I thought this kind of odd...I ran 3dMark05 on my new PNY 6800 AGP card at stock for comparison, and initially got 3080, which is pretty good for my box. Then I used RivaTuner to unlock the other other four pipes (it's apparently successful), and watched my score drop by over 120.
Certain things such as fill-rate multi-texturing and pixel shader saw significant gains, some saw small gains (CPU, vertex shader complex, etc) while just a few things saw small decreases...so it's weird that I lost over 100 points. I'm not a math major, but it doesn't add up. The only thing I can think of is that the additional pipes over-saturated the AGP bus...I'm running 4x.
I have not yet overclocked the card.
Does my assumption sound correct?
Specs:
Asus P4B533, Northwood 1.6A@2.4, 1G of Crucial PC2700 (2-2-2-5), Enermax 350W, WD 120G, Deskstar 40G (the 60 series), NEC 3500A, Plextor 4824A, Audigy2 ZS Platinum, etc.
On a positive note: water in HL2 looks lovely (I used to have a Ti-4200)
Certain things such as fill-rate multi-texturing and pixel shader saw significant gains, some saw small gains (CPU, vertex shader complex, etc) while just a few things saw small decreases...so it's weird that I lost over 100 points. I'm not a math major, but it doesn't add up. The only thing I can think of is that the additional pipes over-saturated the AGP bus...I'm running 4x.
I have not yet overclocked the card.
Does my assumption sound correct?
Specs:
Asus P4B533, Northwood 1.6A@2.4, 1G of Crucial PC2700 (2-2-2-5), Enermax 350W, WD 120G, Deskstar 40G (the 60 series), NEC 3500A, Plextor 4824A, Audigy2 ZS Platinum, etc.
On a positive note: water in HL2 looks lovely (I used to have a Ti-4200)
