How to measure input lag of 2 screens in dual screen or mirror mode?

I have a computer with a Radeon HD 7870 graphics card which is plugged to two screens:

my main screen (a 24" LCD Hyundai W240D v2) through DVI
my TV (50" LCD Samsung UE50ES6100) through HDMI

When I configure the 2 screens in dual screen (extended or mirror mode), I observe an input lag: when I move the mouse cursor on the main screen it feels OK but when I move it to the TV it feels laggy, but this is a subjective observation. How can I measure or at least estimate the input lag between the 2 screens?

I have a Canon EOS 550D (aka Digital Rebel T2i in some countries) reflex camera that I can use to take pictures from 1/4000 to 30 seconds. My first thought was to display a video showing a time count with micro seconds on the 2 screens then take photos and see if the time is the same on the 2 screens. Will it show me the input lag? Is this protocol correct?

Update: I know that I won't be able to measure the absolute value of the input lag because I have no absolute reference, but I'm still interested in knowing the input lag between these 2 screens.

A.L

Posted 2014-12-01T23:56:12.157

Reputation: 509

Your description is a little unclear. Are you saying that everything feels "laggy" when you are running in dual monitor mode? Or that the TV feels "laggy" but the main monitor doesn't? – Jason C – 2014-12-02T00:27:13.580

@JasonC: I wasn't sure if laggy was the correct word. In other words, when I move the mouse cursor on the 24e screen it seems ok but when I move it on the TV is feels different. – A.L – 2014-12-02T00:28:46.513

Answers

Your description is slightly unclear. I will assume you mean the TV feels "laggy" while the main monitor does not.

This would not be describable as "input" lag since your main monitor feels fine. The mouse does not care where it is on the screen and there would be no change in input response that would be dependent on the location of the pointer.

This would be more describable as "output" lag, where the image displayed on the TV is presumably lagging a little behind the main monitor.

To that end, the following quick experiment should make any difference in display times evident:

Start an application, like notepad or something, and move the window to the border between monitors so that half is on the main display and half is on the TV.
Drag the window along the border. Observe if the half of the window on the TV is "trailing behind" the other half. You can take a video of it if you want, or just observe with your eyes.

If you want a more accurate measurement then, yes, your idea of displaying a timestamp on both screens (in mirror mode) and taking a video will give you a relative time for the display output.

I suppose if you wanted to measure absolute time from input to display, you could also make sure your mouse or keyboard is visible next to your monitors, and take a video and measure the time between your physical action and its effect on the screens. This might be tricky and unreliable for a number of reasons, though. And, in any case, if the main monitor "feels OK" and the TV does not, just knowing a relative time between the two (above experiments) will at least be enough to help confirm that.

Jason C

Posted 2014-12-01T23:56:12.157

Reputation: 8 273

Thanks, you're right it's output lag. About the trailing behind, a video is limited to 30 Hz (with my camera), so it will capture 2 frames of the screens for each frame in the video, I don't think a video will be usable. About displaying a timestamp, do you have an idea to do that? – A.L – 2014-12-02T17:38:02.007

For the trailing behind; I meant just eyeball it. If it "feels" different as you describe, you should be able to visually spot any noticeable difference as well. Just put a window on the border, drag it up and down, and see if it looks funny. To display a timestamp; hmm, if you search around I'm sure you can find an app; or you could use something that displays high resolution timestamps, such as your favorite audio editing application. Are you on Windows? – Jason C – 2014-12-02T21:15:35.677

(Dunno if it helps but here is a JavaScript timer that's accurate to about 5ms on Chrome, Windows 7 -- which is not very accurate, but is still higher than your video refresh rate).

– Jason C – 2014-12-02T21:27:36.357

You know, I don't know if this helps, but I just purchased a monitor that has a specific "game mode" for its digital inputs. It states that enabling game mode will "increase response but sacrifice quality", which it does. I have no idea what it is about or how it works, but it has an effect and does imply that there is at least one device out there that admits to having laggy response at high quality, so there must be something to it. So perhaps check for a similar setting or try other monitors. – Jason C – 2015-02-03T23:05:48.657

I used your JSFiddle and adapted it, it really helped me, thanks a lot. I used in order to take photos of my 2 screens side by side. The results are surprising: the PC monitor displays timestamps shifted by 70 ms from the TV, in other words, the timestamp is lower on the TV. I'll post images tomorrow.

– A.L – 2015-02-04T01:19:14.510

I was wrong, that's the TV which display images after the main monitor, for example: 9422 ms on the monitor versus 9338 ms on the TV. – A.L – 2015-02-04T23:28:01.773

I used the Javascript timer from Jason C's answer and here are the results:

With the PC directly connected to the monitor and the TV, output lag is about 40 ms between the two displays:

With the PC directly connected to the monitor, and connected to the TV through a Home Cinema Receiver (Onkyo TX-SR333), the output lag is about 100 ms between the two displays:

A.L

Posted 2014-12-01T23:56:12.157

Reputation: 509