Top
570°

Sony admits Playstation Move has 22ms delay lag

During a recent playtest, a Sony rep admitted that Playstation Move has a 22 millisecond lag. The lag has been described as a 'little noticeable'.

Read Full Story >>
maxconsole.net
The story is too old to be commented.
sixaxis2745d ago

"On the plus side however the accuracy of movement puts this motion controller above the Wii experience."

awesome

whateva2745d ago (Edited 2745d ago )

this is like saying Sony admitted that Blue-ray can hold 50GB of data.

this is a achievement not something that they was trying to hide.

"The device has a faster response time than human reflexes to ensure precise gesture controls"

blitz06232745d ago

22ms is neglectable. That's why Sony is advertising it as no lag, which obviously attracts more people.

They can obviously tell everyone it has 22ms delay, but why tell everyone if 22ms is not noticeable? Anyone who makes a big deal out of this is stupid.

nix2745d ago

that's how long is 22 milliseconds. (approx)

so get over it haters!

Chubear2745d ago

makes it seem like "Ohhh they've been caught! and found out!" lol

sikbeta2745d ago

lol that's so true, this was made to sound like a "bad thing"

"OMGz da 22ms delay, da bad lagz :("

Dumb lol

hay2745d ago (Edited 2745d ago )

Wow, I expected something in range 60-100ms. 22ms lag is small enough for fluid 30fps gaming without intense action.

DarkSpawnClone2745d ago

isn't that like less then a second LOL ?, i know i wont notice it.

HolyOrangeCows2745d ago (Edited 2745d ago )

"apparently who ever named this News article don't know how short 22 milliseconds is!"
Exactly.
That's better than Wii, and a HECK of a lot better than Natal.

EDIT: I'll bet that the first comment is made by Bungie. What a freaking loser.

WildArmed2744d ago

lol yeah. I wa slike.. there's nothing to admit there.

That's an impressive feat.
Thats nearing controller response time.

LinuxGuru2744d ago

22 thousandths of a second is freaking nothing at all to even begin to be concerned about.

Totally insignificant.

+ Show (6) more repliesLast reply 2744d ago
krisq2745d ago (Edited 2745d ago )

[edit]
You're right guys. I took 1s as 100ms :/

Azerach2745d ago

its less than a frame of lag at 30fps. 1000ms/30 = 33,3ms

wicko2745d ago (Edited 2745d ago )

That is NOT 6-7 frames in a 30fps game. 22ms is less than 1 frame, 30 fps is 1 frame every 33ms.

kneon2745d ago

Your math is off. At 30fps a frame lasts 33ms. So the Move lag is under 1 frame duration at 30fps. They can do that because the camera is running at 60fps. There is up to 16.6ms lag just due to the capture rate of the camera, that means all the processing is being done in under 6ms, that's pretty impressive.

This is one of my beefs with Natal. If the rumors are true, then the Natal camera is running at 30fps, so the minimum lag is double the move. Plus there is far more processing to do with Natal, and since they've removed the on-board CPU that processing will take even longer.

IdleLeeSiuLung2745d ago (Edited 2745d ago )

To be entirely correct, a frame doesn't define a time frame. You could have 60fps, and 22ms is more than one frame.

What you really are talking about is refresh rate or frequency. Both define a time frame for a cycle to re-occur. So you could say 30fps or 60fps is a frequency (although they are measured in Hz).

Frame != Frequency

Halo 3 is 30fps game and Modern Warfare 2 is 60fps.

Hideo_Kojima2744d ago

That means that in a 30FPS game there is 0-1 lag because by the time the screen refreshes the movement would have already been captured.

So 2/3 of the time you will get 1 frame of lag and 1/3 of the time the lag will be 0 depending on wether you moved the controller right after a new frame refreshed so that the controller has time to calculate before the new frame has been desplayed...

Why the hell am I going on and on...

Basically there is not lag.

+ Show (2) more repliesLast reply 2744d ago
Wrathman2745d ago (Edited 2745d ago )

is it not a problem when you do continuously fast movements?for example, when you play that table tennis game with a capable competitor,wont you see the lag when the game gets fast?TT is a very fast game.

in comparison to the wii..wot is it latency?

and just before anyone starts bitchin about natal lag..no-one knows the lag of natals final build yet.but im sure it will be more than 22ms.how much more tho?

krisq2745d ago (Edited 2745d ago )

But it will be noticable only on very fast, continous movement.

Azerach2745d ago

At the moment natal lag is about 7 times more than that of the move.

kneon2745d ago

Well the rumors say the camera only runs at 30fps, so you've got an average lag of 16.6ms just due to the camera, and up to 33ms if the action happens just after a frame capture. That doesn't include any of the processing, which is far more complex than what Move has to do. With no on-board processing anymore I expect a minimum 3-4 frames lag at 30fps.

Then you also need to consider error rate. What Natal is trying to pull off is quite tricky and it is essentially making a best guess at what it is seeing. It will occasionally make mistakes. While the colored spheres on Move may look goofy to some, it does offer a very well defined tracking target making it far less likely to make mistakes.

DaRazorback2745d ago

I am not to concerned about it. I mean think about it most TVs have a min. of a 5 ms response time and some are closer to 15 ms. So it shouldn't be that big of a problem.

Rock Bottom2745d ago

If the final lag is about 22ms, it wont be noticeable at 30fps , it will be on frame of delay which is the perfect lag(even if it capture the coordination for the move wand movement in less than a frame time, it cannot show it until the next frame) but in reality things aren't as simple as that, games has their own delay time, which is in a lot of cases higher than 60ms and might go as high as 120ms.

Ju2745d ago (Edited 2745d ago )

@keon, if the camera scans with 30fps or 33ms a frame, then that's the minimum lag you can ever reach. Without errors and 0 time for processing. So its 33+ ms at least. 16.6ms for a 60Hz scan rate.

IdleLeeSiuLung2744d ago (Edited 2744d ago )

In terms of lag, I believe about 133ms is where people start noticing lag in games. So 22ms still leaves a lot of room for other processing using PS Move.

The PS Move technology is still very simple at it's core. The ball determines the position in space captured by the camera and I'm sure the gyroscopes tell the orientation of the PS Move as a vector. Add the origin (i.e. the ball location) + the vector and combined with the accelerometer reading and you got a complete description of the PS Move. Repeat this process as fast as the gyroscope, accelerometer and camera can capture the data and that is it!

I could probably make one in my home albeit less professional....

Natal on the other hand reads your environment like a topography map face on of your living room. Then tries to determine what the various parts means and maps it to a skeleton to determine your movements and even in some instances predict your movement. Wetter the technology is suitable for gaming is yet unproven and it might be like the Wii motion control, barely good enough for casual gaming, but the technology non the less is nothing sort of amazing and a much more difficult problem than what PS Move has implemented.

kneon2744d ago

@Ju

Well not exactly. you're assuming it takes a whole frame to capture a frame. It takes less than a frame to capture a frame, but we don't know how much less. you can think of it as taking a series of photos with a pause in between. If the motion to be detected occurs right after a frame has been capture then you have a full frame of lag. On the other hand if it occurs during a frame capture then it's less than that.

So the camera induced lag is variable and depends not only on when the action occurs but also on how long it takes to capture each frame. the ratio of capture time to frame duration will affect the amount of variability.

Ju2744d ago (Edited 2744d ago )

@Kneon, this is quite true, however, this implies the scan frequency is way above the 30Hz. I assume, if the specs say 30Hz scan, that is the fastest frequency you can read that signal. No matter how it's composed (full frame or not - 1/33sec to get some useful data). And, especially because you mention the "lag induced by motion", to be accurate, this scan frequency must be higher than the say "processing" frequency (general rule, IMO, is about twice as fast).

So, again, with a 30Hz scan frequency, and neglecting all errors, and assuming you can extract all information out of exactly one frame (which you can't) the minimum lag (under this considerations) cannot be less then 33ms.

+ Show (6) more repliesLast reply 2744d ago
FragGen2745d ago (Edited 2745d ago )

None of this changes the fact that motion controllers basically suck at their core. LOL.

I played our Wii with my daughter last weekend and it really struck me how I had forgotten how truly gay MC is and how badly it can ruin a game it is not suited to. It just feels so gimmicky and tacked on in so many Wii games and that thing has MC from the ground up.

The PS3 and XBOX fanboys are so busy worrying about who's d*ck is bigger on the move versus Natal thing that they're not uniting against the common enemy: retarded casual games and poor control schemes.

Hopefully, developers will make these an asset rather than the curse they've been to the Wii.

MazzingerZ2745d ago (Edited 2745d ago )

it took me longer to read '22ms'

Droid Smasha2745d ago

inb4 "Move has no lag"

oh that right sony just CONFIRMED there IS lag

KeiKei2744d ago

your comment is a waste of time

bjornbear2744d ago

the article is saying Move has no lag, not humanely detectable lag anyway

you fail so epicly =D

NFGC2745d ago

To put that in perspective, the best DSL ping time is around 20ms which is nearly no lag. 22ms for a motion controller is a significant achievement

SeNiLe9112745d ago

so 22 millisecond's is what, One Th...?

UltraNova2744d ago (Edited 2744d ago )

Human brain response time from image processing to motor function signaling, in other words how long it takes us to see something and move either our head,legs or arms according to that image is an average of 200 ms.

So the 22ms it takes Move to register a gesture from us is basically untraceable by our brains. Even if you see it your brain will discard it as non-data (data that will not be stored in our memory cortex for later processing).

So either Sony is lying or Move is a near perfect Motion controller. Why would they, to get caught later?

This is a reference for anyone who is willing to indulge themselves in some advanced neuroscience.

http://www.fmri.org/pdfs/gr...

marinelife92744d ago

Is barely perceptible.

Go to this site if you want to see the lag. It has a slider that you can go all the way to 80ms with.

http://img228.imageshack.us...

Sheikh Yerbouti2744d ago (Edited 2744d ago )

Twenty-two thousandths of a second...

That is equatable to ~45 frames per second. Human perception is 13 frames per second and able to perceive lag at a periodicity of every 76 ms. I guess this means its impact on lag is minimal. Actual results may vary by game.

+ Show (9) more repliesLast reply 2744d ago
sid4gamerfreak2745d ago (Edited 2745d ago )

thats not a lot of lag actually...

infact, its barely noticeable considering human reflexes are 30-100ms

deafwing2745d ago (Edited 2745d ago )

jk = I'm sure you can make coffee in that time, pour milk, add sugar and be back on your sofa in time to do a ShoRyuKen. (couldn't help it).

But seriously, 22ms ... I guess it could add up depending on how the game is designed. If it does, and the lag starts hitting seconds, and then tens of seconds, then "Houston we have a problem."

Azerach2745d ago

Thats why computers these days have multiple cores, its easy to design a game so the lag doesnt add up. A bit unrelated but there was an interesting interview about GoW3 MLAA and how they reduced the lag from 120ms to 20ish by utilising the SPUs.

xg-ei8ht2745d ago

Does anyone know the lag of natal or wii.

deafwing2745d ago

I don't think Nintendo ever reported on that, nor did M$ ... but I'm sure if someone asked nicely they might divulge.

Strange_Evil2745d ago

Natal is somewhere around 100-150ms (the last one, no news on the upgraded one)... 22ms is nothing.. Even a normal controller has that much lag... For example, GTA4 has a lag of 166ms.. Not only that, but even TVs have a LAG factor which adds up..

All in all, if it's 22ms or even 30ms, then it's 1:1 since human eye cannot perceive a lag of that sort.

Hotel_Moscow2745d ago

i hope i dont notivce this because my eye perceives things at a higher pace than my left eye fore some reason

Gun_Senshi2745d ago

natal has ovcer 1 second lag according to tony hawk

Qui-Gon Jim2744d ago

The things I have seen put Natal's lag at 100 ms or more, but that is total lag (including game lag), whereas this 22 ms is controller lag. The best latency possible in games is 66.6ms for a 60Hz framerate, or about 100ms for a 30Hz framerate, with some games going as high as 150 or even in the 180's (GTA IV). If Natal does stay at 100 ms of lag, it puts it at a pretty standard amount of lag. Keep in mind, though, that everything that has been seen is for Ricochet, a pretty simple game to process and render.

Move's inherent lag is 22 ms. Compare that to the 16 ms lag of the DualShock and you see that it increases game lag by only 6 ms. So most games using Move will have lag of 72 ms to 106 ms.

Check out this article for more information on controller latency:
http://www.eurogamer.net/ar...

I think the lag of both motion controllers looks worse than it is simply because of the visual feedback of seeing your movements echoed on screen rather than just pushing a button to fire a gun.

+ Show (1) more replyLast reply 2744d ago
BeaArthur2745d ago

Still more laggy than my controller.

xabmol2744d ago (Edited 2744d ago )

Move is 5.5ms slower than a standard wireless controller.

5ms is nothing.

BubbleSystemSuck2745d ago

Good job Sony.

Sony knows about hardware.

BTW... is something strnge the low Stock of PS3 on USA... something like this happen before the New model of PS3... is a hint?

PCnPS3Gamer2745d ago (Edited 2745d ago )

in north america they are shipping the ps3 slim with the upgraded smaller more efficient 40nm rsx chip soon so maybe their letting the old 70nm rsx chips or something like that get low then send the 40nm rsx chip ps3's out.thats just my thought tho....

i know 22ms isnt bad its just so hard not to think it is considering my hd tv is 5ms, but it also has less to process. Also i'm thinking maybe the move delay is including the move device aka controller, bluetooth factor,ps3 process delay, tv delay if that is the case thats not to shabby.

also i really hope the natal fanboys or ps3 bashers aka the media dont spin this good news and make it sound horrible.

SuperStrokey11232745d ago

A a die shrink is coming (already here with all of the newer ps3s), 40nm i believe. They have switched their process over to this so there should be more units available shortly.