Old Discussion Thread for all 3 motion controllers

[Arwin]

That sounds great, really looking forward to that feature. And absolutely, people can get used to a fair amount of latency ... you should try to measure how much latency most cars have. Also with racing games you have to think a considerable amount ahead already - if you see a slow corner running, you may have to brake several seconds in advance. What's 200ms of lag in that context?

However, I do think you'll find that a lot of popular games have reduced that lag significanly, and it may well be one of those secret ingredients that makes a game popular, just like most people aren't aware of intonation in language, but it still greatly affects their interpretation.

 

[patsu]

Nintendo filed a patent for a Wiimote American Football controller:
http://www.siliconera.com/2009/08/24/nintendo-developing-a-squishy-football-controller/

A recently filed patent application by Howard Cheng, Vice President of Research and Development at Nintendo of America, reveals a plan for a soft football controller for use with the Wii and the remote.

Throwing – Players firmly strap their hands to the ball and make a throwing motion. The pitch angle and force of the throw determines the trajectory arc of the throw. Side to side motion determines the yaw angle. Pressing buttons on the Wii remote can adjust other options.

Running – The football controller sense body motion, shifting left and right to dodge tacklers and even jumping motions by raising the ball. Running speed is determined by players making a jogging motion.

 

[Shifty Geezer]

Burnout might appear more responsive, but it may well be the case that it is only scanning a reduced skeleton - hands and feet - versus the full body scan of the Breakout demo.

I can't see that making a difference. That would suggest it's processing time causing latency, but if that were true, the processing time would overrun frames constantly, and the responsiveness who go down the toilet. Especially when we're comparing to a single player at the moment. When it has to deal wtih multiplayer, the processing requirement would be even greater, and the low input exaggerated.

I guess it could be an issue if the processing is happening on a PC perhaps. But then that doesn't bode well for the custom chip. If a meaty PC can't handle full motion capture at 30 fps, they'd need a monster processor in Natal!

What other cause could there be for an information delay? Network issues seems unlikely. Big buffers? USB is fast enough to deliver 12 MBs in 1/5th second, so if the lag is caused by that, we're looking at 12 MBs of data from Natal! Perhaps at this stage there is lots of data for debugging, and the final can just send image and skeleton info?

 

[grandmaster]

Well what is interesting is that there was a PC present at the demo as well as the 360. Burnout was run by double-clicking on a Burnout icon on the PC desktop. The tech screen with the RGB camera feed and the IR image was also PC based.

 

[patsu]

I wouldn't worry too much about lag. In the worst case, they can add custom controllers (e.g., driving wheel). Users will need them for core gaming anyway for extended action and triggers. Once there are custom controllers, they can simplify the tracking.

In the mean time, we can't really judge the "hands free" performance using the current PC setup. It's too early (and probably pointless) since the final solution will be hardware based. OTOH, the final solution needs to fit into a smaller budget and casing (Have to fight overheating issue !).

That said, it is much more critical for them to work on the gaming concepts. Without them, the technology may be "useless".

 

[Arwin]

Well what is interesting is that there was a PC present at the demo as well as the 360. Burnout was run by double-clicking on a Burnout icon on the PC desktop. The tech screen with the RGB camera feed and the IR image was also PC based.

Yeah, that's what I guessed. Now Shifty, don't make too much of the performance of a PC unit. A custom chip can be much better at this particular work. In theory a single SPE type core could do a better job at this than an 8 core Intel processor, never mind that it's not at all clear whether their debug software even uses more than one core at a time for the crucial parts. Or maybe they are already using a programmable chip in the processor, but it's running in debug mode. Hard to say at this point. All I know is if they are targeting fall 2010, then they would have obvious reasons for waiting that long.

 

[NeoTechni]

For Sony to do the opposite, to track full body movements in 3D is essentially impossible with their current design.

and yet theyve showed it doing that already

I have not seen this "severe lag" you're talking about. Sources?

his source is every single video
there is almost a full second of lag
and it loses track of the user at least once a video

PS3MC requires that you hold their controller for it to figure out positional data. Is this accurate?

down to the sub pixel according to every video
though natal isnt accurate at all which is why psmc is shown painting with brushes and natal is shown throwing paint at a wall

urther, it seems like Natal tracks multiple limbs, so to do the same on PS3MC, wouldn't you need to hold two controllers, and have two attached to your legs?

no since pseye can do that

 

[Shifty Geezer]

and yet theyve showed it doing that already

Full body, 3D tracking? you'll have to link to that because if they've shown it, I don't remember seing it.

there is almost a full second of lag
and it loses track of the user at least once a video

It's a fifth of a second, nothing like a whole second, and the more recent demo's haven't shown any tracking faults that I've heard about.

I don't think you're impressions of Natal are at all fair. It has some disadvantages versus PS3MC, but PS3MC certainly can't do anything and everything Natal can do either. Maybe you should read this whole thread to catch up on what's been discussed so far to get up to speed?

 

[patsu]

Behind the scene with Richard Marks:
http://blog.us.playstation.com/2009...scenes-with-scea-research-development-part-1/

No new info, but he explained some of the rationales behind the design.

EDIT:
Oh, rumble confirmed @ 7:10

 

[Arwin]

Behind the scene with Richard Marks:
http://blog.us.playstation.com/2009...scenes-with-scea-research-development-part-1/

No new info, but he explained some of the rationales behind the design.

EDIT:
Oh, rumble confirmed @ 7:10

Well, and something I noticed in the gamescon trailer, the analog stick seems to be gone and instead replaced by a large pressure sensitive button. He stresses that they developed this to be fairly precise as to give you a sense like picking up objects and letting them go. It's very interesting and clever.

What I also saw in the trailer and what I expect them to discuss in the next part, is among others the features of the glowing ball. I've seen it burn really bright as well as take various colors. All in all this device is starting to look really good.

The final thing I think I got from this is that the precision may not be extreme in the sense that the gyro is approximate, and that the sub-pixel precision relates to the 60fps 640x480 image you're getting from the PS Eye. It's still very/plenty precise however, as the demo shows. They basically show them fooling around with the demos that we've already seen, but now with two people doing it at the same time.

The combination of the pressure button and the rumble is very good though. Me like.

And they look on track. From what I'm seeing I'm confident they'll make Spring.

 

[Shifty Geezer]

The bricks demo had me immediately thinking about LBP and opening up creation in that. Could be an ideal application.

 

[archangelmorph]

Got a prototype wand in the office recently.. looking forward to trying it out as soon as we can get a devkit setup for it..

 

[patsu]

If you can talk, does the prototype have an analog stick ?

 

[Arwin]

From what I've seen, I'm thinking the new pressure button has replaced it. But this button may be able to sense direction, not sure. We'll find out sooon.

 

[Brad Grenz]

The most recent video talks about a pressure sensitive trigger, but that's not where you'd put a stick. They also mention "buttons" so I'm assuming that means face buttons on top with a trigger on the bottom. Perhaps the face buttons will double as something like a directional pad.

 

[Quaz51]

The bricks demo had me immediately thinking about LBP and opening up creation in that. Could be an ideal application.

and with stereoscopic display for better perception of distance when manipulate object with wand

 

[Shifty Geezer]

Given LBPs awesome lighting, 3D and natural interaction would be truly mindbogglingly good.

 

[patsu]

Yes ! LittleBigPlanet plus its ability for people to create motion sensing mini-games should be top of the priority.

 

[whome0]

Was it 2010 spring they wanted to bring PS3Wand games on market?
I believe we see Wii ports and party games in a first wave. EA Tiger Woods golf game definitely gets PS3Wand support but must wait for next year version.

PS3Wand lit bulp should work pretty well in dark rooms as well, most(?) games don't need a bluescreen keying.

 

[Asher]

Yes ! LittleBigPlanet plus its ability for people to create motion sensing mini-games should be top of the priority.

As awesome as LBP was, its sales were not. Sony's priority should be on making a new killer-app game, a la Wii Sports, not retrofitting old games that didn't meet sales expectations with the control scheme.