• 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.

Poor 3870x2 benchmark

Soldato
Joined
13 Dec 2002
Posts
8,005
Location
Newcastle upon Tyne
after finally getting it installed heres what i got, first one is 3870x2 with q6600 at stock

3DMark Score 11717 3DMarks
SM 2.0 Score 4991 Marks
SM 3.0 Score 4962 Marks
CPU Score 3524 Marks


second is 3870xt x 2 CF with q6600 @ 3.6ghz

3DMark Score 18348 3DMarks
SM 2.0 Score 7487 Marks
SM 3.0 Score 8478 Marks
CPU Score 5038 Marks

obviously i understand the cpu score being much higher in the second one but th sm2 and sm3 scores?
 
You are CPU bound. The more you push the CPU speed up the more of the power of the card will be released.

It's because of the resolution of 1280x1024. The CPU is more in control at this resolution and the 2.4Ghz is limiting what the card can release.

The faster that CPU is pushed the more SM2 and SM3 scores will raise.
 
but if he as 3dmark pro he could higher the res....

Sorry I should have answered that better.

I mean the default resolution to compare scores is 1280x1024 and when trying to see if you are scoring right then you refer to the 3dmark06 scores thread and all benchmarks are done at the default low resolution.

I've stated in another thread that as the average screen size is 1680x1050 here at OCUK (from a poll) that we should raise the bar on the 1280x1024 as we are moving forward all the time in technology but still this low resolution is getting used and I think it's pointless.
 
Cool, at stock speeds on the default 3dmark 06 test I pulled 13100 on a stock Q6600 on my p5e and asus 3970x2.

Glad to see that once I get to grips with clocking the system the score will leap up.

Do you mind telling me what sort of settings you had to change to get the cpu overclocked to 3.4ghz?
 
Cool, at stock speeds on the default 3dmark 06 test I pulled 13100 on a stock Q6600 on my p5e and asus 3970x2.

Glad to see that once I get to grips with clocking the system the score will leap up.

Do you mind telling me what sort of settings you had to change to get the cpu overclocked to 3.4ghz?

just increased the fsb and also vcore to 1.4
 
Depending on your motherboard, but normally on Core2Duo's/Quad's you are increasing the FSB in the bios. With the extra frequency, you need a tad more voltage (normally about 0.1v extra), and sometimes a little on the northbridge (don't ask me for a figure as I don't know on voltage for NB).

This should EASILY see you to 3Ghz :) Just make sure you have sufficient case cooling and cpu cooling, and stability test your system (give it a good long 8 hour thrashing with some utilities).

Matthew
 
I've stated in another thread that as the average screen size is 1680x1050 here at OCUK (from a poll) that we should raise the bar on the 1280x1024 as we are moving forward all the time in technology but still this low resolution is getting used and I think it's pointless.

I agree in terms of doing casual comparisons on forums etc, but from Madonion's perspective, they couldn't really set 3dm06 to have a higher default res because then it would be out of reach for many people at the time, e.g. those with TFT monitors of 19" and smaller.
 
I agree in terms of doing casual comparisons on forums etc, but from Madonion's perspective, they couldn't really set 3dm06 to have a higher default res because then it would be out of reach for many people at the time, e.g. those with TFT monitors of 19" and smaller.

Very true and good point. I was just wanting pure GPU scores without anything at all holding it back and even through the majority have 1680x1050 capable monitors it would exclude those who don't and I can't say that's fair.

It was just an idea as not everyone wants to OC and yeah in the instances that this occurs then someone helping could run at higher resolution with a comparable system to help take the CPU out of the equation.
 
Back
Top Bottom