

 

[cakefoo]

Honestly, my experience with WiiMotion+ has led me to believe it's not as drift prone as you think it is.

But I never said WM+ was very drift-prone. I said the accelerometers were. You need to differentiate the two.

And yes, while I agree I'm generalizing here, I don't think the difference is large enough for me to say that the Wii Remote w/ motion+ doesn't do 1:1 while the Move does.

Play Sports Champions. 3D position tracking means that if you want to block right, you bring your hand to your right side and orient the sword appropriately. You know how WSR does blocking? You have to hold a button, and the pivot point snaps up to head level and your orientational movements suddenly have a blocking behavior. Play Sports Champions table tennis. If your opponent is prone to smashes, you literally have to take a step back in your room. If they hit it to the left section of your side of the table and you want to do a forehand, assuming you're right-handed you have to literally take a step left so your forehand swing is happening on the left side of the table. If they lob the ball to your side, you have to swing high to reach the ball. These are all things you can't do on the Wii.

As for the WSR implementation of tracking your movements, I was disappointed that it wouldn't use the accelerometer at all and incorporate linear motion, but that doesn't mean it couldn't be done. It's obviously possible as seen in the LiveMove2 demo.

By AILive's own admission, that accuracy is only good enough for the duration of one motion. You can't keep doing motion after motion uninterrupted. The accelerometer drifts too much, its job is not to track position, but inertia and 3D gestures, and, at best, they've used it for aiming in archery, but it only works there because the movements are calm and the duration of the aiming mode is very brief.

If the Move doesn't use the compass to calibrate the gyro, then I don't really see how it's doing it. The ball is only one point, which is not enough to determine the orientation of the controller. If that's the case, it may just be that the gyro sensor in the Move is less prone to drift than the WiiMotion+ sensor..

Compass = magnetometer, and it definitely is being used to keep the orientation calibrated, working in concert with the gyroscope. The gyro is better than the one in Motion+, by the way.

 

[cakefoo]

So the wiimote lags less?

The Shoot can run with the big dogs:

Latency is 2, sometimes even just 1 frame. Lower than that and you're entering deep into the realm of diminishing returns.

Bonus gif

And that was only like my third playthrough ever, and first time playing it since Friday morning. The pointer feels incredible.

It's true though that some Move launch games don't have as responsive a cursor- this can be due to a number of things, like the intentional design to smooth input, but it can also come down to how well optimized the game is for Move controls. It's certainly not a limitation of the hardware itself.

 

[poppabk]

The gyro is better than the one in Motion+, by the way.

I wasn't aware that we knew what the gyro is yet. I think it probably is at the very least equivalent, but I thought the teardowns haven't identified the chip yet.
PS nice demonstration of the pointer, does indeed seem to work well.

 

[spwolf]

even reading about motion gaming makes me hungry

 

[Redbeard]

If Killzone 3 uses The Shoot's pointer it would give an incredible advantage to move users

 

[DangerousDave]

Bonus gif

And that was only like my third playthrough ever, and first time playing it since Friday morning. The pointer feels incredible.

I hope that I won't find you ever in Killzone 3 online. At least in opposite teams.

 

[Xellos]

That's nonsense - I played RE5 by resting both hands on the couch/lap, doing basically all aiming with my wrist, and I never needed a recalibration.

I think this may be in reference to a post I made in the general Move thread. Maybe recalibration isn't the right word but RE5 does constantly recenter the reticle. The game puts the reticle in the center of the screen when you pull the T trigger (the "draw gun" button) regardless of where on the screen you are actually trying to point. You always have to drag the reticle from the middle of the screen to the target, and if you weren't aiming at the middle of the screen to start with it can feel somewhat awkward. I wish the setup was more like The Shoot where the reticle is on screen at all times (and can be re-centered if via a button press if needed).

 

[AndyMoogle]

Hey, I'm not disputing Move clear superiority in Absolute position tracking. .

And many people are reporting the pointer to be as good or a bit worse than the Wii's.

I certainly feel that the pointer is overall worse with Move. It's got a sluggish feel to it that I don't get at all with a Wiimote. For most games you have to aim the whole controller directly at the screen rather than just do minor movements with your wrist. But that's just developers implementing it all wrong. Hopefully it will get better as they get more experience with Move.

Move works fine, but it's definitely not the massive improvement over the Wiimote w/ M+ that some gaffers like to think.

 

[Fafalada]

Latency is 2, sometimes even just 1 frame. Lower than that and you're entering deep into the realm of diminishing returns.

All 3D games are (at least)double-buffered - making minimum input latency One-Frame, regardless if your input device is a controller or telekinesis. So there's no real way to go lower then that

And most games running on multiple cores (since PS2 era) will add 1-2 extra frames of latency, making 1-frame latency a very rare thing even with traditional controllers.

I wish the setup was more like The Shoot where the reticle is on screen at all times (and can be re-centered if via a button press if needed)

Yea, but that would require a departure from original broken RE5 controls, and after 5milions or whatever sold Capcom probably thinks those controls were actually good...

 

[cakefoo]

For most games you have to aim the whole controller directly at the screen rather than just do minor movements with your wrist

I don't know what you're referring to? The Shoot, Sports Champions, Time Crisis, Flight Control HD, Start the Party, Tumble, MAG- they all have the traditional pointer behavior you speak highly of... the only game fitting your description is Beat Sketchers.

Move works fine, but it's definitely not the massive improvement over the Wiimote w/ M+ that some gaffers like to think.

What Move games have you played?

 

[KZObsessed]

I think the pointer controls can be excellent with Move. I had some huge calibration bugs initially with The Shoot, in that calibration just wouldn't work. Quitting and restarting fixed it and I was very impressed. It was really accurate, it felt like i could aim at every pixel and there was hardly any latency at all. Upon loading the demo again I got a bug where the cross hair wouldnt appear after calibration, I tried it and I managed to get through the entire first section of the level without any cross hair, was really fun that way

Just because you find some pointer implementation isn't all there on certain launch titles which aren't even shooters doesn't mean we should just blame the tech. MAG apparently controls excellently with Move and I'm sure SOCOM and KILLZONE 3 will as well. Hopefully the upcoming Beta will allow us to test it out.

The most impressive aspect of Move tho is obviously the positional tracking, just playing Table Tennis in Sports Champions shows such a big improvement.

Bonus gif

And that was only like my third playthrough ever, and first time playing it since Friday morning. The pointer feels incredible.

Great work with the gifs cakefoo.

 

[poppabk]

cakefoo can you do a few little experiments.
In shoot what happens if you rotate the controller while keeping the ball in the exact(ish) same position, ie rotating around the axis of the ball? What happens if you move the controller while keeping it perfectly level? I am guessing that they are using both rotation data and the ball position to determine pointing so both should move the pointer independently as well.

What happens in table tennis if you step forward while keeping the remote stationary and what happens if you move just the remote forward very slowly?

 

[KoruptData]

cakefoo can you do a few little experiments.
In shoot what happens if you rotate the controller while keeping the ball in the exact(ish) same position, ie rotating around the axis of the ball? What happens if you move the controller while keeping it perfectly level? I am guessing that they are using both rotation data and the ball position to determine pointing so both should move the pointer independently as well.

What happens in table tennis if you step forward while keeping the remote stationary and what happens if you move just the remote forward very slowly?

I can anser the Table Tennis question. If you step foward while keeping the Move stationary, it is still tracked and represented on screen. Testing it out one time, I wanted to see if I could volley by keeping the thing still and just moving my body around and it worked perfectly.

 

[Jokeropia]

Where are the images?

Here's the entire patent application. There's a link to the images at the top of the page. I've noticed they can be hard to view in some browsers, so I re-hosted a few I think are interesting here:

Fig 21.

Fig 22.

Fig 23.

 

[malingenie]

Move works fine, but it's definitely not the massive improvement over the Wiimote w/ M+ that some gaffers like to think.

Speaking form a hardware-ignorant stance, my predilection for Move over WMP probably stems from three reasons:

-HD Vs. SD
-Big TV for PS3 (52") and Small TV for my Wii (32")
-Surround sound (PS3) vs. crappy tv sound Wii

These factors may be tricking me into the following statement but:

Playing Sports Champions has been the biggest jaw dropping experience of both solutions. The first time I saw my paddle getting closer and farther away from the net, or picking up a low shot to the left, I converted. Wii Motion Plus (WSR) had a total of 4 hours playtime, Sport Champions already tripled that and I got my package friday.

 

[marc^o^]

It is a fact wii Motion controller remains the most versatile controller ever released.

Still the Move trounces it as far as Motion is concerned. No competition, no doubt to have. It just works in full 1:1 3D glory.

 

[distrbnce]

Playing Sports Champions has been the biggest jaw dropping experience of both solutions. The first time I saw my paddle getting closer and farther away from the net, or picking up a low shot to the left, I converted. Wii Motion Plus (WSR) had a total of 4 hours playtime, Sport Champions already tripled that and I got my package friday.

I think what's really hooking us is the direct, obvious translation of your real skill into the game. How you can feel and know you're getting better at it. That's what leaves me thinking about SC all day when I should be working. I just want to go home and unlock.

Playing gladiator duel with a friend is unbeatable... until an even better version of this idea comes.

 

[LM4sure]

Seems that it is disputed which is better: the wiimote or move. Interesting. Either way, better accuracy isn't going to sell the wii or the move. they both seem to work well enough.

since I already have a Wii, there is no reason for me to get the Move. The game lineup is lacking and I can't see there being a killer app that only supports Move.

 

[Vinci]

Seems that it is disputed which is better: the wiimote or move. Interesting. Either way, better accuracy isn't going to sell the wii or the move. they both seem to work well enough.

since I already have a Wii, there is no reason for me to get the Move. The game lineup is lacking and I can't see there being a killer app that only supports Move.

I'll get Move when something really awesome comes out for it. Till then, yeah, I'm in wait mode.

 

[alterno69]

I love love love Table Tennis in SC, the precision is awesome, playing the silver level tournament and wow with some assits off it is even more impressive, i have to keep an eye on my grip to give my balls the intended effect. It is quite impressive, i have never played on Wii but Move is really really accurate.

 

[Dead Man Typing]

I think this may be in reference to a post I made in the general Move thread. Maybe recalibration isn't the right word but RE5 does constantly recenter the reticle. The game puts the reticle in the center of the screen when you pull the T trigger (the "draw gun" button) regardless of where on the screen you are actually trying to point. You always have to drag the reticle from the middle of the screen to the target, and if you weren't aiming at the middle of the screen to start with it can feel somewhat awkward. I wish the setup was more like The Shoot where the reticle is on screen at all times (and can be re-centered if via a button press if needed).

The cursor is always on screen when playing RE4 on Wii. I'm surprised to hear it isn't the same with Move.

 

[RaySpencer]

I find it weird that The Shoot doesn't actually track the ball.

 

[donny2112]

I can't see there being a killer app that only supports Move.

That is one thing that Nintendo really went whole hog on. Making the sequel to the most influential game of the generation exclusive and taking a huge IP and making the next game in the series exclusive. That's some big supporting points on Motion+ from Nintendo.

 

[marc^o^]

I'll get Move when something really awesome comes out for it. Till then, yeah, I'm in wait mode.

You'd be a fool not to believe SC is really awesome. If you bought a wii for titles such as wii sports resort, well, there's no excuse. Same difference as in wiimote > motion+ transition. Pure joy. The future in your hands. No hyperbole, I'm known as a wii fan on this forum!

 

[malingenie]

I think what's really hooking us is the direct, obvious translation of your real skill into the game. How you can feel and know you're getting better at it. That's what leaves me thinking about SC all day when I should be working. I just want to go home and unlock.

Playing gladiator duel with a friend is unbeatable... until an even better version of this idea comes.

I want to be rich and play PvP with two controllers each!

 

[cakefoo]

cakefoo can you do a few little experiments.
In shoot what happens if you rotate the controller while keeping the ball in the exact(ish) same position, ie rotating around the axis of the ball?

I wedged the ball onto a tripod and it works as a pointer should, tracking orientation. The old "ball being used to maintain orientational calibration" theory was old speculation based on us not knowing there was a 3-axis magnetometer present.

What happens if you move the controller while keeping it perfectly level? I am guessing that they are using both rotation data and the ball position to determine pointing so both should move the pointer independently as well.

You're correct. Also, when you cover up the ball and do position movements, the pointer moves too. And it can't be the gyro misinterpreting lateral movements as orientational, because a) it's a really good gyro and b) the 3-axis magnetometer is a rock.

What happens in table tennis if you step forward while keeping the remote stationary

In TT it tracks the remote alone. That's all that's onscreen. However, moving your body is absolutely crucial to getting the paddle to go where you want relative to the table's position. For instance, if you're a righty, you can't do a forehand on a ball coming to the left half of your table, unless you step to the left a couple feet so that your physical swing is occurring on the left side of the physical play area, and mirrored in game in the same virtual position of the table 1:1.

what happens if you move just the remote forward very slowly?

The paddle moves forward very slowly.

 

[distrbnce]

Seems that it is disputed which is better: the wiimote or move. Interesting. Either way, better accuracy isn't going to sell the wii or the move. they both seem to work well enough.

since I already have a Wii, there is no reason for me to get the Move. The game lineup is lacking and I can't see there being a killer app that only supports Move.

It has a killer app... I just think people are hesitant to accept this if they've only played Wii Sports. Sports Champions sells it to anyone I've let play it. Even my most jaded anti-waggle friend posted on facebook the next day "my arms are dying from that game... i want it"

 

[balladofwindfishes]

"selling it to anyone" is often not enough to be a "killer app"

I can tell just from the buzz of move that it's not even close to being a "killer app" and certainly nowhere near any of the Wii's many true killer apps.

 

[distrbnce]

"selling it to anyone" is often not enough to be a "killer app"

I can tell just from the buzz of move that it's not even close to being a "killer app" and certainly nowhere near any of the Wii's many true killer apps.

True, but the dramatic change in perception I've seen people have after using it is something that can't be ignored.

Has anyone shared SC with someone that HASN'T been impressed? It does what you expect and you can instantly feel that connection. That has never been available before.

 

[marc^o^]

"selling it to anyone" is often not enough to be a "killer app"

I can tell just from the buzz of move that it's not even close to being a "killer app" and certainly nowhere near any of the Wii's many true killer apps.

I thought a killer app was a reason enough to buy a console? By this definition, SC is a killer app, and then some if you are into table tennis.

 

[cakefoo]

Here's the entire patent application. There's a link to the images at the top of the page. I've noticed they can be hard to view in some browsers, so I re-hosted a few I think are interesting here:

Fig 21.

Fig 22.

Fig 23.

Ah, I see. It's measuring the diameter of the IR light. So it's going to be limited by the effective XY resolution of the camera and the diameter of the IR light.

I did a test on the Wii sensor bar calibration screen that shows IR lights as white circles. First of all, the higher the sensitivity setting, the more accurate it gets, but the more prone to IR confusion from foreign sources. That said, the highest sensitivity (5) was only able to detect a change in IR diameter after about 18 inches of movement on the Z-axis. At sensitivity setting 1 it took about 30 inches to trigger the diameter change. I was standing about 4 or 5 feet away in both cases.

For the horizontal position test, I was able to move about 10-12 feet left to right, before it detected any change in diameter between the closer and further IR light. I was standing about 7-8 feet away.

That is not good at all. Move can detect 1-2cm of depth and 1mm on XY.

I did that test because it was easier than figuring it on with math. But since we can't know for sure if that was using the Wiimote to its full potential, let's do it mathematically, using Move's dimensions and accuracy ratings as a guide.

Wii camera effective resolution: 1024x768
Move camera effective resolution: 6400x4800 (link - look for "tenth of a pixel")

Move's advantage: 6.25:1

One Wii IR light array diameter: 2cm (estimate)
Move sphere diameter: 4.5cm (estimate)

Move's advantage: 2.25:1

Total advantage (multiplied together) 14.1:1

Since the Wii method of determining horizontal position relies heaviest on depth, we'll take the 1.5cm accuracy rating for Move's Z-axis and multiply it by 14.1 to get 22.2cm, or 8.5 inches, for my guess for the Wiimote's accuracy in terms of single orb depth tracking. The hands-on test with the calibration screen indicated only 18 inches of sensitivity, so we will compare that to the calculated accuracy ratio of 8.5 inches and come away with the conclusion for the Z-axis sensitivity that it is 2.11x better than what the sensor bar calibration test indicated.

So let's take that same magnitude of 2.11x and apply it to the real world horizontal position accuracy results of 10-12ft.

Drumroll....

That's still only 4.7-5.7ft of X positional accuracy. Plus we need to remember that at this ideal sensitivity, the Wiimote may be confusing other light sources as IR light, so the sensitivity would have to be reduced, which on the sensor bar screen is indicated by a shrinkage in the diameter of each IR lightsource.

edited for the last sentence

 

[poppabk]

Ah, I see. It's measuring the diameter of the IR light. So it's going to be limited by the effective XY resolution of the camera and the diameter of the IR light.

I did a test on the Wii sensor bar calibration screen that shows IR lights as white circles. First of all, the higher the sensitivity setting, the more accurate it gets, but the more prone to IR confusion from foreign sources. That said, the highest sensitivity (5) was only able to detect a change in IR diameter after about 18 inches of movement on the Z-axis. At sensitivity setting 1 it took about 30 inches to trigger the diameter change. I was standing about 4 or 5 feet away in both cases.

For the horizontal position test, I was able to move about 10-12 feet left to right, before it detected any change in diameter between the closer and further IR light. I was standing about 7-8 feet away.

That is not good at all. Move can detect 1-2cm of depth and 1mm on XY.

I did that test because it was easier than figuring it on with math. But since we can't know for sure if that was using the Wiimote to its full potential, let's do it mathematically, using Move's dimensions and accuracy ratings as a guide.

Wii camera effective resolution: 1024x768
Move camera effective resolution: 6400x4800 (link - look for "tenth of a pixel")

Move's advantage: 6.25:1

One Wii IR light array diameter: 2cm (estimate)
Move sphere diameter: 4.5cm (estimate)

Move's advantage: 2.25:1

Total advantage (multiplied together) 14.1:1

Since the Wii method of determining horizontal position relies heaviest on depth, we'll take the 1.5cm accuracy rating for Move's Z-axis and multiply it by 14.1 to get 22.2cm, or 8.5 inches, for my guess for the Wiimote's accuracy in terms of single orb depth tracking. The hands-on test with the calibration screen indicated only 18 inches of sensitivity, so we will compare that to the calculated accuracy ratio of 8.5 inches and come away with the conclusion for the Z-axis sensitivity that it is 2.11x better than what the sensor bar calibration test indicated.

So let's take that same magnitude of 2.11x and apply it to the real world horizontal position accuracy results of 10-12ft.

Drumroll....

That's still only 4.7-5.7ft of X positional accuracy.

You are doing it wrong. You aren't measuring the single IR light you are measuring the IR light pattern. You don't have to detect a change in size of the light, you need to detect a change in the distance between the light patterns, both the individual sets of 3 lights and the group of two lights.
Plus your resolution calculations are off, because the eye has a much larger field of view. How much of that resolution do you actually use? I would estimate a third.

 

[DeadGzuz]

since I already have a Wii, there is no reason for me to get the Move. The game lineup is lacking and I can't see there being a killer app that only supports Move.

The killer app is SC. My few days with SC wowed me and entertained me (and my wife and kids) more than anything in the last 10 months of owning the Wii. I've already got permission by everyone in the family to sell the Wii.

Now before you lable this an anti-Wii troll, we do not play any of the traditional franchises from Nin. No Mario, Metroid, Zelda, etc. So just owning it for Sports/Resort is now silly (probably always was). I can take that $250+ I make on the sale and put it towards games we will play.

 

[poppabk]

I would lay money on the wiimote being able to judge z distance within a few cm's, its x,y is definitely on the few mm range, you can see that in any pointer game. But not to sound like a broken record - its irrelevant, because you can only judge that distance under ideal circumstances, circumstances which are 99% useless for games. Sony didn't go all out tech wise on the Move, its a slightly better wiimote with a ball on top, but they did use their experience with camera tech to make the hardware and software work in a way that makes the experience a lot more accurate that the wiimote/M+.

 

[cakefoo]

You are doing it wrong. You aren't measuring the single IR light you are measuring the IR light pattern. You don't have to detect a change in size of the light, you need to detect a change in the distance between the light patterns, both the individual sets of 3 lights and the group of two lights.

Not if the purpose is to determine horizontal position I don't think-- distance between the two lights can only be used for Z-tracking. The distance from the Wiimote to each of the 2 individual IR arrays on the sensor bar has to be used to gauge horizontal position (see Jokeropia's posts).

Plus your resolution calculations are off, because the eye has a much larger field of view. How much of that resolution do you actually use? I would estimate a third.

How much of the screen area is used for Move is not important, as far as I'm concerned. But yes, field of view should help the Wii's cause by a factor of 2.5 (30 degrees vs 75 for PS Eye).

So modify the horizontal position rating from within 4.7-5.7ft of accuracy down to a still unimpressive 1.89-2.3ft.

That's not even good enough to know if the remote is on your left or right side. Plus you have to be pointing at the sensor bar, which isn't practical in a motion game. Nobody plays ping pong pointing at the screen after every swing, nobody wants to deal with that.

 

[cakefoo]

I would lay money on the wiimote being able to judge z distance within a few cm's, its x,y is definitely on the few mm range, you can see that in any pointer game. But not to sound like a broken record - its irrelevant, because you can only judge that distance under ideal circumstances, circumstances which are 99% useless for games. Sony didn't go all out tech wise on the Move, its a slightly better wiimote with a ball on top, but they did use their experience with camera tech to make the hardware and software work in a way that makes the experience a lot more accurate that the wiimote/M+.

What are you even basing that on?

sorry, DP~!