• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

does limiting FPS in game make Video card work less?

A

AnMig

Golden Member
Noticed in Borderland you can set a cap in the frame rate to 30-60, 72, 120 and no limit.

I play at 1440 (Default Cat leap) everything maxed out except fro AA (4x).

FRAPS reports FPS at 120's but I can hear my cars working (fan noise).

I figure setting cap at 72 FPS would be more than enough for smooth game play.
Will this make my card work less ergo cooler and less fan noise. Will my cards still work at max FPS while software in game merely displays 72?

I mean if it still rendering max frames then might as well set no limit.

System
I-5 @ 4.5ghz
7970 x2 Crossfire (1050mhz)

Peace
 
Yes, its like lifting weights. If you can bench 3 sets of 12 at 225, but then suddenly you decide that you want to bench 2 plates (135) instead you are doing less work. The effort required is much less. Similarly, if your video card can do 500 frames per second in Darksiders and you turn vsync on, the GPU usage % will be at around 10-20%.
 
Your card will only work less if it can process frames faster than the cap rate.

For example, if your card is limited to 60 fps, then setting the rate at 60 and above will see no reduction in workload.

A slower graphic card will need a lower frame cap to see a workload reduction.

This is a quick reply.
 
You could also just turn off crossfire. Half the video cards running should be half the noise--maybe even less as one won't have to deal with increased temps of the surround air to the same degree.

I leave my SLI off and only turn it on for games I know can benefit.
 
So when you turn off Crossfire/SLI the second card's fan will not spin its fan above idle speeds assuming the GPU temperature stays down, is that correct?
 
AnMig said:
.......
I figure setting cap at 72 FPS would be more than enough for smooth game play.
Will this make my card work less ergo cooler and less fan noise. Will my cards still work at max FPS while software in game merely displays 72?

I mean if it still rendering max frames then might as well set no limit.

System
I-5 @ 4.5ghz
7970 x2 Crossfire (1050mhz)
Peace
Click to expand...
Should be the case. There is a utility (fps limiter) which actually cap framerates in order to save power/heat and cut down on noise for apps which can't do that. You could always compare the gpu utilization and temps to see if it works.
 
In use can anyone tell me what's the difference between capping your framerate and adaptive v-sinc? Assuming you set the limit to the same as your refreshrate with both.
 
Thanks, capped it 72 fps
Also did discovered I can have phsyx even though I have a AMD card, off loaded to the cpu, have it set to medium hasn't affected my fps so far.
 
Last edited:
3DVagabond said:
In use can anyone tell me what's the difference between capping your framerate and adaptive v-sinc? Assuming you set the limit to the same as your refreshrate with both.
Click to expand...

Adaptive vsync turns itself off temporarily whenever framerates dip below 60 (or whatever you set the vsync to) in in order to minimise stuttering that would happen with normal vsync (because it has to be in sync with the monitor's refresh rate).
 
bononos said:
Adaptive vsync turns itself off temporarily whenever framerates dip below 60 (or whatever you set the vsync to) in in order to minimise stuttering that would happen with normal vsync (because it has to be in sync with the monitor's refresh rate).
Click to expand...

Wouldn't capping the framerate at the refreshrate accomplish the same thing?
 
3DVagabond said:
Wouldn't capping the framerate at the refreshrate accomplish the same thing?
Click to expand...

Well there is an additional limitation to vsync, in that when your card can't keep up with the limit (say, 60 FPS), instead of dropping to whatever it can handle, it will drop to a specific factor (e.g., 30 FPS).

In contrast, capping would not have this liability; when the card can't keep up with your cap, it will just perform at whatever it's capable of, without needing to leap-frog between specific factors like vsync does.

Adaptive vsync attempts to combine the best of both of these scenarios, capping when needed, but avoiding the leapfrogging.
 
KingFatty said:
Well there is an additional limitation to vsync, in that when your card can't keep up with the limit (say, 60 FPS), instead of dropping to whatever it can handle, it will drop to a specific factor (e.g., 30 FPS).

In contrast, capping would not have this liability; when the card can't keep up with your cap, it will just perform at whatever it's capable of, without needing to leap-frog between specific factors like vsync does.

Adaptive vsync attempts to combine the best of both of these scenarios, capping when needed, but avoiding the leapfrogging.
Click to expand...

This is why I'm wondering what the difference would be between capping and adaptive v-sinc. Does capping the framerate not sync the signal with the monitor's refresh, maybe?
 
It is a waste anyways to have anything over 60fps,only time vsync should be disabled is if your minimums are hitting lower then 60,but if the game can hold 60 fps,enable it.

My 7850 still hits in the freaking 40 fps area in BF3 with high 4xmsaa or ultra with 2x msaa and that is on my temporary 17'' 1280x1024 native lcd,only way i could enable vsync 60fps and get a smooth experience is running all medium with 2x msaa,and even some maps like gulf of oman still hit the lower 50's🙄 .

So off goes vsync unless i avoid gulf of oman,which won't happen lol.
 
AnMig said:
Thanks, capped it 72 fps
Also did discovered I can have phsyx even though I have a AMD card, off loaded to the cpu, have it set to medium hasn't affected my fps so far.
Click to expand...

PhysX run at reduced CPU rate...not the same as PhysX high(NVIDIA GPU only) FYI.
 
3DVagabond said:
This is why I'm wondering what the difference would be between capping and adaptive v-sinc. Does capping the framerate not sync the signal with the monitor's refresh, maybe?
Click to expand...

No, capping doesn't sync the signal, you can still get tearing with capping.
 
3DVagabond said:
This is why I'm wondering what the difference would be between capping and adaptive v-sinc. Does capping the framerate not sync the signal with the monitor's refresh, maybe?
Click to expand...

Correct. You can/will still have tearing by doing this. v-sync ensures that the frame buffer that is displayed always contains a fully completed image.
 
Yes, if the frame rate cap is lower than what your card can achieve running uncapped then the video card will only have some fractional load, this will lower temperatures and with dynamic fans will keep noise levels lower.

Capping the frame rate isn't enough to eliminate tearing, while you might generate frames at the right rate it doesn't guarantee they're in sync with the refresh of the monitor, you need to force vsync on to eliminate tearing.

Vsync caps your frame rate at the refresh rate of the monitor, or some fraction of it, if the card cannot hit the frame rate needed.
 
PrincessFrosty said:
Yes, if the frame rate cap is lower than what your card can achieve running uncapped then the video card will only have some fractional load, this will lower temperatures and with dynamic fans will keep noise levels lower.

Capping the frame rate isn't enough to eliminate tearing, while you might generate frames at the right rate it doesn't guarantee they're in sync with the refresh of the monitor, you need to force vsync on to eliminate tearing.

Vsync caps your frame rate at the refresh rate of the monitor, or some fraction of it, if the card cannot hit the frame rate needed.
Click to expand...
To be more precise, it's not just any fraction.

These are the frame rates with Vsync on.

Max sync rate plus the following: 1/2 max : 1/3 max : ..... 1/N max where N is a whole number
 
Thanks to the people on this forum with reading comprehension skills for answering my question. Both of you. 😉

Seriously, thanks to those who responded.
 
PrincessFrosty said:
Except where ((1/N)*refresh) is not a whole number 😉
Click to expand...
On further thought,I'm going to change my agreement to the above statement.

Take a 60Hz refresh.

Frame rate jumps in descending order are:
60
30
20
15
12
10
8.5714 (60/7)
7.5 (60/8)
etc

Frame rates do not have to be whole numbers. That is merely an artifact of any number system, in this case base 10.

The only important info is that it takes 1, 2, 3, 4, 5, 6, 7, 8, ......... refresh cycles to update a new screen image.
 
maddie said:
On further thought,I'm going to change my agreement to the above statement.

Take a 60Hz refresh.

Frame rate jumps in descending order are:
60
30
20
15
12
10
8.5714 (60/7)
7.5 (60/8)
etc

Frame rates do not have to be whole numbers. That is merely an artifact of any number system, in this case base 10.

The only important info is that it takes 1, 2, 3, 4, 5, 6, 7, 8, ......... refresh cycles to update a new screen image.
Click to expand...

I thought you'd come back to this, I was kind of deliberately leading you in that direction 🙂

Actually it's not that it's an artefact of the number system, it's more about your assumptions about how vsync works. The way it's talked about in forums and between gamers suggests that people think it has a kind of "target" frame rate, but that's not really what is happening, it doesn't try and target for example 1/7th of your refresh rate...

The video card is still rendering frames as fast as possible, if the rendering isn't done in time for the next refresh on the monitor then the old frame is simply repeated on concurrent refreshes until the next whole frame is done rendering. You can still only have 1 full frame per refresh and a whole number of refreshes in 1 second.

A lot of this comes down to frame rate being an odd way of measuring performance, there's a great article on this here which is eye opening, especially when you learn that FPS is a non-linear performance measurement - http://www.mvps.org/directx/articles/fps_versus_frame_time.htm

Remember that frame rate is actually an average, if you average over 1 second which is what your unit measurement is in (frames per 1 second) then you'll always get a whole number, essentially you're measuring discreet frames. Anything more than 1 second and the measurement is an average.
 
Last edited:
You must log in or register to reply here.

TRENDING THREADS

Back
Top