AVS Forum banner

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

1M views 4K replies 818 participants last post by  Foxbat121 
#1 ·
Input Lag


What is Input lag?

Input lag is a delay from the moment your output device (PC, Game console, DVD, etc.) sends an image to your TV to when it is finally displayed on your screen to see.

more info can be found here http://en.wikipedia.org/wiki/Input_lag


It is generally accepted that ALL display devices that are digitally based, LCD, Plasma, DLP, LCD projectors, anything thats not CRT based Have input lag, not to be confused with pixel lag which manufactures often advertise.

ONLY CRT displays due to their purely analog nature are Lag free analog signal goes in analog signal is displayed.


Digital Displays must convert any NON-digital signal (everything except HDMI and DVI are analog) into a digital signal to be processed and displayed, ALL LCD's even PC LCD's suffer from this delay. PC LCD's that are manufactured for gaming and general use primarily tend to be the fastest, offering near CRT speed of Display. many however can lag just as much if not even more than some HDTV LCD's. This is why using a CRT for these test is so important to insure that your baseline display is truly lag free. Using another LCD even a fast one can make a TV's input lag results appear to be LOWER (faster response) than it Really is. how much lower will depend on the LCD monitors lag (the fastest LCD's have been tested at 2-8 milliseconds with the average LCD monitor around 16ms witch is a whole frame delay, many LCD monitors can have lag in excess of 30ms (2+ frames delayed)

that being said, most LCD PC monitor have low input lag because they are cheap and cheap means they are TN based (Twisted Nematic) most TN monitors do not use RTC or Overdrive and if they do they are not a good display to use as a control for these tests unless you can disable this feature in its menus (most newer PC monitors will allow this so check for it if your monitor has Overdrive mode)

Does this lag Matter to me?


Input lag matters to many HDTV users, Mainly Gamers that like to play fast twitch FPS games, Fighting games (street fighter style), Music games (guitar hero/rock band) any game where timing is important. It can also effect Normal TV users that never play games and watch a lot of movies on their TV, depending on your setup. If you route all sound into your TV and do not have a surround receiver you are unlikely to see lip sync problems however, if you do use a Surround Receiver and it does not have HDMI or you are not using HDMI into the Receiver you may notice that the lip sync on your BD/DVD's is every so slightly off due to the sound output from your DVD player being plugged directly into the Receiver and is being played back when it is supposed to be played back. If you do see that happen in this type of situation it is likely you HDTV is DELAYING the video signal so much that you can notice it visually even in a movie


Generally most People cannot see or feel a difference of less than 30ms roughly 2 frames, some say they can but they are the rare few and likely younger people


3 frames delayed seems to be a Grey area where some people can notice it and others cannot, if you can notice this much delay you can often adapt to it and become used to it but it can very much still impact your gaming performance in games played online most notably FPS games due to the need for leading moving targets in the game etc.


4+ frames delayed is too much delay and should be readily noticeable to just about anyone if they took the time to notice it and is very detrimental to game play


For a visual example of how this effects your game play in FPS games here is a video of a guy playing counterstrike on a Sony EX5 (EX500 in the US)



The purpose of this thread is to Test your HDTV for Input lag VS a CRT display or a fast LCD monitor that has been tested by someone or yourself google for reviews and report back the results with pictures as proof of the results

These tests can never be more accurate than 1 Frame (16ms) and will never be completely accurate as there are many variables when doing these tests including but not limited to, your camera and its settings, your video card and its drivers/settings. LCD blur getting in the way of Clear readings etc. Due to this I will be posting summarized results in this Post of each TV tested and the Measurements will be in Frame Delay with one frame equaling 16ms.


How do I test my TV for input lag?

First you need a PC with a video card that has dual outputs


hook up the TV and CRT monitor to your PC's outputs


Set your video card to output in "Clone mode" so the same image is displayed on both screens at the same time


Get a Digital camera with a high ISO setting (ISO 800 is often the best setting) If it has a shutter delay option set it to 1/60 so they CRT and LCD will both be clear to see in every picture, if this is not set the CRT will have a rolling Horizontal Dark line as you can see in many pictures I've messed up for you below


if your testing with both displays being LCD you can rasie the shutter speed as high as your camera will allow to get clearer and more consitant results


last step is to go to the webpage in the link below in Rule #1 and scroll down till you find the Input lag timer and once you got everything setup the way you want it and camera ready to go, hit start and start snapping photos



post the pictures here with a summery of results if you have time to figure that out (frame delay is more important than figuring MS differences)

rules are


#1

you must post Pictures of your results using the input lag timer here are a few free ones
Lagom.nl
FlatpanelsHD

I also have a more accurate timer called SMTT, i had to pay for it but they allow you to freely distribute it to people just not on a website so if you would like a copy PM me. I currently have copies of Win7/Vista x64 and WinXP 32bit that i can give out

if you wish to Purchase a copy of a different version of SMTT you can do so at their website, if you do buy another version of SMTT please contact me if you wish to give me a copy so i can distribute it to future people who wish to test their displays and are not running Vista/Win7 x64, the cost of the program is 5 Euro's
http://smtt.thomasthiemann.com/index_en.html


SMTT will provide you will much more accurate results because it uses DirectX and unlocks the V-Sync of your graphics card's outputs, allthough the timer's down the side of the screen appear to not be in sync they in fact are. all displays scan from top to bottom and as the update happens on your display the timer keeps on advancing and that is the reason for the added accuracy of SMTT provided that your PC has enough CPU power to exceed 1000 FPS for the given resolution your at (my Core 2 Duo E4500 that is oc'd to 3.4ghz clocks around 2000 fps at 1920 x 1080)


when using this program you always want to take the highest visible number on both displays to use to draw your input lag result and if you do this you will end up with very consistent results. its extremely accurate


#2

tell us what model TV(s) involved in the test and what their relevant settings are for each test (game mode, PC mode etc..)



#3

it is preferred that you have a CRT display as your baseline for the comparison but if you do not a fast PC LCD is ok or another TV that has been tested before by someone vs a CRT is ok BUT, please state that you did not use a CRT and what the average lag is of the display is that you used from another source tested vs CRT if possible, if not thats ok, but results should be taken as the difference between the 2 LCD's that have been tested in your photos and not as an accurate reading of the input lag of either display obviously



#4

Make sure you also test your video cards outputs for delay by switching the display connections (ex, TV connected to #1 input Monitor on #2, take a handful of shots, then switch the monitor to #1 and the TV to #2 ) The reason for this is so far at least one user using a Geforce 260 has discovered that there is consistently a 1 frame delay from his primary connector to his secondary connector on his video card, all others so far have not experienced this including other Nvidia cards and ATI cards but it should always be checked so we can insure more accurate results


#5 to obtain the best possible results you need to have a control display that is capable of displaying your test subject's (TV) native resolution, for 1080p LCD TV's this is 1920 x 1080. CRT monitor rarely are able to display this resolution but many of them can if you force the resolution on them but you may/can damage your monitor if you do this so be aware. 1920 x 1080 native TN LCD monitors are ideal for these tests and many of them exist these days, if your monitor is lower resolution like 1680 x 1050 (very common for 22" TN monitors) this can work fairly good still at least with Nvidia card's because what happens is when you set clone mode for 1680 x 1050 Nvidia's by default recognize that your TV will not accept that resolution and automatically upscale it to 1920 x 1080. this does result in some inaccuracy of results but in my own tests the Nvidia scalier in my GF 8800 GT is very fast and only adds around 5ms of lag by doing this (i also have a CRT that will display 1920 x 1080 when its forced to do it)


Panel type and how much they lag typically


IPS = low input lag 0-16 ms(IPS-Pro made by IPS-Alpha (panasonics) S-IPS (made by LG LCD display)


xVA = 30+ ms input lag often in the 45-60ms range in game modes for most TV's that use this panel type(S-PVA, MVA type panels found in sony, samsung and many major brand TV's sold today. they are manufactured for the most part by S-LCD (sony/samsung owned company) AU Optronics and Chi Mei


TN = low input lag but not found in 95% of LCD TV's due to poor picture quality and not made in large sizes etc.

120hz sets often increase input lag. this does not necessarily mean that a 120hz set will always have high lag, there have been more than one 120hz sets that have very low input lag but it all depends on how they designed the TV to work. some models like the Samsung A650's and the Toshiba xv645u's actually disable 120hz completely when in a special mode made for PC's/gaming and then and only then they have low input lag.


the reason for VA's wide usage is simple. its cheaper to manufacture in large sizes and offers better static contrast ratios over IPS panels making for the best picture in LCD's. The problem is VA LCD panel type by its nature has a very poor pixel response speed in excess of 16ms natively with many G2G transitions so in order to fix this major flaw these panels must employ RTC short for Response Time Compensation or Pixel Overdrive as it was called by many LCD monitor manufactures when it first came out. The problem for gamers with RTC is in order for it to work the panel's TCON must buffer a few frames to analyze them for changes then tell the pixels to change state faster via over voltage the control transistors for the pixels by a certian amount depending on how much the change was


That being said there are VA displays that can produce good lag numbers at or under 30ms, samsung A series and toshiba 40xv645u are examples of this as well as sharp TV's with Vyper Drive (these sharp TV's use ASV panels and are a variation of VA tech) from what i've seen though these displays achieve low input lag by reducing how much RTC is used in the signal processing and it can create more pixel blur than is normal esp with dark images like moving a dark gray box over a black background will result in visible smear and ghosting yet more common transitions are still fast. this is a trade off us gamers will have to live with if we want TV's that perform good for gaming but can also produce great TV/movie picture quality


IPS panels naturally have a good response time and do not require the use of RTC to bring their response time down to the level that a person cannot see pixel smearing but they cannot even begin to compare to VA panels with black levels without resorting to backlight dimming tricks (panasonic calls it A.I. Picture) that being said it is not to be assumed that an IPS does not use RTC and is lag free, only Panasonic's with IPS-Pro panels (many of their 32/37 720/1080p panel's in the past few years have had these panels) have next to no input lag consistently, LG IPS TV's can vary from fast to ok or worse possibly and the 120hz+ models will likely be laggy unless they have a special mode that defeats the 120hz mode


here is some sample photos from LG Display comparing pixel response time of IPS and VA




other factors can be involved in input lag though like the speed of a certian TV's internal scalier when its given a non-native resolution as well as video processing circuitry like noise reduction and most of all 120hz/motion enhancers though all of these things can be overcome by disabling them in most all TV's so this is often not the real problem for most gamers unless you plan on using older consoles/Wii's with your TV (720p scaling can be a concern for PS3 owners as well)


An easy method to see if an LCD is VA is to press on the screen when a black/dark image is on screen see photo below also from LG Display



Test Results section


here i will divide the tested TV's into 3 sections as follows

GREEN TV's have no more than 2 frames delayed when in a certain mode that reduces lag

BLUE TV'S will be in the middle section, what i call the gray area, the TV's that end up in between 2 and 3 frames delayed and casual gamers will have no issues with most of these TV's

RED TV'S are for TV's with more than 3 frames delayed, some of the faster ones maybe ok to some gamers but overall these TV's will have very noticeable input lag to the majority of gamers out there today unless they only play games that are not sensitive to input lag but even then they just don't realize its there, ignorance is bliss as they say



for results on Plasma TV's please visit Orta's Plasma Input lag thread

BEST gaming TV's tested so far


Samsung A650's '08 40" and bigger ONLY
(results from 650 gamers lag thread ) S-PVA panel is likely for A series Samsungs, these are the only VA's that have achieved low input lag due to a special PC mode that hurts Picture Quality dramatically by owners accounts (it may very well be turning off RTC and is much of the cause of the PQ hit)

VGA mode / 1080p.


Average lag: 0-1 frame delay

VGA mode / any PC resolution accepted by the 650. 480P and 720P will not work in VGA.


Average lag: 0-1 frame delay

HDMI2 PC mode / 1080p.


Average lag: 0-1 frame delay

HDMI GAME mode / 1080p.


Average lag: 1-2 frame delay

HDMI Movie Mode / 1080p / No image enhancers activated.


Average lag: 3-4 frame delay

HDMI Standard Mode / 1080p / No image enhancers activated.


Average Lag: 3-4 frame delay

Component Standard Mode / 1080p / No image enhancers activated.


Average Lag: 4-6 frame delay

Component GAME mode / 480p.


Average lag: 1-2 frame delay

Notes: results are taken from Samsung 650 Input lag thread ; tests done on these displays in this thread have not been vs CRT's but PC LCD's some of the PC LCD's have been tested vs CRT's and have been shown to be near 0 lag say in the 2-8ms lag range, however most of the tests done have been on an untested Dell LCD monitor that I have yet to find any input lag info on, but appears to be a fast TN panel. Actual Input lag may be 10ms or so higher than they are shown here due to this ; the absolute fastest response on these displays are ONLY on HDMI 2 Labeled as PC and VGA input (any res. supported 1080P & normal PC resolutions NO 720P/480P on VGA) Go read that thread for more detailed results and info on these displays Including what modes disable what picture settings etc. the fastest modes disable many picture controls but not all of them

Panasonic 32" LZ800 HDMI 1080P IPS-Pro panel

1080P Standard mode


0-1 frame delay

1080P Game mode


0-1 frame delay

480i Game mode

1-2 frame delay (closer to 1 than 2 but definitely a bit more lag than 1080p)

notes: this is one the very fastest LCD TV's ever tested, would love to see what panasonic's current TV's test at, it should be around the same as these models. Tested by Mr Deap

LINK TO PHOTOS

Panasonic TC-L32S1 IPS-Pro panel

1080P HDMI


0-1 frame delay in game mode or out does not matter

notes: was tested vs a CRT monitor with an average of 14ms delay of picture with many displaying 0 ms delay, one of the best gaming TV's today for sure, the smooth motion feature appears to just be a scanning backlight and does not impact input lag but appears to show some visible flicker. Tested by hidefLoans

LINK TO PHOTOS

Panasonic TC-L37S1 IPS-Pro panel

1080P HDMI


0-1 frame delay just like its little brother
720p HDMI vs CRT


1 frame delay (13ms) very good result for a non native resolution

notes: this is my new main PC monitor, text is not perfect 100% of the time. these Panny TV's have a strange problem with displaying red and blue text ONLY on dark/black backgrounds, text otherwise is very good once you get the picture settings set right and running cleartype tuner in windows helps some more as well. optimal Sharpness on these Panasonic's is 0-10 at most, the sharpness setting adds edge enhancement at anything over 0. Tested by Myself

LINK TO PHOTOS

Panasonic 32" TC-L32X1 (720p) IPS-Pro panel

720p HDMI vs LCD monitor tested to have 2ms lag


0 input lag

notes: this is a very well done test using a good camera and can show how consistent your results can get once you have a good camera setup correctly for these tests. very good job Saiyanjin2. If you take a look at the you tube video in his post and you pause it you can see that the rotating frame ticker to the right of the counter is actually AHEAD of the TN pc monitor indicating that the panasonic is slightly faster than the TN in pixel response time! Tested by Saiyanjin2

LINK TO PHOTOS


Here's a youtube video of his tests




Toshiba 40xf550u likely to be a VA panel just like my Toshiba but is unknown

Game mode on via HDMI


1-2 Frame Delay

Game mode OFF via HDMI


3-4 Frame Delay

Notes: this set does let you make pq adjustments in game mode over hdmi/component. Results are vs a CRT display TV was not at native resolution but was not upscaling the picture so it does not matter as the scalier was not active. Tested by kumquatsrus

LINK TO PHOTOS

Sharp LC52E77U Sharp's ASV Panel a variation of VA tech only found in Sharp TV's

480i GAME MODE ON

AV Mode: Game

Fine Motion Enhanced: On

Active Contrast: On

Lag: 2 frames

1080p GAME MODE ON

AV Mode: Game

Fine Motion Enhanced: On

Active Contrast: On

Lag: less than 1 frame

results taken from this webpage , photos of the test are there as well ad more tests out of game mode and it appears he used a laptop LCD as his control but that should be very fast as they are normally TN panels, this model Sharp does not have their "Vyper Drive" game mode like some other model's so its probably safe to assume that the models with Vyper Drive will perform as good and maybe better with lower resolutions but i've yet to come across evidence that shows this so be aware


Panasonic 50X1 Plasma

1280 X 720


2 frames delayed

notes: TV is 1366 x 768 native, it will display this resolution over HDMI but unfortunately my CRT monitor would not display it so i had to use the lower resolution

35ms input lag, actual input lag could be lower at the panel's native resolution but as most 720p video games will be outputting 1280 x 720 it does not really matter

as usual with Panasonic TV's they do not reduce lag in game mode, it is simply another picture mode on the TV, test was done VS a CRT and performed by Myself


LINK TO PHOTO

NEC 3215 (32" LCD commercial display)pretty likely to be an IPS panel based on performance and its 178 degree rated viewing angles

1440 x 900 (native res is 1366x768)


1 frame delay

notes: tested vs an LCD monitor model Samsung 942B, This is not a mainstream TV set but rather a Commercial display/TV its 720p (fine for 32" unless you plan to use as a normal PC monitor) and it has no speakers and only a few inputs (VGA and DVI input i believe) tested by Zenith DTT900

LINK TO PHOTOS

Vizio 32L10A 720p LCD TV

720p HDMI 1 3 frames delayed

720p HDMI 2 & 3 2 frames delayed

notes: tested vs a CRT. HDMI 1 apparently has 1 frame more lag than 2 and 3, never seen this happen before on a TV but when your dealing with lower end brands you never know what kind of odd stuff comes up. tested by bahnzo

LINK TO PHOTOS

LG 37LH20 720p IPS panel

720p 2 frames delayed (25ms roughly)

480i 3 frames delayed (48ms)

notes: tested vs CRT's. lots more information on this TV is availible in the excelent report posted by wondras the link is below this line.

LINK TO PHOTOS

The following models are European TV's tested by an Italian review site


Lag numbers are listed in Standard mode then Game mode


Remember 1 frame= 16.6ms and they performed these tests on a laptop with an inaccurate timer program so +10ms input lag should be added for a more realistic number on these results



Blue area TV's may be ok for some people but not good enough for the most demanding gamers

Panasonic TC-L37U22 LG IPS panel

1080p 2-1/2 frames delayed (42ms)

notes: tested vs a CRT by igomadness. iamdanwilliamson has also tested this model vs an LCD and came up with 32ms, further evidence that Tests done with PC LCD monitors need 10ms added to them to be close to accurate

LINK TO PHOTOS

LG LH90 LGD IPS panel more than likely user did not state the size of his TV but it should not matter

1280x1024 not scaling the image


2 frames delayed

tested vs an LCD monitor so lag is likely to be 1/2-1 frame more than tested. his picture settings in the test were "ISF/Expert 1 with Local Dimming ON" , this TV should perform at least 1 frame better if he put it into game mode i would guess but he does not want to find out for us
thanks for the test result either way though rahzel


LINK TO PHOTOS


Samsung 55C7000 S panel (SQ01) S-LCD S-PVA panel

1080p HDMI/DVI 1 labeled PC 9 frames delayed 156ms input lag!

1080p HDMI/DVI 1 labeled PC LED motion plus ON! (not AMP) 2 frames delayed! (31ms)

notes: does show potential for having very good input lag with the correct settings at 1080p, input lag is likely to be worse, possibly much worse if not in 1080p or not in the special samsung PC mode that only works on ONE HDMI input so be warned. for this reason and the lack of using a CRT for the test i'm sticking it in the blue section tested by DarkGSR

LINK TO PHOTOS


Sony 32" XBR6 (205r fw) 1080p HDMI S-LCD made S-PVA panel confirmed by myself (its my TV!) i took the back cover off to check it

Photo mode 1080p


2-3 frames delay

Video mode W/game mode 1080p


2-3 frames delay

Video mode W/O game mode 1080p


2-3 frames delay

VGA @ 1280 X 1024 (TV is not scaling the picture)


2-3 frames delay

720p via Component Game mode on


2-3 frames delay

480i Svideo game mode on


2-3 frames delay


Note: The Sony's WILL allow changing of more picture controls in game mode, only setting Photo mode will disable cinemotion/motionflow and noise reduction settings, VGA mode is the most limited input on the display but you can adjust backlight, brightness and contrast in VGA, this TV was tested vs a CRT monitor. Tested by Myself

LINK TO PHOTOS

Samsung 46" A550 720P HDMI VA panel of some sort it could be a S-PVA or a MVA due to samsungs use of 2 different brand panels in their TV's

Normal mode 720p


2-3 frame delay

Game mode 720p


2-3 frame delay

Normal-labeled as PC 720p


2-3 frame delay

Notes: display is 1080p native, user couldn't test vs CRT at 1080p due to CRT resolution incompatibility

LINK TO PHOTOS

Samsung 46A550 1080P

1080P (game mode on? does not say in post)


2-3 frame delay

Tested by clicq

LINK TO PHOTOS

Toshiba 40xv645u HDMI 1080p in game mode took the back cover off my set to see what panel it had it is made by samsung (not S-LCD) and is made in china definitely a VA panel

Average lag 1-2 frames delayed (27ms averaged over 11 pictures 34/20 HIGH/LOW results) see notes for reason why its not in the green section

Game mode disables 120hz (duh) most picture controls are active still except NR stuff and Resolution+ and the nice thing is the TV will save separate picture mode settings for the same input making it easy and fast to switch from a movie mode to a game mode for PS3 owners/PC users wanting to have 120hz sometimes Text on this TV when used with a PC is some of the best i've ever seen very sharp and zero noise. In game/PC mode color gradient tests on all 3 primary colors show extreme color banding problems, color accuracy is definitely impaired greatly when in these modes on this TV making it OK for gaming but bad for PC use in these modes if you care about color accuracy furthermore having 120hz turned off this TV exhibits visible pixel smearing with dark images moving over an even darker background. interestingly enough i was only able to see this happen using a test designed to create smearing, in real world use I never saw this happen but if you play any games that are very dark I have no doubt you will see this happen if the set is in game mode/out of 120hz mode and for this reason i'm sticking in into the Gray area rather than the best section even though its input lag performance is good enough to be in the best section Tested by Myself

LINK TO PHOTOS


LG LH30 AUO VA Panel

1080P


2-3 frames delayed (30-50ms lag)

LINK TO PHOTOS

LG used to put their own S-IPS panels in these TV's in models manufactured in the first half of 2009 but sometime in the middle of 2009 they stopped doing that and put AUO VA panels into them and the input lag went up to sony/samsung levels, if your lucky enough to find one with an LGD made IPS panel they have very low input lag like panasonic IPS TV's


here is proof of this from forum member Odysseus18

The following models are European TV's tested by an Italian review site


Lag numbers are listed in Standard mode then Game mode


Remember 1 frame= 16.6ms and they performed these tests on a laptop with an inaccurate timer program so +10ms input lag should be added for a more realistic number on these results

WORST Gaming TV's tested so far

Vizio SV472XVT likely uses an LG 47" IPS panel because thats the only LCD made in that size AFAIK

VGA 1080p - min 54ms (3.25 frames), average 69ms (4 frames).

HDMI 1080p all advanced image processing options OFF - min 61ms (3.5-ish frames), average 67ms (4 frames).

HDMI 1080p all advanced image processing options ON and maxed out - min 59ms (3.5 frames), average 65ms (4 frames).

Composite 480i - I didn't test this with SMTT but using the Wii in this mode was almost unplayable. Horrible input lag. Avoid at all costs (use the component cable instead).

Component 480p - As soon as I switched the Wii to use a component cable and ran in 480p it felt just like the XBox 360 in 1080p (i.e. about 3-4 frames of input lag).

notes: Tested vs a CRT. it appears that even though these Vizio TV's use an LCD panel that has been shown to have low lag Vizio's poor implementation of the mainboard hardware/software is creating non-defeatable input lag because this panel when in LG and probably Toshiba TV's that use them as well have much less input lag and preform good. Tested by ninjapowered

LINK TO PHOTOS

Samsung LN52B630 tested vs a 17inch CRT monitor VA panel of some sort it could be a S-PVA or a MVA due to samsungs use of 2 different brand panels in their TV's

VGA input 1920 x 1080


3-4 frame delay

HDMI1 DVI game mode OFF


6 frames delayed

HDMI1 DVI game mode ON


3-4 frame delay

notes: "I was running my pc with a millisecond clock in clone mode. MSI 9400GT video card with HDMI,VGA,DVI. I was using a Canon A570Is camera. I found I had to use a 125 second exposure (ASA800) to get the picture. With a higher shutter speed (500) the CRT monitor displayed a moving black bar which usually covered the clock.


AMP set to Off,Dynamic Contrast set to Off"


LINK TO PHOTOS

Sony 52EX701 Appears to use Sharp's new X-Gen LCD panel

VGA in at 1024 x 768


5 frames delayed


HDMI in at 1024 x 768


5 frames delayed


owner tested it vs a CRT monitor and for the moment at least only tested it at this low resolution so its part of the cause of the lag but it's still lagging too much to account for simple scalier lag that normally averages at 10-20ms input lag at most, hopefully he will attempt to test at a higher resolution soon so we can see if lower numbers are possible but i doubt it will get into the gamer friendly range at this point. Tested by MetalAlien


here's his youtube video of the test, he used a high speed capable camera for the test


link to the owners posts regarding this test

Samsung 32C450 720p CMO Panel

1360x768 HDMI


3-4 frames delayed (50-60ms lag)

was tested vs a 16" CRT monitor by Odysseus18

LINK TO PHOTOS

Sony EX500 likely to be a VA panel either AUO or S-LCD

not sure what resolution was used 5.5 frames delayed, 94ms input lag


here's a youtube video showing how bad the input lag is on this TV



Insignia NS-32LCD-09 who cares what panel it had it was terrible



Avoid at all costs, I owned this TV briefly before i Knew about input lag and how to test it but I can tell you that when i had it connected via VGA at 720P (its native res) it looked ok picture wise but input lag in windows with mouse movements were EXTREMELY apparent. If i had to guess it lags at least in the 80-100 MS range maybe even higher it was very bad !! not fit for gaming ever

It should pretty much go without saying that you should never buy a cheapo brand TV if you want a TV with low input lag, major players like sony and samsung cannot even get good lag numbers so its pretty unlikely that any cheap off brand or best buy house brand model TV is going to perform good input lag wise, Vizio may be the only exception to this rule but i've heard very mixed results on how they perform input lag wise ranging from good to terrible


The following models are European TV's tested by an Italian review site



Lag numbers are listed in Standard mode then Game mode


Remember 1 frame= 16.6ms and they performed these tests on a laptop with an inaccurate timer program so +10ms input lag should be added for a more realistic number on these results

 
See less See more
4 11
  • Like
Reactions: EmmaDarknet
#27 ·

Quote:
Originally Posted by clicq /forum/post/16098528


The problem with running multiple timers at once is that they're all vying for CPU attention. Think of it this way, all the timers ask the computer to update them every millisecond -- the CPU probably can't, considering it's got 3 timers to pay attention to, plus whatever stuff is running in the background, so maybe it skips one...


I guess what people should do when running these timers is increase the priority level to attempt to make the timers more consistent, though there's only so much you can do
.


I wonder if we need to start listing computer specs with these tests as well
.


Actually, here's an idea for a timer -- why not make a 60 fps video that just has a timer? Sort of like the videos at http://www.mediacollege.com/downloads/video/timecode/ but at 60 fps instead of 30... and you'd just have to make sure the media player isn't dropping any frames.

CPU speed is not an issue, the problem is that were talking about very fast numbers here and this program is written in flash and cannot sync the output to the refresh rate of the display, i found a timer program that is not flash or java based, and is not setup right by default but you can adjust the frame rate down to 60hz (60fps) and you can even tell it to count the sub seconds in 0-60 rather than 0-100 or 0-1000


program can be downloaded here http://www.stopwatch-timer.com/

It's shareware but just has nag screen and no limitations


if we use this program at 0-60 setting then we are talking full frames and the numbers are alot closer. i did some quick tests on my display at my normal PC settings,


alternatively there is the timer on this page that kumquatsrus posted a few posts up http://www.lagom.nl/lcd-test/response_time.php


i'm uploading my pics right now of results with these timers more in a bit
 
#28 ·
here is the first program from http://www.stopwatch-timer.com/


I ran 2 instances of it one set at 1000 sub second mode (milliseconds) and one at 0-60 per frame mode.








and here is the other web based timer program, this one is sort of neat because you can change the number and background colors to get easier to read numbers consistently









my dumb flash was gettin in the way some of the time on these and it was really late last night so i might re-do my test with this timer idk


but anyways both programs vary just like the app we have been using for awhile, give me a few min to figure out the results and ill edit this post with them


ok, here they are, the 2nd set I really need to re-do that timer seems to be the most consistant but i got some real blurry pics + the flash messing it all up so I will re-do them later but here are the results

Stopwatch App

MS F


31 3

74 3

16 1

47 3

16 3

http://www.lagom.nl/lcd-test/respons...ponse_time_gif
MS F

28 2

53 3

42 2 blurry - had to do some guessing

42 3 blurry - had to do some guessing

22 2


1-2 last photo flash obscured the millisecond reading and frame reading showed both numbers on tv lol
 
#29 ·
Sony 40" v4100 VS Samsung 40" A750


Sammy AMP on high, Sony MotionFlow on High, No Dynamic contrast, No black level adjustments, no Cinemotion/Film Mode, All conneted via DVI/HDMI.


Samsung Primary:



















Sony Primary:













 
#30 ·
Setup:

Samsung 46A550 on DVI->HDMI, 15" CRT on VGA. Resolution set to 1080p.

Shutter speed set to 1/500th of a second.


For the following, Game mode refers to not naming the input, but using the "Game" entertainment mode. Normal mode is again not naming the input, and PC mode is the input named as "PC".


Game mode:

(Lagom timer) 3 / 2 / 2 frames

(stopwatch) 37ms / 31ms (couldn't read one image)


Normal mode:

(Lagom) 3 / 2 / 3 frames

(stopwatch) 37 / 47 / 47ms


PC mode:

(Lagom) 3 / 2 / 2

(stopwatch) oops forgot to run it


I don't want to clutter the thread up with pictures (I think it makes it hard to read), plus I'm too lazy to make links, so they're all viewable in this album .


I guess the bottom line is that the 550 has 2-3 frames of lag, which is in the ballpark of my previous tests...


I'm a bit confused about the stopwatch.com timer -- setting the refresh rate to 60 didn't seem to make it sync on refresh... I ended up just using 100Hz. Though that would seem to suggest that it's only accurate to 10ms... so confusing. I'm about to try making a 60 fps timecode video, maybe...
 
#31 ·

Quote:
Originally Posted by KVW /forum/post/16101822


Sony 40" v4100 VS Samsung 40" A750


Sammy AMP on high, Sony MotionFlow on High, No Dynamic contrast, No black level adjustments, no Cinemotion/Film Mode, All conneted via DVI/HDMI.

Just wanted to be clear, but the only difference between these two sets of pictures is which TV was connected to which input on your video card? (in other words, this is something people should probably do as well
)
 
#32 ·

Quote:
Originally Posted by clicq /forum/post/16102113


Just wanted to be clear, but the only difference between these two sets of pictures is which TV was connected to which input on your video card? (in other words, this is something people should probably do as well
)

Yeah it is, Me and KVW have been talking about it on msn


oddly enough I tried the same thing with my video card and saw no difference


considering that his results are much more consistent than mine I'm really thinking this POS kodak camera is a major part of the problem i'm having getting consistent results in my tests any my results vary by more than 1 frame as well so that must be it


Edit: he is using a GForce 260, i'm using GForce 8800 GT so it could also be video card specific but that is unlikely as they are the same main type and use the same drivers as well who knows
 
#34 ·

Quote:
Originally Posted by clicq /forum/post/16102113


Just wanted to be clear, but the only difference between these two sets of pictures is which TV was connected to which input on your video card? (in other words, this is something people should probably do as well
)

Yah, I retract my previous once-said-statement that their's isn't any difference between switching cables (primary vs secondary display). This makes testing even more time consuming.
But the difference is only by roughly half a frame. Still, it's disturbing none the less because I'm kinda tired of doing these tests. I just want to find the one tv set that fits my needs at this point. I didn't even bother opening up the 42pz85u plasma I rented last night. After reading up on the plasma forums last night, it's just too impractical to use as a full time PC monitor. sigh... I might just keep this over-price sammy (a750).
 
#35 ·
I actually just redid the test using a 60fps video I made since I thought video players are already pretty well optimized for timing and they should honor vsync and all that... and I get the same results as the lagom timer. Again, I have a Samsung 46A550, this is testing in 1080p, normal mode. Here's a few examples:







Image 1: 1118 - 1116 = 2

Image 2: 313 - 310 = 3

Image 3: 702 - 699 = 3

Image 4: 1116 - 1114 = 2


These are frame numbers, so each frame is 1/60s = 16.67ms. There looks to be a 2-3 frame delay, which matches up with the previous tests I did. My own feeling is that all the timers have flaws, but the lagom timer is probably the best one to use because you don't get lulled into trying to make your measurements more precise than they can be.


I also tried switching inputs but didn't see any noticeable difference either -- I'm using an ATI Radeon 2600XT.


Here's my own summary on things:

1) any test we do is subject to some variation, probably on the order of a frame (16ms or so), so differences less than that are probably not significant

2) some video cards may treat the outputs differently, so it'd be a good idea to swap inputs if possible


Anyway, time to go back to enjoying my TV instead of endlessly testing it
.
 
#36 ·
haha yeah nice work on the video, I agree that if you are doing it correctly your results should not vary by more than 1 frame (16ms)


it is interesting though that a A550 that does not have 120hz or alot of the extra processing that they a6,7,850's have is slower than them according to the a650 lag thread I would really love to see one tested with the lagom timer vs a CRT if anyone has one that could do it.


i'm also glad to see that the A550 has the same amount of lag as my sony 32xbr6 becuase when i was buying my TV my only other real option was the A550 and i went with the sony because it was cheaper! and had a better picture imo
 
#37 ·
Cross posting KVW's results of a comparison of lag between a samsung A750 and the new B650

these results are ONLY a show of how much slower the new B650 is than the A750 these are not accurate total lag numbers!!

Quote:
Originally Posted by KVW /forum/post/16147007


For some time now I've been in dire search for a display that can be a PC monitor for gaming/web-browsing and still be able to sit back and enjoy a movie in a big screen fashion. So factors that are important to me are:


a) Picture quality (of course this is first right)

b) Motion smoothing (because i'm extra sensitive to low FPS - I actually like SOE)

c) Low input lag (for the online twitch games)

d) 40-42" (once you use a huge screen for a PC, you can't go back)

e) Windows text quality (hard to live with fuzzy text)


Probably in that order although they all weigh heavily in my decision making.


Now after going through several displays (through the help of Fry's/Best Buy return policy), from Samsung, Sony, Mitsubishi, Panasonic, Toshiba and even Hitachi , I've narrowed to two Samsungs. Now mind you, this isn't because they are the best displays out there, not by any means. Just that they seem to be close enough to what I'm looking for and frankly, I'm getting pretty tired of the endless searching, returning and testing.


Now ever since I got the B650, I must admit that I was becoming attached to it, despite its ugly chin design (what were they thinking). Its whites are just a tad whiter than the a750, the blacks a bit more black. There's just a bit more POP to the picture, perhaps because of slightly better contrast. Note the words, 'just a bit more' - we're not taking a huge difference here! Both have the lovely gloss screen, although the b650 does suffer from slight flash lighting from all four corners, it makes up for it with its ability to have a darker overall black. I was also liking the new customizable AMP settings.
I thought for sure the b650 was going to be my keeper. That was until... I did the lag tests.



Yes, that's right folks. I'm sad to report, It looks as though Samsung has made their displays SLOWER this year than last.



Before I show you my screenshots, Let me first explain why I post primary vs. secondary screenshots. It's because, at least in my setup, it makes a difference. The Primary display (in clone mode) throughout all my testing always seems to have nearly a frame advantage (about 16ms). So always take note of that when you see my screenshots! The primary display has a slight advantage in speed. So if the primary is losing to a secondary, then you KNOW it's bad!


Without further ado, here are the numbers (B650 on left, A750 on right on ALL pics!):



Samsung b650 hdmi1 AMP sliders maxed vs Samsung a750 AMP on High: Both no Dynamic contrast or black enhancements - b650 primary









Samsung b650 hdmi1 AMP sliders maxed vs Samsung a750 AMP on High: Both no Dynamic contrast or black enhancements - a750 primary









Samsung b650 Hdmi1 named "DVI PC" vs Samsung a750 HDMI2-PC - b650 primary









Samsung b650 Hdmi1 named "DVI PC" vs Samsung a750 HDMI2-PC - a750 primary









Samsung b650 hdmi1 named "HDMI1-PC" vs Samsung a750 named HDMI2-PC - b650 primary









Samsung b650 hdmi1 "Game Mode" vs Samsung a750 HDMI2-PC - b650 primary









Samsung b650 VGA Mode vs Samsung a750 HDMI2-PC - b650 primary










So it looks as though I'll be returning the b650. I just can't in good conscious choose the b650 over the a750 knowing the PQ is only slightly better but at a cost to input latency in every mode. Perhaps to some of you guys however, it may be worth it. Also, I should note that the b650 would cost me $150 buxs more than the a750.


BTW just so everyone knows, EVERY panel i lag tested these past few months was ALWAYS behind my GF's panny 32lz800 as far as lag, even a samsung in VGA or HDMI2-PC mode. As far as I'm concerned it IS the lag king. Now if only Panny would make them in larger sizes with 120hz.
 
#39 ·

Quote:
Originally Posted by Orta /forum/post/16149983


$1,000,000,000 video processor in action.

yep more video processing = more lag
 
#41 ·

Quote:
Originally Posted by d34dl1n3r /forum/post/16151572


I bet you wouldn't be able to even tell a difference between lag in an A750 vs a B650 if you didn't have numbers. Not to mention that beings the testing methods are flawed... all of this means nothing.

LOL yes yes, it's all nothing! why even try anymore... We should all just accept whatever and let the chips fall where they may right?


Right. Have fun with that...



Edit: Sorry don't mean to come off as a jerk but whether or not there is variance in this measurement, there IS repeatability. And with repeatability, that means it IS still an affective form of measurement. Period. Now whether you can feel the difference of a frame or 2 is entirely up to the individual. Sure it's nitpicky but that's why we're here for right? Obviously, if we come to an AVS forum and actively post, we're all are not the general joe-blow who walks in to a store without any knowledge of what he's buying. We are enthusiast. And minor details, such as input lag, matter to us.
 
#42 ·
Question for all of you guys running these timed tests through clone mode to detect latency: Have any of you actually tried playing an actual game? For example, a twitch based game like Street Fighter IV? Because tonight I turned the game on, went into training mode that displays input and kept switching between gaming mode/PC mode/standard HDMI and I found no tangible difference. As soon as I'd press a key on the controller in any of these modes, the action would be immediately displayed on the screen. I had no trouble executing combo moves in training mode so do you guys think maybe this whole lag input issue is blown out of proportion? Unless all of you on AVS are superstar twitch game players, I really doubt you will notice any input lag while gaming. Just my two cents...
 
#43 ·

Quote:
Originally Posted by 5150Joker /forum/post/16152532


so do you guys think maybe this whole lag input issue is blown out of proportion?
YES!!!

But, I will say this: I think some people (very few) ARE sensitive to 'input lag'. For them it's a legitimate concern.

For the other 99% of gamers, it's not a big deal. It's just something to obsess about on the AVS forums.


Quote:
Unless all of you on AVS are superstar twitch game players, I really doubt you will notice any input lag while gaming. Just my two cents...

Agreed.

I'm an avid gamer. I hooked up my 360 to my Sony 52W4100 and didn't notice ANY lag.

If anyone's that worried, get a CRT.
 
#44 ·

Quote:
Originally Posted by 5150Joker /forum/post/16152532


Question for all of you guys running these timed tests through clone mode to detect latency: Have any of you actually tried playing an actual game? For example, a twitch based game like Street Fighter IV? Because tonight I turned the game on, went into training mode that displays input and kept switching between gaming mode/PC mode/standard HDMI and I found no tangible difference. As soon as I'd press a key on the controller in any of these modes, the action would be immediately displayed on the screen. I had no trouble executing combo moves in training mode so do you guys think maybe this whole lag input issue is blown out of proportion? Unless all of you on AVS are superstar twitch game players, I really doubt you will notice any input lag while gaming. Just my two cents...

It IS more or less is blown out of proportion. For 95% of people out there, perhaps even me, it won't matter. Infact most ppl adapt to it without even knowing it. However for me it's just the principle of it. I just want the least amount of lag, that's all.



The only time it matters in competitive play is in very few scenarios.


Take this for example:

Lets say the avg person has a reaction time of 150ms.

Lets say you are playing against someone in a first person shooter.

Lets say your internet lag is the same.

Lets say you both round a corner and run in to each other and see each other at the exact same time - you both react to pull the trigger first.


Because your competitor is using a CRT and you chose to use an HDTV, your reaction is anywhere from 40-50ms (game mode) to 100+ms (standard mode) behind his reaction because you saw him 40-100+ms LATER than he did, thus he pulled the trigger first, you die first. Big deal? No. But we should strive for better should we not?
 
#45 ·

Quote:
Originally Posted by 5150Joker /forum/post/16152532


Question for all of you guys running these timed tests through clone mode to detect latency: Have any of you actually tried playing an actual game? For example, a twitch based game like Street Fighter IV? Because tonight I turned the game on, went into training mode that displays input and kept switching between gaming mode/PC mode/standard HDMI and I found no tangible difference. As soon as I'd press a key on the controller in any of these modes, the action would be immediately displayed on the screen. I had no trouble executing combo moves in training mode so do you guys think maybe this whole lag input issue is blown out of proportion? Unless all of you on AVS are superstar twitch game players, I really doubt you will notice any input lag while gaming. Just my two cents...

You have to realize that the lag here is really small -- blinking your eye takes on the order of 300ms, so 60ms is really short. Most people don't have reflexes that fast.


To add to what KVW is saying, it becomes a problem only in games where timing really matters. For example, in Dead or Alive 4, you have about a 12 frame window to do a reversal, which is 200ms. This is about the average human reaction time . Now, if you have, say a TV with 100ms of lag, you've just cut down your "effective" time window in half. Of course, you can adapt by starting the reversal earlier.


Is this going to matter to most people? Probably not, but for some people, it's the difference between being able to consistently pull off an attack and only being able to do it some of the time.
 
#46 ·
i agree, input lag is impt. if you are a competitive gamer that plays alot of time sensitive games. i notice a difference when i play games like halo3. minimal or less input lag can make the difference between a missed headshot with a sniper and instant death. could also make a difference in heated br battles. an opponent and i could be "one shot" and if he has less input lag (assuming equal accuracy for both of us), he will live and i will die. online lag comes into play as well. that's why the less lag there is overall, the better it is for a gamer in these kinds of situations. in a perfect world, we would have 0 latency in online games and sd crt/ crt monitor-like response with our hdtvs in regards to input lag. but we don't live in that world...yet, lol, so i for one would like to minimize lag as much as possible; if it is within the realm of my control, of course.
 
#48 ·
btw, i'm gonna go ahead and post the results from my testing last weekend on my toshy. figure it might help broaden the testing results. i'll post some pix with game mode on and off vs a dell crt. ill also summarize results with other comparisons between my zenith sd crt, toshiba laptop, samsung lcd pc monitor, etc. since i did extensive testing overall. so without further ado...


game mode on via hdmi1 on a toshiba 40xf550u

Attachment 138254

Attachment 138255

Attachment 138256

Attachment 138257


btw 16-32ms, so let's say 24ms avg.



 
#49 ·
yep, it really is pretty subjective and depending on what your doing it may or may not matter a whole lot to you.


online games esp FPS prolly get the most impact out of it because the input lag basically adds to your connection lag


I've been playing FPS games sense age 13 or so back when Quake came out, I was playing it online, in the days of modems and when 200 was a good ping on a modem and 300 was the norm.


I can tell you that anything over 100 ping is very noticeable and affects FPS gameplay quite a bit when everyone else is sub 100 ping even on a CRT monitor, so if you have a ping of 50 lets say, and your tv has an input lag of 50 as well, you now effectively have a 100 ping because they stack this is why its important to some people



now yes you can become used to 100ms latency total but the guy with 50ms is going to have an advantage over you in the game
 
#50 ·

Quote:
Originally Posted by clicq /forum/post/16153012


There's actually a good example of the difference input lag could make at the bottom of this page (to save you from reading, the LCD on the left has about 4 frames of lag, while on the right is a CRT).

wow great article pretty much sums it up for those who haven't quite grasped what all the hoopla is about.
 
#51 ·
game mode off

Attachment 138258

Attachment 138259

Attachment 138260

Attachment 138261


so 3-4 frames, avg is 56ms.


so there you have it, game mode does make a big difference on this set, yielding me 2 frames less input lag on avg. trust me, in halo3 and other time sensitive games, that makes a world of difference. especially when it is kill or be killed, lol. one caveat is like kvw, i didn't adjust for primary/ secondary between displays, so the actual difference might actually be a bit more (since i use my toshy as secondary on the vid. card) i won't post more pix, since i have 400-500 worth of testing pix, lol, so that's not gonna happen.


i will summarize comparisons with other equipment i tested. i compared a dell crt monitor with a zenith crt tv: it was a wash most of the times with both screens showing the same exact times most of the time. i compared the dell crt monitor with my toshiba laptop. the laptop would show the same time as the dell crt half the time, with 16ms showing the other times, so maybe 8ms? i then compared my samsung syncmaster lcd pc monitor with the dell crt. i was surprised since this monitor is freakin old, about 8 years old now. it was only behind the crt by 1 frame most of the time! pretty damn good i would say.


so there you go, not bad overall for the toshy 40xf550u. it's as fast or somewhat faster than other sets in game mode, but probably still not as fast as hdmi/pc or vga for a samsung. though the toshiba has a vga input as well that i didn't really test, so i imagine it would be faster. unfortunately, you can't make many picture adjustments in vga mode, and the max res. is 1280x1024, so i wasn't gonna bother using that anyway. otoh, this set does let you make pq adjustments in game mode over hdmi/ component.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top