![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Navigation
Install the app
How to install the app on iOS
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
More options
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Latency testing at UHD 4k
- Thread starter jts888
- Start date
- Joined
- Mar 14, 2013
- Messages
- 2,743
Flod has posted his methodology here
that's still just using the hdmi output on an arduino board, which I'm highly doubtful can do UHD@60Hz w/ 4:4:4 chroma, etc.
I'm starting to strongly suspect that most 4k latency figures are VERY hand-wavey, since you'd need either a LB-like device with 18Gb/s or whatever HDMI throughput or both a reference display and full-bandwidth HDMI 2.0 splitter with either zero or know constant timing offsets between the outputs.
I'm starting to strongly suspect that most 4k latency figures are VERY hand-wavey, since you'd need either a LB-like device with 18Gb/s or whatever HDMI throughput or both a reference display and full-bandwidth HDMI 2.0 splitter with either zero or know constant timing offsets between the outputs.
- Joined
- Mar 14, 2013
- Messages
- 2,743
I think you've misunderstood his device. His device simulates a USB mouse command, which is interpreted by software to display a single line on his CRT. He has a photodiode that measures the light from this line on the CRT, and he measures the time difference between the time the mouse command was sent and the time the photodiode registered the light event. So this measures system lag in addition to display lag.
Last edited:
I think you've misunderstood his device. His device simulates a USB mouse command, which is interpreted by software to display a single line on his CRT. He has a photodiode that measures the light from this line on the CRT, and he measures the time difference between the time the mouse command was sent and the time the photodiode registered the light event. So this measures system lag in addition to display lag.
Yeah, I definitely didn't read through his description well enough initially, but I struggle to see how this is a good solution for trying to compare LCDs' latencies.
- Joined
- Mar 14, 2013
- Messages
- 2,743
If you test two displays with identical software environments, you can directly compare the input lag. Not sure what's so complicated about this.
Flod has posted his methodology here
not really ideal for this
easist way is getting a cheap 1000fps camera
If you test two displays with identical software environments, you can directly compare the input lag.
- software/systems aren't that deterministic
- you have to work within 8-17 ms timing windows for synchronous display refreshes
- you need to have a second display on hand with with very consistent latency characteristics
- it's still a pain in the ass to use compared to a dongle you plug into an HDMI port and hold up to the screen
it's interesting, but it's still a very poor solution unless you're specifically interested in finding whole-stack latencies.
- Joined
- Mar 14, 2013
- Messages
- 2,743
yea i just posted that link so you'd get some insight into input lag testing methodology.
as for the 8-17ms window, you take multiple measurements and look at the distribution. But you could also just work without vsync and not have to worry about that.
Aside from the distribution of possible refresh start times and USB polling intervals (which you can account for by taking hundreds of measurements), I don't think other sources of error are going to be significant when the display lag itself is in the order of multiple milliseconds).
as for the 8-17ms window, you take multiple measurements and look at the distribution. But you could also just work without vsync and not have to worry about that.
Aside from the distribution of possible refresh start times and USB polling intervals (which you can account for by taking hundreds of measurements), I don't think other sources of error are going to be significant when the display lag itself is in the order of multiple milliseconds).
Got some disappointing news from Leo Bodnar:
>> Does Leo Bodnar have any plans to make an HDMI 2.0/UHD 4k video lag tester in the near future?
>> It seems as though a large number of UHD displays have substantially different (usually higher) latency at their native resolutions than at 1080p, but there is no convenient UHD testing solution in the marketplace at present.
>> Thanks!
> Thank you for your enquiry.
> Unfortunately have no fixed plans to make this device, at least within the near future. The technology required to output this resolution is significantly more complicated than the current version so it will not be an easy or cheap project.
> Best Regards
> John
So at this point it looks like latency numbers floating around the internet for 4k will be somewhat suspect, and quite possibly for some time.
>> Does Leo Bodnar have any plans to make an HDMI 2.0/UHD 4k video lag tester in the near future?
>> It seems as though a large number of UHD displays have substantially different (usually higher) latency at their native resolutions than at 1080p, but there is no convenient UHD testing solution in the marketplace at present.
>> Thanks!
> Thank you for your enquiry.
> Unfortunately have no fixed plans to make this device, at least within the near future. The technology required to output this resolution is significantly more complicated than the current version so it will not be an easy or cheap project.
> Best Regards
> John
So at this point it looks like latency numbers floating around the internet for 4k will be somewhat suspect, and quite possibly for some time.
I've still found http://tft.vanity.dk/inputlag.exe to be very accurate and useful when used with high shutter speed cameras.
I'd really like to see someone put this next to an established 1080p/1440p display and compare the time. Spare us the downsides of flash-based testers, please.
I'd really like to see someone put this next to an established 1080p/1440p display and compare the time. Spare us the downsides of flash-based testers, please.