CPU Scaling - 4.0Ghz vs. 4.5Ghz Gaming Benchmarks

[Tempered81]

Many people tend to think a 3.4-3.6Ghz cpu is just fine for gaming- Dual or Quad. This is true if you're running a single 8800GT or slower video card, or if you're bound to resolutions below 1280x1024. The faster CPU the better - and here are the benches to prove it (using the two fastest Gfx cards to test the CPU).


http://www.evga.com/forums/tm.asp?m=530527

Up first is Call of Duty 4:

In this game I decided to use the "war Pig" level. The test was run from the level's start to the part where the American tank shoots the enemy tank through the building. I chose this level because it has a nice mix of open areas, lots of enemies and lots of action.

Settings are everything in game maxed 4x AA

4Ghz E7200

HD4870X2 ---- 146 fps avg
GTX280 ------ 89 fps avg

4.5Ghz E7200

HD4870X2 ---- 169fps avg
GTX280 ------ 104fps avg




Next up is BioShock:

In this game I decided to run the benchmark from when the biosphere opens at the start of the game on through to the part where the glass tunnel floods with water. Along that time we have some nice volumetric lighting, complex water effects, and real moody shadows.

All in game settings maxed

E7200 @ 4Ghz

HD4870X2 ---- 184 fps avg
GTX280 ------ 112 fps avg

E7200 @ 4.5Ghz

HD4870X2 ---- 216 fps avg
GTX280 ------ 138 fps avg




Moving on to Crysis:

The benchmark was run from the first time you see the sun rise from behind the trees on the first level to disabling the beacon on the beach.

DX10 all settings maxed

E7200 @ 4Ghz

HD4870X2 ---- 38 fps avg
GTX280 ------ 35 fps avg

E7200 @ 4.5Ghz

HD4870X2 ---- 42 fps avg
GTX280 ------ 39 fps avg




Up on the block next is GRID:

Benchmark was run on the "Long Beach" track from start to finish 3 laps. Nothing really important about the level, I just really like that track

All settings maxed 4xAA

E7200 @ 4Ghz

HD4870X2 ---- 110 fps avg
GTX280 ------ 98 fps avg

E7200 @ 4.5Ghz

HD4870X2 ---- 139 fps avg
GTX280 ------ 111 fps avg




World in Conflict:

This is the only game that I decided to just use the in game benchmark. Preset chosen was "Very High" 4xAA

E7200 @ 4Ghz

HD4870X2 ---- 47 fps avg
GTX280 ------ 46 fps avg

E7200 @ 4.5Ghz

HD4870X2 ---- 54 fps avg
GTX280 ------ 51 fps avg




Oblivion:

Benchmark was run from Fort Wooden Hand to Skingrad East Gate. Along that path was see a dense red wood forest, a thick grassy plain, and we can see for a very long distance.

Settings are 8xAA, using Capt Kills 4096 texture mod, and all settings maxed:

E7200 @ 4Ghz

HD4870X2 ---- 144 fps avg
GTX280 ------ 85 fps avg

E7200 @ 4.5Ghz

HD4870X2 ---- 155 fps avg
GTX280 ------ 98 fps avg




Half Life 2: Episode 2:

In this one I ran the bench from the games start where you are on the destroyed train, to the part where the portal storm occurs and takes out the bridge behind you.

All settings are maxed and using 8x AA

E7200 @ 4Ghz

HD4870X2 ---- 110 fps avg
GTX280 ------ 74 fps avg

E7200 @ 4.5Ghz

HD4870X2 ---- 121 fps avg
GTX280 ------ 83 fps avg

What's really surprising is how much these cards gain in performance as you ramp up the clock speed. Remember, the cards were left at stock speeds, the only overclocking was done to the CPU. Both of the cards showed appreciable and noticeable gains even going from 4Ghz to 4.5Ghz. That is simply amazing! Even at 4Ghz CPu speed these cards are not "tapped" and can still go faster.

http://www.evga.com/forums/tm.asp?m=530527

 

[dguy6789]

Good work, impressive cards indeed.

 

[Tempered81]

Yeah, i really liked it. Btw, The op has two R700 CFX 800/4000 & an E8600 @ 5Ghz in his SIG on Evga Forums.

 

[badnewcastle]

Yeah, Craptacular has put together some amazing systems.

 

[scheibler1]

wow I thought anything over 3ghz didn't bottleneck a gtx 280....and I just ordered a Q6600

 

[myocardia]

Originally posted by: scheibler1
wow I thought anything over 3ghz didn't bottleneck a gtx 280....and I just ordered a Q6600

Well, it's a case of dimishing returns. After a certain poit, probably around 3.2-3.4 Ghz, you're going to see your framerate go up alot less per Mhz/100 Mhz. And of course, the slower the video card, the sooner you'll see those diminishing returns.

 

[VirtualLarry]

Originally posted by: jaredpace
What's really surprising is how much these cards gain in performance as you ramp up the clock speed. Remember, the cards were left at stock speeds, the only overclocking was done to the CPU. Both of the cards showed appreciable and noticeable gains even going from 4Ghz to 4.5Ghz. That is simply amazing! Even at 4Ghz CPu speed these cards are not "tapped" and can still go faster.

I wonder what this means for future GPU developments. Especially if modern GPUs are now outrunning CPUs. What benefit will faster GPUs bring to the table, if they are CPU-limited? Who will buy them? Will the high-end GPU market collapse?

I just picked up another pair of HD4850s for my quad-core rigs, now each of them can have a crossfire 4850 pair. I'm wondering how CPU-limited I will be.

 

[Tempered81]

Crossfire 4850 performance isn't that far behind 4870X2 / GTX280 performance, so I'd guess the more speed, the merrier.

 

[soccerballtux]

Originally posted by: VirtualLarry

Originally posted by: jaredpace
What's really surprising is how much these cards gain in performance as you ramp up the clock speed. Remember, the cards were left at stock speeds, the only overclocking was done to the CPU. Both of the cards showed appreciable and noticeable gains even going from 4Ghz to 4.5Ghz. That is simply amazing! Even at 4Ghz CPu speed these cards are not "tapped" and can still go faster.

I wonder what this means for future GPU developments. Especially if modern GPUs are now outrunning CPUs. What benefit will faster GPUs bring to the table, if they are CPU-limited? Who will buy them? Will the high-end GPU market collapse?

I just picked up another pair of HD4850s for my quad-core rigs, now each of them can have a crossfire 4850 pair. I'm wondering how CPU-limited I will be.

What it means, I think, is that we have graphics cards outpacing game graphics.

 

[soccerballtux]

Originally posted by: scheibler1
wow I thought anything over 3ghz didn't bottleneck a gtx 280....and I just ordered a Q6600

If it means anything scheibler1, I still will be going quad myself when I upgrade.

and @op-- Thanks, great benchies.

 

[imported_Scoop]

So, what to gather from the results? That 1650x1080 isn't enough to make games GPU limited.

 

[scheibler1]

i fell better now...didnt realize these results where with 1680x1050...I game at 1920x1200 and use as much AA/AF as the hardware can push....cpu will be less of a factor

 

[ArchAngel777]

It would be more worth while to see if these gains on on the high end of the frame rate, or the low end. I have a feeling that minimum frame rates would not be affected by your CPU test, but that the high points are, thus skewing the average.

Another factor, when you have that much GPU power, you will want to enable 4X AA, TSSAA and 16XAF. I can promise you that the minute you enable TSSAA you will watch the bottleneck shift right onto the GPU. IMO, TSSAA is very important in games... Regular multi sampling won't hit the transparancy textures and you can see nasty aliasing... So, TSSAA is a must, IMO if you have spare GPU power.

Also, someone brought up diminishing returns, something that also needs to be factored into the equation. Plus, if you simply move from a max frame rate of 100 to 105 in certain areas, you do increase the average, but the actual fluidity of the game doesn't change between the two systems.

 

[apoppin]

http://forums.anandtech.com/me...=2230839&enterthread=y

i posted a couple of 19x12 benches with 4870x2 and my e8600 at 3.33Ghz and [almost] 4.0Ghz
- i just could not get +1Mhz higher than 3.99Ghz
- my p35 MB's issues, i think - vcore is stock!!
- when i upgrade to x48, i think i can get higher; core temps are good


the minimum frame rate IS higher with the faster cpu, generally

 

[MarcVenice]

Look, all these benches show is that a faster CPU gets you even more ridiculous framerates. To get everything out of a gtx280 or HD4870X2 an e8400 still suffices, because there's no difference between 80 and 100 fps. If the minimum framerates however would go up from 25 to 35, then you'd have a good point. But I don't think they do.

 

[VirtualLarry]

Originally posted by: soccerballtux

Originally posted by: VirtualLarry

Originally posted by: jaredpace
What's really surprising is how much these cards gain in performance as you ramp up the clock speed. Remember, the cards were left at stock speeds, the only overclocking was done to the CPU. Both of the cards showed appreciable and noticeable gains even going from 4Ghz to 4.5Ghz. That is simply amazing! Even at 4Ghz CPu speed these cards are not "tapped" and can still go faster.

I wonder what this means for future GPU developments. Especially if modern GPUs are now outrunning CPUs. What benefit will faster GPUs bring to the table, if they are CPU-limited? Who will buy them? Will the high-end GPU market collapse?

I just picked up another pair of HD4850s for my quad-core rigs, now each of them can have a crossfire 4850 pair. I'm wondering how CPU-limited I will be.

What it means, I think, is that we have graphics cards outpacing game graphics.

And yet Crysis won't budge much above 30FPS.

 

[aigomorla]

Originally posted by: Scoop
So, what to gather from the results? That 1650x1080 isn't enough to make games GPU limited.

thats what im trying to make out also.

Also the fact that less then 1% of the world computers are even clocked at 4ghz + making it very difficult to achieve.

And i dont think nvidia or ATI would be dumb enough to make a bottle necked card @ 4ghz. :X

 

[myocardia]

Originally posted by: apoppin
the minimum frame rate IS higher with the faster cpu, generally

Yeah, I've noticed that myself. Depending on how much CPU power you have, adding a faster video card can make your max framerates go through the roof, but your minmums not go up much, while adding a faster CPU while keeping the same video card will usually make either the minimum and maximum go up by roughly the same amount, or sometimes even make your minimums go up more (higher percentage) than your maximum framerate.

 

[spinejam]

tx for the info op!

 

[tonyou]

What resolution did you use for these benchmarks?

 

[Tempered81]

its in the link. 1680 x 1050 8xAA or 4xAA, 16XAF

 

[SlowSpyder]

Originally posted by: MarcVenice
Look, all these benches show is that a faster CPU gets you even more ridiculous framerates. To get everything out of a gtx280 or HD4870X2 an e8400 still suffices, because there's no difference between 80 and 100 fps. If the minimum framerates however would go up from 25 to 35, then you'd have a good point. But I don't think they do.

Agreed 100%. I think sometimes people forget to put what they're looking at into perspective. The extra .5GHz did bring some measurable differences, but rarely impacted real world game play, at least not significantly. I don't think anyone is going to argue with the 'the faster the better' mentality, but I'd certainly still pick up a faster/better GPU over a faster CPU (generally speaking, within reason) to improve my overall gaming performance. My 2.8GHz Phenom with a GTX280 is still going to provide a better gaming experience then a QX9650 @ 4.5GHz with an 8800GT probably 99.9% of the time.

 

[RussianSensation]

Originally posted by: jaredpace
Many people tend to think a 3.4-3.6Ghz cpu is just fine for gaming- Dual or Quad. This is true if you're running a single 8800GT or slower video card, or if you're bound to resolutions below 1280x1024. The faster CPU the better - and here are the benches to prove it (using the two fastest Gfx cards to test the CPU).

The games tested are all at 1680x1050 resolution...a resolution no one uses with GTX 280 and especially 4870 X2. On top of that, in situations where 4.0ghz was sufficient it was already getting really high frames so to say 4.5ghz "improves" gaming is NOT true. In situation where 4.0ghz was not sufficient (i.e below 60 frames), 4.5ghz did NOT make the game any more playable either.

As far as I am concerned these benchmarks are not really meaningful to test the impact of a CPU for high end graphics cards because he chose a pointless resolution (i.e. at least should have used 8/16AA throughout) and did not provide numbers for minimum framerates. Secondly, they fail to show that a faster cpu provides any more real world playability. So I am not sure you can disprove that 3.4ghz isn't sufficient.

 

[sgrinavi]

nice~ thanks

 

[Zstream]

Originally posted by: RussianSensation

Originally posted by: jaredpace
Many people tend to think a 3.4-3.6Ghz cpu is just fine for gaming- Dual or Quad. This is true if you're running a single 8800GT or slower video card, or if you're bound to resolutions below 1280x1024. The faster CPU the better - and here are the benches to prove it (using the two fastest Gfx cards to test the CPU).

The games tested are all at 1680x1050 resolution...a resolution no one uses with GTX 280 and especially 4870 X2. On top of that, in situations where 4.0ghz was sufficient it was already getting really high frames so to say 4.5ghz "improves" gaming is NOT true. In situation where 4.0ghz was not sufficient (i.e below 60 frames), 4.5ghz did NOT make the game any more playable either.

As far as I am concerned these benchmarks are not really meaningful to test the impact of a CPU for high end graphics cards because he chose a pointless resolution (i.e. at least should have used 8/16AA throughout) and did not provide numbers for minimum framerates. Secondly, they fail to show that a faster cpu provides any more real world playability. So I am not sure you can disprove that 3.4ghz isn't sufficient.

This is one of the worst arguments I have ever ever heard. Hell even the most red/green fan boys provide a better argument.

In the testing it shows that the faster CPU improves gameplay rather you like to admit it or not. It means that the faster the CPU you will achieve greater FPS.

He also used 4xAA which is one of the most common AA settings and benchmark settings. Finally, last but not least it means we are finally at a point where the CPU is not fast enough to keep up.