6800GT + 3DMark03 overclocking tests

[cornelious0_0]

I honestly don't know why i sat down and wrote all this out.....I guess I was bored.....I figured I might as well do something with it and post it up. At the very least you'll be able to see how much the scores increase after being raised every 5MHz and draw your own conclusions. Unless otherwise noted, everything was done using my sig system, at stock 2.8GHz.

Here we are:

350/1000: 10,826
355/1000: 10,918 (+92)
360/1000: 11,032 (+144)
365/1000: 11,114 (+82)
370/1000: 11,238 (+124)
375/1000: 11,366 (+128)
380/1000: 11,427 (+61)
385/1000: 11,506 (+79)
390/1000: 11,583 (+77)
395/1000: 11,655 (+72)
400/1000: 11,735 (+80)
405/1000: 11,806 (+71)
410/1000: 11,903 (+97)

410/1010: 11,924 (+21)
410/1020: 11,924 (+5)
410/1030: 11,959 (+30)

410/1030 + CPU @ 3.43GHz: 12,526

I have since worked on getting the overclock a little higher.....and with my cooling mods I've got planned for the card I'm sure I'll soon be seeing more jumps in the OC and Benchies.

Again, take this for what you want, just thought I might as well post it up. It is however interesting to see after I hit 375 on the core performance kept going up.....but the gains dropped off significantly.....again.....w/e.

 

[tamislan]

Interesting post. I think the most noteworthy aspect of it is how little difference overclocking the RAM makes in comparison to overclocking the core. You're getting a very similar max clock to me.

 

[cornelious0_0]

Interesting post. I think the most noteworthy aspect of it is how little difference overclocking the RAM makes in comparison to overclocking the core. You're getting a very similar max clock to me.

Yeah, I did find the ram quite interesting.....but dont worry....everything will be taking a nice leap forward when I'm done the cooling mods I'm doing to the card.

I'll update this thread or start another one when the mods are done.

 

[tamislan]

Yes, I read your mods thread - quite ambitious and I look forward to seeing the results .

 

[CraftyOtter]

Thanks for confirming this with 3dMark03, I wouldn't have the patience to run it 17 times like you did. I ran Aquamark benchmarks a couple days ago and also saw little improvement by upping the memory core. Sorry, don't have my numbers handy but the results made me consider running my memory at stock speed to cut down on unnecessary heat generation.

 

[cornelious0_0]

Thanks for confirming this with 3dMark03, I wouldn't have the patience to run it 17 times like you did. I ran Aquamark benchmarks a couple days ago and also saw little improvement by upping the memory core. Sorry, don't have my numbers handy but the results made me consider running my memory at stock speed to cut down on unnecessary heat generation.

I know what you mean.....it really doesnt have that much of an effect on scores.....3DMark03 or otherwise.....and even raising it to 1030 raises my ambient temp on the card by a couple C....which in turn raises the core.