Show me what yo can do Punk !Due to the fact Asus had foreseen them presets I wanted to test them out to see what effect they had on games and co. Since the 3D clocks are practically the same and only a small boost over the reference clocks I didn't expect any miracles. My own voltmodded 285 was run at its default reference speed to compare out of the box performance.
I tested the 3Dmarks 01, 03 and 06 at their original default setting. For the games I opted to test them at 2 common resolutions of 1280 x 1024 and 1600 x 1200. All games and bench tests were repeated 3 times and the average score was implemented in the below charts.
Here's a rundown on the test rig :
Windows XP with SP3 was used as OS, Forceware drivers from the CD were used : naimly 185.85 WHQL ones, for the ATI card we used the latest Catalyst 9.7 I added the 4890 card just as a quick reference. Keep in mind that the main objective is to compare the 2 NVIDIA cards here, this is by al means not intended to be a common graphic cards roundup.
Game tests run at 1280 x 1024 and 1600 x 1200 resolution with the integrated benchmark tool, High detail settings selected
Farcry 2Tom Clancy's HawXLeft 4 DeadWe took the average FPS from the game tests. Maximum FPS means nothing if your minimum FPS drops too much, making a game unplayable. Since I'm used to playing First person shooters my own personal limit is 60 FPS. If I can't keep that steady I'm lowering detail levels. (Absolutely not the case with these 3 games)
Let's move on to the Out of the Box results :
Out of the box performance 3DMark wise shows tiny differences, pretty logic as the OC Asus selected is how.. tiny...
Game wise, we see the same small increase of a few fps, if you get it for free, why not use it... Most users will run the optimized mode as it gives the best of both worlds... max out of the box game performance and one of the better power saving profiles in 2D. The ATI card wasn't able to run the Left 4 Dead benchmark, it would always revert back to the desktop before posting results. I didn't go deeper into this, as I kept the focus on the two 285's (this due to limited time for the review)
All in all not much to go for there, but we all know the ROG lineup is not designed for standard usage. Some of you guys are extreme and can't wait to get your soldering iron out and voltmod this card. But many users maybe want to find out what speeds they can expect while keeping the FAN setting on AUTO.
I tried to get the max clocks out of the cards, making the below clocks pass the 3 3Dmark tests and also successfully complete the integrated benchmarks of the 3 games. This means that for e.g. for 3D01 I could run higher clocks, yet if it failed on e.g. 3D03. So I stepped down... Note that the achieved values can differ from card to card; some clock less, some clock better...
My vanilla eVGA card was pretty average at stock volts of 1.16 GPU, topping at 705Mhz GPU core, 1605 Shader clocks and 1300mhz on the rams. Boosting the GPU volts via the voltmod to 1.26 increased my OC potential big time. Yet take note that the fan is left on AUTO, if I increased the fan speed to eg 80% or even 100% I can clock higher, but my wife won't appreciate all that racket coming out of my PC. The goal here is to have a nice blend of fan speed/noise and extra power.
As you can see the Matrix card, even with the GPU at only 1.2, can beat the modded card. This really shows some potential. The better power circuitry and cooling seem to give some nice results.
3dmark scores are close, what can you expect of just a mere 10Mhz more on the GPU, the vanilla OC'ed doesn't really get spanked :)
Same counts for the game scores...pretty close.
Follow us on Twitter
Follow us on Facebook
