• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

Kobun Heat's "Ask Me Stuff About The Revolution" Thread

Status
Not open for further replies.
Kobun Heat said:
We didn't have extensive hands-on time with the controller. It was about five minutes each with a variety of demos. Standing, mostly, although some of the guys sat down briefly.

I'd say it's about as immediately understandable as a computer mouse or the DS touch screen. As Iwata said in his keynote, everybody is at the same "start line" -- you shouldn't need prior gaming experience to figure out how to use the controller. It's intuitive -- point and shoot.


All of the points you made about the controller are exactly what I imagined and hoped they would be(don't understand why people are having a hard time grasping this. You were pretty thorough). Anyway, thanks for clarifying everything for us. I can't wait for this to launch. :)

Kobun, did anyone drop a hint as to when we will hear/see some software, tech specs, finalized name etc for the rev? Surely before E3 no?
 
Oh, it only works with Nintendo brand Revolution televisions which cost $599.
Well, I guess that would explain the "No-hdtv" policy. :/
 
Leatherface said:
Kobun, did anyone drop a hint as to when we will hear/see some software, tech specs, finalized name etc for the rev? Surely before E3 no?
Things may have been discussed. Do you think I'm going to say anything about them on GAF, though?
 
Is Iwata going to discuss more Revolution details at the Wi-Fi event in October, Kobun?
 
Kobun Heat said:
Things may have been discussed. Do you think I'm going to say anything about them on GAF, though?
So that's a yes then? From one to ten how would you rate rate the awsomeness of this information?
 
To clarify, kobun just confirmed what I've been trying to say all along. and no I didn't say it shoots lasers, but it works as if its a laser pointer i.e. direct contact with screen/imaginary line etc. Obviously its all done by sensors and nothing to actually do with the screen technically, and there are many other options, but I would hope this is the main method of interfacing.

Why some of you failed to understand this despite everything kobun said I don't know - thank you so much for pictorial evidence kobun, I was about to give up the ghost, even I was wondering if I was making sense or not.

Why did we care if it was direct aim vs controlling on-screen cues? The two have vastly different impacts on our gaming experience and have very different potentials - at least from a game design point of view. Firstly I'm of the belief that direct pointing = revolutionary whilst controlling an onscreen cursor is not, despite the new control. that would merely be a direct replacement for using the analogue stick or mouse to aim in say Goldeneye or Halflife and would be quite boring, if perhaps abit more accurate - no new doors are opened, no gameplay experiences are changed. The new method suddenly makes you feel like your hand is actually holding the gun as you stalk down those dark corridors in Metroid, and you aim using your own line of sight, not the onscreen gun or the cursor in the middle of the screen.

Milhail, surely you won't be tilting it upwards that much? I'm pretty sure most games will offer multiple control configurations anyway, especially FPS games.
 
maharg said:
I like how to you, a light gun -- technology that worked very well 20 years ago -- is revolutionary.

I'm sorry, what light gun. Where?

You ever played an FPS where you actually feel like you're holding the gun? Which is er...the whole frickin point of FPS.
 
From my understanding, you're not pointing at anything on screen with the remote anymore than you're literally pointing at icons when you move a mouse. You are manipulating a cursor using the remote. The cursor is what actually points at stuff on screen. The movement of your hand is translated in some immediately obvious way to the movement of the cursor (left moves cursor left etc.)

If you're using the remote like a light gun where you literally have to point at the screen I can see where the rev would need some way of knowing where the screen is. However, that's a different application altogether.
 
I'm sorry, but relying on "laserpointing" opperation and accuracy for movement would suck for most purposes. Plus its just not possible without acutually using the screen.
 
Last post on the topic...seriously guys, kobun just confirmed how it worked in the demos and theres 2 pages of people still getting it wrong or not believing him.

Gahiggidy said:
I'm sorry, but relying on "laserpointing" opperation and accuracy for movement would suck for most purposes. Plus its just not possible without acutually using the screen.

Its not going to be used for most games, and neither is the whole "control cursor like mouse" thing. These two functions will probably be competing over stuff like RTS, FPS, Nintendogs. A 10 secod calibration won't hurt anyone.

Have you actually considered carefully how a 3D mouse would work? You do realise you would have to hold your hand in the same position all the time? This is kind of against Nintendo's philosophy where movement is natural. If an enemy leaps out at me, I (or a casual gamer) might lean to the side and point at the enemy to kill it, as if I was trying to dodge it.

If I did this with a 3D mouse, what the hell is it going to think? It would go haywire. Plus any sudden or quick movements (i.e. instinctive) will result in me losing where the cursor is (simply cos my eyes didn't keep up) and having to regain my orientation by finding the cursor position visually in relation to the enemy before I can manipulate the control, this could get messy when an enemy is all over you.

With a pointer I could whizz about the screen like a light gun and always know where I am. I can rest it on my lap and if I get ambushed by 20 enemies I can stick my arm out straight and wave it around like a lunatic.

Think it through carefully, if it was like a mouse - you either cotrol the camera like a pc FPS (meaning you gotta keep your hand very still) or as in the metroid game you control the cursor, it would be like in Goldeneye when you hold down the R button and a cursor comes up on screen. For accuracy and sensitivity 3D mouse gets crushed imo.
 
Calibration would not work. As soon as you move your arm a couple feet to the side you'd have to re-calibrate all over again.
 
Gahiggidy said:
Calibration would not work. As soon as you move your arm a couple feet to the side you'd have to re-calibrate all over again.

The way I see it, if you tell the system or the sensors where the TV is, thats all it needs to know. It already knows which angle, which direction, which position in 3D space the pointer is, now all it needs to know is if the pointer is landing on the screen. I could be wrong on how it can work, but the 3D mouse sounds awful, why do you want it?

kobun has said how it works in detail, all the major websites have called it a laser pointer, Iwata called it a "DIRECT POINTING DEVICE" (come on now...) and the demos show clearly that people can point directly at the screen. Flashlight demo being the perfect example, people are off their chair, approaching the TV and pointing at it as if it was a real flashlight and they are looking into some window into another world.

Are you just pulling my leg?
 
maharg said:
I like how to you, a light gun -- technology that worked very well 20 years ago -- is revolutionary.
Light-gun technology wouldn't be applicable for games that need constant input.
If the game read the gun position at every frame the screen would flash like crazy!

Gahiggidy said:
Calibration would not work. As soon as you move your arm a couple feet to the side you'd have to re-calibrate all over again.
That would depend how the calibration worked. You could define the dimensions of the screen using the top-left and bottom-right corner. The position of each corner in 3D-space can be expressed as the interesection of two rays. To do this you would have to shoot at each corner from two different positions in your room (of course this method makes some assumptions about the pitch and roll of the screen).
 
Shao said:
Last post on the topic...seriously guys, kobun just confirmed how it worked in the demos and theres 2 pages of people still getting it wrong or not believing him.

No, Kobun made observations based on a limited amount of time using the device with no extraordinary effort to determine the mechanism (not that anyone should have expected him to). His observations fit many possible theories.

It's not him we don't believe, it's you. You're more sure of exactly how it works than he is, and he's the one who was there.

Also half the things you say make no sense.
 
How about a "3D Joystick"?


Here's how it could work: The sensors tell the remote where the plane of the TV is located. It does this by "looking" at the two sensors to register its relative location to each point (do we even know if they are powered in any way?) Much in the same way an analogue-stick has a spring to pull the stick back to center, the sensors act as an anchor in the 3D Joystick (Virtual Joystick a better decriptor?). the further you pull away from the center of the "sensor bar" the faster/further you move in the game. As you can see, it'd be very easy to program a "dead zone" with this method. This would be more "realistic" than a 3D mouse as you would be required to return the remote back to face the tv set more less perpendicular. A 3D mouse without a "dead-zone" could lead to frustration in a game environment. Its much easier to operate than a "laser gun" as you wouldn't require to use a "sight" to find the target (TRY playing DuckHunt while shooting from the HIP.) Not to mention it wouldn't matter much har far away you sit. Obviously, the close you are to a target, the eassier it is to hit targets with a gun (real or play).

Makes sense?
 
Gahiggidy said:
How about a "3D Joystick"?


Here's how it could work: The sensors tell the remote where the plane of the TV is located. It does this by "looking" at the two sensors to register its relative location to each point (do we even know if they are powered in any way?) Much in the same way an analogue-stick has a spring to pull the stick back to center, the sensors act as an anchor in the 3D Joystick (Virtual Joystick a better decriptor?). the further you pull away from the center of the "sensor bar" the faster/further you move in the game. As you can see, it'd be very easy to program a "dead zone" with this method. This would be more "realistic" than a 3D mouse as you would be required to return the remote back to face the tv set more less perpendicular. A 3D mouse without a "dead-zone" could lead to frustration in a game environment. Its much easier to operate than a "laser gun" as you wouldn't require to use a "sight" to find the target (TRY playing DuckHunt while shooting from the HIP.) Not to mention it wouldn't matter much har far away you sit. Obviously, the close you are to a target, the eassier it is to hit targets with a gun (real or play).

Makes sense?
No
 
Gahiggidy said:
Calibration would not work. As soon as you move your arm a couple feet to the side you'd have to re-calibrate all over again.

It would work. Once the system knows where the TV is, roughly, and how big it is, in relation to the sensors, it would allow for a pretty big range of movement. Though I'd guess this would work best if the sensors were as close to the TV as possible. If the system can't assume the sensors are right near the TV, theres the chance that it might not know if, when you are calibrating, you are point at the 4 corners of a small TV one foot in front of you or a huge TV 50 feet from you. The sensors should prevent this problem, though.
 
maharg said:
No, Kobun made observations based on a limited amount of time using the device with no extraordinary effort to determine the mechanism (not that anyone should have expected him to). His observations fit many possible theories.

It's not him we don't believe, it's you. You're more sure of exactly how it works than he is, and he's the one who was there.

Also half the things you say make no sense.

I didn't say I'm sure how "it" works, I'm quite positive I know how a laser-pointer style game would work though, as opposed to a 3D mouse. Frankly the two are very different and his observations simply do not amount to a 3D mouse at all. His observations couldn't be clearer, he even drew a diagram and you still do not agree - so how is it I'm the one you don't "believe"? I haven't said what it is at all.

i just say it would be better if it was a pointer and not a mouse and try to explain why. didnt expect the barrage of naysayers though.

From a game design point of view, from a gamer point of view, considering stuff like human nature, game design philosophy, technology etc, these were simply the opinions I formed. Sorry if I confused you kid. ;)
 
rawk said:
It would work. Once the system knows where the TV is, roughly, and how big it is, in relation to the sensors, it would allow for a pretty big range of movement. Though I'd guess this would work best if the sensors were as close to the TV as possible. If the system can't assume the sensors are right near the TV, theres the chance that it might not know if, when you are calibrating, you are point at the 4 corners of a small TV one foot in front of you or a huge TV 50 feet from you. The sensors should prevent this problem, though.
And how does the "calibration" method know if I'm actually pointing at the corner of the screen? What, am I gonna have to get on my knees and peer over the face of the remote to line it up just right? Calibration works for touch-screens because you are touching the screen. The remote doesn't even have a sight to line it up. Nevermind the computing power needed to keep track of EVERY movement of the controller the moment after you calibrate it.
 
Gahiggidy said:
How about a "3D Joystick"?


Here's how it could work: The sensors tell the remote where the plane of the TV is located. It does this by "looking" at the two sensors to register its relative location to each point (do we even know if they are powered in any way?) Much in the same way an analogue-stick has a spring to pull the stick back to center, the sensors act as an anchor in the 3D Joystick (Virtual Joystick a better decriptor?). the further you pull away from the center of the "sensor bar" the faster/further you move in the game. As you can see, it'd be very easy to program a "dead zone" with this method. This would be more "realistic" than a 3D mouse as you would be required to return the remote back to face the tv set more less perpendicular. A 3D mouse without a "dead-zone" could lead to frustration in a game environment. Its much easier to operate than a "laser gun" as you wouldn't require to use a "sight" to find the target (TRY playing DuckHunt while shooting from the HIP.) Not to mention it wouldn't matter much har far away you sit. Obviously, the close you are to a target, the eassier it is to hit targets with a gun (real or play).

Makes sense?

Why wouldn't you require a "sight"? I assume you mean on-screen cursor?

I would like to see the stick mimic a joystick for games like starfox, ace combat etc.. How they achieve this technically I have no idea. But those would benefit from having some kind of onscreen targeting system like most plane combat games do.
 
Gahiggidy said:
And how does the "calibration" method know if I'm actually pointing at the corner of the screen? What, am I gonna have to get on my knees and peer over the face of the remote to line it up just right? Calibration works for touch-screens because you are touching the screen. The remote doesn't even have a sight to line it up. Nevermind the computing power needed to keep track of EVERY movement of the controller the moment after you calibrate it.

Well, it's not like I have all the answers for you, but here's my take: if a developer makes a light gun-esque game that uses the revolution controller, it's gonna have to work within a margin of error taking into account that people won't be able to hit the very corner pixel with exact precision. It would probably have you hit X number of targets, cross check that info with what it knows about TV sizes, account for a little human error, and do with that info what it can. At this point the game has a basic 3-D view of the play area: the TV screen, the sensors, and the controller, all in a 3-D void. You wouldn't have to track every single movement -- just the angle and position at the time of the button press. Same way lightguns only need to check if you hit when you pull the trigger.

The games would have to be designed with all of this in mind, though. It wouldn't be pixel precise, by any means, but if they added a little aiming reticule thing to the screen, I think it would work just fine.
 
No, by sight I mean a little plastic fin on the end of the controller.

application_gunhand.gif


Like a said, try playing a lightgun game by shooting from the hip. Not very accurate.
 
Gahiggidy said:
And how does the "calibration" method know if I'm actually pointing at the corner of the screen? What, am I gonna have to get on my knees and peer over the face of the remote to line it up just right? Calibration works for touch-screens because you are touching the screen. The remote doesn't even have a sight to line it up. Nevermind the computing power needed to keep track of EVERY movement of the controller the moment after you calibrate it.

wow, gahiggidy, you've defeated the revolution controller
 
Replying to some stuff from the previous page, if the sensors track the controller in 3D space, and the orientation of the controller is tracked as well, then no calibration would ever be needed, simply because the Revolution would always be able to adjust given the controller's location in 3D space. You would still point at the left side of the screen if thats where you wanted to shoot, no matter where you were. At least for the Metroid demo.

Seeing as how the controller knows how far its tilted/twisted, a programmer could easily make an FPS that follows a more conventional FPS aiming style, where the view follows the cursor exactly. A better way to hold the controller in that instance would be more like an ice cream cone, to avoid undue strain on the wrist. Metroid Prime 3 may even have an option for that type of aiming system.

The only thing we need is a confirmation that it does indeed track the actual location of the controller. I skimmed the last page and a half, so it might already be there, I'll go look.

Thanks for all the information Kobun, sorry if I've been a pain in the ass at all. :)

EDIT: Haha, say Peer's confirmation on the last page. Well, that's good news. I assume he means full 3D (x, y, z) location, and not just distance.
 
quadriplegicjon said:
by using a cursor.
And what does that do? The cursor may be pointed at the corner of the screen but the remote could be pointed at the corner of the room, not the tv set.
 
Did Kobun mention any calibration that went on during the demos? He mentioned that journalists stood up, and then sat down at certain points as well - yet the controller still worked.
 
When I go to E3 I'm gonna bring a laser-pointer and tape it to the Rev. remote to prove you guys wrong. The cursor will not line up with the laser.
 
Gahiggidy said:
And how does the "calibration" method know if I'm actually pointing at the corner of the screen? What, am I gonna have to get on my knees and peer over the face of the remote to line it up just right? Calibration works for touch-screens because you are touching the screen. The remote doesn't even have a sight to line it up. Nevermind the computing power needed to keep track of EVERY movement of the controller the moment after you calibrate it.

Yeah this is a stumbling block, but if they've done it they've done it....

I've no idea how it would calibrate. It certainly wouldn't involve shoooting at the corners of the screen, maybe you can like stand up and press it against the 4 corners of the screen and press A and the system will remember the co-ordinates. Maybe calibration wont need the remote at all. Could still be wrong I guess, at which point I will cry cos the prospect really excites me.

by the way, some guy on digitaltv website says in the laser shooting demo, you dont need an on-screen cursor to shoot stuff cos its just like a light gun game. *shrugs* Thats two people who played it who give me the impression its a pointer, and they dont sound unsure of it.

One way it could work and mimic a light gun. The sensors detect your distance from the set, when starting the game up, you could be asked to point at a target on screen and press A. This could tell the system where the centre of your TV is in terms of 3D space. From then on, calculating the position in 3D space and the angles and distance of the remote etc, it could theoretically mimic a laser pointer.

gahiggidy - you dont need a "sight" if you have a cursor on screen showing where you are pointing, like some guns have laser-sights. Getting away from the gun comparison, a torch has no sight yet its extremely easy to aim. Shining a torch on your TV can give an insight into how a pointer game will work.
 
i'm nor sure i understand this concern over 3d mouse vs pointer here.

all that matters is that it worked really well... right?

i haven't seen it described as a laser pointer anywhere, but i have heard several sights use the term "3d mouse"...

nintendo themselves said there is no need for calibration. and nintendo said it uses blue tooth sensors set near the tv about a foot apart. That has nothing to do with lasers (unless i'm mistaken.... i'm not a techie)

if it works so smoothly that you can aim in an FPS at the speed of thought, does it matter if it's using 3d mouse tech to do that?

i have to admit i am confused by this argument..

edit/ but i am leaning towards the idea of 3d mouse, because the size of the tv should not effect how far you have to move to aim, and i'm betting it won't.
 
Gahiggidy said:
When I go to E3 I'm gonna bring a laser-pointer and tape it to the Rev. remote to prove you guys wrong. The cursor will not line up with the laser.

lol I look forward to it. I'd like the whole thing cleared up.
 
It would work. Once the system knows where the TV is, roughly, and how big it is, in relation to the sensors, it would allow for a pretty big range of movement. Though I'd guess this would work best if the sensors were as close to the TV as possible. If the system can't assume the sensors are right near the TV, theres the chance that it might not know if, when you are calibrating, you are point at the 4 corners of a small TV one foot in front of you or a huge TV 50 feet from you. The sensors should prevent this problem, though.

I think the sensors at the center top and bottom of the TV will eliminate the need for calibration. So that anywhere you go, it is still using the Sensors as a reference point, therefore the Revolution will know where the controller is at all times. The only thing that puzzles me is the width of the TV. How does it know how big or wide the TV is?
 
Krowley said:
i'm nor sure i understand this concern over 3d mouse vs pointer here.

all that matters is that it worked really well... right?

i haven't seen it described as a laser pointer anywhere, but i have heard several sights use the term "3d mouse"...

nintendo themselves said there is no need for calibration. and nintendo said it uses blue tooth sensors set near the tv about a foot apart. That has nothing to do with lasers (unless i'm mistaken.... i'm not a techie)

if it works so smoothly that you can aim in an FPS at the speed of thought, does it matter if it's using 3d mouse tech to do that?

i have to admit i am confused by this argument..

IGN, gamespot, 1up, all describe it as laser pointer. it doesnt use lasers naturally, but it you imagine the red dot from a laser pointer is where the cursor will be. kobun and mark macdonald, editor of egm (guy on digitaltv) describe it as a laser pointer.

Do I really have to keep this up? No...I'm off to bed.
 
Any word on whether the sensors are powered in anyway? Are they perhaps just points for reference and not actually siginalling to either the console or remote?
 
Frankly, the difference is virtually moot. So long as the game provides appropriate feedback to your motion, your brain will be tricked into believing whatever it needs to to play. However, the central issue here is the statement by Nintendo themselves that calibration is not needed, which conflicts with the idea of the cursor knowing exactly where on the screen you are pointing (as opposed to merely making it look like it does through visual cues). There is quite simply no magic bullet that can make it not require calibration for that with everything we have been told by Jim Merrick.

Which is not to say that some games might not actually require and use calibration to achieve that effect. In this way, the so-called 'laser pointer' effect is really a subset of the '3d mouse' functionality. This is imo the most realistic view of things.
 
maharg said:
Frankly, the difference is virtually moot. So long as the game provides appropriate feedback to your motion, your brain will be tricked into believing whatever it needs to to play. However, the central issue here is the statement by Nintendo themselves that calibration is not needed, which conflicts with the idea of the cursor knowing exactly where on the screen you are pointing (as opposed to merely making it look like it does through visual cues). There is quite simply no magic bullet that can make it not require calibration for that with everything we have been told by Jim Merrick.

Which is not to say that some games might not actually require and use calibration to achieve that effect. In this way, the so-called 'laser pointer' effect is really a subset of the '3d mouse' functionality. This is imo the most realistic view of things.
why.

do you need to calibrate a mouse ?

you move your mouse, your mouse knows which way you are going, cursor goes that direction.

same theory here with Revolution's controller

you move your arm, the revolution controller know which axis you are going (X Y Z), your item/character on screen moves that direction.
 
Status
Not open for further replies.
Top Bottom