Locked FPS stability question in Source engine games

Everything about displays and monitors. 120Hz, 144Hz, 240Hz, 4K, 1440p, input lag, display shopping, monitor purchase decisions, compare, versus, debate, and more. Questions? Just ask!
Post Reply
User avatar
A Solid lad
Posts: 317
Joined: 17 Feb 2018, 08:07
Location: Slovakia
Contact:

Locked FPS stability question in Source engine games

Post by A Solid lad » 04 Nov 2018, 12:17

I see a lot of CS:GO streamers with locked fps numbers to 300, getting 299 fps nearly all the time, but mine is fluctuating between 280-295 whenever I lock my framerate.
I have an 8700k+980 Ti+16gb ram PC, and if I set fps_max 0, I constantly get above 450 fps (500-600 most of the time) on dust2, so what's the reason behind this?
I've already tried turning HPET on and off.
(The same was happening with my old setup i7 6700+different mobo ofc+gtx 1060)
Discord | Youtube | Twitch
Steam with a list of monitors & mice I've used/use.

User avatar
Chief Blur Buster
Site Admin
Posts: 11653
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Locked FPS stability question in Source engine games

Post by Chief Blur Buster » 04 Nov 2018, 17:34

Are you using dual-capping? Combining two frame rate cappers (example: 305fps in RTSS and 300fps in CS:GO) can sometimes create this effect, because RTSS is kind of effectively a microsecond precision frametime capper while CS:GO builtin framerate cap is not microsecond accurate.

Picture this: Let's say, perhaps, In game caps may have frametime fluctuations from 1/250sec through 1/350sec frametimes (that averages out ~300fps). This may be tripping-up frametime capping algorithms in a second framerate capping utility, to force a situation of "1/250sec through 1/300sec frametimes" (that averages out to ~275fps).

This may not be what is happening, but this is one theoretical cause of how two framerate caps can combine (with different capping algorithms) to create a lower framerate.

So frame-count capping can interact with frametime-measurement capping if you're using both capping simultaneously (e.g. RTSS capping simultaneously with CS:GO capping at the same time), and create a slight lowering effect like this. There are some use cases of using two framerate cappers, but it can produce some side effects similar to this.

Try disabling everything else that may be capping (uncap RTSS, and/or uninstall) relying only on the CS:GO in-game capping.
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
A Solid lad
Posts: 317
Joined: 17 Feb 2018, 08:07
Location: Slovakia
Contact:

Re: Locked FPS stability question in Source engine games

Post by A Solid lad » 05 Nov 2018, 03:24

Just using the built-in fps_max command, no other capping methods.
Discord | Youtube | Twitch
Steam with a list of monitors & mice I've used/use.

tacocheese
Posts: 2
Joined: 13 Dec 2021, 19:11

Re: Locked FPS stability question in Source engine games

Post by tacocheese » 13 Dec 2021, 19:16

Change preferred refresh rate from app controlled to highest available in the nvcp. I had same problem for past few weeks. I had to cap my my 12th gen i9 to 225 fps to get good frametimes lol. Now I use the above and the frame cap in nvcp. Afterburner also works now and I haven't tried the in game limiter but it probably also works now. And I was getting like 1k fps in empty bot levels. That has since dropped to around 360. It won't go much further past that which I'm assuming is due to the refreshrate setting. But it no longer dips into the 200's for no reason. No input lag added either, tested with nvidia reflex monitor/mouse.

User avatar
A Solid lad
Posts: 317
Joined: 17 Feb 2018, 08:07
Location: Slovakia
Contact:

Re: Locked FPS stability question in Source engine games

Post by A Solid lad » 16 Dec 2021, 08:46

I found out afterwards that the reason for this was me disabling HPET in BIOS.
It should be enabled there and disabled in windows, that's the ideal combo, not the other way around.

Also, for some reason 250 fps locked feels much smoother, with less tearing than 300 on CS and Quake.
You might think it's due to it being closer to my monitor's refresh rate (240hz XL2540), but I don't think that's the case, since locking it at 240 or 235-6-7-8-9 still can't touch the smoothness of 250.
I think it has something to do with the engine... ppl always used to lock their fps 125 back in Quake, and CS's engine has quake as a base, so I guess multiples of 125 work great.
Quake Champions and Diabotical also have 250 as the default cap, that's why I tried it in CS and it works great.
Discord | Youtube | Twitch
Steam with a list of monitors & mice I've used/use.

User avatar
kyube
Posts: 129
Joined: 29 Jan 2018, 12:03

Re: Locked FPS stability question in Source engine games

Post by kyube » 16 Dec 2021, 10:37

Don't mess with HPET in Windows, nor in BIOS, it leads to wonky behaviour.
However, I've seen many claims that the fps_max command in CSGO is very wonky, so I wouldn't be surprised if your claim of 250fps feeling better is correct. It's what I rock with (125, 144 166, 250, 500, 1000 caps to be exact)

User avatar
axaro1
Posts: 627
Joined: 23 Apr 2020, 12:00
Location: Milan, Italy

Re: Locked FPS stability question in Source engine games

Post by axaro1 » 16 Dec 2021, 10:38

A Solid lad wrote:
16 Dec 2021, 08:46
Also, for some reason 250 fps locked feels much smoother, with less tearing than 300 on CS and Quake.
Maybe it might be due to Asynchronicity-Induced jitters being reduced when using a mouse with a polling rate that is a multiple of the fps that the engine is producing (250*4=1000hz), 1 every 4 polling events being effectively processed should give you the right amount of sync between your input and the ingame simulation.

With high polling rate mice set to 4000+hz I noticed that it almost feels like Vsync is on due to severe reduction in temporal jitters, this is very noticeable especially when paired with strobing.
XL2566K* | XV252QF* | LG C1* | HP OMEN X 25 | XL2546K | VG259QM | XG2402 | LS24F350[RIP]
*= currently owned



MONITOR: XL2566K custom VT: https://i.imgur.com/ylYkuLf.png
CPU: 5800x3d 102mhz BCLK
GPU: 3080FE undervolted
RAM: https://i.imgur.com/iwmraZB.png
MOUSE: Endgame Gear OP1 8k
KEYBOARD: Wooting 60he

User avatar
A Solid lad
Posts: 317
Joined: 17 Feb 2018, 08:07
Location: Slovakia
Contact:

Re: Locked FPS stability question in Source engine games

Post by A Solid lad » 19 Dec 2021, 18:16

axaro1 wrote:
16 Dec 2021, 10:38
A Solid lad wrote:
16 Dec 2021, 08:46
Also, for some reason 250 fps locked feels much smoother, with less tearing than 300 on CS and Quake.
Maybe it might be due to Asynchronicity-Induced jitters being reduced when using a mouse with a polling rate that is a multiple of the fps that the engine is producing (250*4=1000hz), 1 every 4 polling events being effectively processed should give you the right amount of sync between your input and the ingame simulation.

With high polling rate mice set to 4000+hz I noticed that it almost feels like Vsync is on due to severe reduction in temporal jitters, this is very noticeable especially when paired with strobing.
That sounds very likely!
I just remembered reading about this like a year ago when the Viper 8k came out.
Discord | Youtube | Twitch
Steam with a list of monitors & mice I've used/use.

Post Reply