How these settings are the lowest input latency for CS2?

Everything about latency. This section is mainly user/consumer discussion. (Peer-reviewed scientific discussion should go in Laboratory section). Tips, mouse lag, display lag, game engine lag, network lag, whole input lag chain, VSYNC OFF vs VSYNC ON, and more! Input Lag Articles on Blur Busters.
Vad1us
Posts: 10
Joined: 04 Oct 2014, 10:28

Re: How these settings are the lowest input latency for CS2?

Post by Vad1us » 29 Jan 2025, 17:19

Slender wrote:
22 Jan 2025, 23:49
reflex gonna lower my frames from 360 to 310-320, better disable that and use fps_max 333/0
It happens because reflex requires 0.3ms to work.
So:
1000ms/360hz=2.777777777777778ms
2.78ms+0.3ms=3.08ms
1000ms/3.08ms=324hz - max possible refresh with reflex enabled.

I also wonder best possible scenario for CS2.
Based on this post I changed my RTSS frame cap from Async to NVIDIA reflex.
And I think I like the gameplay most with:
GSYNC ON + VSYNC NVCP ON + ULL NVCP ON + RTSS Frame limiter NVIDIA Reflex + VSYNC in game OFF + Reflex in game OFF + fullscreen

It could be just placebo, but I rly do not like in game settings of CS2. They feel strange.

Also I disabled DSC for my monitor, and without DSC max hz for HDMI2.1 10bit 2560x1440 is 330hz. I can alt+tab much faster. :D :D :D

lithi0m
Posts: 1
Joined: 21 Feb 2025, 15:32

Re: How these settings are the lowest input latency for CS2?

Post by lithi0m » 21 Feb 2025, 15:48

This is just my experience, a few weeks ago i had 13900kf, 4080, 64 GB of ram and i tested out Vsync fast, and Gsync on -Noreflex in LO and game was working fantastic, week ago i switched to 9800x3d, and i try using the same setting, i didn't notice any smoothness in the game until i tuned off vsync in game and NVCO, i turned off gsync and removed -noreflex from LO and my game is working like a space station.

I also noticed HAGS off is a terrible idea on 9800x3d for huge fps loss(for me) but it worked great on my intel setup.

It also makes no sense to me to put fps in NVCP limiter if you have a monster setup, that's like if you drive v10 lambo and you shut down few cylinders and now you have v6 lambo

cuaderno
Posts: 9
Joined: 02 Feb 2025, 20:26

Re: How these settings are the lowest input latency for CS2?

Post by cuaderno » 22 Feb 2025, 10:52

lithi0m wrote:
21 Feb 2025, 15:48
This is just my experience, a few weeks ago i had 13900kf, 4080, 64 GB of ram and i tested out Vsync fast, and Gsync on -Noreflex in LO and game was working fantastic, week ago i switched to 9800x3d, and i try using the same setting, i didn't notice any smoothness in the game until i tuned off vsync in game and NVCO, i turned off gsync and removed -noreflex from LO and my game is working like a space station.

I also noticed HAGS off is a terrible idea on 9800x3d for huge fps loss(for me) but it worked great on my intel setup.

It also makes no sense to me to put fps in NVCP limiter if you have a monster setup, that's like if you drive v10 lambo and you shut down few cylinders and now you have v6 lambo
Hi, are you running the game fully uncapped (fps_max 0)? what res and msaa? what are your ir CPU temps in-game?

The theory was that -noreflex + NVCP limiter gives a much more stable framepacing compared to having reflex enabled or using in-game fps_max. This is easily measurable in capframex. But I can see why uncapped with a ridiculously high fps number from a 9800x3d would also give the same result, as in even if the framepacing is bad you have more than enough fps above refresh rate to even care. In this case the one issue would be CPU temps, so I am curious if you had any troubles there.

Bucking
Posts: 6
Joined: 09 May 2024, 17:10

Re: How these settings are the lowest input latency for CS2?

Post by Bucking » 19 Mar 2025, 18:06

So for G-Sync these settings should be good right

G Sync on in NVCP
V Sync on in NVCP
Low Latency mode on in NVCP

V Sync off in game
Frame limiter off in Game
Reflex off in game (not sure about this one)

FPS cap with RTSS

sk1p
Posts: 132
Joined: 21 Mar 2020, 01:21

Re: How these settings are the lowest input latency for CS2?

Post by sk1p » 20 Mar 2025, 01:57

phaze wrote:
22 Jan 2025, 21:10
So, I've been seeing that people and even CS2 publishing an article recommending to enable: G-Sync + V-Sync + Reflex.

However I tried it, and the game feels smooth and all but the mouse feels a little heavy or slow, with a bit of motion blur introduction.

So which settings are really the best for the lowest latency achievable? I need answers based on facts and testings please, because this is so headaching.


G Sync ON + V Sync OFF + REFLEX ON?

OR

G Sync ON + V Sync ON + REFLEX ON?

OR

G Sync OFF + V Sync OFF + REFLEX ON?


PC Specs: i9 14900K (Disabled HT for minimzing the latency jitter) + Enabled Low latency option for RAM (32GB 7200Mhz CL34)
GPU: 4070 Super
Monitor: XL2546K
Mouse: Razer Viper V3 Pro 4000HZ
Keyboard: Wooting 60HE
I'll put in my 2 cents )

I have level 10 faceit and 22000 in premier. I played on Deathadder v3(wired) last year and I like game feeling and input lag with my settings very much, but week ago I bought Viper v3 pro and I was dissapointed, expected better perfomance in CS, but it became worse. I play with 1.3 sens, 2000hz polling rate, I always delete Razer Synapse after settings as it gives strange feeling of mouse for me, I think without synapse more responsive and better feeling of mouse in game. And what I found with Viper v3 pro - that form I really like, good feeling without mouse cabel, but I don't feel my sens so responsive as with wired Deathadder v3 and I found that I played with Deathadder V3 better !) Because with Viper v3 pro after DA v3 wired I played like a bot sometimes and can not killed enemies in simple cases for me before.. Don't know why, Viper form is better for me, but I connected it only with one case that wireless is good and HyperSpeed pretty responisve, but anyway wired version of 3950(as I know this sensor in theses mouses) with 2000hz and 800 dpi faster.... so I returned Viper 3 pro in the shop and stay with DeathAdder v3 wired ..

Post Reply