The New Definitive HDTV Lag FAQ

[Ragnorok64]

Is there any remote I can buy for this monitor? I dont feel like getting up all the time to put it down or something.

I doubt there’s an IR port on the thing for any remote to work with it.

[KoshTheKoala]

Is there an ongoing up to date list of HD TV lag? There’s a lot of info out there spread around but I never seems to be condensed into one post or list anywhere. There’s a list on avsforum that I saw but it didn’t seem entirely up to date.

Right now it seems the LG XXLH30 and the Panasonic XXLZ800 are/were some of the best out there. I see that there has been testing done in this thread on various sets, but at 53 pages it’s a bit spread out.

I’m not averse to doing research but honestly I don’t know what sets to even begin with without an updated list. I would think that someone somewhere would have put on together by now, and I just may not be finding it.

digitalversus.com has a lot of tests on it…

I guess someone could compile it. :tup:

Have fun! Naw, I might do that one day… maybe.

Shinshoryuken - it’s pretty accurate, but you need multiple tests to make sure that the camera and screen are synchronized, otherwise you have a 1 frame error of margin. You also need a 60 fps camera, and hopefully you’ll get a better one than mine. The canon sd630 I used records at 60fps, but only at 320x240 resolution.

[Mystic]

I doubt there’s an IR port on the thing for any remote to work with it.

:sad: Still happy with the purchase though.

[shinshoryuken]

digitalversus.com has a lot of tests on it…

I guess someone could compile it. :tup:

Have fun! Naw, I might do that one day… maybe.

Shinshoryuken - it’s pretty accurate, but you need multiple tests to make sure that the camera and screen are synchronized, otherwise you have a 1 frame error of margin. You also need a 60 fps camera, and hopefully you’ll get a better one than mine. The canon sd630 I used records at 60fps, but only at 320x240 resolution.

I got a canon sd550 which I think can do 60fps at 320x240.

but I’m clueless on how to use a camera. should be similar to a sd630…
can you tell me what settings I have to do?

[KoshTheKoala]

I got a canon sd550 which I think can do 60fps at 320x240.

but I’m clueless on how to use a camera. should be similar to a sd630…
can you tell me what settings I have to do?

Just set it to video mode and 60fps mode? The icon should look like a man running.

[shinshoryuken]

Just set it to video mode and 60fps mode? The icon should look like a man running.

THANKS!!!

I just did the test!!

Im mind blown!!:wow:

[KoshTheKoala]

THANKS!!!

I just did the test!!

Im mind blown!!:wow:

I look over the data using the open source pc video editing app Virtualdub

it lets you easily and quickly open video files and advance frame by frame

you can even use virtualdub to export the movie as a series of images then upload them or whatever… or make them into a video

[Mihai]

I just played on the biggest TV I’ve ever seen today…and the input lag was hilariously bad. Tube TV’s > all when it comes to SF.

[KoshTheKoala]

I just played on the biggest TV I’ve ever seen today…and the input lag was hilariously bad. Tube TV’s > all when it comes to SF.

not necessarily true

all standard definition crts are as good as lagless lcds
hd crts are more often than not worse than lcds i think

[shinshoryuken]

I look over the data using the open source pc video editing app Virtualdub

it lets you easily and quickly open video files and advance frame by frame

you can even use virtualdub to export the movie as a series of images then upload them or whatever… or make them into a video

I was using mediaplayer classic.

Ive downloaded Virtualdub too…and I got the same result!! :tup:

[NAP]

Any more word on the LG 32LH20? Its the cheapest 32" lcd I’ve seen that also might actually have acceptable lag.

Also, what are some good commercial grade lcd tvs? 32" preferably.

[shinshoryuken]

Any more word on the LG 32LH20? Its the cheapest 32" lcd I’ve seen that also might actually have acceptable lag.

Also, what are some good commercial grade lcd tvs? 32" preferably.

I don’t think “commercial grade” LCDs are free from lag…you still have to test it out to find out. People who buys these screens won’t actually test them out for input lag and just go with the manufactured specs.

Most so called commercial grade LCD are also called “digital signage”, most of them are S-IPS type panels which are good for viewing angles but are not as fast as the TN panels.

So far most LCDs thats been tested “lagless” are TN panels that are 26" and smaller.

[System_1]

I was thinking of getting a 37’ LCD HD tv. Anyone knows one that’s lagless?

[shinshoryuken]

I just did another test this time not on my home set up:

HP LP3065 > DVI-D> PC SF4

using KoshTheKoala method

It took Ryu 12 frames to hit with jab after pressing the button, It’s rated at 3 frames, so 12 -3 = 9 frames of input lag.

so 9 frames - 6 frames (inherent lag) = 3 frames of lag

I’ve actually tested this method with a CRT too and it confirms the 6 frames inherent lag.

Ive tested the monitor before using the CRT vs LCD stop clock test and got 2~3 frames of lag.

So this method might be the most accurate method to use!! its more clear cut than comparing ##s on a clock and its actually using real input PLUS your using sf4 for the test.

The only thing I’m concern is if the PC version has more than 6 frames inherent lag compare to the XBOX360/PS3 version…can’t test it cause it Im not home.

[KoshTheKoala]

I just did another test this time not on my home set up:

HP LP3065 > DVI-D> PC SF4

using KoshTheKoala method

It took Ryu 12 frames to hit with jab after pressing the button, It’s rated at 3 frames, so 12 -3 = 9 frames of input lag.

so 9 frames - 6 frames (inherent lag) = 3 frames of lag

I’ve actually tested this method with a CRT too and it confirms the 6 frames inherent lag.

Ive tested the monitor before using the CRT vs LCD stop clock test and got 2~3 frames of lag.

So this method might be the most accurate method to use!! its more clear cut than comparing ##s on a clock and its actually using real input PLUS your using sf4 for the test.

The only thing I’m concern is if the PC version has more than 6 frames inherent lag compare to the XBOX360/PS3 version…can’t test it cause it Im not home.

The PC version has 6 frames too, I checked that a few days ago.

The arcade version of SFIV actually ran on PC hardware - Taito Type X2, which is basically Windows XP + Intel C2D E6400 + NVIDIA 7900GS, so the PC version, if anything, should be faster than the ported ps3/360 versions. But it really isn’t.

[shinshoryuken]

The PC version has 6 frames too, I checked that a few days ago.

The arcade version of SFIV actually ran on PC hardware - Taito Type X2, which is basically Windows XP + Intel C2D E6400 + NVIDIA 7900GS, so the PC version, if anything, should be faster than the ported ps3/360 versions. But it really isn’t.

Thanks KoshTheKoala for bringing up this method of testing.

Even though the LCD vs CRT stopwatch test was the norm for testing input lag, I still think there alot of factors to that can give inaccurate results. ex. ports not synchronize, camera shutter speed etc,…

I’ve been testing the 60fps method with various moves with different characters and I was able to get constant results. You just need to know the the specific move frames and adding the magic 6 frame inherit lag. :tup:

[ulovemikeroch]

not necessarily true

all standard definition crts are as good as lagless lcds
hd crts are more often than not worse than lcds i think

No shit, keyword is lagless. Any lagless display, regardless of the type of display it is, is a good one.

The PS3 supports 1080p output via component also.

It doesn’t.

[Ragnorok64]

It doesn’t.

I’m pretty sure it does. You just won’t get any source that has HDCP on it (BD movies) to output at that resolution, but games will.

[System_1]

OK, so for those who are thinking taking the word of someone testing an LCD which we don’t know the input lag of against another LCD to determine the latter LCD’s input lag seems valid enough, I present the following argument/counterpoints to show that it is indeed a fallacy.

So you’re saying your monitor has a low input lag by basing it upon another monitor that has an input lag of “reasonably low”

That is neither accurate nor precise.

Have you read the first post of this thread? Allow me…

(although, technically, response time has LITTLE to do with lag. after all, a monitor with a 16ms response time cannot have 0ms input lag. Nonetheless, the point is a monitor can have a 2ms response time and yet still have a 60ms input lag so the general point is still valid)

That said, even if the processing of the monitor takes 0ms, the LCD panel of your monitor itself takes 8ms to react. 8ms is roughly half of 16.67ms which is the time of one frame.

So, your monitor is already behind by a half a frame without even adding in the processing time yet!

Interesting typo: “if fact”

While I didn’t find INPUT LAG information on a 19inch NEC, I did find lag information very quickly and easily on several other NEC LCD monitors. Why would your 19 inch monitor be necessarily immune to the same input lag related “flaws” that these other monitors have? And why test based on it unless you’d previously ruled out that your monitor is not affected?

http://www.digitalversus.com/duels.php?ty=6&ma1=52&mo1=95&p1=969&ma2=52&mo2=345&p2=3251&ph=12

Well, the NEC monitor on the right in the page I just liked to shows a input lag ranging from 10 to 50ms. That’s a variance of 40ms. Which is a variance of 2.4 frames on a 60hz monitor. Your monitor we don’t have data on so it could very well have the same variance. It a forum where we are sweating 1 or 2 frames of lag, you cannot correctly say that a monitor that might have a 2.4 frame variance “gives a decent estimate”

I already said it was not precise, just that it had reasonably low input lag for an hdtv. considering alot of hdtvs have horrendus input lag, in the order of 100 - 250 ms (http://www.avsforum.com/avs-vb/showthread.php?t=558125), even the worst case of 50 isn’t that bad.

Jesus Christ you fail on so many levels.

At least the previous poster had somewhat constructive arguments, not just trying to be the e-bad ass.

Anyway I tested vs crt to add some precision to my statement

The clearest picture I had showed a difference of 13 ms, which was pretty typical, but the HDTV was not in native resolution (crt didnt support it) so the input lag would likely be even less.

http://img4.imageshack.us/img4/6162/dynexvscrt.jpg

edit: so I decided to do some more testing, as input lag can be pretty variable. If your doing this make sure to set the refresh rate to 60, I was getting some pretty high numbers before I set them both to 60hz (30-40 ms at times).

Anyway here’s the difference for 10 pictures in a row (in ms): 17, 18, 20, 14, 0, 13, 14, 0, 12, 0. The zeros kind of surprised me, but I double checked. To sum up:

Max: 20
Min: 0
Avg: 10.8

This is for the 42" Dynex DX-L42-10A

[shinshoryuken]

edit: so I decided to do some more testing, as input lag can be pretty variable. If your doing this make sure to set the refresh rate to 60, I was getting some pretty high numbers before I set them both to 60hz (30-40 ms at times).

Anyway here’s the difference for 10 pictures in a row (in ms): 17, 18, 20, 14, 0, 13, 14, 0, 12, 0. The zeros kind of surprised me, but I double checked. To sum up:

Max: 20
Min: 0
Avg: 10.8

This is for the 42" Dynex DX-L42-10A

I suggest to try out KoshTheKoala/NKI method. Just as long as you have a camera that can do 60fps, the result is much more accurate and simpler to do.

I’ve been trying out this method to whatever TV I see, no matter if its a CRT or LCD and the result were mind blowing. Always consistent!!

LCD vs CRT stopwatch test results can be too wacky.