HELP: testing input lag in "clone mode" on 30" monitor @ native rez against 19" CRT

howdy all,

I want to test input lag on my new monitor (gateway XHD3000) @ native resolution = 2560x1600 first and foremost - then, heck I'd even wanna test non-native input lag too, just to see... I never found any hard #s for this monitor, and now I got one to play with --- I'm new to display input lag testing, but the process sounds easy enough according to this [H]ardForum post.
the problem I'm running into is that the instructions say I need to put my video card (EVGA nvidia GTX 260 SC w/nvidia control panel) drivers into clone mode with my XHD3000, and my 19" test CRT connected. sadly, my 19" CRT will NOT run @ 2560x1600! the nvidia display drivers when configured to use "clone mode" appear to only allow me to use the same resolution on the 19" CRT as the XHD3000, or vice versa. I cant figure any way of using staggered/different display resolutions per monitor when in "clone mode". so my main test I wanna run is foiled... can someone make suggestion on howto get around my 19" CRT limitation of no 2560x1600 resolution? or possibly how one could "enable" nvidia clone mode with differing resolutions per monitor, if possible? or... I'm open to suggestions!

the HTML timer seems to be a great way of testing input lag... but I was wonder if I could one could just say... use a timer app that starts BOTH timers on each monitor @ the same time - could I do that instead of the whole clone mode?? does anyone have suggestions on how I could resolve this issue, and test input lag @ native resolution on XHD3000 + testing with my 19" CRT monitor (max capable resolution = 1600x1200)

fnord0

IIRC there is an option that lets you sale 1:1 with a lower rez monitor, basically means only a portion of the screen will show up 1600x1200 of it will, I don't remember how to do this though.

Digital Viper-X- said:

IIRC there is an option that lets you sale 1:1 with a lower rez monitor, basically means only a portion of the screen will show up 1600x1200 of it will, I don't remember how to do this though.

Click to expand...

hrrmmmm, I cant seem to locate this setting anywhere in the newer nvidia drivers (I am on v182.06 and I've now tried v182.08) - does anyone else here know where I can locate this setting, and how to implement? I still keep running into the same issue = when I enable "clone mode" in the nvidia diplay driver, I can only set resolution per 1 single monitor, I cannot get around this.

little help, please? TIA!

Use 1280x800 -- the LCD panel should "natively scale" that directly to 2560x1600. That's the only 2 resolutions it supports, iirc.

I guess I will just try that, but to me it sounds like my monitor is going to UPSCALE like yr saying to 2560x1600... doesn't that introduce some form of input lag itself?? anyways I'll give it a go and see how it turns out...

any other ideas of how I could test 2560x1600 input lag with my CRT, or is this just impossible with my current setup?

It's supposed to simply simply send 1 input pixel to 4 output pixels.
edit: okay, so I misunderstood what this 2560x1600 panel does...

so my test @ 1280x800 shows ~90ms latency! ha, that is horrid.. I can literally see the input lag at this resolution, BUT when I shut down the "clone mode" via nvidia control panel and use 2560x1600 on the XHD3000 30" monitor, and 1600x1200 on the CRT = I can tell you the input lag seems to disappear, and this 30" monitor behaves like it has no input lag whatsoever @ native resolution in games, and @ vista x64 desktop (i can shake any of my open windows/applications with half on one monitor, and half on the other, and it seems like there is ZERO lag - but when doing this @ 1280x800 in clone mode or normal mode the input lag is literally visable). I believe u and all, but damn it sure seems like there is a scaler in the monitor that is introducing lag, plain and simple. hence my desire to test input lag @ native rez = 2560x1600... is there any other way I can test native input lag? puuuuhhhhleeeze say yes

Try doing clone mode with the nVidia card set to "use NVIDA scaling" or "use NVIDIA scaling with fixed aspect ratio" under "flat panel scaling" in the NVcontrol panel and then rerun the test at 1600x1200. The image will be blown up to full screen on the XHD3000, but it will be the video card doing the upscaling, so it should introduce no lag because the XHD3000 is effectively getting a 2560x1600 screen from the vid card.

What you've done so far is confirm what everyone is saying, that the XHD3000 has horrid input lag when the scalar is employed at all.

10e said:

Try doing clone mode with the nVidia card set to "use NVIDA scaling" or "use NVIDIA scaling with fixed aspect ratio" under "flat panel scaling" in the NVcontrol panel and then rerun the test at 1600x1200. The image will be blown up to full screen on the XHD3000, but it will be the video card doing the upscaling, so it should introduce no lag because the XHD3000 is effectively getting a 2560x1600 screen from the vid card.

Click to expand...

I was thinking the same thing, BUT when going into the "Flat Panel Scaling" screen, EVERYTHING is grey'ed out =(

10e said:

What you've done so far is confirm what everyone is saying, that the XHD3000 has horrid input lag when the scalar is employed at all.

Click to expand...

sadly I agree, but I tell u this monitor RIPS at native rez, so I try to play all games @ native resolution and it's been great... I just wanna measure it @ native just to see, but I keep running into this CRT-resolution wall. any other thoughts how I could proceed?

The 1280x800 test won't mean anything, because the XHD3000's built-in scalar introduces input lag at lower resolutions - the lower the resolution, the worse the input lag. The test needs to be done at 2560x1600.

The XHD3000 has been out for a while now, and I am not able to find input lag numbers ANYWHERE, particularly at the few sites that actually report input lag numbers. It's as if nobody has been able to do this test.

Probably cuz only a few CRTs could push that res... Sony 24" FW900? Then again if there's sites showing specs on the other 30"ers then I dunno what gives.

The FW900 only goes up to 2304 x 1440. I'm having trouble finding any evidence that any CRT has ever done 2560x1600. A user review for the Eizo F 775, here, suggested that it does 2560x1600 but the specs seem to say it's a 1600x1200 monitor so I'm left confused.

I found reviews that listed the input lag for the Dell 30" monitors, but they don't go into detail about how they did the test or what monitor they used. I'm beginning to think that the sites that report input lag in their LCD monitor reviews haven't reviewed the XHD3000 because they can't measure the input lag at 2560x1600. This in turn makes me wonder if the input lag tests for the other 30" monitors were done at 2560x1600; if not, would they then be valid?

This App by FlatpanelsHD is excellent at testing input lag: http://www.flatpanelshd.com/focus.php?subaction=showfull&id=1229335064

mistar said:

This App by FlatpanelsHD is excellent at testing input lag: http://www.flatpanelshd.com/focus.php?subaction=showfull&id=1229335064

Click to expand...

yep, that's the app i been using... the problem is when i put videocard (EVGA GTX 260 SP216 SC) into clone mode, the CRT doesnt allow anything over 1600x1200 res - yet I wanna test input lag on my 30" monitor @ native rez (2560x1600).

fnord0 said:

I was thinking the same thing, BUT when going into the "Flat Panel Scaling" screen, EVERYTHING is grey'ed out =(

sadly I agree, but I tell u this monitor RIPS at native rez, so I try to play all games @ native resolution and it's been great... I just wanna measure it @ native just to see, but I keep running into this CRT-resolution wall. any other thoughts how I could proceed?

Click to expand...

Try booting up first with only the XHD3000 on, and see if the flat panel scaling options are there. You should be connected via DVI to it (which I'm assuming you are due to dual-link) to get the options. I find it weird that it won't let you change the scaling options at all.

EDIT: It looks like you only had it connected and they aren't there. Ugh. As dumb as it sounds, you may want to clean out/sweep out the drivers first and then re-install them and see if it fixes it. Sorry for making you do all the work

10e said:

Try booting up first with only the XHD3000 on, and see if the flat panel scaling options are there. You should be connected via DVI to it (which I'm assuming you are due to dual-link) to get the options. I find it weird that it won't let you change the scaling options at all.

EDIT: It looks like you only had it connected and they aren't there. Ugh. As dumb as it sounds, you may want to clean out/sweep out the drivers first and then re-install them and see if it fixes it. Sorry for making you do all the work

Click to expand...

mr. 10e, your reputation here @ [H]ardForums preceeds u - I'm actually tickled pink yr one of the people replying to this forum post. I'm more then happy to take yr advice... I just so happen to have my trusty copy of Driver Sweeper and a fresh download of NVidia Drivers v182.08 so I will give this a go = uninstall/reboot, fresh driver install/reboot and see if this fixes that FPS option! thanks for the tip!

I really am anxious to see the input lag #'s @ native res. I just bought this monitor back on 2/16/2009, so I have 30days total to decide if I wanna keep it... that final decision needs to happen real soon. I will update ya with results of uninstall/reinstall of drivers.

fnord0

fnord0 said:

I just so happen to have my trusty copy of Driver Sweeper and a fresh download of NVidia Drivers v182.08 so I will give this a go = uninstall/reboot, fresh driver install/reboot and see if this fixes that FPS option! thanks for the tip!

Click to expand...

well, after multiple uninstall/reboots, reinstall/reboot - and many variations of that process, even a couple trips to safe mode. I tried suggestions from other websites and forums and mixed in all suggestions from this thread... I tried to boot with one monitor connected via dual-dvi connector+cable on port 1, and nothing connected to DVI port 2 of my EVGA GTX 260 SC - dropped resolution to 800x600 (even tried booting @ 800x600) then changing upto 2560x1600 (and other variations), etc, etc. ALL THIS, and I am still seeing all FPS (Flat Panel Scaling in NVidia Control Panel) options all greyed out! grrrrrrr

I swear I'm not doing anything wrong here, I'd like to think of myself as quite competent/proficient with Vista x64 (and nearly all windows + several linux flavors - I've been doing this shit since I was 10years old = ~22years now). I just can't get FPS working... I am beginning to think that my monitor's built-in scaler comes with NO WAY to shut it off, wether hardware or software?? (my scaler comment being completely base-less, just my own oppinion).

little help here?

thanks for everyone's input so far!
-fnord0-

PS. digging thru [H]'s NVidia GPU forum, I found this informative post titled "When is Nvidia going to fix their flat panel scaling?" so I am currently reading each post (4pages) to see if there is a workaround... fingers crossed. I am still open to other suggestions on how to test 2560x1600 for input lag, if there is another option besides playin' with Flat Panel Scaling.

PPS. what do others think of this idea :: If I can locate an application that involves multiple stop watches that can be started at the same time, and keep the 2+ watches synchronized - so that I can position one stopwatch on the CRT monitor and another stopwatch on the 30" screen (without using "clone mode" in NV CP, which I am hoping should allow me to use 2560x1600 on 30", and 1600x1200 on the 19" CRT) - then just hit START on one stopwatch, and take pictures of the 2 monitors as normal (@ 1/100th, and ISO 400)? would that work??

fnord0 said:

PS. digging thru [H]'s NVidia GPU forum, I found this informative post titled "When is Nvidia going to fix their flat panel scaling?" so I am currently reading each post (4pages) to see if there is a workaround... fingers crossed.

Click to expand...

no luck there, it's looking like if one uses "clone mode" you are required to use the same resolution, no if, and's, or but's = no options from everything I am seeing. I could be wrong, but everything I have read indicates just this.

fnord0 said:

PPS. what do others think of this idea :: If I can locate an application that involves multiple stop watches that can be started at the same time, and keep the 2+ watches synchronized - so that I can position one stopwatch on the CRT monitor and another stopwatch on the 30" screen (without using "clone mode" in NV CP, which I am hoping should allow me to use 2560x1600 on 30", and 1600x1200 on the 19" CRT) - then just hit START on one stopwatch, and take pictures of the 2 monitors as normal (@ 1/100th, and ISO 400)? would that work??

Click to expand...

I still don't know if this is a valid test or not, but I've found a couple apps that might do just as I am suggesting here = Multi-Timer and Jumbo Timer. yet these 2 apps only display 1/10 of a second, but the website says it is accurate up-to 1/100,000 of a second.

I'm thinking this app = StopWatch 2009 is probably the best of the bunch, as it shows up-to 1/1,000,000 of a second - BUT I need to find a way of rotating the application vertically, so I can have 1 stopwatch display on the 30" screen, and another stopwatch display on the 19" CRT.

can someone tell me ;

1. is the above idea of mine in anyway a valid way to test input lag on my 30" monitor @ 2560x1600? (my CRT only goes up-to 1600x1200, which is why I ask at all)
2. is there a way to ROTATE applications in Windows Vista x64? so that instead of viewing the application in the traditional (vertical?) view
   
   I'd like to display it so that the application lies on it's side. (horizontally?)
   

thanks to one and all who have helped. I look forward to further correspondance with you all.

-peace-
fnord0

no one, eh? I located this application = StopWatch.de "BASIC" - it shows 1/1,000th seconds, works perfect for 2x multi-monitor setup! one stopwatch is used, with a single start button. the stopwatch runs on monitor #1, and is fully zoomable with CONTROL+Z (u can blow it up full-screen if wanted) -- then on monitor #2 (CRT) it acts as a scoreboard, with a BLOWN UP view of the current time on the stopwatch from screen #1, and "LAP" time, but that isn't used... the 2 monitors keep EXACT same time, and appear to be updating synchronously.

I am able to take pictures with my camera as documented in the [H]ardForum Input Lag Measurement post - but I am NOT using "mirror mode" in NVidia Control Panel, so this allows me to use 2560x1600 (native) resolution for my 30" XHD3000 as primary monitor, and I can run the 19" CRT as monitor 2.

would using StopWatch.de "BASIC" as described above be a legitimate way to test input lag @ 2560x1600? (since my 19" CRT monitor tops out at 1600x1200)?

fnord0

-bump-

well, well... bringing back the dead thread! I am hoping someone here @ [H]Forum can tell me if this is a legit way to test my XHD3000 @ 2560x1600, while my CRT monitor is @ 1600x1200... this time I think I got it right.

I found 2 applications = UltraMon + StopWatch 2009

If one sets up 2x monitors, setup so that the desktops are "on top of each other", meaning the primary is on TOP, the secondary is on the bottom.


then using the StopWatch 2009 application, start-up all 8 stopwatches, so that you can span them acorss both screens like I have shown here
what do u all think? is THAT a valid way to test for input lag?

-OR-

If you dont think that'll work, what about UltraMon's capability of MIRRORING a single application that is running on monitor #1 TO monitor #2, it seems to work great.

please fellow [H]Forum goers, let me know if this may be a good solution for my perdicament! thanks!

fnord0

[duplicate post removed]

The stopwatch test works, but if you use multimonitor, there's software-induced screen refresh lag, as some graphics drivers will refresh one then the other. But this is a nonissue for very fast graphics cards on very fast systems.

There's another way too:

I found that JudderTest (normally used for testing framerate smoothness, motion blur, etc) is an excellent test of input lag for same-sized computer monitors, if you stack two monitors top-by-top.

JudderTest is a screen with horizontally-moving vertical-bars. The screen with the bars that move ahead, is the monitor that has the least input lag.


For differently sized computer monitors, watch for bars that show up at the left side. The monitor that has the bars that appear onscreen first, is the one with the least input lag. (But the difference will be pretty small)

Use a high step value (Say, 10 pixels instead of 1 pixels)

JudderTest can be found on Google (from a site called AVSFORUM)

Why not set both screens to 1024x768, and the XHD3000 to 1:1 mode (no scaling, i.e. no additional lag) and test in clone mode? Seems to be the simplest measurement solution. Also you may notice slightly different numbers with the timer at the top of the screen versus the bottom, because of the way the screen refreshes.