• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Anyone interested in Q6600 - 2500k comparisons?

DPD should be here in the next hour :).

Not sure that bad company 2/left 4 dead has a benchmarking facility?

May not be a very accurate representation of real fps differences as both games can be varied, and although based on average frame rates, may provide misleading results.

Will run an x264 conversion now.
 
If you can get some rough figures or benchmarks for games, I'd be very interested in clock for clock figures :-)
 
I've currently got a QX6700 and would like to know how it compares out of curiosity if you have the spare time :)

Some more built-in benches to try are: GTA IV, FarCry 2, Mafia 2 defo has built-in bench, and Street Fighter IV. Maybe a bit of Stalker: CoP bench for DX11.

That should be a broad spectrum of games to compare with. Make sure you run the games 3 times each and show the median results to allow for slight changes.

And if your looking for CPU scores in 3dmark11 (otherwise known as Physics Score), its listed under the tab Detailed Scores. Here is my bench with a very average score to see what I'm referring to:
-http://3dmark.com/3dm11/385026
 
Last edited:
Heres mine:

TF2 1680x1050
Qx6700 @ 3.2ghz
150fps max

2600k @ 4.4ghz.
299fps max (wont go any higher as TF2 is not designed to go above 300fps)
 
i just run ever ones favorite game 3dmark06
with a gtx 460
heres results

Q6600 3.2ghz 400 x 8

overall: 16862
SM 2.0: 6571
HDR/SM 3.0: 8053
CPU: 4686

2500K @ STOCK

overall: 21891
SM 2.0: 9014
HDR/SM 3.0: 10012
CPU: 6035

ill post some more down the line if people wish
just playing aound with the settings trying for 4.5ghz+ atm
 
This are the ones i've ran on the sandybridge rig, compared to 775 rig. Will run other tests 2moro. Games run on same settings on both rigs.

Q6600 @ 3.3Ghz
3D Mark 2011 - P3865 - http://3dmark.com/3dm11/479461

CINEBENCH R10
OpenGL Standard Test - CB 4579 (0)
Single CPU Render Test - CB 3704 (0)
Multiple CPU Render Test - CB 12572 (0)

GTA 4
Average FPS: 60.36
Duration: 37.29 sec
CPU Usage: 77%
System memory usage: 52%
Video memory usage: 82%

Resident Evil 5
Average - 62.8 fps (A)

i5 2500k @ stock
3D Mark 2011 - P4223 - http://3dmark.com/3dm11/502424

CINEBENCH R10
OpenGL Standard Test - CB 6325 (0)
Single CPU Render Test - CB 6689 (0)
Multiple CPU Render Test - CB 22949 (0)

GTA 4
Average FPS: 65.40
Duration: 37.13 sec
CPU Usage: 63%
System memory usage: 51%
Video memory usage: 79%

Resident Evil 5
Average - 86.7 fps (A)
 
It seems that 2500k is about 45% faster in cinebench single and multiple cpu render test while about 10% faster in 3D MARK 2011.

Also I wonder why 2500K gives only 5 FPS more in GTA4 while in RE5 it gives about 24 FPS.

So it seems that 2500k performance gain over Q6600 at same clock speed really depends on the software application that is being used. But very nice comparison OP. Run crysis and metro2033 and it will be interesting to see the results :).
 
WingZero30 said:
It seems that 2500k is about 45% faster in cinebench single and multiple cpu render test while about 10% faster in 3D MARK 2011.

Also I wonder why 2500K gives only 5 FPS more in GTA4 while in RE5 it gives about 24 FPS.
Click to expand...

By monitoring temps + CPU/GPU workloads on my second monitor whilst these tests are running, it seems that the games which show the smallest improvement appear to be GPU limited rather than CPU limited.

For example, my Mafia II test showed a hardly worth mentioning improvement, but that appears to be the same case with the i5 2500k clocked to 4.4Ghz.

Just Cause 2, Counter Strike Source VST (fps_max 800) and Unigine Heaven v2.1 appear to follow the above pattern (barely noticeable results)

On my q6600 system, it probably would have made more sense to have run the games on their default configurations and then run the tests. Because i used the maximum settings, and as a result the benchmarks appear to be affected by the GPU limitation.

On the sandybridge system, throughout the benchmarks the GPU is utilising 99% - 100% GPU and the CPU graphs show that there is plenty of work load left.

Metallifux said:
Just out of curiosity what settings did you use to get the Q6600 stable at 3.3GHz?
Click to expand...

I used:
412 FSB
8 Multiplier
1.425vcore

That was on a p45 board, with 4x 2GB sticks of 1066 RAM (which could argue was restricting the clock speed), and B3 q6600 with 1.285 vid
 
3t3P said:
hmmm, my q6700 may be ok for a while then...

shame I wanted to upgrade BAD
Click to expand...

Bear in mind as i mentioned that my results could have been more relevant/reliable had i used lower graphical detail levels on the tests. The benchmarks show GPU limitations rather than CPU, which cant be a bad thing as the 470 is considered a mid-high end card.

Also bear in mind, thats the sandybridge at clock, I now have it at 4.4GHz, with a small vcore increase, temps hitting 60oc at full load. (q6600 would touch 75)
 
I am averaging 90FPS in Black ops and BFBC2, using a [email protected] coupled with a HD6970.

I really don't feel the need to upgrade to SandyBridge if I'm getting those sort of frames, anything over 60hz is a bonus.

I think a lot of people are putting to much emphasis on their processors and not enough on there graphics cards.

SandyBridge is great if you are doing a lot of video or photography rendering, in HUGE amounts. My Q6600 keeps up with everything I throw at it, Lightroom rendering/exports.

If you want to get good frames in a game, get the best graphics card, not the best processor. Obviously certain cpu's will bottleneck all the goodness of the gpu if you start using dual cores etc.
 
Buchanan0204 said:
If you want to get good frames in a game, get the best graphics card, not the best processor. Obviously certain cpu's will bottleneck all the goodness of the gpu if you start using dual cores etc.
Click to expand...

I think too. I have [email protected] and 5850 and it plays all games fine and don't think I need to upgrade my cpu either.

SB is only good mainly if doing lots of 3d rendering, video editing, photoshop work, cadwork etc. For gaming a good decent quad cpu will do and as mentioned people will notice more improvements with better graphics card :).
 
Scougar said:
^^^ Depends on the game. Generally yes, but examples like Arma II really need a fast CPU.
Click to expand...

Of course there are going to be the odd exception, but is a £500 bundle really worth it to crank up the textures a notch or two on the odd game. I know i'd much rather spend £500 on a week in the sun or a posh hotel for the weekend.
 
matthab said:
Heres mine:

TF2 1680x1050
Qx6700 @ 3.2ghz
150fps max

2600k @ 4.4ghz.
299fps max (wont go any higher as TF2 is not designed to go above 300fps)
Click to expand...

Using the old fps_max 999 command and it will lift it but oddly with Ati not more than 999fps, this also works in CS:S and HL series of games. I ran CS:S but I need to know what settings we are all using as I kept hitting 999fps. Are we maxing out the game to try to push the GPU which in turn will mean we are more likely to hit the CPU ceiling?

The average in CS:S was 113.135fps and max was exactly 994fps where there wasn't much to render. Thats not accurate as the average I seen on Fraps throughout was always around 440fps.

Anyway, here is the screenshots of the settings used and result (Jpeg 80% quality for forum):-

hl22011012515294227.jpg
hl22011012515314676.jpg


As you can see, even though my CPU was working away on all 4cores @75%, the QX6700 is limiting the maximum framerate of the GPU, otherwise I'd expect to easily hit at least 500fps on the Test Results. I know Nvidia users always score much better in this test (even my old 8800 did) so if I anyone wants me to run a particular game at certain settings (I can run up to 1920x1200) then let me know :)
 
You must log in or register to reply here.
Back
Top Bottom