Blur Busters Forums

"I was kind of scared though of going from 1ms to 5ms as it is what most IPS are but I've seen the new one at 165Hz are 4ms"

The response time numbers are completely useless for telling you anything useful. It's way important to know the overshoot and the response times of different shade transition like they do at TFTcentral. The number isn't even accurate at all, it tries to cram everything into a number which isn't possible.

"and my guess is that since it is IPS it is prolly both the same panel behind and something from LG"

They are from AUO but your right, most likely the same.

"Now is it true that it is possible on all G-Sync monitors that supports ULMB, to use both at the same time with some custom resolution/timing tricks or is it a huge pain in the ass and very complicated to do, buggy, and not worth it?"

It's pretty easy to do but i don't think anyone can say for certain if it will work with those monitors specifically.

Please keep in mind that IPS ULMB has more crosstalk than TN.

lexlazootin wrote:"Now is it true that it is possible on all G-Sync monitors that supports ULMB, to use both at the same time with some custom resolution/timing tricks or is it a huge pain in the ass and very complicated to do, buggy, and not worth it?"

It's pretty easy to do but i don't think anyone can say for certain if it will work with those monitors specifically.

Hmm, then how about we get an "offical list" of supported monitors for (G-Sync + ULMB) on that link below? I'm sure it would interest a lot of people!
www.blurbusters.com/faq/120hz-monitors

RealNC wrote:Please keep in mind that IPS ULMB has more crosstalk than TN.

Hmm, there is more but I guess as long as it is within acceptable levels it should be alright. Well there's more you say but how much more approximately? Also if it is possible to adjust and improve the crosstalk with the "Vertical Total" trick, then it should compensate for the little extra crosstalk. I've got a TN panel but perhaps the only ******** retarded monitor from BenQ that can't be adjusted through VT so an IPS with ajustable VT would prolly be better.

PanzerIV wrote:

RealNC wrote:Please keep in mind that IPS ULMB has more crosstalk than TN.

Hmm, there is more but I guess as long as it is within acceptable levels it should be alright. Well there's more you say but how much more approximately? Also if it is possible to adjust and improve the crosstalk with the "Vertical Total" trick, then it should compensate for the little extra crosstalk.

It's quite a bit more. It's clearly visible in games. And it's not adjustable with VT tweaks. The gsync module does not allow it.

RealNC wrote:It's quite a bit more. It's clearly visible in games. And it's not adjustable with VT tweaks. The gsync module does not allow it.

1- You mean that if I modify the VT with a custom res, then G-Sync will get disabled and won't work anymore until I set it back to default VT???

2- Anyhow to be honest I was much more looking at the Acer XB271HU for it's ULMB feature that comes with the G-Sync module as I can already handle high fps with my "GTX1080 & i5 8600K" and at least ULMB can work at 85Hz and 100Hz unlike mine that is 120Hz minimum which is plain retarded from BenQ. So is it possible to only use ULMB with a modified VT or changing this value will not just disable the use of G-Sync but ALSO the use of ULMB completely?! O_o

PanzerIV wrote:1- You mean that if I modify the VT with a custom res, then G-Sync will get disabled and won't work anymore until I set it back to default VT???

ULMB will get disabled.

So is it possible to only use ULMB with a modified VT or changing this value will not just disable the use of G-Sync but ALSO the use of ULMB completely?! O_o

From what people have reported so far, modifying VT will disable ULMB or result in the display not accepting the signal and turning off. This is also true for me (using an XG2703-GS.)

Anyhow to be honest I was much more looking at the Acer XB271HU for it's ULMB feature that comes with the G-Sync module as I can already handle high fps with my "GTX1080 & i5 8600K" and at least ULMB can work at 85Hz and 100Hz unlike mine that is 120Hz minimum

ULMB can work lower too with CRU timing tweaks. Mine works at 60Hz:

See:

viewtopic.php?f=4&t=3501

Yes, ULMB 60Hz hack works well for PC 60Hz (emulators etc).

Just not for consoles as it requires a custom mode.

No, VT tricks do not work with the NVIDIA implementation of ULMB. It already uses maximum possible VT internally automatically (internal scanrate conversion, via partial-buffer-and-burst-accelerated-scanout). IPS can only be reduced in strobe crosstalk to a certain extent and cannot be then further improved, 5ms GtGis extremely close to 8.3ms refresh cycle scanout, so scant time between refresh cycles to keep strobe crosstalk down.

Chief Blur Buster wrote:No, VT tricks do not work with the NVIDIA implementation of ULMB. It already uses maximum possible VT internally automatically (internal scanrate conversion, via partial-buffer-and-burst-accelerated-scanout).

Hmm, I know it is a year old thread that haven't been updated lately but you've said that on ULMB the VT trick could improve the input lag even if it had zero impact on the crosstalk. What would be the final word as of now in 2018 on the latest G-Sync 165Hz monitors?

viewtopic.php?p=25890#p25890

ULMB is very picky about the dotclocks you use -- monitors with 85Hz, 100Hz and 120Hz ULMB that are flexible with the ULMB tricks (e.g. 60Hz ULMB tricks) are the easiest ones.

See these:
ULMB 60 Hz Hack
ULMB + GSYNC Hack

So one uses the 100Hz ULMB timing to create 85 Hz ULMB (+ large VT) with lower-lag than native 85 Hz ULMB. Or even, the 120 Hz ULMB timing to create 85 Hz ULMB. For this, 85Hz ULMB using 120Hz ULMB timings would technically have approximately a ~1.5ms lag savings for screen centre, and ~3ms lag savings for screen middle. Maximum possible bottom-edge input lag savings from faster scanout = (1/85sec - 1/120sec) = ~3.43ms less lag for bottom edge. You're essentially creating a lower-Hz ULMB (e.g. 85Hz) signal from the full-Hz ULMB dotclock (e.g.120Hz) to hack a "QFT" onto ULMB. This is the lag savings from faster scanout.

Unfortunately flexible VT's are out of the question when it comes to trying to increase VT's at the highest-Hz ULMB -- this is a monitor firmware restriction. So you can't reduce strobe crosstalk of higher-Hz ULMB, because you're already at the maximum allowable pixel clock. NVIDIA puts some rather severe restrictions on the dotclock of a signal for ULMB, so there's no "Quick Frame Transport" (QFT) style tricks possible manually.

This would be a suggestion to NVIDIA to add QFT (Quick Frame Transport) capabilities to ULMB to help reduce ULMB lag.

(The new lag-reducing HDMI 2.1 "QFT" protocol is also essentially achieved via an equivalent of Large Vertical Totals too).

LG 27GK750F-B

Well, it does strobe @ 240 hz. Strange thing is, i'ts also really smooth under 240 fps, or even 400 fps.
From what i remember basically anything not being exactly 240 fps , (or 120 fps in those days) was an absolute crap fest when strobing.

It seems that that is no longer the case, i don't understand HOW though. I checked multiple times if the strobe was really on, and it is .
Am i the victim of wishful thinking/placebo effect ?

I think i'm just going to leave it on always.