Blur Buster's G-SYNC 101 Series Discussion

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.
User avatar
jorimt
Posts: 2481
Joined: 04 Nov 2016, 10:44
Location: USA

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by jorimt » 24 Jan 2020, 11:27

Kaldaien wrote:
22 Jan 2020, 21:53
I also would love to know how my design compares with RTSS, nobody in my userbase has the hardware to do FCAT analysis.
My current high-speed setup can't test for frametime performance, but I can get an input lag baseline with relatively few samples. I recently did a couple of (very) casual tests comparing RTSS against Nvidia's NULL auto-limiter in Apex, and they appeared to have the same input lag increase over the in-game limiter in that game:
viewtopic.php?f=5&t=5903&start=110#p45250

The scenario in the above link isolates the input lag difference between limiters by using G-SYNC + V-SYNC + an FPS limit just within the G-SYNC range, and since every frame update first occurs at the top of the scanout, I simply capture the top left of the display and watch for the change on the given vertical element on-screen (in this case, a pole) after strafe (mapped to left click).

I only did 10 samples per for these (compared to the 40 per I did for my article), but it's enough to tell how many more frames of input lag the external limiter has over the internal limiter (without me going to an insane amount of work).

Honestly, if your limiter's method is comparable to RTSS (haven't looked into this myself), then it will likely be the same (up to 1 frame more input lag than an in-game limiter).

External limiters are typically anywhere from up to 1-2 frames more input lag than an in-game limiter, with the worst I've tested for G-SYNC (where input lag is concerned) being the legacy Nvidia Inspector driver-level limiter, with up to a 3 1/2 frames input lag increase over the in-game limiter.

Anyway, if I get the chance (no guarantee when that will be), I'd just do a quick 10 sample test in Apex and add it to the existing charts in the link I posted above. If you have any specific recommendations for optimal FPS limiter parameters in your tool for this purpose, feel free to share them (no rush).
(jorimt: /jor-uhm-tee/)
Author: Blur Busters "G-SYNC 101" Series

Displays: ASUS PG27AQN, LG 48CX VR: Beyond, Quest 3, Reverb G2, Index OS: Windows 11 Pro Case: Fractal Design Torrent PSU: Seasonic PRIME TX-1000 MB: ASUS Z790 Hero CPU: Intel i9-13900k w/Noctua NH-U12A GPU: GIGABYTE RTX 4090 GAMING OC RAM: 32GB G.SKILL Trident Z5 DDR5 6400MHz CL32 SSDs: 2TB WD_BLACK SN850 (OS), 4TB WD_BLACK SN850X (Games) Keyboards: Wooting 60HE, Logitech G915 TKL Mice: Razer Viper Mini SE, Razer Viper 8kHz Sound: Creative Sound Blaster Katana V2 (speakers/amp/DAC), AFUL Performer 8 (IEMs)

1000WATT
Posts: 391
Joined: 22 Jul 2018, 05:44

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by 1000WATT » 25 Jan 2020, 04:54

Kaldaien wrote:
22 Jan 2020, 21:53
*sigh* when will this end?
574.jpg
574.jpg (62.08 KiB) Viewed 7772 times
Pangs of conscience. Raskolnikov is it you? 8-)
I often do not clearly state my thoughts. google translate is far from perfect. And in addition to the translator, I myself am mistaken. Do not take me seriously.

1000WATT
Posts: 391
Joined: 22 Jul 2018, 05:44

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by 1000WATT » 29 Jan 2020, 03:53

Since jorimt did not test the version of the "not recommended" synchronization that I prefer. But now is not about that.
And conscience did not allow me to ask for testing and spend a lot of time for the sake of 1 person. I bought a Nikon 1 j1 for $ 35.
choosing a testing methodology, I ran into problems.

Quote from the article. https://www.blurbusters.com/gsync/preview2/
"Conveniently, one can calculate the difference in frame numbers, between the LED illumination and the first gun reaction (whatever the first on-screen reaction is; such as a crosshairs animation, or gun moving upwards right before muzzle flash)."
whatever the first on-screen reaction is- very logical, let's get started.

Battle (non) sense used flash animation after a shot for its tests.
I tried to repeat but the animation was different every time and the results were very vague.
I also tried using a bullet counter as a marker. Already better, but it is located at the very bottom of the monitor and if you compare the delay g-sync and fixed hz, the difference will only increase. This marker was also abandoned.

jorimt method is "Left click was mapped to strafe left"

Next I tried the ahk script.
(LButton::
MouseMove, 0, -704, 0, R
return
Pause::Suspend)

It looks like this https://www.youtube.com/watch?v=bWHTDE27KM0 In my version, the LED is powered from the mouse and turns off when the left button is pressed. I also removed the lens so that the sight does not move randomly.

Problems arose on the following option. As a marker, I used the menu (https://imgur.com/l7hO2wC) and attached it to the left mouse button.
When in adobe premiere counted frames, I noticed another marker, namely the cursor.
Depending on the limiter (ingame / rtss), the cursor appeared either earlier than the menu or later.
I decided to measure the delay between two markers and between two limiters.

results (40 passes)
jorimt method g-sync
ingame= 32,27083333
rtss==== 45,97916667
diff==== -13,02083333

2 markers
g in menu== 38,70833333
g in cursor= 46,29166667
diff======= -7,583333333

g rtss menu== 52,66666667
g rtss cursor= 46,0625
diff======== 6,604166667

The idea that the cursor marker has the same delay with different limiters pushed me to record video through OBS(well, what if :lol: ). Top ingame Bottom rtss. Tracks are synchronized by cursor.https://www.youtube.com/watch?v=vnnpeLhXkrw
It’s not as accurate as a 1000fps camera but it makes you think.
I often do not clearly state my thoughts. google translate is far from perfect. And in addition to the translator, I myself am mistaken. Do not take me seriously.

1000WATT
Posts: 391
Joined: 22 Jul 2018, 05:44

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by 1000WATT » 29 Jan 2020, 04:09

Forgot to say if someone wants to check. The cursor does not appear when you first press it when entering the game or alt + tab, before recording you need to click 1-2 times and you will see the cursor.
I often do not clearly state my thoughts. google translate is far from perfect. And in addition to the translator, I myself am mistaken. Do not take me seriously.

janos666
Posts: 16
Joined: 15 Dec 2017, 18:50

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by janos666 » 30 Jan 2020, 12:52

I have a stupid question (I think I should know the answer but I seem to have forgotten the details which lead to it):
Do external framerate limiters (nV V3, RTSS, etc) add any delay (simply from being hooked into the chain) while the fps is below the cap (due to CPU and/or GPU performance limitations) or does this delay appear only when they start throttling the framerate?
In other worlds: Is it potentially beneficial to disable the limiter or even V-Sync as well in case the quality settings will keep the fps inside the VRR range >95% of the time anyways (say, a very hardware demanding but slow paced offline game averaging around 60 fps on a 120+ Hz display)?

janos666
Posts: 16
Joined: 15 Dec 2017, 18:50

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by janos666 » 30 Jan 2020, 14:03

Kaldaien wrote:
22 Jan 2020, 21:53
Unfortunately, I have recently fallen out of favor with G-Sync and VRR in general. My experience with LG's 2019 4K OLEDs and their Black Frame Insertion mode has been eye opening. As we are all no doubt aware, these are effectively mutually exclusive technologies due to frame persistence. For me, G-Sync was of interest only for its ability to mitigate frame pacing drawbacks associated with dropped frames at the resolutions and quality settings I play games. I am much too old to be sensitive to input latency, so a locked 60 Hz BFI signal is good enough for me and looks better than trying to brute force motion blur away with more frames :P
Oh. I was a plasma fan for a long time (ehh... I was in love with my Pioneer 9G Kuro!) because of it's motion resolution (it's similar to native 180fps if I recall the active/black duty cycle numbers correctly at ~33% or similar). Then I used the 60+60 BFI on the C8 (with "low latecy V-Sync" or FastSync, depending on the game and the average uncapped framerate my PC could reach in the given title). But on the C9, native 120fps (1440p120 until I get a VGA card with HDMI 2.1) has the same motion resolution as 60+60 BFI while offering VRR at the same time. I prefer VRR because it's easier to keep the framerate between 40 and 120 than at a steady 60 and it's a lot smoother experience when you can average significantly higher than 60 (the best is a steady ~117 or course but that's obviously even harder than a steady 60 --- but with VRR, it's enough to average high, no need to have a lot of headroom for keeping it steady). I am happy to pay a bit of motion resolution here and there for a significantly smoother motion and lower latency (at least when the average is well above 60).

User avatar
jorimt
Posts: 2481
Joined: 04 Nov 2016, 10:44
Location: USA

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by jorimt » 30 Jan 2020, 14:12

janos666 wrote:
30 Jan 2020, 12:52
I have a stupid question (I think I should know the answer but I seem to have forgotten the details which lead to it):
Do external framerate limiters (nV V3, RTSS, etc) add any delay (simply from being hooked into the chain) while the fps is below the cap (due to CPU and/or GPU performance limitations) or does this delay appear only when they start throttling the framerate?
The latter as far as I'm aware.

As for any passive effects of the hook itself when the FPS isn't being limited, some have claimed it may still affect mouse response or system performance, but I haven't found any compelling evidence of that myself.
janos666 wrote:
30 Jan 2020, 12:52
In other worlds: Is it potentially beneficial to disable the limiter or even V-Sync as well in case the quality settings will keep the fps inside the VRR range >95% of the time anyways (say, a very hardware demanding but slow paced offline game averaging around 60 fps on a 120+ Hz display)?
While an FPS limit is entirely optional in this scenario (even though it can be argued you may hit a higher FPS in menus, so it could still be beneficial to retain the FPS limit for reduced input lag during navigation), if you want zero tearing, V-SYNC (with VRR) isn't; with G-SYNC + V-SYNC "Off," for instance, even if your average FPS is well within the refresh rate, it can still tear during frametime spikes.
(jorimt: /jor-uhm-tee/)
Author: Blur Busters "G-SYNC 101" Series

Displays: ASUS PG27AQN, LG 48CX VR: Beyond, Quest 3, Reverb G2, Index OS: Windows 11 Pro Case: Fractal Design Torrent PSU: Seasonic PRIME TX-1000 MB: ASUS Z790 Hero CPU: Intel i9-13900k w/Noctua NH-U12A GPU: GIGABYTE RTX 4090 GAMING OC RAM: 32GB G.SKILL Trident Z5 DDR5 6400MHz CL32 SSDs: 2TB WD_BLACK SN850 (OS), 4TB WD_BLACK SN850X (Games) Keyboards: Wooting 60HE, Logitech G915 TKL Mice: Razer Viper Mini SE, Razer Viper 8kHz Sound: Creative Sound Blaster Katana V2 (speakers/amp/DAC), AFUL Performer 8 (IEMs)

verllar
Posts: 19
Joined: 16 Sep 2017, 11:43

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by verllar » 13 Feb 2020, 14:04

If i enable "max frame rate to 162fps, and In-game 162fps cap on 165hz monitor will it give me more input lag than only In-game?

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

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by RealNC » 17 Feb 2020, 10:52

verllar wrote:
13 Feb 2020, 14:04
If i enable "max frame rate to 162fps, and In-game 162fps cap on 165hz monitor will it give me more input lag than only In-game?
Nobody knows. So just don't do that if you care that much about input lag.
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.

hmukos
Posts: 30
Joined: 16 Apr 2020, 15:41

Re: Blur Buster's G-SYNC 101 Series Discussion

Post by hmukos » 16 Apr 2020, 15:58

So I stuck on the G-Sync ON + V-Sync OFF bottom tearing problem and after some googling found this thread. So if I understood correctly, limiters like RTSS measure frametimes just before the GPU starts rendering the frame. So if GPU time varies RTSS won't capture it and as a result frames would output to screen at different frametimes and when the frametime is slightly below display's limit it would tear at the bottom.

My question is, why can't RTSS measure frametimes right after GPU has done rendering and only then delay displaying the frame? And if this is not possible than how do VR games and headsets achieve perfect frametime and why can't it be done the same in flat games?

Post Reply