RTSS VS Ingame limiter, GPU Usage question And using Gsync

Everything about input lag. Tips, testing methods, mouse lag, display lag, game engine lag, whole input lag chain, VSYNC OFF vs VSYNC ON, and more! Input Lag Articles on Blur Busters.
Post Reply
Daviii97
Posts: 3
Joined: 01 Nov 2019, 12:10

RTSS VS Ingame limiter, GPU Usage question And using Gsync

Post by Daviii97 » 01 Nov 2019, 12:36

Hello everybody, Now i know that RTSS is input lag neutral and some Ingame limiters are actually better to lower input lag but my question is, is there A list or did somebody test already which ingame limiters are actually worth using over RTSS ? for example in the new COD MW the ingame frame limiter doesnt work at all so its a no brainer but in bf5 rainbow six siege and other games when i tested both i couldnt really tell a difference so which one should i use ?

Also another Question that has been bothering me since i watched the Battlenonsense input lag video, Is the change in input lag even noticeble if you cap your framerate so your gpu is under 90% ? I honestly tested it many times and in COD for example if i set the cap at 162 for Gsync it varies from 110-140 and my GPU usage is at 97% constant even with all max it never gets to 99% which is the first thing thats weird becasue i run a 980ti and with everything on max i get like 70fps but not 99% gpu usage.My i9 9900k is at around 20%. If i set my fps cap at 110 my usage drops to around 83-85 but it feels worse then having the framerate unlocked.So i didnt really get what the best case for low input lag gsync is after watching that video, cap fps for low gpu usage or it isnt that big of a deal? i also run the new NULL ULL mode and was wondering if it is even worth at 95-97% Gpu Usage ? i read somewhere its only worth at 99% and adds input lag at lower then that ?

And lastly since i got this monitor a year ago i went through periods where i was debating to use or not to use Gsync, I really like the tear free exeprience and also less microstutter with Gsync But its weird that i can feel a slight worse input lag with gsync on then with off Like its the Feeling that my sensitivity of my mouse is a bit faster and Mouse feel a bit lighter with gsync off then on, is that even input lag that im talking about ? and with fixed refresh the picture quality just doesnt look as steady and i can feel weird micro stutter sometimes. im so confused with all of this And i feel like im wasting half of the monitor if i play with fixed refresh. Sorry for a lot of different Questions in one post but ive been trying to tweak this for months and cant get it "feeling right"

User avatar
jorimt
Posts: 830
Joined: 04 Nov 2016, 10:44

Re: RTSS VS Ingame limiter, GPU Usage question And using Gsy

Post by jorimt » 01 Nov 2019, 13:54

First off, to greatly simplify this for you (and anyone else not familiar with these concept that may happen to be reading this), here's a slow-mo video depicting what the display is doing per frame:

phpBB [video]


That's quite literally it.

Each of those sweeps you're seeing are the display completing a single refresh. Each of these refreshes are called a "scanout cycle." The higher the refresh rate, the more of these are repeated per second.

Anything rendered from the system has to complete display in the span of one of these cycles, repeat.

With any syncing method (V-SYNC, G-SYNC, FreeSync, Fast Sync, Enhanced Sync, you name it), the goal is to match a single rendered frame to a single scanout cycle per.

However, with V-SYNC OFF, depending on how high or low the framerate is, more than one rendered frame will be scanned into a single scanout cycle, which causes the tearing you see, as the partially rendered frame(s) below the previous frame(s) is/are offset visually due to being more recent, while, again, all being displayed within that same single scanout cycle (e.g. where the frames(s) should have begun at the top of the monitor, they've begun anywhere between the top and the bottom of the monitor instead).

Optimally a single rendered frame should display per single scanout cycle, in perfect sequence and in real-time; render, display, render, display, render, display, and so on.

However, many things get in the way of this, and the "why" is almost infinitely varied, so I won't break it down here, but, basically, a rendered frame can ultimately be:

  • Interrupted (aka "torn")
  • Skipped
  • Dropped
  • Repeated
  • Delayed
And that, as well, is quite literally it.

Anyway, to specifically get to your questions:
Daviii97 wrote:Now i know that RTSS is input lag neutral and some Ingame limiters are actually better to lower input lag but my question is, is there A list or did somebody test already which ingame limiters are actually worth using over RTSS?
Short answer:
No.

Long answer:
Most in-game limiter's have lower input lag, but have more frametime drift than RTSS. Some have this drift without affecting G-SYNC performance negatively, while others do (and some only have presets; 30, 60, 120, etc). In other words, if the in-game limiter in question is visibly (and frequently) causing more stutter than RTSS (by eye, not by graph, mind you), even with G-SYNC, then it's probably not a very stable/good in-game limiter.

Beyond that, there isn't an easy way to tell. That said, RTSS (while, yes, neutral) has, at worst, 1 more frame of input lag than the best in-game limiters, and even then, it's usually more stable.
Daviii97 wrote:Is the change in input lag even noticeble if you cap your framerate so your gpu is under 90% ?
When your system is GPU-bound (this is separate of and not counting sync-induced input lag), you can, at best, typically further reduce your input lag by 1 frame when using a framerate limiter (by keeping it at a framerate where the GPU usage doesn't max out) and/or Low Latency Mode (in cases where the current framerate is, typically, both uncapped AND causing max GPU usage).

Regarding the pre-rendered frames queue specifically, it's all highly situational, and the effects of Low Latency Mode settings (previously known as "Maxium pre-rendered frames"), as well FPS limiters, often depends on the system specs, the system's load, the game engine, the game API (DX11, DX12, OpenGL, Vulkan, etc), and the framerate at any given point.
Daviii97 wrote:i can feel a slight worse input lag with gsync on then with off Like its the Feeling that my sensitivity of my mouse is a bit faster and Mouse feel a bit lighter with gsync off then on, is that even input lag that im talking about ?
Within its range, and at the same framerate, the only "lag" G-SYNC causes over "V-SYNC OFF" is the lack of tearing.

As previously mentioned, G-SYNC displays a single frame per scanout, and V-SYNC OFF shows (usually "parts" of) more than one frame (thus the previous mention of "interruption") per scanout (the simultaneous amount of which is dependent on the refresh rate/framerate ratio). As such, yes, there can be a different feel.
Daviii97 wrote:and with fixed refresh the picture quality just doesnt look as steady and i can feel weird micro stutter sometimes. im so confused with all of this And i feel like im wasting half of the monitor if i play with fixed refresh.
If G-SYNC is available to you, and you're not playing with standalone V-SYNC OFF, then fixed refresh is certainly inferior, especially if you're using standalone V-SYNC, which will cause the micro-stutter you mentioned due to recurring repeated frames with framerates below the refresh rate (and sync-induced input lag due to delayed frame delivery with framerates above the refresh rate).
Author: Blur Busters "G-SYNC 101" Series

Display: Acer Predator XB271HU OS: Windows 10 Pro MB: ASUS ROG Maximus X Hero CPU: i7-8700k GPU: EVGA GTX 1080 Ti FTW3 RAM: 32GB G.SKILL TridentZ @3200MHz

Daviii97
Posts: 3
Joined: 01 Nov 2019, 12:10

Re: RTSS VS Ingame limiter, GPU Usage question And using Gsy

Post by Daviii97 » 01 Nov 2019, 14:41

Thank you for this highly detailed Answer this is why i love this website so much. So i guess it is just getting used to the tearing free experience gsync offers. Also what i meant with playing fixed refresh is just uncapped frames all vsync off and no gsync.

But if Gsync doesnt input more lag at the same framerate why do so many people still dont reccomend Gsync for competitive games ? isnt it theroetically better to have no microstutters and a clear picture at 162 frames a second at 165hz then lets say having a few miliseconds less input lag from running a game at 300FPS that you cant notice anyway? is it just general misinformation regarding this topic?

Also to sum it up the Best Settings are Gsync ON + Vsync ON in nvidia control panel + RTSS at -3 from refreshrate + ULL Ultra if your system is capable enough to eliminate gpu buffer (since i dont feel any stutters ingame with it on Ultra i guess mine is).

User avatar
jorimt
Posts: 830
Joined: 04 Nov 2016, 10:44

Re: RTSS VS Ingame limiter, GPU Usage question And using Gsy

Post by jorimt » 01 Nov 2019, 15:14

Daviii97 wrote:Also what i meant with playing fixed refresh is just uncapped frames all vsync off and no gsync.
Tearing with V-SYNC OFF (especially with FPS high above the refresh rate) can mimic the look/feel of micro-stutter.
Daviii97 wrote:But if Gsync doesnt input more lag at the same framerate why do so many people still dont reccomend Gsync for competitive games ? isnt it theroetically better to have no microstutters and a clear picture at 162 frames a second at 165hz then lets say having a few miliseconds less input lag from running a game at 300FPS that you cant notice anyway? is it just general misinformation regarding this topic?
Some of it is probably due to misinformation and/or a lack of familiarity with G-SYNC, but either way, much of it comes down to preference and conditioning; again, G-SYNC can simply feel different and be hard to adjust to after being accustomed to V-SYNC OFF + uncapped FPS, regardless of whether the former can be a more consistent experience.
Daviii97 wrote:Also to sum it up the Best Settings are Gsync ON + Vsync ON in nvidia control panel + RTSS at -3 from refreshrate + ULL Ultra if your system is capable enough to eliminate gpu buffer (since i dont feel any stutters ingame with it on Ultra i guess mine is).
Pretty much, though I'd note "RTSS" OR (good) in-game limiters "-3 from" refresh rate. As for Ultra, it's game-specific, but yes, if you're not experiencing increased stutter when using it (otherwise, try dropping it to "On").
Author: Blur Busters "G-SYNC 101" Series

Display: Acer Predator XB271HU OS: Windows 10 Pro MB: ASUS ROG Maximus X Hero CPU: i7-8700k GPU: EVGA GTX 1080 Ti FTW3 RAM: 32GB G.SKILL TridentZ @3200MHz

Post Reply