Support NeoGAF

[buckfutter]

Uh, didn't Harrison lift the controller directly up to lift the object during the tech demo at the E3 conference? And haven't we heard about the nunchuck attachment being shunted forward without tilting to open doors in Red Steel? How are they doing that without measuring acceleration through space?

 

[Chichikov]

-when Jedimike says acceleration, he means Translation
-when Jedimike says tilting, he means Rotation

he's just using the wrong terminology to describe his words,

Far be it from me to guess what people mean, but you’re not necessarily right.
Acceleration != translation.
and if the ds3 use acceleratometers, his distinction is actually important.

still: Trans X, Trans Y, Trasn Z, Rot X, Rot Y, Rot Z = 6 axis

It’s still false.

 

[Kangu]

It's not six axis, it's six degrees of motion. We've been over this a thousand times. Look at the wikipedia link posted, those are the types of motion the DS3 can detect. It has the exact same motion sensing capabilities as the wiimote, but it lacks any kind of tracking or pointing functionality.

So project hammer= no problem. Duck Hunt= no way.
Excite Trucks= no problem. Zelda= No way.
Sonic Firesomethin= no problem. Elebits= No way.

 

[Pope Benedict XVI]

Ignorance is bliss... Acceleration is a measure of how fast the speed of something is changing. Accelerometers measure acceleration. In the case of the PS3 controller, the accelerometers measure the acceleration of the gyros.

When you move a controller without tilting it... nothing changes. What is an accelerometer supposed to measure? Gravity is the same. Everything is constant... the only thing you did was move it's relational placement, which requires external sensors to measure.

I'm no scientician, but if I closed my eyes and a machine shook me up/down and side-to-side, I would definitely be able to tell i was moving, and in which direction. I'd know because the liquids and puddings inside me would be jiggling about. They could have some kind of jiggle technology that sort of lags behind the movement of the controller shell, if you know what i mean

 

[gutter_trash]

if everyone is using the game Warhawk alone as example, that is the wrong way to go about it because as FLYING game it will most likely not use all translation Axis just as Excite Truck and Sonic Wii doesn't use all Axis (as a game)

*Sonic Wii only uses 2 Axis of rotation and 0 Translation (i'm not knocking this game, it's on my to buy list)

 

[Kangu]

Ignorance is bliss... Acceleration is a measure of how fast the speed of something is changing. Accelerometers measure acceleration. In the case of the PS3 controller, the accelerometers measure the acceleration of the gyros.

When you move a controller without tilting it... nothing changes. What is an accelerometer supposed to measure? Gravity is the same. Everything is constant... the only thing you did was move it's relational placement, which requires external sensors to measure.

Goddamn this post is so ignorant you should edit it out and replace it with a picture of pokemans cat.

 

[Version 3.0]

Are gofreak and jedimike just talking about the semantics of the thing? I thought I understood this sixaxis stuff till they confused me just now!

edit: the controller surely CAN detect up/down movement. What's his name, the english guy, moved it up and down at the e3 conference

I don't think so. He said "I pick it up, and it flies!", and the Warhawk took off. But that was mapped to a button at E3. I played Warhawk a couple of times, and it had tilt functionality only.

I'm not saying the controller can't detect movement, maybe it can. But Warhawk at E3 could not. I think that bit during the press conference was smoke and mirrors (ie. bullshit).

 

[gutter_trash]

degrees ONLY apply to rotation MAN! not translation (wich is units)

 

[mckmas8808]

For the record, during the 1up podcast, they distinctly said the sixaxis doesn't do directional stuff, a la pulling the controller towards you.

So do you consider up and down directional?

 

[Chichikov]

Ignorance is bliss... Acceleration is a measure of how fast the speed of something is changing. Accelerometers measure acceleration. In the case of the PS3 controller, the accelerometers measure the acceleration of the gyros.

When you move a controller without tilting it... nothing changes. What is an accelerometer supposed to measure? Gravity is the same. Everything is constant... the only thing you did was move it's relational placement, which requires external sensors to measure.

I don’t have any knowledge as to how the ds3 works.
But you can use accelerometers to measures 3d movement.

 

[SnakeXs]

So do you consider up and down directional?

I never said what I consider to be what. I'm not gonna get into a semantics/PR battle over words. When we get details about EXACTLY what the DShake can do, that's what I'll believe. As long as it's vague PR bullshit and technobabble I'll gladly take a back seat.

 

[gutter_trash]

So do you consider up and down directional?

Yes as the Translation Y axis = up/down
Left and Right would be the Translation X axis

about the hot topic debated Towards and Away that is the one everone wants to know: that's Translation Z axis (depth)

 

[Chichikov]

degrees ONLY apply to rotation MAN! not translation (wich is units)

Seriously man, I assume English is not your native language (it's cool, it's not mine either). But please, look up degrees of freedom.

 

[mckmas8808]

I never said what I consider to be what. I'm not gonna get into a semantics/PR battle over words. When we get details about EXACTLY what the DShake can do, that's what I'll believe. As long as it's vague PR bullshit and technobabble I'll gladly take a back seat.

No it does matter because jedimike and the 1upyours crew is saying that you can't move the controller up and down, left and right, or back and forth for movement.

I'm saying you can. So what do you say?

 

[Pope Benedict XVI]

I don't think so. He said "I pick it up, and it flies!", and the Warhawk took off. But that was mapped to a button at E3. I played Warhawk a couple of times, and it had tilt functionality only.

I'm not saying the controller can't detect movement, maybe it can. But Warhawk at E3 could not. I think that bit during the press conference was smoke and mirrors (ie. bullshit).

But what about the rubber duck demo? He made the duck go up and down several times as i remember it.

Maybe there are only certain conditions where you can do a vertical take-off in Warhawk?

 

[sugarhigh4242]

So project hammer= no problem. Duck Hunt= no way.
Excite Trucks= no problem. Zelda= No way.
Sonic Firesomethin= no problem. Elebits= No way.

I think something like Duck Hunt would be possible if you calibrate the controller to a single reference point before you start.

Also, the Eyetoy would be potentially used for that kind of motion-tracking.

 

[Zaptruder]

I never said what I consider to be what. I'm not gonna get into a semantics/PR battle over words. When we get details about EXACTLY what the DShake can do, that's what I'll believe. As long as it's vague PR bullshit and technobabble I'll gladly take a back seat.

From where you can comfortably snipe with FUD.

 

[SnakeXs]

No it does matter because jedimike and the 1upyours crew is saying that you can't move the controller up and down, left and right, or back and forth for movement.

I'm saying you can. So what do you say?

And I'm saying I'll reserve any judgement until it's not based off of vague PR shit and there's actual facts to go bye. What the **** don't you get about that, and why the **** are you pushing so hard? Jesus.

From where you can comfortably snipe with FUD.

Because I have a record of "sniping" against Sony/the Dual Shake? :lol

 

[Version 3.0]

soo... if the wii remote costs $40 and doesnt have analog buttons or as many buttons or a rechargeable battery built in, how much will the ps3 controller cost?

No, you're measuring it against the wrong device. Instead of wondering vs. the Wiimote's price (which has the sensor/pointer function), you should measure against the nunchuck. The PS3 controller, apparently, is a nunchuck.

So start at $20 and add the extra stick, buttons, wireless, battery, and Sony logo to the price.

Spoiler

$74.99

 

[ant1532]

It's not six axis, it's six degrees of motion. We've been over this a thousand times. Look at the wikipedia link posted, those are the types of motion the DS3 can detect. It has the exact same motion sensing capabilities as the wiimote, but it lacks any kind of tracking or pointing functionality.

So project hammer= no problem. Duck Hunt= no way.
Excite Trucks= no problem. Zelda= No way.
Sonic Firesomethin= no problem. Elebits= No way.

I don't understand, are you saying that it can't so those games since the Wii Mote uses a sensor bar and pointing features? Or is the sensor bar used for more then just pointing and aiming?

 

[gofreak]

When Harrison picked the controller up and the controller moved up, in the first demo, I think that was simply a precanned animation. When he put the controller down, the controller did not go down.

The duck bobbing up and down and jumping out of the water, I think was mapped to the controller flicking up and down. Like I said before, you can map things to suggest translation in game, but I don't know if that's coming out of the controller's own translation (I don't think so).

It's possible the controller can detect that, the warhawk diagram suggests it, but in the E3 presentation demos I don't think that was there. I think they would have gone out of their way to show it off more explicitly if it was (e.g. actually move the controller around in the first demo..but it was just in the same position all the time, rotating around a single point etc.).

 

[Chichikov]

I think something like Duck Hunt would be possible if you calibrate the controller to a single reference point before you start.

Also, the Eyetoy would be potentially used for that kind of motion-tracking.

It is theoretically possible (again, assuming the ds3 can detect such movement), but inertia based systems tend to be inaccurate at low speeds and when stationary, I’ll be surprised if they can make it work.

 

[mckmas8808]

And I'm saying I'll reserve any judgement until it's not based off of vague PR shit and there's actual facts to go bye. What the **** don't you get about that, and why the **** are you pushing so hard? Jesus.



Because I have a record of "sniping" against Sony/the Dual Shake? :lol

The fact is WE ALREADY HAVE THE FACTS!!

BAM! This is the first fact.

And the rubber ducky at E3 2006 was the second fact laid out in this thread. We are just trying to educate you here. Where are you getting your info from 1UpYours?

 

[Kangu]

I don't understand, are you saying that it can't so those games since the Wii Mote uses a sensor bar and pointing features? Or is the sensor bar used for more then just pointing and aiming?

Because of the sensor bar and pointing features. Granted you could calibrate the DS3 with some kind of on screen pointer and use it as a 'gyro mouse'. That would probably suck balls though.

The wiimote is also supposedly able to do realtime 3d tracking, which the DS3 could also immitate (albeit very poorly), but I haven't seen it featured in any announced games.

 

[Zaptruder]

And I'm saying I'll reserve any judgement until it's not based off of vague PR shit and there's actual facts to go bye. What the **** don't you get about that, and why the **** are you pushing so hard? Jesus.



Because I have a record of "sniping" against Sony/the Dual Shake? :lol

Record or not, you're doing pretty well here, intentional or not!

 

[mckmas8808]

When Harrison picked the controller up and the controller moved up, in the first demo, I think that was simply a precanned animation. When he put the controller down, the controller did not go down.

The duck bobbing up and down and jumping out of the water, I think was mapped to the controller flicking up and down. Like I said before, you can map things to suggest translation in game, but I don't know if that's coming out of the controller's own translation (I don't think so).

It's possible the controller can detect that, the warhawk diagram suggests it, but in the E3 presentation demos I don't think that was there. I think they would have gone out of their way to show it off more explicitly if it was (e.g. actually move the controller around in the first demo..but it was just in the same position all the time, rotating around a single point etc.).

So my question for you would be why release this picture?

Why even so millions of people somthing about your controller that it can't do at all? Why?

 

[Pope Benedict XVI]

When Harrison picked the controller up and the controller moved up, in the first demo, I think that was simply a precanned animation. When he put the controller down, the controller did not go down.

The duck bobbing up and down and jumping out of the water, I think was mapped to the controller flicking up and down. Like I said before, you can map things to suggest translation in game, but I don't know if that's coming out of the controller's own translation (I don't think so).

It's possible the controller can detect that, the warhawk diagram suggests it, but in the E3 presentation demos I don't think that was there. I think they would have gone out of their way to show it off more explicitly if it was (e.g. actually move the controller around in the first demo..but it was just in the same position all the time, rotating around a single point etc.).

Wasn't the motion control kind of shoe-horned into warhawk at the last minute? We know that very few developers, even major ones like konami, didn't even know about the motion sensing. The warhawk guys may not have had time to implement transitional movement as well as tiliting. The use of tilting is far more obvious in a flight-based game.

The diagrams and sony's rhetoric make it very clear, i think, that the controller can detect movement across space.

 

[Kangu]

When Harrison picked the controller up and the controller moved up, in the first demo, I think that was simply a precanned animation. When he put the controller down, the controller did not go down.

The duck bobbing up and down and jumping out of the water, I think was mapped to the controller flicking up and down. Like I said before, you can map things to suggest translation in game, but I don't know if that's coming out of the controller's own translation (I don't think so).

It's possible the controller can detect that, the warhawk diagram suggests it, but in the E3 presentation demos I don't think that was there. I think they would have gone out of their way to show it off more explicitly if it was (e.g. actually move the controller around in the first demo..but it was just in the same position all the time, rotating around a single point etc.).

Dind ding ding ding. WINNAR. The canned animation was triggered by the accelerometers detecting an upwards motion of the controller. Same with the duck jumping. The ony way the controller could track it's position in space, as has been suggested, is by constantly sampling the length and duration of acceleration to determine some kind of rough positioning in space. Of course moving the controller at a constant velocity would result in it having no ****ing idea where it was. Don't expect to see this in any PS3 game.

 

[Kangu]

So my question for you would be why release this picture?

Why even so millions of people somthing about your controller that it can't do at all? Why?

It can detect motion in all of those directions. It can't track its position in 3d space in an independent manner though. The picture is accurate as far as motion detection goes.

 

[usea]

If you can't understand 6 axes from that controller image, words won't help you.

 

[SnakeXs]

Record or not, you're doing pretty well here, intentional or not!

I hate throwing around labels, but you must REALLY be skewed to see a guy who posts 1 statement from a podcast, NOT giving an opinion about it (saying OMG U GUYS PWNED NO MOVEMENT), stating CLEARLY that he'll await solid info until judging the thing, and seeing anything but that.

And, mckmas8808, stop it. Just as you so valienty believe you have THE FACTS, others believe that THEY have THE FACTS that disagree with you. Where the truth lies I don't know, but the point is there's a grey area, like it or not, and until it's filled in I WON'T GUESS WHAT I THINK THE THING CAN OR CANNOT DO. K? K.

 

[Pope Benedict XVI]

Dind ding ding ding. WINNAR. The canned animation was triggered by the accelerometers detecting an upwards motion of the controller. Same with the duck jumping. The ony way the controller could track it's position in space, as has been suggested, is by constantly sampling the length and duration of acceleration to determine some kind of rough positioning in space. Of course moving the controller at a constant velocity would result in it having no ****ing idea where it was. Don't expect to see this in any PS3 game.

oh shit, ok, NOW i get it!

 

[gutter_trash]

Seriously man, I assume English is not your native language (it's cool, it's not mine either). But please, look up degrees of freedom.

degrees as measurement of circular lines like when you performed at school with a compas, that's measurement for ROTATIONS

maybee your reference of degrees differs from mine, but describing degrees as measurement is for measuring rotations

45degrees, 90 degree, 180 degrees, 360 degree (full circle, can't do that with human wrist)

perhaps I delve too much in 3D space, 3D softwar for decribing a rotation but it's all based from math, so there is no fuzzy lingo here

 

[mrklaw]

Because of the sensor bar and pointing features. Granted you could calibrate the DS3 with some kind of on screen pointer and use it as a 'gyro mouse'. That would probably suck balls though.

The wiimote is also supposedly able to do realtime 3d tracking, which the DS3 could also immitate (albeit very poorly), but I haven't seen it featured in any announced games.

the sensor bar just gives you a good, constant 'center' from which to measure. Its like its constantly calibrating itself.

If you've played wario ware: twisted, you'd know why this is important. That could detect rotational movement, but often it'd drift and think 'up' was when you had the GBA twisted slightly. without the sensor bar the Wii would be the same.

So, in a similar way to the GBA wario ware, the PS3 could use the controller as a mouse/pointing device. But it'd likely lose calibration quickly. eg 'point at the middle of the screen and press X'. Then any relative movements of the controller are used to adjust your 'aim'.



As for whether it detects translational movements... well I assumed from the E3 stuff it did. But maybe not. But Jedimike - you can easily detect acceleration in a plane with an accelerometer. And I thought Phil Harrison touched on that directly? saying as well as tilt it can sense movement in 3D space, but only while its moving (suggesting an intertia/accelerometer based system)

 

[Pope Benedict XVI]

Dind ding ding ding. WINNAR. The canned animation was triggered by the accelerometers detecting an upwards motion of the controller. Same with the duck jumping. The ony way the controller could track it's position in space, as has been suggested, is by constantly sampling the length and duration of acceleration to determine some kind of rough positioning in space. Of course moving the controller at a constant velocity would result in it having no ****ing idea where it was. Don't expect to see this in any PS3 game.

If it was moving at a constant velocity, wouldn't it know where it was based on the last period of acceleration and the lack of subsequent accleration or deceleration?

 

[Kangu]

If it was moving at a constant velocity, wouldn't it know where it was based on the last period of acceleration and the lack of subsequent accleration?

How would it distingush between moving at constant velocity and not moving at all? It couldn't.

 

[gofreak]

Dind ding ding ding. WINNAR. The canned animation was triggered by the accelerometers detecting an upwards motion of the controller. Same with the duck jumping. The ony way the controller could track it's position in space, as has been suggested, is by constantly sampling the length and duration of acceleration to determine some kind of rough positioning in space. Of course moving the controller at a constant velocity would result in it having no ****ing idea where it was. Don't expect to see this in any PS3 game.

Haha, I'm really confused now.

I was making the case for it not being able to detect motion along the axes, but rotation and rotational speed. I don't think anyone was ever making a case for it tracking its position in 3D space (though if it could detect relative motion, it could maybe figure that out by sampling the speed at a certain interval and using distance = time x speed..at least doing that you could measure position relative to an origin).

I think for the duck bobbing up and down it was just us rotation and rotational speed around the x axis. If the speed exceed a certain threshold while rotating 'up' (e.g. a flick upwards, basically), it'd make the duck jump.

 

[Pope Benedict XVI]

How would it distingush between moving at constant velocity and not moving at all? It couldn't.

it would remember that there was some acceleration, and be aware that there had been no subsequent deceleration

 

[mckmas8808]

It can detect motion in all of those directions. It can't track its position in 3d space in an independent manner though. The picture is accurate as far as motion detection goes.

That's exactly what I thought. But jedimike and the 1Upyours crew is saying it CAN'T do motion on the left to right, up and down, and back and forth motion.

 

[gutter_trash]

if it detects movement in space, that's translation.
doesn't matter if it's via sensor bar or recognising accelerated up/down, left/right, away/towards movement from on point to another, it's still movement in space.
though not as acurate as Wii, but it's still there

 

[Kangu]

Haha, I'm really confused now.

I was making the case for it not being able to detect motion along the axes, but rotation and rotational speed. I don't think anyone was ever making a case for it tracking its position in 3D space (though if it could detect relative motion, it could maybe figure that out by sampling the speed at a certain interval and using distance = time x speed..at least doing that you could measure position relative to an origin).

I think for the duck bobbing up and down it was just us rotation and rotational speed around the x axis. If the speed exceed a certain threshold while rotating 'up' (e.g. a flick upwards, basically), it'd make the duck jump.

AFAIK the controler has accelerometers and gyrometers. Gyrometers measure gyration, I assume, using some kind of pendulum system which either has a base calibration of where it is 'centered' or calibrates it on the spot when you turn on the console. Aceleromters detect aceleration, and I assume this could work with pendulums or small weight as well. It's probably a lot more sophisticated than that though.

 

[Kangu]

it would remember that there was some acceleration, and be aware that there had been no subsequent deceleration

Well that's a good point. I guess it could be handled like that. As has been said though, it would probably require constant polling and even then wouldn't be very accurate.

That's exactly what I thought. But jedimike and the 1Upyours crew is saying it CAN'T do motion on the left to right, up and down, and back and forth motion.

Well I haven't heard the podcast, so I don't know what context that was said in. Maybe you could enlighten me?

 

[gutter_trash]

AFAIK the controler has accelerometers and gyrometers. Gyrometers measure gyration, I assume, using some kind of pendulum system which either has a base calibration of where it is 'centered' or calibrates it on the spot when you turn on the console. Aceleromters detect aceleration, and I assume this could work with pendulums or small weight as well. It's probably a lot more sophisticated than that though.

so that settles it, gyrometers to detected the tilting (rotations x 3) and the accelerometers to detect the movement/acceleration (translation x 3)

 

[Pope Benedict XVI]

Well Sony sure did a great job explaining their new controller :lol

 

[Chichikov]

degrees as measurement of circular lines like when you performed at school with a compas, that's measurement for ROTATIONS

maybee your reference of degrees differs from mine, but describing degrees as measurement is for measuring rotations

45degrees, 90 degree, 180 degrees, 360 degree (full circle, can't do that with human wrist)

perhaps I delve too much in 3D space, 3D softwar for decribing a rotation but it's all based from math, so there is no fuzzy lingo here

Trust me, I know what arc degree means.
That word can mean quite a few things.
Press that link and you understand what I meant by degrees of freedom.
And that 6 axes is not based on any math I know of.

 

[mckmas8808]

Well I haven't heard the podcast, so I don't know what context that was said in. Maybe you could enlighten me?

Okay. Well they first laugh at the name "Sixaxis", which I laughed at too because it think it sucks. But then they start to question if the controller can even do all 6. They say something about maybe at TGS Sony might surprise us with some "extra" funtionality like being able to move the controller backwards and forwards for motion in PS3 games.

 

[Razoric]

Well Sony sure did a great job explaining their new controller :lol

it doesnt matter if they did or not... there is so much FUD around PS3 that it's not even funny. this is just another drop in the bucket.

 

[Chichikov]

so that settles it, gyrometers to detected the tilting (rotations x 3) and the accelerometers to detect the movement/acceleration (translation x 3)

Again, movement != acceleration.
Those sensors detect force, there is no force applied to them at constant speed movement.
You can extrapolate 3d position based on known starting position and acceleration (like in inertia based navigation systems) but those tend to be expensive and inaccurate at slow speeds.

 

[Zaptruder]

I hate throwing around labels, but you must REALLY be skewed to see a guy who posts 1 statement from a podcast, NOT giving an opinion about it (saying OMG U GUYS PWNED NO MOVEMENT), stating CLEARLY that he'll await solid info until judging the thing, and seeing anything but that.

Actually, IIRC the 1upyours guy were making comments about the sixaxis; why sixaxis? There's only 5 DOF, no backwards and forwards.

They acknowledge that there is translational movement in that comment, but they've simply screwed up their facts - there IS forward/backward movement detection, just as there is for the other 5 axis.

Maybe it was a seperate comment from the one you heard; in your case, I'd suggest that the 1up guys simply haven't had any hands on experience other than with warhawk... which had the motion sensing put in on short notice... and indeed, would have little need for translational movement (likely they didn't get the chance to land the plane either, which would've meant up/down movement as demonstrated in the infamous Sony conference demo).

Given all those facts, you still insist on peddling your 'no opinion' line... it seems to me that you just want to be uncertain about it.

 

[davepoobond]

there's no such thing as "sixaxis" that's why they MADE UP THE NAME to name the particular aspect of the controller this since you'll never hear anyone else in the goddamn world referring to anything as a "sixaxis"