G-Sync vs ULMB vs Fixed frequency

Talk about NVIDIA G-SYNC, a variable refresh rate (VRR) technology. G-SYNC eliminates stutters, tearing, and reduces input lag. List of G-SYNC Monitors.
Sayu
Posts: 6
Joined: 14 Oct 2018, 21:50

G-Sync vs ULMB vs Fixed frequency

Post by Sayu » 14 Oct 2018, 22:23

Hello everyone, I have read a lot of articles on your site. Congratulations on your work.
I don't speak too much English, I help myself with translators, so I allow myself to ask several questions to shed light on my lantern: lol:

Currently my configuration is:

CPU: I7 8700k oc
GPU: GTX 1060 6GB oc
RAM: 3600 Mhz oc
SCREEN: ROG Swift PG258 (https://www.asus.com/ROG-Republic-Of-Ga ... FT-PG258Q/)

And I would like to know what is best between G-Sync at 240hz vs ULMB at 144hz vs Fixed frequency at 240hz for H1Z1 games (Now dead, I have hope for the future) ? I only play this one and I haven't seen any articles about it.

Knowing that:

I am between 120/144 fps in town and 240 fps in the countryside.
That someone told me that the game was badly optimized and therefore G-Sync/ULMB are to be discarded. Really?
I tried the ULMB at 10 and 50 it is unplayable because it's too dark. The only thing left is the Pulse Width at 100. Is it still interesting?
It's a game where you need the most reactivity, while remaining beautiful for the fights from far away.
I don't think I'm using V-Sync for as little latency as possible. Forcing a cap fps in game to 141 or 237?

Sincerely, thank you, I am eager to learn.
Last edited by Sayu on 15 Oct 2018, 20:43, edited 1 time in total.

User avatar
RealNC
Site Admin
Posts: 3741
Joined: 24 Dec 2013, 18:32
Contact:

Re: G-Sync vs ULMB vs Fixed frequency

Post by RealNC » 15 Oct 2018, 10:31

"Fixed frequency" is never a good choice when you've got a g-sync monitor.

ULMB is there to make animation look more clear (less blur.) But it's not that good if the game you play has too unstable FPS. If you can play a game with vsync ON and constant 144FPS, then ULMB is good. (Except that it can be too dark for some people.)

G-Sync is good in all cases. It doesn't matter if the game is optimized or not. G-Sync removes tearing, and it also decreases stuttering. It does not have input lag like vsync does.

There is really no reason to disable g-sync, unless you want ULMB. So the question is not "fixed refresh vs g-sync vs ULMB." It's just "g-sync vs ULMB." Fixed refresh is not useful.
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

Sayu
Posts: 6
Joined: 14 Oct 2018, 21:50

Re: G-Sync vs ULMB vs Fixed frequency

Post by Sayu » 15 Oct 2018, 20:31

Thank you very much for your answer.
Last edited by Sayu on 15 Oct 2018, 20:57, edited 2 times in total.

Sayu
Posts: 6
Joined: 14 Oct 2018, 21:50

Re: G-Sync vs ULMB vs Fixed frequency

Post by Sayu » 15 Oct 2018, 20:55

RealNC wrote:"Fixed frequency" is never a good choice when you've got a g-sync monitor.
Because the fixed frequency causes stutters and tearing if the fps are below or above 240hz? Or just below, maybe? The input lag gained unlike G-Sync is almost zero, right?
ULMB is there to make animation look more clear (less blur.) But it's not that good if the game you play has too unstable FPS. If you can play a game with vsync ON and constant 144FPS, then ULMB is good. (Except that it can be too dark for some people.)
Yes I noticed that the ULMB is very interesting because during close fights which are quite nervous, the image is much more fluid and less blurry but I have a little more microstutters.

Are these statements below true about the cause of micro stutters?

1/ A weak gpu unlike a big cpu? Because I noticed that I had more when I'm in low quality so use of the cpu unlike in high quality so use of the gpu at 100%, what do you think?

2/ The fact that the fps drop below or above 144? Or only underneath, again?

3/ The fact of putting phys x on the cpu or auto? I put on the cpu because it is more powerful but is it a good idea?

I read this: https://www.blurbusters.com/faq/motion-blur-reduction/ It answers my questions but I would like like confirmation, thank you :D

This negative effect here: https://www.blurbusters.com/wp-content/ ... icates.png is only applicable to ULMB? G-Sync/fix are not concerned?

Which means that I have to put my screen in 144hz with a fps cap at 144hz in game? or 141hz to have the -3 to gain input lag? or the latter corresponds to G-Sync? A little lost:/
G-Sync is good in all cases. It doesn't matter if the game is optimized or not. G-Sync removes tearing, and it also decreases stuttering. It does not have input lag like vsync does.
Ok so even the worst game, G-Sync is better than the fixed frequency.

User avatar
RealNC
Site Admin
Posts: 3741
Joined: 24 Dec 2013, 18:32
Contact:

Re: G-Sync vs ULMB vs Fixed frequency

Post by RealNC » 15 Oct 2018, 21:55

Sayu wrote:Because the fixed frequency causes stutters and tearing if the fps are below or above 240hz? Or just below, maybe? The input lag gained unlike G-Sync is almost zero, right?
Fixed frequency has will have stutter when your FPS is not equal to your refresh rate.
Yes I noticed that the ULMB is very interesting because during close fights which are quite nervous, the image is much more fluid and less blurry but I have a little more microstutters.
Correct. With ULMB, it's best to use vsync and make sure your FPS is the same as your refresh rate. If play play at 120Hz ULMB, your FPS needs to be 120 and you need to use vsync to eliminate microstutters.
1/ A weak gpu unlike a big cpu? Because I noticed that I had more when I'm in low quality so use of the cpu unlike in high quality so use of the gpu at 100%, what do you think?

2/ The fact that the fps drop below or above 144? Or only underneath, again?
I don't understand these questions :P
3/ The fact of putting phys x on the cpu or auto? I put on the cpu because it is more powerful but is it a good idea?
I don't think it makes a difference in most games. "Auto" should be fine.
This negative effect here: https://www.blurbusters.com/wp-content/ ... icates.png is only applicable to ULMB? G-Sync/fix are not concerned?
Yes. This is a problem that happens only when using ULMB.
Which means that I have to put my screen in 144hz with a fps cap at 144hz in game? or 141hz to have the -3 to gain input lag? or the latter corresponds to G-Sync? A little lost:/
With ULMB, if you don't use vsync, then you need as much FPS as you can get. No FPS cap. However, if you want less microstutter, then you need to use vsync, and cap your FPS to 144. Or even better, cap your FPS to 143.8FPS, to improve input lag. See my guide here.

With g-sync, you should use 240Hz and cap to whatever you want, up to 237FPS. So you can cap to 30, 50, 60, 80, 120... up to 237. It's up to you what you want.
Ok so even the worst game, G-Sync is better than the fixed frequency.
Yes.
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

Sayu
Posts: 6
Joined: 14 Oct 2018, 21:50

Re: G-Sync vs ULMB vs Fixed frequency

Post by Sayu » 16 Oct 2018, 02:28

If I understand everything:

1/ RESUME

Fixed Frequency :


So to avoid having tearing, stuttering & microstuttering in fixed frequency, it is only necessary to synchronize the refresh rate to the frame rate? But it requires a very powerful graphics card because i have to use a minimum of 240 fps and not an average or maximum of 240 fps. Hello the RTX 2080ti at 1300€ :cry:

G-Sync :

But since I have an mid-range graphics card, the G-Sync is actually useful. So if I understood correctly the best would be G-Sync + cap 237 for the least inputlag without V-Sync or cap 240. According to the video below, the inputlag is the same as V-Sync OFF. Beautiful. https://youtu.be/rs0PYCpBJjc?t=310

ULMB :

Monitor at 144Hz + Cap at 144Fps and not 141 fps this time. If you have micro-stutters, activate V-Sync or no cap/fps max.

Legend :

Based on these sources :
https://www.blurbusters.com/faq/motion-blur-reduction/
https://www.testufo.com/stutter

Microstuttering : Need to have a better graphics card (probably to make sure you're not below the refresh rate or not have a bottleneck. I mention this in point number 2), a better gaming mouse (it's already done), synchronize frame rates to the display's refresh rate (using frame capping or using VSYNC ON).
Stuttering : Do not be below the refresh rate.
Tearing : Do not be above the refresh rate.

2/ I wanted to ask if having a very powerful cpu and a weak gpu and putting everything on the cpu could make microstuttering. A story of a bottleneck. I am interested in that.

3/ Okay, thanks for PhysX on auto, the duplicate images, the worst of optimized games & your guide that I will apply for the ULMB. For G-Sync you don't have to be as precise as I can cap as I want.

4/ frame rate (fps)= refresh rate (hz) = strobe rate (What is it?)

5/ For G-Sync we aim at -3 fps of the refresh rate to lose inputlag. Why not do it for the ULMB as well?

6/ The inputlag decreases more and more with a high refresh rate (240) and capfps (237) even if my fps are lower (120/180)?

Thank you infinitely for all these answers.

User avatar
RealNC
Site Admin
Posts: 3741
Joined: 24 Dec 2013, 18:32
Contact:

Re: G-Sync vs ULMB vs Fixed frequency

Post by RealNC » 16 Oct 2018, 13:33

Sayu wrote:So to avoid having tearing, stuttering & microstuttering in fixed frequency, it is only necessary to synchronize the refresh rate to the frame rate? But it requires a very powerful graphics card because i have to use a minimum of 240 fps and not an average or maximum of 240 fps. Hello the RTX 2080ti at 1300€ :cry:
You can avoid tearing in fixed frequency by using vsync or fast sync. But there is no reason to do that, since you have a g-sync monitor.
But since I have an mid-range graphics card, the G-Sync is actually useful.
G-Sync is useful regardless of what graphics card you have. You can have 1050 Ti or a 2080 Ti. G-Sync is just as useful in both cases.
So if I understood correctly the best would be G-Sync + cap 237 for the least inputlag without V-Sync or cap 240.
Not 240. Always 237 or lower. Your FPS limit can be somewhere between 1 and 237.
Tearing : Do not be above the refresh rate.
No. Tearing will always happen if you don't use g-sync, vsync or fast sync. Even if you get 144FPS at 144Hz, if you don't use vsync or fast sync, you will have tearing.
2/ I wanted to ask if having a very powerful cpu and a weak gpu and putting everything on the cpu could make microstuttering. A story of a bottleneck. I am interested in that.
If you have a CPU bottleneck, you can get stuttering. It's called "CPU stutter." This is usually not the case with GPU bollenecks. So it's better to have a GPU bottleneck rather than a CPU bottleneck.
4/ frame rate (fps)= refresh rate (hz) = strobe rate (What is it?)
Strobe rate is the rate at which the backlight is flickering in motion blur reduction (MBR) mode. At 120Hz, the backlight will flicker 120 times per second. That means Hz=Strobe rate. There are some monitors out there that cannot strobe at 60Hz. In these monitors, if you try to use 60Hz + MBR, they will instead use 120Hz backlight strobing. So Hz is NOT equal to strobe rate. This is not good.

This can only happen at 60Hz though. Also, ULMB does not have this problem. G-Sync monitors do not allow you to use ULMB at 60Hz to begin with, so there is not problem. Even if you do tweak the settings to enable ULMB at 60Hz, it will be OK because you will 60Hz strobing, not 120Hz strobing.

Long story short: don't worry about it.
5/ For G-Sync we aim at -3 fps of the refresh rate to lose inputlag. Why not do it for the ULMB as well?
Because 141FPS at 144Hz ULMB has microstutter. If that's OK with you, then you can use a 141FPS cap.

G-Sync does not have this problem.
6/ The inputlag decreases more and more with a high refresh rate (240) and capfps (237) even if my fps are lower (120/180)?
Yes.
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

Sayu
Posts: 6
Joined: 14 Oct 2018, 21:50

Re: G-Sync vs ULMB vs Fixed frequency

Post by Sayu » 17 Oct 2018, 16:14

You can avoid tearing in fixed frequency by using vsync or fast sync. But there is no reason to do that, since you have a g-sync monitor.

No. Tearing will always happen if you don't use g-sync, vsync or fast sync. Even if you get 144FPS at 144Hz, if you don't use vsync or fast sync, you will have tearing.
Yes, I was asking the question for the others people having only the fixed frequency option, i want to learn, again ^^

To summarize, in permanency FPS = HZ can fix the micro-stuttering & stuttering but not tearing even with a fps cap. I thought the opposite because on the site here: https://www.testufo.com/stutter#demo=te ... 00&pps=720 It created the tearing by having 150 fps, either above the refresh rate. The site have wrong ?
G-Sync is useful regardless of what graphics card you have. You can have 1050 Ti or a 2080 Ti. G-Sync is just as useful in both cases.

Not 240. Always 237 or lower. Your FPS limit can be somewhere between 1 and 237.
Personally, I think it helps the little gpu more because having 144fps at all times with ULMB or 240fps in fixed frequency are very hard without paying 700/1300€. So yes, the G-Sync
is fabulous because we don't have to worry about fps losses, tearing, stuttering, etc... even with a small gpu while having a small inputlag thanks to the 240hz/237capfps. But I agree that it is also useful for big gpu because G-Sync has other positive effects.

Yes, I misspoke, cap 237fps have less inputlag than V-Sync ON or OFF with cap 240fps.
Strobe rate is the rate at which the backlight is flickering in motion blur reduction (MBR) mode. At 120Hz, the backlight will flicker 120 times per second. That means Hz=Strobe rate. There are some monitors out there that cannot strobe at 60Hz. In these monitors, if you try to use 60Hz + MBR, they will instead use 120Hz backlight strobing. So Hz is NOT equal to strobe rate. This is not good.

This can only happen at 60Hz though. Also, ULMB does not have this problem. G-Sync monitors do not allow you to use ULMB at 60Hz to begin with, so there is not problem. Even if you do tweak the settings to enable ULMB at 60Hz, it will be OK because you will 60Hz strobing, not 120Hz strobing.

Long story short: don't worry about it.
Ok so whether I am in 144hz or 120hz in ULMB my strobe rate will be the same as my refresh rate. So I just have to worry about the cap fps + don't go below of the refresh rate with the fps.
If you have a CPU bottleneck, you can get stuttering. It's called "CPU stutter." This is usually not the case with GPU bollenecks. So it's better to have a GPU bottleneck rather than a CPU bottleneck.
80 or 100% gpu use have the same 0 impact ?

40 to 80% cpu use have the same 0 impact ? Is it from 100% use that I get the cpu shutter or before ?

PS : Should I use V-Sync or Fast Sync with G-Sync ? Why yes ? Why no ?
PS² : But if I have 160 fps with G-Sync, it's my monitor that goes down to 160hz ? So I'll have the inputlag of a 160hz ?
PS3 : On your guide with the frequency at -0.01 of the true refresh rate, RTSS is mandatory or the one on game works ?

THANK YOU VERY MUCH AGAIN <3
Last edited by Sayu on 27 Oct 2018, 17:10, edited 1 time in total.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: G-Sync vs ULMB vs Fixed frequency

Post by Chief Blur Buster » 17 Oct 2018, 16:38

Sayu wrote:Yes I noticed that the ULMB is very interesting because during close fights which are quite nervous, the image is much more fluid and less blurry but I have a little more microstutters.
For ULMB, you want to optimize all microstuttering weak links as much as you can.

Including mouse microstuttering and framerate-vs-refreshrate harmonic/beatfrequency microstuttering.
Fixing Amplified Microstuttering Visibility From Blur Reduction.

Normally the tiny microstutters (of fixed-Hz) are hidden by motion blur. But with blur reduction (ULMB), all the tiny microstutters become visible.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
RealNC
Site Admin
Posts: 3741
Joined: 24 Dec 2013, 18:32
Contact:

Re: G-Sync vs ULMB vs Fixed frequency

Post by RealNC » 17 Oct 2018, 18:01

Sayu wrote:To summarize, in permanency FPS = HZ can fix the micro-stuttering & stuttering but not tearing even with a fps cap. I thought the opposite because on the site here: https://www.testufo.com/stutter#demo=te ... 00&pps=720 It created the tearing by having 150 fps, either above the refresh rate. The site have wrong ?
What do you mean? Tearing is always there. FPS doesn't matter.
But I agree that it is also useful for big gpu because G-Sync has other positive effects.
No GPU on the market can give you 240FPS in the vast majority of modern games. Not even an RTX 2080 Ti. Only very few modern games can run at 240FPS, unless you play at low resolution and/or low graphics settings.
80 or 100% gpu use have the same 0 impact ?
100% GPU load will usually not result in stutter. 100% CPU load might have stutter.
40 to 80% cpu use have the same 0 impact ? Is it from 100% use that I get the cpu shutter or before ?
It depends on the game. In some games, 100% CPU load might produce stutter.

There are no fixed rules here. Different games are programmed differently. Some work fine even at 100% CPU load, some are not.
Should I use V-Sync or Fast Sync with G-Sync ? Why yes ? Why no ?
Use either vsync ON, or vsync OFF. Do not use fast sync, since it does not make any sense when you are using an FPS limiter.

Some games work better with vsync OFF (like Witcher 3), some work better with vsync ON (like Prey.) It depends on the game. If you see some small tearing, use vsync ON. If you never see tearing, you can use vsync OFF.

As a general rule, I start with vsync OFF when playing a new game. If I start seeing tearing, I switch to vsync ON.
But if I have 160 fps with G-Sync, it's my monitor that goes down to 160hz ? So I'll have the inputlag of a 160hz ?
No. You will have the input lag of 160FPS. The monitor will still display each frame at the speed of 240Hz (4.1ms per frame), not 160Hz (6.25ms per frame.) So the monitor will run at 160Hz (it will display 160 images per second), but each image will only need 4.1ms to be displayed.

Monitors do not display the whole image at once. It takes time. The monitor starts displaying pixels at the top left of the screen, and keeps displaying pixels line by line, until it reaches the bottom of the screen. At 240Hz, this takes 4.1ms. In G-Sync mode, the monitor will always take 4.1ms to display a frame, even if FPS is lower than 240FPS.

This is why a G-Sync monitor with higher Hz has lower input lag even while playing 60FPS or even 30FPS games.
PS3 : On your guide with the frequency at -0.01 of the true refresh rate, RTSS is mandatory or the one on game works ?
You need RTSS. In-game limiters are not accurate enough.
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

Post Reply