3dmark score

[phenix]

Originally posted by SaulTurnedPaul
I disagree... a stock 9800XT runs at 412/365 and MarBaS's 9800P is at 411/360. In other words, he's getting much better scores with a slower proc, slower memory with half as much, and a slower graphics card. Heck, I even get nearly 5400 with my 9800P at 405/351 with crappy SDRAM and a 2100+. Something's gotta be wrong. I'd guess that its a driver problem.

MarBaS, how's Redline working for ya?

Oh...i thought a pro was faster than an xt. I stand corrected.

[Moraelin v2.0]

Fishbet, if your card doesn't even use the AGP as more than a PCI slot, yeah, that would explain the score. In my case it's using 8x AGP.

Normally I'd guess that your AGP drivers are either not installed or not working right, for whatever reason.

[SexyMF]

FYI,

I just got my 9800P. With last rev of ATi drivers on default settings (balanced) I get 5612. My rig is overclocked:

2.08 T'Bred @ 2400Mhz (200x12)
RAM in sig @ 200Mhz 7,3,3,2.5


The CPU does play a bigger part than people like to think.

I noticed considerable improvement by using the ATi drivers over whatever came with the card.

Also, AGP 8x does nothing noticable for me. I ran my first tests at 4x. I think that this may be a case where the bandwidth of 4x is underutilised anyway.

Turn of 6xAA and see how your score diminishes!

[wrathchild_67]

Originally posted by SexyMF
The CPU does play a bigger part than people like to think.

Not really. The only CPU dependent score is the first game test. It accounts for exactly 5.6% of the total score.