The Game Rundown: Finding CPU/GPU Bottlenecks, Part 2

[Guest]

We're back with Part 2 of our bottleneck exploration in a mainstream gaming PC equipped with a Core i5 and GeForce GTX 460, ready to dive into test results from ten additional games. All of the results are summarized and analyzed in our conclusion.

The Game Rundown: Finding CPU/GPU Bottlenecks, Part 2 : Read more

 

[nativeson8803]

It's disappointing to see that devs still aren't taking advantage of multiple cores like they could.

 

[KingArcher]

Would there be any performance difference between windows 7 32bit and 64bit?
Assuming you use the same amount of RAM [4GB].

 

[slayvus]

Great run down guys. I would of liked to see a three core benchmark thrown in there, but what can you expect from a dual core with HT.

@KingArcher Unless the program was designed for x64, you had more than 4GB of unused RAM that the game could use and could use more than 4GB I highly doubt whether or not your on x86 or x64 would matter.

@nativeson8803 What difference would it make on games that are still GPU limited at two cores? Even then, there were games limited by the GPU at four cores. Taking advantage of a CPU with four cores or more is going to be hard when you need more GPU to get more frame rates.

 

[palladin9479]

@Slayvus almost true. The difference would be @2 GB of available memory. NT x86 kernel only allows an application to access 2GB of address space with the other 2GB reserved for kernel use. In NT x86 world each application can only access 2GB of memory, with the application getting rather unstable once it goes over 1.8 GB. This limitation also applies to the NT x64 kernel running a 32-bit problem through WOW64. Its something left over from the NT 4.0 / 5.0 world.

 

[archange]

Just keep in mind that these result are relevant in the context of using the GTX 460 768 MB, which is the real bottleneck here. Better GPUs and/or multi GPU setups would shift the bottleneck towards the CPU.

In the end, it's all about finding the perfect balance in hardware. If you want my advice, pair the GTX 460 768 MB with a fast dual core, but get a good quad with higher end model GPUs.

 

[deisu]

This methodology should be used to new cpu/gpus/games reviews.

 

[Guest]

I'd like to see them test the first Supreme Commander.

It still taxes the most powerful systems

 

[Guest]

They should have used Phenom II X6 and test with 1/2/3/4/6 cores enabled. This made me want more of this because there aren't any 6-core benchmarks.
Make a Part 3.

 

[Mobutu]

+1 deisu

 

[lashabane]

[citation][nom]archange[/nom]Just keep in mind that these result are relevant in the context of using the GTX 460 768 MB, which is the real bottleneck here. Better GPUs and/or multi GPU setups would shift the bottleneck towards the CPU.In the end, it's all about finding the perfect balance in hardware. If you want my advice, pair the GTX 460 768 MB with a fast dual core, but get a good quad with higher end model GPUs.[/citation]
I totally agree with archange. I understand that you're trying to go for the "norm", but if you're looking for bottlenecks, you need to remove said bottlenecks to see where it truly lays. Start with a quad core clocked at ~3.5ghz and swap out gfx from there. Push and push to see where it truly lays instead of going for a baseline. Although you can use that baseline to see what the "average" person might use. If the article is truly about the bottleneck, use something beefier than the 460.

 

[lashabane]

[citation][nom]roffle[/nom]They should have used Phenom II X6 and test with 1/2/3/4/6 cores enabled. This made me want more of this because there aren't any 6-core benchmarks.Make a Part 3.[/citation]
And overclock the crap out of the 1055, 1075, 1090/whatever you use.

 

[RabidFace]

[citation][nom]roffle[/nom]They should have used Phenom II X6 and test with 1/2/3/4/6 cores enabled. This made me want more of this because there aren't any 6-core benchmarks.Make a Part 3.[/citation]
Was thinking the same thing And maybe throw in some multi-CPU tests as well

Lets see GTA4 with a dual-socket 6-core CPU in SLI with dual GTX 280s

 

[RabidFace]

EDIT: 480s

 

[archange]

Yes, but for such a setup, TH should first get one of those compact, autonomous Nuclear Reactors... And a lifetime supply of depleted Uranium. xD

 

[xizel]

in BFBC2 i doubled my frames going from E8400 to Q9550 with a HD4870 1gig crossifre setup

 

[Chris_TC]

"Mass Effect 2 needs at least four cores."
"With this game, you should grab a quad-core chip [...]"

With 63.7 fps on a single core there's absolutely ZERO reason to buy anything more expensive for this game.

 

[nevertell]

[citation][nom]archange[/nom]Just keep in mind that these result are relevant in the context of using the GTX 460 768 MB, which is the real bottleneck here. Better GPUs and/or multi GPU setups would shift the bottleneck towards the CPU.In the end, it's all about finding the perfect balance in hardware. If you want my advice, pair the GTX 460 768 MB with a fast dual core, but get a good quad with higher end model GPUs.[/citation]

THEN WHY THE HELL DID THEY NOT USE A 480 GTX 4way sli to rule out gpu bottlenecks ?

 

[gamerk316]

Again, this is a bad test. You're testing for CPU usage...using max graphical settings. As such, you're seeing the result of a GPU bottlenecking teh system.

Set everything to low, THEN test the CPU. This type of test reveals nothing except that the GPU is more often then not the primary limiting factor in gaming.

 

[Onus]

I like the fact that power usage at the wall never cracked 300W. That tells me my SG-650 should be adequate essentially forever.
The points I take from this are:
1. Any [new] GPU bought for serious gaming at high resolutions should have 1GB.
2. As a requirement for gaming performance, overclocking is unnecessary. This means that massive coolers and lots of noisy fans are also unnecessary.
3. Data loading issues that show the benefit of additional cores would undoubtedly show a benefit from a SSD, e.g. less stuttering.
4. A hugely expensive edong is not required to get decent performance in most games.
It would be a lot of extra work and involve arbitrary choices, but it would be interesting to see the effects of reducing settings, even a little. At high resolutions, I'm not sure my eyes are even good enough to notice whether or not AA is on. How much does a reduction in settings shift the bottleneck off the GPU?

 

[Guest]

Seems like most games run fine on a tri-core CPU between 2,5 and 3GHz.
I also believe if the tests where done with a better graphics card (eg a 1GB VRAM Radeon 5770) you could run 80-90% of the games with a 3GHz 2core CPU.

 

[Guest]

I would like to know if DX9 games, ran on DX11 hardware sees any CPU relieve.

 

[Wheat_Thins]

Fantastic articles; Thank you!

 

[Guest]

I would love to see an AMD test setup.

 

[superflykicks03]

[citation][nom]gamerk316[/nom] Set everything to low, THEN test the CPU. This type of test reveals nothing except that the GPU is more often then not the primary limiting factor in gaming.[/citation]

Wasn't that the point of this article? I have so many friends that ask for upgrade advice and think that they need to upgrade to an i5/i7 and overclock the heck out of it before they can upgrade their graphics cards. Then I ask what card they are looking to get and it usually falls into the range of what was tested here: equal to or less than a 460 or 5850.

This article shows that with DX 11 games a very nice midrange gfx card will generally be the bottleneck, and most of the time its best to upgrade your GPU first. Heck, before this article if i was running a nicely oc'ed C2D and a 9800GT, I wouldn't have necessarily upgraded the graphics card first. Nobody runs games on low settings; this article makes the real world argument that if you have limited funds, 95% of the time your money is best spent on changing out your GPU first.