I decided to benchmark my 9800M GTX with different forms of overclocking, including Core only, Mem Only, and Shader only, along with different combinations of overclocking. My benchmarking platform was 3DMark06. Driver version 179.24, specifications in sig.
Default clocks: 500 / 799 / 1375
Overclock:
Core was set to 600
Mem was set to 900
Shaders set to 1600
Download this Excel file to view the results:
http://www.mojaveunderground.com/3dm6ben.xlsx
http://www.mojaveunderground.com/3dm6ben.xls
It is interesting that the memory has the least amount of difference on performance, but is needed to get the scores up real high. Core overclock is the most effective.
View the Comparison tab to see side by side results
-
Thanks for the effort, this could prove useful in my future attempt. It would be better if you could upload version that non-office 2007 can open, thanks again!
-
If you are sharing Microsoft Office documents and you use 2007, make sure to save it as a 97-2003 document so that people with Office 97 to Office 03 can read the document.
-J.B. -
I did. The link is under the 07 link...
-
Interesting to note with my 8800 gts is that my stock clocks are 500/799/1250 and I OC to 650/900/1650, which is highest I have found stable. Do you/can you OC this high? I thought is interesting because I gain about 1500 points and you get less than a 1000. No biggie either way, just something I noticed. Nice work BTW.
-
I would guess since the 9800 GTX has a lot more pipelines (112 to 64), it's already running fairly hot and has less room to increase clocks.
Core/Mem/Shader Overclocking difference results
Discussion in 'Gaming (Software and Graphics Cards)' started by emike09, Dec 20, 2008.