input lag at 97% GPU usage vs 90% GPU usage with 240hz

Everything about latency. Tips, testing methods, 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.
Post Reply
forii
Posts: 218
Joined: 29 Jan 2020, 18:23

input lag at 97% GPU usage vs 90% GPU usage with 240hz

Post by forii » 14 Apr 2020, 11:35

Hello,

I can't figure out one thing.

How is the input lag or response time(?) on 240Hz monitor if:
a) we cap FPS at 218 fps and have max 97% gpu load in game

vs

b) capped fps at 140 fps and have max 90% gpu bound in game

Capping at any bellow 218 fps give same max 97% gpu load, so either I have cap fps at 200, 180, 166 in game I still reach 97% gpu usage anyway, that is why I can use 140 fps cap to don't go above 90% gpu usage and have probably less input lagg but also.... less HZ or.. at 218 fps have slight more input lag but have higher Hz..

What is better?

Also...
my avg fps depends of the game mode is 185, and in other mode = 150

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

Re: input lag at 97% GPU usage vs 90% GPU usage with 240hz

Post by Chief Blur Buster » 14 Apr 2020, 11:40

It's not easily answerable without testing -- varies hugely by game, by system, by CPU load, by CPU-vs-GPU load, etc.

Some games can achieve near-100% GPU load without noticeable latency side effects, while other games can suddenly lag down when GPU load falls below 10% in eco mode (because of automatic GPU power management creating slight latency delays). These types of things are ferociously "black box" without testing.

Different games have different ways of dealing with high-load situations, and some do so pretty gracefully, while others do not.

General rules of thumbs are often offered online -- and by YouTubers such as Battle(non)sense and others.

Also, frame rate caps aren't GPU-load-caps. Sudden high-detail areas can push a GPU load to practically 100% no matter how low your frame rate cap is. So a frame rate cap is sometimes only a rough/coarse method of reducing GPU workload. One can use RTSS to try to analyze this data, and do a best-effort cap.
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!

forii
Posts: 218
Joined: 29 Jan 2020, 18:23

Re: input lag at 97% GPU usage vs 90% GPU usage with 240hz

Post by forii » 14 Apr 2020, 11:50

Well, CPU Load is not over than 50%, basically we talk about GPU bound game.

Ah. And I use NULL at ultra, because I heard it is good for gpu bound games, so actually it might be good for 97% gpu usage.

How can I easy measure input lag without any camera?

knypol
Posts: 76
Joined: 18 Aug 2016, 03:40

Re: input lag at 97% GPU usage vs 90% GPU usage with 240hz

Post by knypol » 14 Apr 2020, 11:52

I found pretty interesting video about input lag.

https://www.youtube.com/watch?v=8ZRuFaF ... e=emb_logo

knypol
Posts: 76
Joined: 18 Aug 2016, 03:40

Re: input lag at 97% GPU usage vs 90% GPU usage with 240hz

Post by knypol » 14 Apr 2020, 11:54

to have lowest input lag u need as much as u can fps but not exceeding 95% of gpu load.

forii
Posts: 218
Joined: 29 Jan 2020, 18:23

Re: input lag at 97% GPU usage vs 90% GPU usage with 240hz

Post by forii » 14 Apr 2020, 11:56

knypol wrote:
14 Apr 2020, 11:54
to have lowest input lag u need as much as u can fps but not exceeding 95% of gpu load.
Yes I know this... but bellow 95% gpu load is only at 144 fps... 144 fps at 240hz its not that good, even with g sync.

I also heard that only 99-100% is HUGE input lag, 97% is pretty fine compare to 99%
Image

Based on that photo above and tests
Imo 218 fps with g sync (withouyt v-sync ofc) seems better at 240hz, even with slight more input lag.

Edit.

A lot of streamers on twitch.tv says that they use uncapped fps + all settings in game at low, that means they will never reach gpu usage at 99-100% because of the lowest settings in game + nvidia, that is also the best way to get lowest input lag - they say.

I will stand with that, because i didnt find any competive guy who play shooter games and he cap his fps,

I will consider use NULL + g sync thought in that case, i never reach more than 240 fps anyway, so I do not cap my fps + I will use ELMB-Sync, i think thats the best combo, g sync doesnt add input lag if i checked

skylit
Posts: 15
Joined: 25 Sep 2019, 10:09

Re: input lag at 97% GPU usage vs 90% GPU usage with 240hz

Post by skylit » 15 Apr 2020, 01:01

You could be experiencing EMI issues. Not an expert on the topic, but capping the FPS does feel a lot better in this situation and mitigates what could be perceived as lag. It's likely not lag but interference of the signal which I assume makes your cursor feel muddy and harder to aim at a setting you're used to using correct?

If you can, try moving your entire PC setup to a different physical location and go from there. Your house or electricity might not have the best integrity. There are band-aid solutions to this if you do have this issue, but its more or less trial and error.

TTT
Posts: 253
Joined: 28 Jul 2018, 14:17

Re: input lag at 97% GPU usage vs 90% GPU usage with 240hz

Post by TTT » 10 May 2020, 10:22

skylit wrote:
15 Apr 2020, 01:01
You could be experiencing EMI issues. Not an expert on the topic, but capping the FPS does feel a lot better in this situation and mitigates what could be perceived as lag. It's likely not lag but interference of the signal which I assume makes your cursor feel muddy and harder to aim at a setting you're used to using correct?

If you can, try moving your entire PC setup to a different physical location and go from there. Your house or electricity might not have the best integrity. There are band-aid solutions to this if you do have this issue, but its more or less trial and error.
His post has nothing to do with EMI or muddy mouse, hes talking about whether higher hz with high usage or lower hz with less usage has more/less lag..

Post Reply