AVS › AVS Forum › Display Devices › LCD Flat Panel Displays › Input lag wars!, post your input lag results of your LCD display here for reference
New Posts  All Forums:Forum Nav:

Input lag wars!, post your input lag results of your LCD display here for reference - Page 10

post #271 of 4031
2/25/10 at 12:31pm
Thread Starter 
Lol it really takes some serious abuse to burn in crt projectors

It took my sister bout a year of exessive 4:3 watching on my phillips crt rptv before I noticed the pillar burn it caused, at least it taught her what happens and now that we replaced it with a plasma she knows better
Reply
Reply

Gear mentioned in this thread:

post #272 of 4031
2/25/10 at 5:08pm
The Panny TC-L32X2 (32U22 "backordered") is now available through their site. I'm debating on purchasing it now or waiting until it hits Amazon due to their awesome customer service and fast returns/refunds (if needed). However, I'm impatient and may make a decision tomorrow :P
Reply
Reply
post #273 of 4031
2/25/10 at 9:14pm
Quote:
Originally Posted by wondras View Post


It sure would be nice to know how the RB2 test accounts for sensor/controller/scaler lag.

A guy at Shoryuken (forums) did some pretty scientific tests, and concluded that wireless adds no lag at all versus wired control. Nice to know, anyway, since wireless controllers are standard these days. And scaler lag..well we want to know that, right?

Just looked at the back of my video card, which is a pretty nice one (factory 100mhz overclocked HD 4890) and it does have two DVI outs...

I wonder if a DVI-HDMI adapter could introduce any lag itself? Adapters are frequently known to do so, arent they?

My current monitor is 1680X1050 resolution, though, unfortunately not 1080P.

Now, even though I'm no longer thinking it's accurate though, I'm still concerned that RB2 showed my monitor to have almost 50 ms of lag in some tests This would make me unsure of any results using it as the "low lag baseline" monitor.
Reply
Reply
post #274 of 4031
2/25/10 at 9:20pm
Thread Starter 
Quote:
Originally Posted by sharky974 View Post

A guy at Shoryuken (forums) did some pretty scientific tests, and concluded that wireless adds no lag at all versus wired control. Nice to know, anyway, since wireless controllers are standard these days. And scaler lag..well we want to know that, right?

Just looked at the back of my video card, which is a pretty nice one (factory 100mhz overclocked HD 4890) and it does have two DVI outs..

scaler lag can vary pretty greatly from 8 to 16 or more MS depending on the quality and design of the scaler

XB360 is the only console with a built in scaler and its nvidia so its very fast but when other consoles are used its up to the TV if the display settings are not set to native or cannot be as in the case of the Wii with most HDTV's (pretty much all current ones)

as i've said before way too many variables to take into account making for bad reliability in results

at least the camera method sticks to an error rate of at most 16ms and if you take enough photos you can rule out the bad ones easily and what your left with is the lag of the display.
Reply
Reply
post #275 of 4031
2/26/10 at 4:25am
Quote:
Originally Posted by sharky974 View Post

I wonder if a DVI-HDMI adapter could introduce any lag itself? Adapters are frequently known to do so, arent they?

i highly doubt a dvi to hdmi adapter introduces lag since no scaling or transcoding is involved, it's just some re-routing of copper. If it were the case though then a straight dvi-to-hdmi cable would fix that.
Reply
Reply
post #276 of 4031
2/26/10 at 6:46am
Has anyone tried a frame-based test like the ones mentioned in the LG xxLH30 thread?

I'm going to try it on my 37LH20-UA to see if I can get a more accurate figure for its lag.
Reply
Reply
post #277 of 4031
2/26/10 at 10:01am
Quote:
Originally Posted by wondras View Post

Has anyone tried a frame-based test like the ones mentioned in the LG xxLH30 thread?

I'm going to try it on my 37LH20-UA to see if I can get a more accurate figure for its lag.

I private messaged that poster (Jensend), hopefully he gets back to us soon, i was curious about what software he used as well.
Reply
Reply
post #278 of 4031
2/26/10 at 10:40am
Quote:
Originally Posted by frito View Post

scaler lag can vary pretty greatly from 8 to 16 or more MS depending on the quality and design of the scaler

XB360 is the only console with a built in scaler and its nvidia so its very fast but when other consoles are used its up to the TV if the display settings are not set to native or cannot be as in the case of the Wii with most HDTV's (pretty much all current ones)

as i've said before way too many variables to take into account making for bad reliability in results

at least the camera method sticks to an error rate of at most 16ms and if you take enough photos you can rule out the bad ones easily and what your left with is the lag of the display.

The 360 uses ATI hardware....
Reply
Reply
post #279 of 4031
2/26/10 at 11:13am
Thread Starter 
Quote:
Originally Posted by bpmurr View Post

The 360 uses ATI hardware....

you are correct, i was thinking of the original xbox's hardware
Reply
Reply
post #280 of 4031
2/26/10 at 12:02pm
I made a lag tester that is easier to use and more accurate than the number based timers:

http://lcdlagtester.googlepages.com/

Someone else independently made a very similar tool:

http://www.hardforum.com/showthread.php?t=1423433


Each bar represents 16.7 ms of lag (1 frame). It takes about that much time for a black pixel to turn white, so a lot of the Lag reported on the counter based tests are just because of contamination between two different frames.
LCDs refresh from top to bottom, so the solid white bars look like a gradient as they turn on.
Reply
Reply
post #281 of 4031
2/26/10 at 12:17pm
Interesting thread here. I own the 37lh20 with the IPS panel.
Reply
Reply
post #282 of 4031
2/26/10 at 2:14pm
When doing tests I would recommend testing your video card to make sure the ports output synchronously. On one of my machines, with an 8800GTS, one port lags behind the other by about one frame.

Just rerun the test and switch which monitor goes to each video port.
Reply
Reply
post #283 of 4031
2/26/10 at 2:27pm
Quote:
Originally Posted by jensend View Post

When doing tests I would recommend testing your video card to make sure the ports output synchronously. On one of my machines, with an 8800GTS, one port lags behind the other by about one frame.

Just rerun the test and switch which monitor goes to each video port.

My two available options for testing are:

- A laptop with DVI output (MacBook Pro with 8600GT)

If I compare first against a CRT, then against the TV in question, any timing difference between the laptop LCD and DVI output should be factored out. This does have the cumulative margin of error of two tests, though.

- Desktop with a GeForce3 video card

This card doesn't do dual display; the DVI and VGA connectors are just physical interfaces that always output exactly the same thing. (I may be able to connect an oscilloscope to verify that they are in exact sync, but I don't see why they wouldn't be -- it would take a more complicated design with no real benefit to make them different.) I'm not sure yet if it is fast enough for the tests, but that should be pretty easy to spot as hiccups in the flashing frames.

I'll probably try both, actually. I can also drag home a dual-DVI PC if the tests are inconclusive. More datapoints are a good thing, I think...
Reply
Reply
post #284 of 4031
2/26/10 at 3:33pm
Quote:
Originally Posted by wondras View Post

Looks like my LG 37LH20-UA does in fact have an S-IPS panel...



It's a shame they don't still use IPS on the LH30 series. It was a great gaming set.

I did some more in-depth lag tests on the LH20, and it seems to be somewhere between the Panasonic S1 and X1 IPS-Alpha LCDs and the Toshiba XV645U.

I've got a few more days to decide if I want to keep it or order a Toshiba instead. Not sure what to do yet...


Thanks for doing the work for me. AWSOME!!!!!!
Reply
Reply
post #285 of 4031
2/26/10 at 3:38pm
Thread Starter 
Quote:
Originally Posted by jensend View Post

I made a lag tester that is easier to use and more accurate than the number based timers:

http://lcdlagtester.googlepages.com/

Someone else independently made a very similar tool:

http://www.hardforum.com/showthread.php?t=1423433


Each bar represents 16.7 ms of lag (1 frame). It takes about that much time for a black pixel to turn white, so a lot of the Lag reported on the counter based tests are just because of contamination between two different frames.
LCDs refresh from top to bottom, so the solid white bars look like a gradient as they turn on.

Cool as soon as it get my POS camera to work I'll play around with your timer, its definitely the best way I've seen to test for input lag

yeah you are also right about testing the outputs on your video card for delay
Reply
Reply
post #286 of 4031
2/26/10 at 5:30pm
Just got my HDMI 1.3b cables from Amazon, cheap but work perfect only 6ft long, connected 360 to TC-L32S1.

Compared to my XL2370 which has no Input Lag, and been tested by Pros as such.

http://www.pureoverclock.com/review.php?id=841&page=9

Now that my 360 is connected via HDMI to my 360 I can detect no Lag and no difference between the two. I don't do fancy Photo tests, If I can land my 4 hit combo with Ken, in SSF4 HD Remix every time, thats as good as I can want.

I have tried various, other HDTV's and the L32S1 is great for a few reasons, no motion problems or blur, no input lag that bothers me or affects me, I don't think its even 1 frame behind, when connected via HDMI.

And best of all Thanks to Amazon, and Amazon Prime for a great HDMI 1.3b cable and 2 day shipping.

I think that the Panasonic TC-L32S1 also has HDMi 1.3 ports, and I can set my 360 to xvYC Deep Color when HDMI is hooked up.

I tried some DVD's and the 360 Upscaled, them to 1080p and they looked great upscaled.
Reply
Reply
post #287 of 4031
2/26/10 at 6:10pm
Thread Starter 
yeah street fighter is a good real world test for input lag if your an experienced player of those games and used to playing them on CRT displays etc.

a too also believe that the S1's are as input lag free as you can expect from an LCD, it doesn't get any better than this
Reply
Reply
post #288 of 4031
2/26/10 at 6:24pm
I made my lag tester quite a long time ago, and it didn't seem to be working correctly when I tested it today. I recompiled it with the newest version of Processing. It is now working on 3 Windows PCs I tested it with.

Don't use it if you are sensitive to flashing lights.

http://LCDLagTester.googlepages.com/

These were taken with a compact digicam. I have a DSLR, but it isn't really needed for the test.







My monitors seem to have a slightly different refesh rate, as the monitor on the left has a tear that slowly moves down the screen. It still covers exactly the same amount of area per column, though.

Reply
Reply
post #289 of 4031
2/26/10 at 6:48pm
Thread Starter 
camera shutter speed matters a lot though in the accuracy of the results, if the shutter stays open for too long the stripes just like the numbers will smear on the CCD of the camera

I cannot get my crappy kodak camera to work with the batteries i have right now so just for an example of what a crappy camera can do i took some pictures with my cell phone (3.2mp G1)

this was one of the worst ones but they were all useless

Reply
Reply
post #290 of 4031
2/26/10 at 6:59pm
Ok, because I'm very interested to know what my LH90 really has for input lag, I decided to connect an old AOpen 17" LCD monitor to my HTPC. I downloaded the stopwatch test and currently have my HTPC in clone mode with my LH90 connected via HDMI and my AOpen LCD connected via VGA, both running at 1024x768 (does it matter what resolution?).

edit: nm, will post the results in a sec.
Reply
Reply
post #291 of 4031
2/26/10 at 7:03pm
Thread Starter 
Quote:
Originally Posted by rahzel View Post

Ok, because I'm very interested to know what my LH90 really has for input lag, I decided to connect an old AOpen 17" LCD monitor to my HTPC. I downloaded the stopwatch test and currently have my HTPC in clone mode with my LH90 connected via HDMI and my AOpen LCD connected via VGA, both running at 1024x768 (does it matter what resolution?).

Anyway, I ran the test and I must be doing something wrong. According to my pictures, my LH90 is reading numbers lower than my monitor. Does my TV have lower input lag than this old LCD, or am I doing something wrong?

yes the resolution matters

you want to do 1920 x 1080 if you can in clone mode, if your PC monitor will not allow it due to its max resolution being lower then go to the highest you can and hopefully your TV will say its getting a 1080p signal (you video card will be upscaling the image to 1080p then)

with timer tests the monitor displaying the lower number is behind the one displaying the higher numbers. the difference between the numbers is how much lag the display that has the lower number has roughly

if both displays show the same numbers then they either have no lag or have the same amount of lag (unlikely with a TV vs a PC monitor)

also lastly if your PC monitor is an LCD set the shutter speed on your camera to the highest possible 1/1000 is perfect if it can go that high
Reply
Reply
post #292 of 4031
2/26/10 at 7:07pm
Frito: here is one with 40 lines instead of 10:
http://lcdlagtester.googlepages.com/lagtest40.zip
Reply
Reply
post #293 of 4031
2/26/10 at 7:15pm
Thread Starter 
Quote:
Originally Posted by jensend View Post

Frito: here is one with 40 lines instead of 10:
http://lcdlagtester.googlepages.com/lagtest40.zip

still get massive blur on my cell phone, its not the programs fault its a crappy cell phone camera

you last test program though the 40 row will actually work to visually see input lag if you have the control monitor directly to the right of the TV

if the scrolling bar smoothly transitions into the next display you have pretty much zero input lag or close at least

i'll have to see if i can get my hands on another camera or some new batteries for this kodak, the kodak is kinda broken. it only works with fresh alkaline batteries and lasts long enough to take a dozen pictures or so before it stops taking pictures
Reply
Reply
post #294 of 4031
2/26/10 at 7:23pm
Ok, first of all, the monitor I'm using is a very old AOpen F2705 17" monitor which I believe is a TN panel. The highest resolution that both monitors support is 1280x1024. My LH90 is set to ISF/Expert 1 with Local Dimming ON. You can't adjust the shutter speed on my SD780IS, but I set the ISO to 1600 (I heard you should set this as high as possible). A lower ISO setting made readings unreadable. I took 3 pics that were the easiest to read.




So I got 31-32ms (edit: actually, the first shot looks like it would be 27ms). Local Dimming being ON might have caused a tad bit more lag.
Reply
Reply
post #295 of 4031
2/26/10 at 7:43pm
Thread Starter 
not bad, yeah shut off local dimming and retest.

also turn on game mode if you have one or if there is a picture mode called game or PC set it to that.

you want to disable as much pre-processing your TV does as possible to get the best results
Reply
Reply
post #296 of 4031
2/26/10 at 7:46pm
Thread Starter 
ok so i got my hands on a low end nikon that unfortunately you cannot control the shutter speed on.

i could only manage to get it to auto detect a shutter speed of 1/30, its not ideal but heres one of the better shots i pulled off.



shutter speed definitely effects how good this test comes out and any other test because when this same nikon auto set shutter speeds of 1/15 or 1/8 i saw blurred photos much like the one i posted from my cell phone camera

this photo shows that my TV is about 1 frame behind the TN LCD monitor, its likely behind 1 frame because my PC monitor is only capable of 1680 x 1050 so my video card is scaling the output to my Panasonic to 1080p resulting in lag.

i'm off to my buddy's house to help him setup his new plasma he just bought and he has a 1920 x 1200 PC monitor i may barrow from him if it will display 1920 x 1080 without scaling the image
Reply
Reply
post #297 of 4031
2/26/10 at 7:51pm
Quote:
Originally Posted by frito View Post

not bad, yeah shut off local dimming and retest.

also turn on game mode if you have one or if there is a picture mode called game or PC set it to that.

you want to disable as much pre-processing your TV does as possible to get the best results

The LH90 does have a game mode, but I already moved the monitor away and I'm too lazy to move it back right now. Plus Canada is playing Slovakia in Olympic hockey. :P

I wanted to test the settings I would use when I play games. The AOpen F2705 is so old (it doesn't even have DVI!) I doubt it has much (if any) input lag. So yeah, 31-32ms of lag with local dimming on and not using game mode is quite good IMO. A noticeable improvement over my old A650.
Reply
Reply
post #298 of 4031
2/26/10 at 8:30pm
I made a new version with white lines between every testing line, it should make it much easier for auto-exposure.

http://lcdlagtester.googlepages.com/lagtester3.zip
Reply
Reply
post #299 of 4031
2/26/10 at 8:52pm
I hooked up a HTIB tonight, now I'm going to hook my 360 up to the component in of the HTIB, then hook the HTIB to the TV with HDMI.

Will this introduce lag?

I'm not sure the TV will pass through 5.1 audio, so this may be the only way to do it.
Reply
Reply
post #300 of 4031
2/27/10 at 6:22am
would this be useful ?
http://www.online-stopwatch.com/full-screen-stopwatch/
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: LCD Flat Panel Displays

Gear mentioned in this thread:

AVS › AVS Forum › Display Devices › LCD Flat Panel Displays › Input lag wars!, post your input lag results of your LCD display here for reference