2 monitors side by side input lag test help please

[ahmedpc03]

20180924_201455.jpg (101.4 KiB) Viewed 5941 times

hi

i would like to test two monitors connected to a pc mirrored.

and i would like to know which one has lower input lag than the other one

what is the best method for doing this pls ?

i have tried this method on youtube
https://www.youtube.com/watch?v=3Xz9Ugd78Vc&t=2s

but actually iam confused if its working or not

cause i have got the same exact numbers on booth monitors , booth running on 60hz

these two monitors is the lg 27mp59g
and the philips 323E7QDAB

please help i want to take a decision between these two

thanks

[ahmedpc03]

please help guys
just want to make sure my test is ok or not ?

[Sparky]

This type of test needs to be run many times. The refresh rate of the monitors and framerate of the timer have a big impact, so you need a lot of samples to figure out exactly how big the difference is. Keep taking photos until you get several where the number on both monitors is different, the proportion of those to the ones where they're the same, along with the framerate of the timer, will tell you how big the difference is between the two monitors.

[ahmedpc03]

Thanks for your reply

Yup i captured about 20 pictures and all of it gived me the same thing ... They booth exactly the same
And i tried the video method and the timer one

But actually i was little bet confused ... How are they exactly the same and they are on a different manufacutrer and different sizes

Is there something like a feature on nvidia sync the monitors lag together ?

Is this test is ok to check if there is a difference ?

Iam confused

[Sparky]

The timer you're using might have a very low framerate when compared to the refresh rate of the monitors.

[ahmedpc03]

Do you meam if the frame rate higher there well be an input lag difference ?

[Sparky]

The difference in input lag between the two monitors doesn't change with framerate, you'll just be able to see it more consistently. Say you have two monitors at 60hz, and the timer is running at 30fps. Depending on when the frame completes, a monitor will be displaying the image for 33ms, or it might bounce between 16 and 50ms. If the difference in latency between the two monitors is only 1ms, you'd need about 15 samples to even have a 50% chance of seeing differing values.

Seeing differing values at least once can also be used to figure out the framerate of the timer. If you don't know that, a lower framerate timer could be hiding a much bigger difference in latency.

Even if the timer runs at 1khz, if v-sync is on, it will effectively reduce it to the refresh rate of the monitor.

[open]

This website has a tool that works well for this. It alternates black and white frames which lets you see what part of the frame is being scanned out on each monitor.

Remember that your camera also has a scanout speed so you want to configure it to have the fastest scanout possible to eliminate errors or artifacts. Look for iso settings, shutter speed, exposure settings. Also experiment with different orientations of your camera. If your camera is scanning out left to right then the monitor on the left will have an advantage. You want top to bottom or bottom to top scanning. I would prefer bottom to top since your monitors will scan out top to bottom.

Ideally you want to use the same type of output running to each monitor. Like displayport for both. In my testing I found out that my hdmi had about 1 extra frame of lag at 60hz compared to my displayport connection. Finnally you may want to swap the monitors to eachothers output ports on the computer just to make sure that each is outputting with the same lag. Not 100% sure that will be the case if they are both displayport but you never know.

[Chief Blur Buster]

That's www.testufo.com/flicker used by www.blurbusters.com/lightboost/video

The scanout differential (top-to-bottom scanout) means numbers will be different for top edge than bottom edge. That is because a screen takes time to refresh top-to-bottom.

If you want to measure subframe input lag differences, you have to account for the scanout -- that is what SMTT 2.0 lag-tester technique can do, though I'm eventually going to release a TestUFO replacement for SMTT eventually. (I have a prototype but it's not yet idiotproofed enough, since it requires a lot of babysitting to train to use correctly). Accuracy will depend on LCD GtG error margin but technologically it is possible.