• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

~14100 3DMark2001se any help squeezing more?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

mwhitwor

Registered
Joined
Jan 15, 2003
Location
Dallas, TX
First, let me thank everyone who participates in these forums. There's a lot of good knowledge here. Anyway, a little background. In my most recent upgrade cycle, I bought a Gigabyte GA-7VAXP Ultra, a Tbred 2100+ B, and some PC2700 2.5 Kingston Valueram to prop up my existing Deskstar drives and Sapphire Radeon 9500 128<B NP w/ Infineon 3.3 chips. After some initial success, I found that a box that would run 10x180 at +10% voltage and 12x170 or 15x133 at +7.5% voltage clean would tend to lock up or crash out of 3DMark2001se. After adding an SLK800, an Antec TruePower 430W PS, and messing with core, RAM, and CPU voltages, it seemed that it was largely due to the KT400/AGP8x phenomenon. After 10 days of spending way too much time with this, and based on previous experiences with my Via KT133A chipset, I kissed VIA goodbye.

Now I have an Abit NF7-S Rev 2., and all I can say is WOW. I have pushed it VERY little, but I am currently stable at [email protected] actual, and AGP at 1.7 and DRAM at 2.8 with my memory running sync using the 3/3 divider and Aggressive settings. In my quest to find just a little more performance, I've begun focusing on the 9500 and the RAM. I suppose the reason that I probably can't run my CAS 2.5 RAM at 2.5 (have to run at 3) and that I can't run "turbo" timings is because I am running it at 400 instead of 333? Anyone think that running at 333 at turbo, which I think would be async based on my current 10x200 would be better than running aggressive at 400 sync? Just thought I'd solicit opinions because of how many settings there are to test with.
 
I don't mind running benchmarks, but the point is that I would really like to get some insight into which configs might give the best results in some statistically significant way. EG, given the choice to run 333 cas 2.5 or 400 cas 3 which would you choose? Maybe there are other significant variables. Then I can run select benchies.
 
Back