LCD input Lag problems: new tests of 2405fpw and 244t

docmal

Limp Gawd
Joined
Feb 13, 2006
Messages
313
Background

Recently, I had access to a dell 2405fpw, a samsung 244t a sony G520P CRT and a hitachi CRT. I purchased the 244t because I didn't like the white ghosting I was getting while playing games on the Dell 2405fpw. I thought the samsung would be about the same screen but with a faster rating so it would fix my problem. At this point the input lag on the Dell 2405fpw, once I set the screen to 100hz, was barely noticible if at all.

When I recieved the Samsung I was immediatly unhappy. I did not know why but I found myself not wanting to use my computer at all. Then I realized the input lag was driving me crazy. What bothered me most was normal use like surfing the web, writing code or doing everyday tasks. Gaming was just as bad but since my mind was more on the game I noticed it much less. Finally I decided to find out why I had all of a sudden become so sensitive to this input lag. I decided to do some experiments.

Testing Methodology

I used a program called stopwatch pro 3.1 to produce a large digital clock display, and I cloned my screens using nvidia control panel. I photographed using the sport mode of my Canon Digital Rebel XT (Digital SLR) 8 megapixel camera. This sportmode is one of the fastest on the consumer market taking ~3 pictures per second. The sport mode is engaged by just holding down the shutter button and letting the camera go at it. For each test I took about ~30 pictures, released the button and took ~30 more in order to get two different intervals for a total of ~60 photographs in each test. The reason for so many photographs is that many of the photos result in unreadable numbers because the photo is seeing a number change over, for instance 8->9 looks VERY close to 3->4.

Test Results

I tested with a variety of configurations, I switched inputs, and I switched primary and secondary designations in the nvida control panel. I tested around 10 times producing around 600 photographs. Here are my conclusions: (which by no means should be taken as fact since my methodology was not near scientific enough and the test was done mostly for my own purposes in discovering why I was experiencing these quirks, I share this with you all because I think you should know)

  • CRTs are the fastest and the two I tested produced nearly identical times.
  • The port used on the videocard (7800gt) does not matter much if at all.
  • The primary/secondary designation in the nvida control panel does not matter much if at all.
  • The Dell 2405fpw @ 100hz refresh has an input lag of 25-45ms
  • The Dell 2405fpw @ 60hz refresh has an input lag of 45-50ms
  • The Samsung 244t @ 100hz has an input lag of 45-65ms
  • The Samsung 244t @ 60hz has an input lag of 55-65ms
  • In 6 tests comparing the Samsung to the Dell there are no photographs of the Samsung being faster than the Dell.

When comparing both LCD displays (Samsung and Dell) the differences seemed to follow a wave pattern. The Samsung and Dell would start off very close but then the Samsung would fall behind by about 15ms, then it would fall behind the Dell by 25ms and finally would max at about 35ms behind the Dell. Then the Samsung would fall back to 25 then 15 then back to 0 again. I cannot tell you which monitor's input lag is fluctuating or if they both are fluctuating because I could not figure out how to clone 3 or 4 displays. Without a baseline reading from a CRT we cannot gage which monitor is fluctuating.

Subjective Conclusions

Subjectively speaking I will tell you that after being unhappy with the Samsung LCD for a week and a half and then switching back to the Dell it was clear that I did not notice the input lag on the Dell as much or at all. There are several factors that might explain this. My theory is that the Dell is 10-20ms faster than the Samsung, but also the Dell showed significantly more range being behind the CRT. This leads me to beleive that that the Samsung's lag, although greater, is also more constant and therefore more noticable. Whereas the Dell may frequently fall below a noticible Lag level for periods of time. It seems my tolerance for noticing Lag while moving the mouse sits around 30ms, I am sure this varies from person to person though.

Again, this is my own subjective testing. Take from it what you want. But I can give you my opinions on the two panels in question summed up like this:

Samsung:
  • better picture and color representation
  • blacker blacks
  • more motion blur in video and games (motion blur is the bluring or same color ghost trails that appear on the lcd during fast movement)
  • less white ghosting in games (white ghosting is the light colored ghost that trails objects in a video game)
  • less flicker and tearing
  • Much more noticable input lag

Dell:
  • good picture, color and blacks but samsung is a little better
  • less motion blur in video and games (not much less though)
  • Much more noticable white ghosting
  • More flicker in games
  • Much less noticable input lag

Now, after I send back my Samsung do I buy another Dell, do I wait for 2407 or do I say forget LCD and get the sony fw900 CRT from ebay. Hmmmm maybe I should get both.

Pictures
Samsung 244t vs Dell 2405fpw Pic 1 - 0ms diff
Samsung 244t vs Dell 2405fpw Pic 2 - 10ms diff
Samsung 244t vs Dell 2405fpw Pic 3 - 15ms diff
Samsung 244t vs Dell 2405fpw Pic 4 - 15ms diff
Samsung 244t vs Dell 2405fpw Pic 5 - 25ms diff
Samsung 244t vs Dell 2405fpw Pic 6 - 15ms diff
Samsung 244t vs Dell 2405fpw Pic 7 - 16ms diff
Samsung 244t vs Dell 2405fpw Pic 8 - 0ms diff
Samsung 244t vs Dell 2405fpw Pic 9 - 15ms diff
Samsung 244t vs Dell 2405fpw Pic 10 - 25ms diff
Samsung 244t vs Dell 2405fpw Pic 11 - 20ms diff

Dell 2405fpw vs Sony g520p CRT Pic 1 - ~40ms diff
Dell 2405fpw vs Sony g520p CRT Pic 2 - ~50ms diff
Dell 2405fpw vs Sony g520p CRT Pic 3 - 46ms diff
Dell 2405fpw vs Sony g520p CRT Pic 4 - 45ms diff

Samsung 244t vs Sony g520p CRT Pic 1 - 55ms diff <-- neat cause I cought the timer at 1 minute exactly on the CRT
Samsung 244t vs Sony g520p CRT Pic 2 - 64ms diff
Samsung 244t vs Sony g520p CRT Pic 3 - ~50ms diff

[Edit: for fancy formating so it is easier to read]

[Edit: pictures added]
 
Since it seems you are not broke i'd give the FW900 a shot. I love mine as most on the board have probably noticed. You are already trying numerous LCDs. The FW900 is pretty cheap for what you get and then you can wait for whatever you want or maybe keep a LCD around for work. Kind of thought we'd see some evidence, like some of your pics and stuff. Thanks for all the effort though.
 
Im sorry, I forgot to mention that I will post pics when I get home from work... I'm trying real hard not to do any real work today at the office so I have been trolling the boards.

I will gather some good photos and put them up around 4pm cst.
 
Just wondering what the CRT lag times were, when compared to the Dell or Samsung?
 
I used the CRT as the base since they are known to have delay measured in nanoseconds.
 
I think my hand to mouse movement has a delay of 3 seconds. Can you actually see a delay in the cursor motion across screen, or delay in your clicking, etc..
 
I can physically see the difference on the clock between the LCD and the CRT in real time while they are running.

The most noticable mouse lag is when the sensitivity is low-medium and you change direction or go from stopped to start. You cannot really see it, but I can definintly feel it.

My current subjective test is I start from the windows start button and move all the way to a window close button at the top tight of the screen and click it. I attempt to do this in one swift motion. On the CRT I hit it almost every time. On the LCD when my brain tells me to stop the cursor either floats by the target or I stop short trying to compinsate. It is so bad that I set my $1400 samsung display aside and am using the CRT again. In fact when the samsung was setup I really didn't want to even use my computer at all for a week.
 
fishy1234 said:
Does anyone else experience this kind of lag with LCDs?

Yep, several posts on this already. Good stuff docmal...more hard evidence. I wish we had tests on more LCDs.

From what I've pieced together, it seems to be related to size as well (the 24" monitors are exhibiting this worse). I wish I had the financal resources to prove it. ;)

Navy
 
It looks like I will be getting 2 benq 19 inch LCD displays this week, and ordering either a 2405 or 2407. So I will test the new displays when I recieve them.

I am worried that the new 2407 will have similar results to the samsung, in that case I will want a 2405 and I hope they will still be available.
 
docmal said:
My current subjective test is I start from the windows start button and move all the way to a window close button at the top tight of the screen and click it. I attempt to do this in one swift motion. On the CRT I hit it almost every time. On the LCD when my brain tells me to stop the cursor either floats by the target or I stop short trying to compinsate. It is so bad that I set my $1400 samsung display aside and am using the CRT again. In fact when the samsung was setup I really didn't want to even use my computer at all for a week.

I was just doing this moments ago and felt like just leaving my computer. It's unbelievably frustrating.

I'm using the Acer AL2416W-DS (a twenty-four inch widescreen) on DVI and it's basically ruining my computing experience in every way. A monitor is supposed to monitor what is happening in your computer, hopefully realtime. If I wasn't looking for realtime I would have purchased a $1200 printer. (Yeah, I'm frustrated.)
 
That is a lot of delay. I am assuming when you say input delay, you're talking about when the information is actually "cycled" -> (input delay) -> it gets drawn on the screen?

I used to play FPS games competitively (at a high enough level where $10,000 first prizes were awarded at live competitions).

If you're playing a game online and your ping time is 15-30ms (which is very possible if you have a good connection and the server is somewhat close to you), having an extra 20-30ms delay tacked on due to your LCD not displaying the information fast enough would end up translating to 50-60ms total (ping time delay + input delay).

There's a large difference between 20ms and 60ms when it comes down to FPS gaming. I had no it was this bad in LCDs, I just thought the delay was either there, or not (depending on the model).

My entire outlook on LCDs changed from this post. I was eager to finally make the switch to LCD, but now I am strongly considering sticking with a CRT.
 
This is great post docmal .

Very good stuff to proove what we know about LCD and what others HIDE .
 
IMO this lag/delay is a combination of things:

1) Response time
2) DSP's
3) PVA/S-PVA panel technology

the S-PVA panels seed to have the worst lag most likely because of the massive use of overdrive.
 
Guys,

I'm shopping in the 21-24" range for a general use + gaming monitor. The 2405(7)fpw still seems to be the best size 'bang for the buck' at ~$750. There are two issues holding me back however. One, the subject of this thread is the input lag. Two, is the worry of having to keep up with video hardware to drive 1900x1200 for new games.

Now, my question may just relate to both. How many people run their monitors in a reduced resolution (i.e. 1680x1050)? Does this fare will vs. 1900x1200 in both general desktop usage as well as gaming being that it's not the native resolution? I've heard these panels scale down pretty well and the numbers don't strike me as a huge reduction. Bear in mind I'm coming from a 19" Trinitron running @ 12x9 for desktop and 16x12 for gaming. My video card until next year's upgrade will be an X850 (pro->xtpe), so I know 1900x1200 won't be feasible for very long.

And, assuming the 16x10 resolution still looks good enough, is it possible that having to drive less pixels will result in less input lag? Or has anyone ever investigated if resolution has any impact on the perceived lag. I do play FPS a lot (BF2) so I have a feeling the input delay would drive me nuts if it's always there.

Actually if someone feels like doing a huge favor, could you post pics of the same screen running in the different resolutions to compare? Any other input or advice is appreciated!
 
Do the smaller LCD's have this problem too? I have a 17" and don't notice it, but I'm looking to go with a 20" widescreen in the near future. Has anyone done this sort of test with a 20" panel?
 
The Dell 2405fpw @ 100hz refresh has an input lag of 25-45ms
The Dell 2405fpw @ 60hz refresh has an input lag of 45-50ms

How can this be? LCDs have a fixed refresh rate. Don't they?

As for the clock, I would not trust that windows and the video card are able to redraw the screen with the level of accuracy that you are looking for. I think this gets even worse with the display cloning. I use dual LCDs at home and in the office and if I drag a window in between the two and movie it up and down I can see one display is always just behind the other as the screen is redrawn by windows. The Hz refresh rate has little to do with the redraw rate of the desktop. The main benefit of going above 60 or 70Hz was to get rid of the flicker on CRTs. If the refresh rate is high enough it will be out side of what your eyes can perceive and you won’t see the flicker. That has nothing to do with how fast windows and the video card update or redraw that image with new information.

You need some way of sending the very same signal to both displays. Is the problem only with the DVI inputs? The easiest way I can think of is the split a composite video feed with a Y-adaptor to a TV and the 2405. I think I’ll try that when I get home just for kicks. I have the 2405 and a little TV in my computer room. I would not bet my life on it but my guess is I can snap a picture anytime I want and I’ll see the same frame or image on both displays. In fact I could even run the same clock program from a second PC I have that has composite out. I think it does anyway. :)
 
Tanquen said:
The Dell 2405fpw @ 100hz refresh has an input lag of 25-45ms
The Dell 2405fpw @ 60hz refresh has an input lag of 45-50ms

How can this be? LCDs have a fixed refresh rate. Don't they?

As for the clock, I would not trust that windows and the video card are able to redraw the screen with the level of accuracy that you are looking for. I think this gets even worse with the display cloning. I use dual LCDs at home and in the office and if I drag a window in between the two and movie it up and down I can see one display is always just behind the other as the screen is redrawn by windows. The Hz refresh rate has little to do with the redraw rate of the desktop. The main benefit of going above 60 or 70Hz was to get rid of the flicker on CRTs. If the refresh rate is high enough it will be out side of what your eyes can perceive and you won’t see the flicker. That has nothing to do with how fast windows and the video card update or redraw that image with new information.

You need some way of sending the very same signal to both displays. Is the problem only with the DVI inputs? The easiest way I can think of is the split a composite video feed with a Y-adaptor to a TV and the 2405. I think I’ll try that when I get home just for kicks. I have the 2405 and a little TV in my computer room. I would not bet my life on it but my guess is I can snap a picture anytime I want and I’ll see the same frame or image on both displays. In fact I could even run the same clock program from a second PC I have that has composite out. I think it does anyway. :)

It's not a Clone issue, I tried cloning two CRT's and the timers were identical on both digital camera shots ,as shown here:

Shot 1

Shot 2 - This shot is the most telling, you can see the numbers were in mid transition yet the effect is identical on both monitors.
 
I also did what mathesar has said and found them identical. During my tests I also swapped inputs and reran the tests to find the same results.

LCDs do have a locked refresh but windows does not. There was a post describing why this worked, if someone has the time to find it that would be great. The gist however was that setting windows to 100hz makes more data flow to the monitors DSP which only outputs 60hz. But the fact that having faster speed to the dsp reduces the lag about 15-20%. This has been tested, screenshotted and documented so I did not replicate those tests in my work. It would be great if someone had that link handy.

As for the question about reducing the resolution and getting less lag. Most people don't seem to see a difference here either but some say they can feel it. I did not actually test this however since I have no need to run lower rez.

As for running lower rez and how it looks, a good rule of thumb is that if the resolution is a multiple of the native rez, meening they have the same screen ratio then things will look ok. Its when you are stretching things unevenly like 16:10 - 16:9 that makes it look ugly.
 
I'm so happy! I just got a "go" on returning the monitor and getting in store credit... hopefully I can push a little more for a full refund. *comes crawling back to CRT's*
 
Now that the pics have been added you have a very well done conclusive report that only the most rabid !!!!!!s could deny. Good work.

If i was forced to choose I guess I'd take consistent. I might not motice lag that much anyway ( 40 and 50ms I might not NOTICE but it can't be good for accuracy), but the hope is there that behind the scenes your body is adapting. Not much hope in that if not consistent.
 
Well if it is worth anything, I did not really notice the input lag on the Dell 2405fpw (inconsistant). The 244t (consistant) made me want to cry though.

The inconsistant portion of the theory is very inconclusive. It could be that both of the units are fluctuating in delay or that the dell is and the sammy isnt or vice versa.

I think it mostly comes down to the fact that Dell's average input delay, although present, is significant lower and less noticible than Samsung's average input delay.
 
Could the lag be a driver issue? this guy cleared up his lag problem on a 244t it sounds like.

http://forums.anandtech.com/message...eyword=&STARTPAGE=2&FTVAR_FORUMVIEWTMP=Linear

I received the montior last Saturday and plugged it straightaway to a machine with a low end graphics card. I think it was a 9600 Pro but I am not sure now.
It didn't feel sluggish, mouse was as responsive when computer was plugged to a 17 inch LCD.
I didn't play any games on that monitor so I can't say for definite whether the lcd lag issue was definitely not there on this particular computer.

I later plugged this monitor to a 9700 Pro and straigthaway the mouse felt weird under Windows. It was as if the mouse was always lagging behind movements I made. The impression it felt was that the mouse was running at say 40hz and on top of that there was a noticeable delay. Weird..
I fired up Quake 3 and I again noticed the issue, no way I could use the good old Railgun. Movements were always imprecise, it was as if my graphics card couldn't cope with it when I know for certain it can easily do 1600*1200 so 1920 shouldn't be too difficult.

So....I had a look at these DVI settings on the ATI drivers control panel in the Options tab and I ticked the following boxes :
Reduce DVI frequency on high-resolution displayes
Alternate DVI operational mode

I didn't play with the full set of combinations with these buttons but at the moment it clearly feels much better than the first time I plugged this monitor to the R9700 Pro.
I launched Quake 3 and then it was a completely different game, graphics were smooth and mouse was fairly responsive.

Now I can't tell if it is as responsive as on my previous 22 inch CRT (Mitsubish 2040u) because of the following two reasons :
-I changed my wired mouse to a wireless mouse a while ago and the wireless mouse never felt as good as the old one for gaming
-I never played in 1920*1200 on the CRT but only at 1600*1200 so this might have an impact.


PS : I tried changing the frequency from 60hz (normal DVI mode) to 75 or more but I felt that it never helped to fix the issue.

To conclude this post, I was initially thinking of sending this monitor back because it felt it was completely impractical but now thanks to the new settings I'm definitely going to keep it.
 
I'd like a link to that refresh rate "fix" so I can try it on my 2405.
I tried setting the rate through the nvidia "advanced timing" menu, without any success.

Not that it matters though, since there's no lag on my LCD. ;)
 
fishy1234 said:
Does anyone else experience this kind of lag with LCDs?
I've never once experienced this problem on any LCD I have owned. Personally I think its just nitpicking. I have never owned a 24" though.
 
Those driver settings are not available on NVidia cards. But on my other comps with ATI graphics and LCD screens I notice no difference regardless of those settings.
 
Strange how they can advertise an 8-16ms response rate, yet here's documentation of 50+ms responses.. I know the manufacturers don't have standards as far as testing the response rate, but that is rediculous.

It's also sad that no one is doing anything about this. Has anyone that owns these screens forwarded this thread to them? Why doesn't everyone here with one of these screens get up in arms about it with the manufacturer? That's the only way to change anything..

:confused: :mad:
 
Lord Twilight said:
Strange how they can advertise an 8-16ms response rate, yet here's documentation of 50+ms responses.. I know the manufacturers don't have standards as far as testing the response rate, but that is rediculous.

It's also sad that no one is doing anything about this. Has anyone that owns these screens forwarded this thread to them? Why doesn't everyone here with one of these screens get up in arms about it with the manufacturer? That's the only way to change anything..

They're not talking about response time. They're talking about an input delay. I've been following this thread as well as others like it, and I'm very interested in what's going on. I have and use quite a few lcds at work and home, and have for some time. Frankly, I have never encountered such a phenominon, but that is likely due to the lower res/smaller size of the monitor.

I would agree with some of the posters here that it sounds more like a PC side problem, likely within windows or the video drivers or something else. It is possible, however, that the DSPs are being overwhelmed at these extreme resolutions, but only if using an analog connection. If you're using a DVI connection to the monitor, there is very little processing being done. It's just a simple mapping exercise, and even with raw calculations (ie without any kind of display equation), you're talking about no more than say 60*1920*1200*24 total bits to process per second. Even if you had 20% overhead, that's still only a little more than 4 billion bits per second. Sure, that's a bunch, but a 32bit DSP would only need to be clocked at around 130MHz to handle that. Hell, even at 100% overhead, you're just talking about a 200MHz or so DSP. My PHONE has a processor running at 416MHz, so I don't see a problem here. Add to that the equations that can be used for lossless, realtime compression, and even a very lowend DSP should be able to handle the display functions of any of these displays.

That said, I understand that you're experiencing a problem. As I don't have a large panel handy, some of you might want to try a few experiments. First off, try different video cards or different PCs. Just make sure to change between board manuf. and chipsets. The next one, and I think a very likely problem, is the cable. Go get a high end DVI cable. If you're losing some of the information and it is being forced to resend, that could account for the problem and the irregularity. Try using a very clean windows install and virgin driver installs if possible.

I think the key here is that some people have the problem and some don't. We just need to find a repeatable pattern, so that we can find a solution. If it does end up being that certain monitors are worse than others, then perhaps a monitor from a different vendor, using the same panel, won't have the problem. At the very least, we'd know what to stay away from and what to buy.
 
kamikazichaser said:
That said, I understand that you're experiencing a problem. As I don't have a large panel handy, some of you might want to try a few experiments. First off, try different video cards or different PCs. Just make sure to change between board manuf. and chipsets. The next one, and I think a very likely problem, is the cable. Go get a high end DVI cable. If you're losing some of the information and it is being forced to resend, that could account for the problem and the irregularity. Try using a very clean windows install and virgin driver installs if possible.

I have actually done most of these suggestions. Becasue I have 4 PCs at hand I could easily try several configurations. The only suggestion I did not try was a name brand DVI cable. I used the cables packaged witht he Dell and samsung panels, Although I did interchange them and noticed no change.

kamikazichaser said:
I think the key here is that some people have the problem and some don't. We just need to find a repeatable pattern, so that we can find a solution. If it does end up being that certain monitors are worse than others, then perhaps a monitor from a different vendor, using the same panel, won't have the problem. At the very least, we'd know what to stay away from and what to buy.

I believe it is more accurate to say that certain monitors have the problem, and some don't. It is a very difficult thing for some people to notice. As I said, I did not notice the lag on the Dell but after testing you can see that it is there. And I did notice the lag on the 244t, and after testing you can see that this lag is 10ms greater than the dell. This just tells you that my threshold for input lag is around 45ms. Anything higher annoys me. Some people may be more resistant to such things. It is similar to monitor refresh rates, some people cannot see the difference between 60 and 100 hz, but it certainly drives me up the wall.

I believe that the input delay is a property of all panels but the amount of Delay varies depending on the overdrive technology, panel technology and DSP technology.

I currently have two Benq FP91g+ LCDs and they have a slight delay that is noticable sometimes. I notice it most when I switch from my sony CRT to the LCD, but once I have used the Benq for a few seconds I adjust to the lag and it does not effect me. The samsung 244t was too much delay to get used to.

kamikazichaser, your explination of DSP is great and it really got me thinking but you forgot one step in the equation. When the image is being placed on the screen some proccessing is done to artificially reduce pixel response times. This is called overdrive. This in combination with the DSP could add up to the 50-60ms delays we are experiencing on large resolution displays.

The common thoughts are that input lag is caused by PVA panel technology, or the overdrive technology. I believe the likely solution is that each step in the equation from image resolution, to DVI bandwidth, to DSP processing, to overdrive and finally to the PVA panel adds its own portion to the input delay. The result being that large resolution displays suffer from a large input delay. I believe LCD manufacturers have to take steps to reduce this delay on each portion of the process.
 
I'm surprised some people want to go on about how there's no lag in a conclusive test thread. Sound like a bunch of babies to be honest.
 
docmal said:
kamikazichaser, your explination of DSP is great and it really got me thinking but you forgot one step in the equation. When the image is being placed on the screen some proccessing is done to artificially reduce pixel response times. This is called overdrive. This in combination with the DSP could add up to the 50-60ms delays we are experiencing on large resolution displays.

The common thoughts are that input lag is caused by PVA panel technology, or the overdrive technology. I believe the likely solution is that each step in the equation from image resolution, to DVI bandwidth, to DSP processing, to overdrive and finally to the PVA panel adds its own portion to the input delay. The result being that large resolution displays suffer from a large input delay. I believe LCD manufacturers have to take steps to reduce this delay on each portion of the process.

It is true that SPVA and PVA can cause some color inaccuracy, and yes, any amount of processing does cause a bit of lag, but I hesitate to pin it on the processors simply because they should (note the should) be able to process faster than the refresh of the monitor (ie response time). That is to say that the switching time of the pixels should be slower overall than the processing of the input signal. Now, remember that I said "should." It may be that in the real world, we are coming up against more and more of this. There aren't many S-IPS panels in this size category, so it's hard to compare. The giant 3007, however, should be a perfect candidate. I think what we need is to perform these tests on those. If the same PC, using the same DVI cable, can show the problem on a 244t or 2407, but doesn't display the problem on the 3007, then we know the answer with little doubt. Even an older 2335 should provide an effective test.

Anyone out there equiped to perform this test???
 
Your exactly right, we need more tests. The problem is that there are not many people with enough cash and passion to have a 3007 and a 2407 or 244t and also are willing to do these tests. I was very close to getting a 3007 and performing these tests but I think trying to run games at that res would be insane so I didnt get it.

Actually there aren't many regulars here that have the 3007 or the 244t. They are just expensive displays that many people will not purchase.

I guess I could run these tests against my benq 19inch panels... I think they are TN panels.
 
I have a 3007, a spare CRT and a Pentax *istD DSLR, but not a 2405 or a 244T. I can run the comparison tests vs the CRT, but not the others. Would need help setting the camera up, however. Also, what's the clock app used in the pictures?

One issue with the 3007 is this: how the hell would you run cloned with it running at its 2560 x 1600 native? I have a 19-inch SOny CRT that will run at 2,048 x 1,536, but no higher - and even at that res we're talking pretty low refresh. I assume we're going to want to run the dell at native to get conclusive results. Perhaps some kind of scrolling desktop on one of the displays - it is even possible to have a full desktop view on one display and scrolling view on another in clone mode?
 
Kevlarman said:
I'd like a link to that refresh rate "fix" so I can try it on my 2405.
I tried setting the rate through the nvidia "advanced timing" menu, without any success.

Not that it matters though, since there's no lag on my LCD. ;)

this is exactly why this issue has been under wraps for so long, because of stupid comments like this.

I'm sorry, your LCD has lag, YOU JUST DONT NOTICE IT.

some people are more sensitive to these things than others.

I would like to congratulate anyone who doesn't notice the lag on their LCD, it it makes life much easier on them in this reguard.

but some people do notice it, find it unacceptable, and thus would benefit in the future if we call attention to the problem now.

It's like saying "I don't have diabeties, so we should cancel all diabeties research funding!"
----yes a very exxagerated example, just trying to make a point.

I notice lag on my laptop lcd, dell lcd's at school(20",17", other sizes), friends lcd's.
 
The problem is that the CRT won't be connected via DVI, so we won't be isolating that from the experiment. Even so, could you give us a subjective comparison. Perhaps you could, using the same PC, perform standard input tasks that require speed and accuracy (games for the most part) with both displays, then report your general thoughts on the display lag. We know that the lcd will be crisper and, depending on the CRT, will likely have better color reproduction. There will be, however, some lag. The question is: is it annoying after playing on a CRT for a couple hours?
 
I'm worried that SED and other display tech will have the same problem, it seems to me there is no reason for LCD's to have lag other then then the fact they are digital displays, while crt's are analog.

has anyone tested a plasma for lag? :confused:

other comments:

I seriously doubt any LCD has fluctuating lag. the reason you're seeing different deltaT's in different pics is because your caputuring each screen on its partiular frame, with the two lcd's haveing a different refresh rate it could account for a delta(frame)deltaT of up to 17ms, assuimg a ~60hz refresh rate.

to help explain: one pic could have been taken with one screen haveing *just* drawn a frame, while the other screen's frame was *just* about to change, in this case the deltaT lag measurment error would be 1/60 or 17ms.


and to furtur clarify, LCD input lag is COMPLETELYT DIFFERENT and SEPERATE of ghosting lag (which I hate, too ;) )
 
okashira said:
and to furtur clarify, LCD input lag is COMPLETELYT DIFFERENT and SEPERATE of ghosting lag (which I hate, too ;) )

Yeah, let's reiterate that last sentence. We're trying to discuss input lag here, not lag due to response time. Yes, we all hate response lag, but it's a property of LCD panels, so there's nothing to be done.

This new phenomenon of input lag, however, is disturbing. Are these new panel sizes or resolutions reaching a physical limitation in the technology? Are new technologies, such as HDCP among others, causing too much latency in the input signal? Are we simply dealing with crappy hardware/firmware/software designs that cut too many corners? That's what we need to know.

Also, to answer the question about plasma. Yes, plasma has lag, but it's totally different. In my opinion it makes plasma less than ideal for gaming but really bad for text. Remember, the first laptop displays were plasma, not lcd. Yes, the tech has matured massively since that time, but there are inherent problems with plasma for ultra precise work.

One more...remember, that the electronics (dsp, buffer, etc) in a monitor is not a property of the panel. The OEM (Dell, HP, Gateway, Viewsonic, etc) make a monitor out of the panels that Samsung, LG, AU, etc make. Thus, we might find that the 244t is a problem, but the 2407 isn't. I'm just saying that we should keep that in mind.
 
OK, I got the 244T back for comparison with my 3007. These are subjective assessments, but I am very confident they are accurate:

In general, the input lag is very obvious with the 244T, whereas I do no notice any on the 3007. One quick test that makes the difference obvious is lining up the two panels, bezel to bezel, in extended desktop mode and then positioning a window so that it spans both displays. You can then move the window up and down and it's clearly visible that the movement of the window on the 244T lags behind that of the 3007. To beh honest, I don't think it will be necessary to take any photos with the digital timer, since the difference is so apparent to the naked eye.

Hope this helps.
 
Back
Top