Old Discussion Thread for all 3 motion controllers

Status
Not open for further replies.
Foot tracking is real important also. Think beyond fps games, etc, and instead picture stuff that the average casual person would like. For example, a product that teaches yoga or tai chi, and uses Natal to watch your limbs to critique/comment on your form. Or maybe a Karate School product that has a virtual teacher that comments on your moves, again based on limb tracking. Maybe a product that teaches juggling, and it watches how you do it and makes suggestions to improve it. Or maybe a personal trainer program that can watch how you use free weights or doing common exercises, and comments if your form is correct. Who knows what people will think of or what is possible.

Yoga is a lot more subtle than something like Natal could hope to critique, in my experience. I don't think that even the full motion tracking rigs with the little balls velcro-ed to the skin suits would do an adequate job of it.
 
Actually your brain is not easily trick or confused. Your brain does just fine manipulating virtual objects with a few buttons and a couple of analog sticks. Adding in motion detection doesn't suddenly force a more stringent set of requirements when controls become more natural and mimicks the action of the user. Response time is more important than precision mapping.

Isn't precise mapping a function of response time (and other parameters) ? If the timing is off, the manipulate would be off too (Hence, the confusion in the brain: "Why the $^#&$# did it fall again ? I thought I placed it correctly ?").

liolio said:
For MS solution we know really few, in fact we've just learnt today that it's was based on 3DV tech. Clearly Ms have been working a lot on the software layer but how Natal accelerate this software is in the dark. It could be some cheap CPU from the embedded space + a custom dsp + some RAM/ROM but it doesn't say much.
On the PS Eye side of thing educated member here hinted that the possibilities of the ps eye are mostly unexplored which makes me think that MS has significant advance on the software side of thing (hardware side outside of the cameras is not that relevant as cell as processing power in spare where as xenon could not afford too the extra work).

I don't think you can generalize that way. Sony is very advanced in the use of PS Eye software. MS should be more advanced in Natal-like (IR 3D imaging) platform/SDK.
 
As archie4oz pointed out, they were prepared to demo but the confirmation came late. 3DV camera was (close to) going to market, so they should be very ready. That's why I was surpised to hear a 2010 release date, instead of a 2009 one. My sense is MS will push to release it in 2009 with partial functionality.
Natal is not based on 3DV tech, anyway it looks like they have been working on the problem for a while both hardware and software side.
The Sony demoes are actually pretty refined given that they work very well. It's the packaging that s*cks.
I agree the tech is of unmatched precision and works wich lead me to think that the tech has been working on for a while whereas some demo looks like they have been done over night.
That's strange imho and not very sony like, was the project on the verge to be set-aside?
Who knows the important point now is that the tech will reach constumers somewhere in 2010. I can't wait for the next big shows this year and hear more on both sony and MS "motion thingy".
 
Yes, you should definitely watch the full demo. Here's some of it:

http://www.youtube.com/watch?v=n5DHgYz-HpA&feature=related

And don't forget, Microsoft sent out SDKs to publishers - this is much further along than some people seem to think.

They're sending out SDKs to publishers the day of the conference. I said this in the tech forum (the link to the source is there), but according to Keighley this is Fall 2010 at best. It's further along if people think, say, that all MS has is proof of concept. It's not quite as far along if people think this will be releasing this year.
 
Natal is not based on 3DV tech, anyway it looks like they have been working on the problem for a while both hardware and software side.

Ah, I stand corrected. So MS's technologies may indeed be not ready in 2009. Perhaps the mapping to 3D points is the key innovation (I seem to recall 3DV using slices of 2D images, instead of 3D mesh).

I agree the tech is of unmatched precision and works wich lead me to think that the tech has been working on for a while whereas some demo looks like they have been done over night.
That's strange imho and not very sony like, was the project on the verge to be set-aside?

I doubt it. PS Eye researches have been on-going for a while now. Assuming they are not rigged, the live demoes cannot be done from ground up overnight if it must guarantee to work on stage. It's realtime 3D input. They won't even know the stage condition back in the lab. Kaz publicly said that he's interested in augmented reality: http://forum.beyond3d.com/showpost.php?p=1271750&postcount=10
 
Okay, I'm too late to the party so my contributions are already prboably covered among these many posts. Still...

It's funny to see how the worm has turned. I see parallels here with E3'05, where MS showed what they had, and Sony showed what they hoped to have. this time, MS is showing a lot of hopeful lifestyle footage, while Sony are showing what they've got. The presenter really was nervous, wasn't he! He must have been practically crying during MS's conference! They pulled it back in the end though, showing a good case for the pro's of their system.

In summary, my appraisal is MS's system is by far the most elegant and intuitive, and the one people will take most readily to. Not sure when it'll be ready. I'm also not sure how it compares in some aspects. eg. Firing the bow, the camera could track the hands but if the release was occluded by the forward arm, the bow wouldn't fire. It also had quite a lag evident - not a game killer, but still a small barrier to immersion. The PS3 solution does look nerdy, but the tracking was instantaneous. They didn't show how much player motion they were following versus just tracking the widgets. The idea of buttons makes sense - theirs was a good argument about how you need a trigger to fire a gun. Nothing at all stopping MS releasing a versatile remote to work with their system though.

We need more examples of fine-control from Natal to compare motion performance. At the moment, what I've seen places the motion control advantage with PS3, intuitiveness with Natal. Also many of Natal's abilites shown are readily doable on other systems, but MS were the ones to show it and get the cred. Kudos to them! At least they're working on it and showing some benefits. Sony have had years to produce working systems and dared to release concept vids, but they delivered nothing. MS has given them a swift kick up the rear.

As for Nintendo, they're ahead with install base and reputation, but behind in technology. Which doesn't much matter, because they have a product to sell while everyone else is only dishing up promises!
 
Foot tracking is real important also. Think beyond fps games, etc, and instead picture stuff that the average casual person would like. For example, a product that teaches yoga or tai chi, and uses Natal to watch your limbs to critique/comment on your form.
Do we have anywhere a report on what exactly Natal can distinguish/track? They superimposed the skeleton in the demo. Can it follow feet etc.?

Also, it just occured t me, if you can determine a player from their background in an image, you could map a 3D skeleton to it as we know the human proportions, and using inverse kinematics along with some 3D mapping (PS3motes) should be able to derive a good pose. The chief problem there is probably background extraction!
 
Do we have anywhere a report on what exactly Natal can distinguish/track? They superimposed the skeleton in the demo. Can it follow feet etc.?

Also, it just occured t me, if you can determine a player from their background in an image, you could map a 3D skeleton to it as we know the human proportions, and using inverse kinematics along with some 3D mapping (PS3motes) should be able to derive a good pose. The chief problem there is probably background extraction!

Yeah, it follows your total body, but how accurately no one knows.

This blog linked earlier is about the closest I've seen to detailing how it works, but the info is very sparse:

http://procrastineering.blogspot.com/2009/06/project-natal.html

So somehow out of the point cloud they distinguish the human figure and then use a simplified skeletal body to resolve movements and track the user? They pick points of interest out of the cloud to build the skeletal framework. According to the blog, that's the real magic of the solution.
 
If it's a software solution, sounds like MS have trumped everyone! They'll have the working system and patents to limit other people. Still makes me wonder why the development is XB360 first. :???: Then again, Natal will be a PC system won't it, working in Media PCs etc. So as a broad MS interface development, huge RnD towards the MS future, it makes sense.
 
If they made it too sophisticated, they may be vulnerable to a partial solution this fall. This may actually drive Sony and Nintendo to step up their next-level motion sensing marketing and projects this Christmas (Oh goodie !).

Some of the applications can be approximated using other approaches (without the 3D mesh).
 
This article is worse the read (even if I linked him already in the tech section), really worse it
It's still unclear if some work is done in the "thing" or by the 360 it self.
 
erm...OK, watched Milo and the 2 'family' videos - the staged one with the racing game where the dad changes tyres (among others) and the 'live' one where mum does a painting (whilst family are sitting watching). I also watched the live painting demo...not seen anything yet that looks vastly better than eyetoy?

Are there any other demos? So far all I can think is 'eyetoy 3' - from all the fuss I was expecting this to replace controllers? I must be missing something...they also keep stressing 'no more controllers' which would imply no extra buttons?
 
I don't really get the bickering over which of thse is better, they seem to have some fairly stark differences here:

Natal:
- Great for user interface/interaction with the system itself, and media playback.
- Very limited for gaming, no buttons to control anything will totally limit the type of games you can play with it.
- Probably easier to sell as it only requires one periph', and has a cool factor totally unmatched by Sony's offering.

PSMC:
- Much much better for actual gaming
- Far more accuracy/precision and can have actual butttons to control games. Far more potential here from a gaming standpoint
- Will suffer from a major 'been there done that' perception, as it appears very similar to what the Wii is doing. Not much differentiaion here to the casual.
- Faces an uphill battle because of costs (relative to wii), and the fact you need two peripherals for it to work.
- May benefit from it's similarity to the Wii, can port over Wii games.

So, to me:

User interface/cool-factor/probablity of success: Winner Natal

Precision/Control/in-game usage: Winner PSMC

I don't really think PSMC has a great chance to succeed because it's too similar to what is already offered, and it doesn't really do anything that is going to excite people en masse. I think it's an awesome control scheme (both the Archery, and Painting demo's were way better than the MS or Nintendo equivalents), but I don't think it'll take off.

Natal, I think will be succesful, just because of the futuristic interface, and ability to interact with your console as though it's the year 2030 instead of 2010. For that alone, adoption is going to spread pretty quickly. Friends are going to come over, see it in action, and WANT IT!
 
You forgot about PS Eye. Also, I'm not bickering about the controller techs. This is just a discussion.

It would seem that by tracking proxy indicators (e.g., movement, color, 2D shapes), you can achieve many similar applications, may be even cheaper e.g., Instead of pinching, do a larger grain 2D gestures (fist, open palm, V-fingers, thumbs up/down/left/right).

Some applications like recognizing drawing on paper (EyePet) do not need a 3D mesh to begin with.

The missing pieces are (at least):

* Scanning real world 3D objects: The system may not care about the exact shape/size of the object in the mean time. Using a scanned skate board or guitar doesn't make a big difference (Using the real skate board may be more useful). Projecting your favorite skateboard in-game is attractive too, but that can be done in v2.0 next year.

* 3D full body tracking. Can be approximated like what PS Eye is doing now. There are very little use for accurate 3D foot tracking at the moment.

* The big problem is seeing in the dark for some applications. Again, next year.

* Detailed and fast finger tracking. Next time. Not sure if Natal can do this too.
 
You forgot about PS Eye. Also, I'm not bickering about the controller techs. This is just a discussion.

Ya, a discussion formed out of necessity because all the bickering that basically ruined the Press Conference thread.

Not sure how I forgot the PSEye, it doesn't do what Natal does, if it can than that's a HUGE oversight by Sony not to include Gesture/Voice controls in the PS3 XMB.
 
I don't get what is so futuristic? All I see is a load of eye-toy type demos and a staged presentation that could have been completely scripted - and by the sounds of it I'm not alone:

"The Milo demo was partially being manipulated by a developer who was sitting nearby, and I couldn't tell if he was merely calibrating the game or how much he was pulling its strings."

I'd love to see something futuristic, but we already have games with voice recognition, surely the Milo thing could be done on eye-toy? What aspect of is can't?

It's a serious question!
 
*If* this is the case, it may be more because the developers and the logistics need time (e.g., price needs to drop further). Technically speaking, a partial Natal (just the original 3DV functionality) should work.
You only get one chance to make a good first impression. MS would be remarkably stupid if they released Natal with "partial functionality" (eg, not very good compared to its potential). Especially if Sony isn't launching til 2010 themselves.

In a lot of ways, these motion controllers are like a new console launch. I wouldn't be surprised if MS and Sony time it to release concurrently, or close to.
 
I don't get what is so futuristic? All I see is a load of eye-toy type demos and a staged presentation that could have been completely scripted - and by the sounds of it I'm not alone:

"The Milo demo was partially being manipulated by a developer who was sitting nearby, and I couldn't tell if he was merely calibrating the game or how much he was pulling its strings."

I'd love to see something futuristic, but we already have games with voice recognition, surely the Milo thing could be done on eye-toy? What aspect of is can't?

It's a serious question!

In terms of interfacing with your console.

For example, I want to stream some video's off my PC from the Xbox:

- turn on Xbox, sit on couch (don't bother trying to find controller that is probably lodged in my cushions)
- Tell Xbox to "Show Videos" > "PC"
- Using only hand gestures, navigate to video, quick point to begin playing
- Phone rings, say "Pause", get up answer phone
- Sit down, say "Play", enjoy your movie.

Come on... that's cool, that pretty futuristic, that's somethine people will remember when they see it in action.
 
Status
Not open for further replies.
Back
Top