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 » 23 Sep 2017, 09:57

Regarding the recent RTSS conversation, the whole "RTSS adds" or "RTSS reduces" or "RTSS is neutral" delineation is highly relative to what we're talking about in any given scenario, and it's difficult to apply any one of those as a blanket statement.

I think more practical testing needs to be done before the effects of RTSS can more accurately be defined in just words, especially when being conveyed to the layman.
(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)

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

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

Post by Chief Blur Buster » 23 Sep 2017, 11:31

Hey RealNC/Sparky

Einstein is relative.

Elements of both of your post read-out as technically correct depending on "Relative POV".
Simultaneously, both of you can be incorrect if it's a different relative basis.

As Einstien says, it's all relative. While lag does go up, one has to be specific about it: "lag of THIS specific element of chain" or "lag of the entire chain". Our high speed camera tests showed that while RTSS increases lag of a specific part of the chain, but CAN reduce total button-to-pixels lag. One sorta, gotta, be specific about frame of lag reference -- it's all relative.

Our crude, hugely simplified, many-items-omitted, incomplete diagram, shows lag is still a very complex topic.
Image

Now, briefly, on a separate topic: Lag consistency
Anyway, sometimes the absolute lag number isn't the whole story.... For those of us who remember the old bad "SLI"/"CrossFire" frame pacing days -- good frame pacing is also like theoretically using driverless cars to make traffic smoother (and a bit faster too) with cars following closer to each other without constant stop/go. What I'm saying is a situation of "More traffic, but faster/smoother" is technologically possible with clever situations. Anyway in many situations, you can still get better aiming with more input lag (but far less erratic) -- reduced lag error relative to gametime (And frame visibility time) is also an aiming accuracy factor. There are really bad game engines that has erratic lag, making it even "preferably to play CS:GO with VSYNC ON" compared to that, because the erratic lag (even if lower lag) in that other flaky game is so terrible. So we've got absolute-lag and lag-variance as two separate aim-wrecking factors. That said, raising framerate higher (on single GPUs, anyway) often reduces lag jitter, and brute framerate in many games friendly to high frame rate (like CS:GO) is a good hammer to not worry about the difference between "absolute-lag" versus "lag-erraticness". Now, if the game does it properly (GSYNC-friendly games), that's another reason why GSYNC works so well with erratic-frametimes (rapid scene complexity changes, etc). The erratic gametimes/frametimes stay in perfect harmony & sync with erratic refresh-cycle times, as long as the erraticness is not too big (disk access, etc). Seeing erratic framerates be stutterfree, in many cases, often can make it easier to aim in those particular games (obviously, we're talking about games with much more erratic frame rates than CS:GO). But there's cases where lag consistency (including compensating technologies like RTSS or like GSYNC, or even both at same time) fixes things more than absolute average lag. Very game dependant, of course (there's other games than CS:GO that behaves very differently in aiming improvements).
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: Blur Buster's G-SYNC 101 Series Discussion

Post by RealNC » 23 Sep 2017, 12:17

Alright. I'm just saying that there's a trend out there, where people recommend to each other "don't cap your frame rate, especially with RTSS, it adds input lag." And sometimes they link to this thread. And people read it and say "yeah, RTSS adds input lag, leave games uncapped."

(Meanwhile, capping the frame rate with RTSS would actually reduce input lag.)

We're partly the cause of it. If that's OK, I concede. Or maybe you have a better solution than what I propose.
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.

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 » 23 Sep 2017, 12:42

Not to be blunt, but you can't cure stupid.

The paradox is, sometimes the more we attempt to explain or clarify (especially if said clarification is long or wordy, which is what is often unavoidable with this subject material), the less the average user will take the time to read and comprehend it.

I'm constantly repeating or rephrasing myself here and on other forums (which I ultimately don't mind; comes with the territory) on points that would already be perfectly clear if they took the time to thoroughly read the WHOLE article.

It can't always be helped, some people are just lazy (something I can even be guilty of myself when reading other people's material), or prefer their preconceptions to facts, and won't accept anything you say, no matter how technically correct or well phrased it is.

We can only offer so much information before it is up to the individual to get their facts straight with the material already available to them.
(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)

sekta
Posts: 16
Joined: 19 May 2015, 22:19

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

Post by sekta » 23 Sep 2017, 16:31

The way I have interpreted it...

When the bottleneck is the RTSS limiter, rather than a performance bottleneck, Max Prerendered Frames becomes 0, rather than 1+, and so this is the neutral input lag that you should expect.

An ingame limiter does the same, but the limiter can reduce the input lag.

When the framerate is uncapped, and the current fps that you are getting is determined by the performance bottleneck rather than any limiter, you get pre-rendered frames, thus added input lag.

Uncapped = added input lag
RTSS = neutral
Ingame limiter = reduced input lag

If this is correct, then I think this is a much better way to present it. It doesn't portray RTSS as adding more input lag. I've always thought of RTSS as something that lets you have a 'smoother' game, with better frame delivery/presentation, but at the cost of extra input lag (with the extra input lag not being because the fps is lower).

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 » 23 Sep 2017, 18:14

That is the broad general consensus according to our conversations on this thread and my testing of G-SYNC + games with MPRF "1" + RTSS thus far, but much has yet to be learned; I haven't even tested RTSS + V-SYNC OFF above the refresh rate yet, or MPRF values other than "1."

For instance, RTSS only introduces up to 1 frame of delay on average compared to the in-game limiter, which means some samples were showing no added delay over the in-game solution. Further testing would be required, but I'm wondering if, like G-SYNC vs. standalone V-SYNC above the refresh rate (where we are seeing G-SYNC still beat standalone V-SYNC input latency-wise because it drops back within its own range every so often, even with framerates sustained well above the refresh rate), something inversed is happening with RTSS...

Theoretically, what if the delay we are seeing with RTSS is frametime variances causing the framerate to drop below the limit for even mere microseconds at a time, at which point pre-rendered frames kick back in and cause the added delay? One way to test this would be to see if the RTSS delay increased as we up the MPRF value.

That may be a baseless crackpot theory, but it's just a "what if?" example of what we still don't know. "To be continued" basically, and another reason I'm not ready to call or define anything too specifically yet.

What we do know for certain is that the RTSS limiter has more net lag than the in-game limiter, but less than the (current version of the) Nvidia Inspector limiter or uncapped with G-SYNC, V-SYNC, and V-SYNC OFF + MPRF 1 below the refresh rate.

Devoid (even sound) theory, everything else is up in the air until it is actually tested...
(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)

User avatar
kurtextrem
Posts: 41
Joined: 05 Mar 2017, 03:35
Location: Munich, Germany

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

Post by kurtextrem » 24 Sep 2017, 00:19

Sorry, I think I've posted this question at some point, but this new (?) research makes me feel a little uncertain.
You guys are talking about added input lag when NOT using RTSS to cap when having G-Sync on. But G-Sync only has added input lag if fps > refresh rate, right?
So my assumption: If my PC isn't capable of reaching near the refresh rate in a game (e.g. 180 Hz but only 140 fps), no input lag is added even if I don't cap.
Acer XF250Q, R6 competitive player

Sparky
Posts: 682
Joined: 15 Jan 2014, 02:29

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

Post by Sparky » 24 Sep 2017, 01:20

kurtextrem wrote:Sorry, I think I've posted this question at some point, but this new (?) research makes me feel a little uncertain.
You guys are talking about added input lag when NOT using RTSS to cap when having G-Sync on. But G-Sync only has added input lag if fps > refresh rate, right?
So my assumption: If my PC isn't capable of reaching near the refresh rate in a game (e.g. 180 Hz but only 140 fps), no input lag is added even if I don't cap.
In that situation it depends on whether you're cpu or gpu limited. If you're GPU limited, capping framerate just under your minimum framerate can still reduce input lag.

User avatar
kurtextrem
Posts: 41
Joined: 05 Mar 2017, 03:35
Location: Munich, Germany

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

Post by kurtextrem » 24 Sep 2017, 02:42

Capping at 100 fps when I more often than not reach 120-140 is not something I'd see as improvement to be honest (at least not perceived)
Acer XF250Q, R6 competitive player

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 » 24 Sep 2017, 05:26

kurtextrem wrote:Capping at 100 fps when I more often than not reach 120-140 is not something I'd see as improvement to be honest (at least not perceived)
Try capping to 70 when uncapped it's 75. The input lag gets quite lower. Very useful for 1440p and 4K where FPS is much lower.

If there was a frame limiter that did this automatically, that would be awesome. That would mean you'd get an automatic cap to 119 if you reach 120. Unfortunately, nobody ever thought of it before. Maybe we'll see it in RTSS in the future. Suggestions to the author have been made.
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