Benchmarks II
Here's another MadOnion.com benchmark we like to use to judge overall system performance, PcMark2002.
Not a bad score at all and definitely on par with previous results showing the chipset on the DFI board is working well and the board is performing well as a whole.
Let's move on to some gaming benchmarks. All game benchmarks are done at a resolution of 1024x768 unless otherwise specified.
First up is 3DMark 2001SE.
Not at all bad, very nearly 13k with no driver tweaks or card overclocking. This was with the default of best image quality in the graphics control panel with FSAA disabled. But 700marks difference in scores highlights the effect of memory bandwidth here.
On to Quake 3. We are using patch v1.30, and to make it more of a comparison we will run this bench with fastest and high quality settings. This is the Four.dm_66 demo.
Again we see how important the extra bandwidth is for performance.
On to Serious Sam 2. We're benchmarking the publicly available Valley of the Jaguar Time demo, one that is heavily reliant on subsystem speed, at 1024x768x32 Normal preferences.
On to Unreal Tournament 2003.
More than playable in my opinion and again on par with the results we expect.
Next up is Comanche 4 from Novalogic.
Very close to 50fps on this taxing demo, 3.5fps quicker than DDR266 mode and a good showing from the DDR333 mode available from the chipset.
Finally let’s have a look at how this system coped with Code Creatures benchmark. This test runs through 3 different resolutions 1600x1200, 1280x1024, and 1024x768 then gives you a final score.
Not alot of difference here, but this is to be expected in a test that predominantly measures your graphics card.
Enough?
Let's wrap it up now. On to the conclusion.
