Old Discussion Thread for all 3 motion controllers

[Silent_Buddha]

The problem is compounded if the Avatar mimmicks your movements. People might be surprised if there was a demo that showed a large button being pressed after you press a button on the controller. Added to that is the additional latency involved with moving a limb versus pressing a button.

In other words, the time it takes for you to think about pressing a button and the time the button is finished being pressed versus the time you think to move your entire arm, jump, etc. and the completion of moving your entire arm, jump, etc. is going to be much larger.

So the perceptual difference between a button press at 133 ms from completion to output is going to be different than the perceptual difference between a jump at 150 ms from completion to output. Especially if you have done a ton of console gaming and have already adjusted to the control latency on console.

So you have compounding factors which I think will affect regular console gamers far more than those with little to no gaming experience.

Which would help explain why non-gaming media is far more impressed than tradional gaming media.

Then again perhaps, not, as Move with LARGE movements should show similar if lesser perceived lag.

Regards,
SB

 

[boommoob1]

So at this point we can surmise that in all practicality Kinect is a beefed up Eyetoy/Eye camera?. As expected i suppose. If the tech was there at the moment at an affordable price Sony would've gone with it.

 

[liolio]

Well, no - only for 60 fps titles, most of these games seem to be 30 fps, so the extra Kinect overhead would seem to be 33 ms compared to the least laggy 30 fps titles - with games like GTA4 and KZ2 have worse input lag than Natal.

For comparison with Natal, reuslts from DF's input lag features:


I wonder if the lag increases when more than 1 player is playing?

Hum, I don't agree I'm sure Ms has gone with the lowest lag as possible (kinect overhead aside).
The lowest we have seems to 66ms I'm not sure it's relevant to the frame-rate (ie 30fps vs 60fps or two frames vs four ). "Standard" games doesn't face the same constrains as lag is no much of an issue and very few games are optimized to reach low lag.

 

[(((interference)))]

Ok, that's not working:

Star Wars Kinect demo footage:

http://www.youtube.com/v/n_y1-xAXv4g

 

[goonergaz]

Ok, that's not working:

Star Wars Kinect demo footage:

http://www.youtube.com/v/n_y1-xAXv4g

note that that is NOT gameplay but someone 'acting' to a prerender

 

[goonergaz]

Also:
Natal Games will be cheaper

did I miss the prices somewhere?

 

[Rolf N]

Ok, that's not working:

Star Wars Kinect demo footage:

http://www.youtube.com/v/n_y1-xAXv4g

When the on-screen avatar does the motions before the "player", I think we can safely call that smoke and mirrors.

 

[goonergaz]

When the on-screen avatar does the motions before the "player", I think we can safely call that smoke and mirrors.

no, Kinect software now has latency of -133ms lol

 

[(((interference)))]

did I miss the prices somewhere?

Yes

Premium titles featured in Microsoft’s first party launch line-up such as Kinect Sports will retail for £29.99 (about €37 / $44, based on current exchange rates), we were told.

Others might be cheaper, while a family friendly title designed to showcase the device's varied applications will be bundled with Kinect, our source said

 

[obonicus]

I wonder if that means $40 or $50 US.

 

[goonergaz]

Yes

so when you say "Natal games will be cheaper"...you mean "cheaper than current X360 games" I assume? (which is generally the case for these types of games isn't it, I'm sure eyetoy games were cheaper?)

 

[Shifty Geezer]

Isn't Halo 3's lag 133 ms? And KZ2 and GTA4's longer than the 150ms Natal figure?

They're total lag, not controller lag, as others have said, including the screen lag. We're probably looking at about 100ms extra on average to what we're used to.

And how can Move's lag be only 22 ms, isn't that below the minimum lag time for 60 fps titles (IIRC 66ms) never mind 30fps titles (100ms)?

I presmue that's a controller processing time and not lag per se, but the website isn't coming up for me to check. ie. No more lag in 30fps than we currently have because it can be fitted into a single frame. Sony have said this is very adjustable depending on processing, so that figure is maybe just an average anyway, and for something like SOCOM, perhaps there's 50ms, overrunning a couple of frames which is what's detectable in vids. (?)

Does it seem much worse because you're more sensitive to lag in tracking your body movements?

I think that's very true. When you press a button and then see the results, a slight delay isn't out of the ordinary, but tracking yourself, you expect your limbs to move when you move them! Eyetoy had considerable lag which was offputting when seen, although it didn't affect gameplay particularly. Then again the games in EyeToy Play weren't particularly timing sensitive from what I remember. It did have that disco game which didn't feel out of time with the music.

Perfect for a party of kids. "Johnny, let Bobby have a turn now." Out Johnny goes and in Bobby goes without having to wait for an intermission or passing around a controller.

Yes, it's very good for the party type environment, where people can just play. And what the guy said about people loving to see the avatar follow their movements rings true with me. I still love to just see the virtual puppetry in action!

Interestingly the Dance game totally ignores the latency since all it has to do is match your pose and movements to what is displayed on screen.

I heartily disagree. Anyone with a sense of rythmn who goes via the beat of music will find lag distressing. If you can ignore the beat and go via the visual prompts, yes, it doesn't matter. But there's a reason Guitar Hero had calibration and players don't ignore the music to just match the buttons! When I tried GH, at first we missed the calibration, and the buttons being out of time with the music made me rubbish, because I always go via the beat. In a dance game, if the music suggests your arms are up at beat 3, I need to see them up at beat three and not a quarter of the way to beat four!

Remains to be seen whether it'll be a turn off to casuals or not, but I'm pretty sure we'll hear lots of moaning on forums with "elite" gamers.

The biggest complaints should come from girls who take dance classes, the prime audience for a dance game and those who should be most sensitive to lag. But it may be they can ignore it and be out of rythmn.

Also interesting to note how people who aren't console gamers would actually have an easier time with some of these games. Gamers expect almost instant response to a button press to jump. Far faster reflex than actually jumping. Casuals with little to no console experience on the other hand, only have the perspective of actual jumping, which is inherently "laggier" than pushing a button.

Again, I disagree. A 'casual's reactions may not be great when it comes to jumping virtual hurdles, but anyone who can catch a ball or swing a bat is going to notice 200ms of delay. Hitting a ball in a baseball game will be very wrong with the currently shown Kinect, and players will have to adapt their swing from the natural to add some delay compensation, swinging early before they want the response. I dare say the casuals won't grumble because they'll just have a laugh and aren't serious competitors as it were, but I also think if you ask them, they'll say it feels wrong or isn't accurate, but plain don't care.

Basically the same problems and responses with Wii, with it not being at all realistic in Wii Tennis or Baseball, but that didn't stop people buying and enjoying the thing.

 

[goonergaz]

They're total lag, not controller lag, as others have said, including the screen lag. We're probably looking at about 100ms extra on average to what we're used to.

Just from watching the vids released so far (obviously other than the bogus Star Wars one) there's a noticable difference in lag between the 2 control systems.

 

[NavNucST3]

I believe I have gone in the opposite direction from most...I have now upped my willing to buy price to $129.

 

[egoless]

I believe I have gone in the opposite direction from most...I have now upped my willing to buy price to $129.

Interesting. Was this because you compared to EyeToy or Wii? I never purchased EyeToy for my PS2 or PS3, but I have relatives with the Wii. They've purchased wii play for the "free" wiimote, then nunchucks and motion+ for 3 other people. Not to mention the steering wheels, guns, fishing poles and the balance board. Over time, they've spent a fortune. I know the new black wii comes with motion+ and a nunchuck. You sill have to pay 75 US dollars per extra player though for the same control as what's in the package. Ouch.

Unfortunately, I see the same thing happening with PS3 if I go with Move. So for me too, Kinect should command a higher price for my 360.

 

[messyman]

Streaming Kinect demo

http://live.twit.tv/

The girl says Kinect will be available this November.

 

[Nesh]

Ok, that's not working:

Star Wars Kinect demo footage:

http://www.youtube.com/v/n_y1-xAXv4g

Ok that video is freakin lame/. More smoke and mirrors from MS, They run a video of the game and the "performer" acted to it? So,,,,,do you control whats on screen? Or does it control you?
It is so god damn obvious that the character moves before the player.

Its like they are treating the consumer like some kind of idiot who will fall for anything that is thrown at him even if its so obvious.

That reminds me of the 80's 90's when commercials showed stuff that looked 10 times more impressive than the actual product

 

[Scott_Arm]

The Star Wars bit was definitely disappointing. I'm guessing the game really isn't in any state to be shown at all, but staged presentations are lame. I'd rather have some guy come out and say, "And there's going to be a Star Wars game too," and leave it at that. Hopefully it turns out.

I'm probably most interested in the kart racer, and in boxing as part of the sport game. All signs point to this costing less than $200, which means I'll definitely buy it. It could turn out to be a novelty, but its an interesting novelty I want to experience.

 

[liolio]

They're total lag, not controller lag, as others have said, including the screen lag. We're probably looking at about 100ms extra on average to what we're used to.

Hum, I'm sure that DF measurement has been made on display that induce really few lag.
If a tv induce 100ms of lag so you can't take DF value and call it a day. For a gamer with a piss poor tv lag for Halo3 will be 133ms+100ms and so on.
Things have to be taken in account ceteris paribus, ok one can have a shitty tv/display but it won't affect Natal only.

 

[Arwin]

From the impressions I'm getting so far, I'd say that Move lag falls within the render time for a single 60fps frame, and N.. pardon, Kinect within 2 30 fps frames. Since both will overlap for a large part with the rendering pipeline, how much of this will be visible will depend on the type of game. This would mean that the Move controller should be able to function without adding lag even in a 60fps game, whereas Kinect would add between 33 and 99ms of lag.

This seems to match what the Rare guy said - if he says that there is 150ms lag between detection and video out, we know that the best score for this scenario is actually 50ms (measured by DF for the XMB, or 66ms for Burnout Paradise for instance), which means 150-50 = 100ms. If a game performs worse than 50ms, it will be covered partly by the rendering pipeline, so the lag at the video out will typically stay the same. 150ms is also pretty much the limit of what most people find intuitively acceptable, so I think most Kinect games can live with this.

The only uncertain factor here may be how Kinect will scale to multiple players - that could cause additional lag, but it may also be that Microsoft mitages that by just tracking fewer control points and/or making less precise movement predictions.

For Move however, with a maximum of 22ms when using four controllers (which is the maximum), and that should typically fall within a single frame of a 60fps renderer, therefore not causing additional lag even in the fastest games, allowing similar scores as Burnout Paradise and the XMB currently achieve with the sixaxis.

I think it is very clear right now that both systems have their unique strong points, and it will be interesting to see how they'll play out against each other. I thought it was very interesting to see my colleague look at Move in Socom and say, hey, that could make console gaming actually cool - he hates dual analog controls.