Tom's Hardware: CPU/GPU Bottlenecks

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Termie

Diamond Member
Aug 17, 2005
7,949
48
91
www.techbuyersguru.com
Define better

Trust me, you will understand the meaning of a CPU limitation when you pop that new GTX460 in your stock-clocked e8400 and run BC2. By "better" Toyota means that you would get significantly higher frame rates than you do with your 4870, which I'm fairly certain you will not. You will, however, have the option of running higher quality image settings, which may also be "better" if that's what you're looking for.
 

Skurge

Diamond Member
Aug 17, 2009
5,195
1
71
Trust me, you will understand the meaning of a CPU limitation when you pop that new GTX460 in your stock-clocked e8400 and run BC2. By "better" Toyota means that you would get significantly higher frame rates than you do with your 4870, which I'm fairly certain you will not. You will, however, get higher quality images, if that's what you're looking for.

From my experience, going from an X3 to an X4, the difference is in minimum framerates. in Dragon age I would hit 50-60fps most of the time, but it would dip into the 40s as well. With a quad it would stay in the 60s and wouldn't budge. My 4870 could keep up the performance, but the CPU was holding it back, but I wouldn't know that cause there weren't any settings that would ONLY affect CPU performance without affecting GPU performance. In BC2 the average fps was around 40, but jumped to 60 with a quad. Dirt 2 the average stayed the same, but the minimum went from high 20s to mid 30s. The average and minimum FPS in that game became much closer.

A faster CPU will generally give you better minimums.
 

Termie

Diamond Member
Aug 17, 2005
7,949
48
91
www.techbuyersguru.com
From my experience, going from an X3 to an X4, the difference is in minimum framerates.
. . .

In BC2 the average fps was around 40, but jumped to 60 with a quad.
. . .

A faster CPU will generally give you better minimums.

See what stands out here? While I agree that in many games, a better CPU affects minimums, in BC2 it doesn't work that way, as you've just noted. There's no way an e8400/GTX460 combo will ever touch 60fps, and you're getting that on a 5770.

I feel pretty capped on my e8400 in Dirt 2 actually, which you didn't notice in your tests. I get about 45fps regardless of settings, although I should be able to hit 60fps with my GTX460: http://www.anandtech.com/show/3909/nvidias-geforce-gts-450-pushing-fermi-in-to-the-mainstream/12.
 
Last edited:

Skurge

Diamond Member
Aug 17, 2009
5,195
1
71
See what stands out here? While I agree that in many games, a better CPU affects minimums, in BC2 it doesn't work that way, as you've just noted. There's no way an e8400/GTX460 combo will ever touch 60fps, and you're getting that on a 5770.

I actually feel pretty capped on my e8400 in Dirt 2 actually, which you didn't notice in your tests. I get about 45fps regardless of settings, although I should be able to hit 60fps with my GTX460: http://www.anandtech.com/show/3909/nvidias-geforce-gts-450-pushing-fermi-in-to-the-mainstream/12.

I should mention that in BC2 some of my settings were turned down, like AA and HBAO. I did note Dirt2 in my tests. The average FPS in that I was GPU limited but unlocking a 4th core raised my minimums.

I also said this.
http://forums.anandtech.com/showpost.php?p=31006113&postcount=107

That game uses the same engine.
 
Last edited:

Termie

Diamond Member
Aug 17, 2005
7,949
48
91
www.techbuyersguru.com
I should mention that in BC2 some of my settings were turned down, like AA and HBAO. I did note Dirt2 in my tests. The average FPS in that I was GPU limited but unlocking a 4th core raised my minimums.

I can't get 60fps on my e8400/GTX460 with 0AA at 1280x720, let alone at higher settings.

As for Dirt 2, it could be that a triple-core was closer to handling the demands of that game than a dual-core, so you just got the improvement in minimums. Various benchmarks have shown that Dirt 2, and now F1 2010, are very CPU-limited: http://www.tomshardware.com/reviews/gaming-processor-core-i3-athlon-ii,2666-9.html and http://www.tomshardware.com/reviews/sandy-bridge-core-i7-2600k-core-i5-2500k,2833-19.html
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
I can't get 60fps on my e8400/GTX460 with 0AA at 1280x720, let alone at higher settings.

As for Dirt 2, it could be that a triple-core was closer to handling the demands of that game than a dual-core, so you just got the improvement in minimums.
even my E8500 does pretty good for minimums in Dirt 2.

Windows 7 64bit
E8500
GTX260 192sp
4gb


1920x1080 highest settings and 8x AA

E8500 @ 3.16 GTX260 @ 666/1392/2200
avg 64.1
min 46.8

E8500 @ 3.80 GTX260 @ 666/1392/2200
avg 65.5
min 50.5
 

Termie

Diamond Member
Aug 17, 2005
7,949
48
91
www.techbuyersguru.com
even my E8500 does pretty good for minimums in Dirt 2.

Windows 7 64bit
E8500
GTX260 192sp
4gb


1920x1080 highest settings and 8x AA

E8500 @ 3.16 GTX260 @ 666/1392/2200
avg 64.1
min 46.8

E8500 @ 3.80 GTX260 @ 666/1392/2200
avg 65.5
min 50.5

Those are very good numbers. Obviously you're not particularly CPU-limited. I wonder if it could be related to running in DX10 vs. DX11 (note that you're benching higher than a GTX460 in Anandtech's test).
 

Skurge

Diamond Member
Aug 17, 2009
5,195
1
71
even my E8500 does pretty good for minimums in Dirt 2.

Windows 7 64bit
E8500
GTX260 192sp
4gb


1920x1080 highest settings and 8x AA

E8500 @ 3.16 GTX260 @ 666/1392/2200
avg 64.1
min 46.8

E8500 @ 3.80 GTX260 @ 666/1392/2200
avg 65.5
min 50.5

I did the tests in DX11 if that makes any difference. Can't do them now as I'm away from my pc atm.
 

hawtdawg

Golden Member
Jun 4, 2005
1,223
7
81
There are like 5-6 games out there that really take advantage of more than 2 cores at the moment. Even brand-new games like SC2 are still only utilizing 2 cores.

Also, to me, going from 2 to 4 cores in BFBC2 made a pretty noticeable difference in minimum framerates, even with the dual core clocked higher.
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
There are like 5-6 games out there that really take advantage of more than 2 cores at the moment. Even brand-new games like SC2 are still only utilizing 2 cores.
yes BUT games can take advantage of the higher IPC of the newer quads. an i7/i5 beats the crap out of a typical dual core cpu in a game like Starcraft 2.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
Here's a good example of how useless minimums can be: http://www.tomshardware.com/reviews/sandy-bridge-core-i7-2600k-core-i5-2500k,2833-18.html

At 2560x1600 the 2600K has a lowest minimum (4.4 FPS) while the Phenom II X4 970 has the highest minimum (14.5 FPS). How can anyone treat those scores as reliable given the actual pecking order of those processors?
the Metro 2033 is a flyby bench so the minimums can vary from run to run. and yes a single minimum framerate score is meaningless. anybody with the least bit of knowledge knows you need a graph to have a true representation of actual framerates.
 

GaiaHunter

Diamond Member
Jul 13, 2008
3,697
396
126
the Metro 2033 is a flyby bench so the minimums can vary from run to run. and yes a single minimum framerate score is meaningless. anybody with the least bit of knowledge knows you need a graph to have a true representation of actual framerates.

Which generally aren't present, especially on gaming CPU reviews that actually play at real world settings, but somehow the fact that i5/i7 have much better minimums, in a non useless way, is set in stone.

A reasonable article is this one, http://enthusiast.hardocp.com/article/2009/05/19/real_world_gameplay_cpu_scaling , using a 4870x2 CF configuration.
 
Last edited:

Draax

Junior Member
Dec 30, 2010
16
0
0
I never quite bought the CPU limitation theory. I never noticed it with my E6750 @ 3.4 using my 4870x2 nor with my new 6950 moded to 6970.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
I never quite bought the CPU limitation theory. I never noticed it with my E6750 @ 3.4 using my 4870x2 nor with my new 6950 moded to 6970.
do you expect a pop up to tell you that you could be getting better performance? a game isn't going to necessarily bog down just because the cpu is limiting its performance. just because you dont notice something does not mean it isn't there. so is a 412 horsepower Mustang no faster than a 315 horsepower Mustang since I never notice the 315 horsepower one feeling slow?
 

GaiaHunter

Diamond Member
Jul 13, 2008
3,697
396
126
do you expect a pop up to tell you that you could be getting better performance? just because you dont notice something does not mean it isn't there. so is a 412 horsepower Mustang no faster than a 315 horsepower Mustang since I never notice it feeling slow?

If your objective is to go from point A to B and the road only allows 60 km/h or mph (choose your favourite) and both can achieve that, it doesn't really matter.

The core count variable arises when you have to deliver 4 packages and A has 4 vans and B only has 2 vans.
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
If your objective is to go from point A to B and the road only allows 60 km/h or mph (choose your favourite) and both can achieve that, it doesn't really matter.

The core count variable arises when you have to deliver 4 packages and A has 4 vans and B only has 2 vans.
that has nothing to do with what I said. I am saying that just because you dont "notice" your cpu limiting your gpu does not mean it isnt.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
Nor does it mean it is limiting it.
what does that mean? an older Core 2 would most certainly limit a 6970 but again it may not be noticeable in most games. if you actually think your cpu is coming closing to fully pushing a 6970 in all games then you are delusional.
 
Last edited:

GaiaHunter

Diamond Member
Jul 13, 2008
3,697
396
126
that has nothing to do with what I said. I am saying that just because you dont "notice" your cpu limiting your gpu does not mean it isnt.

Of course if the processor is the bottleneck then giving it more graphics power, at same IQ setting, it should have no effect in the speed of the game.

In 5 years we can pick those future processors and pair them with today GPUs and conclude that those 5 GHz water cooled i7-980X were actually bottlenecking current GPUs.

So in fact this "limiting" just means relative performance to the fastest available hardware, which basically is the same as saying the fastest hardware is the fastest.

Crysis is a nice example that even the current fastest hardware is bottlenecking somewhere (either that or it really is badly coded) and so something is limiting something.
 
Last edited:

happy medium

Lifer
Jun 8, 2003
14,387
480
126
I never quite bought the CPU limitation theory. I never noticed it with my E6750 @ 3.4 using my 4870x2 nor with my new 6950 moded to 6970.

Your cpu is enough to keep most games above 30fps, so you might not notice.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
Of course if the processor is the bottleneck then giving it more graphics power, at same IQ setting, it should have no effect in the speed of the game.

In 5 years we can pick those future processors and pair them with today GPUs and conclude that those 5 GHz water cooled i7-980X were actually bottlenecking current GPUs.

So in fact this "limiting" just means relative performance to the fastest available hardware.
yes that is what I was saying. just because he doesnt notice that getting 35 fps where he could be getting 50 doesn't mean his Core 2 cpu is not limiting him. there are several games such as GTA 4, Prototype, Red Faction Guerrilla, Ghostbusters and few others where his 6970 would be barley getting over 50% of what it could with an i7/i5 quad and in some of those it would be noticeable.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |