Weird 3DMark2001 results

[IcedEarth]

This is the deal: I've just assambled these two systems:

system 1:
Duron 750@864 (108*7.5)
256Meg RAM (@108 MHz)
ASUS K7V mobo rev 1.05
Matrox G400 16 DH

system 2:
Athlon 1333 (133*10)
128Meg RAM (@133MHz)
ASUS K7V mobo rev 1.04
Matrox G400 32 SH

3Dmark score for system 1: 1169
3Dmark score for system 2: 960

WTF?!?!?
How is this possible? Both systems have the same mobo (although not the same revision); both run win98 with the latest updates, 4in1 VIA drivers and latest Matrox qualified drivers. BIOS settings for both systems are identical. The only thing system 2 has less than system 1 is 128Megs of RAM.
Can anyone explain this?

I just rebuilt my own system (system 1) and was glad to see that an upgrade from a 100MHz fsb board to a 200MHz fsb board made a hell of a difference. Today I assembled system 2 for a friend of mine and I though it would blow my system away considering the higer fsb speed, memory speed, cpu speed, cpu core and double the amount of video RAM. My friend was able to run Unreal Tournament on his P3 500 at 1024*768@32 while I can only play it at the same resolution in 16 bit. However during the first 3Dmark game I noticed he only got 20 fps where I get 27fps on my system. I doublechecked all the settings and verified if I hadn't forgotten to install the most recent drivers but that wasn't the case.
I really don't get it. Is this difference due to the 128Megs of extra RAM I have? That can't be, can it?

[Guru]

3Dmark score for system 1: 1169
3Dmark score for system 1: 960

Uh I'm guessing sys 2 gets 960 Nvidiot marks?

Do you have UDMA enabled???

Are you using the same BIOS on both systems?

Did you do a clean w98 install on the computer (If you did I hope you installed the 4in1 drivers before dx8 and the Matrox drivers!)?

Are both systems runing the same AGP speed?

Is there a lot of disk "activity" going on in sys 2 while you are runing the benchmark!

etc. etc.

EDIT:What about the rest of the components in the system are they both using the same hard drives etc.

[IcedEarth]

Yeah, I meant sys2; I already corrected it

I'm sure the 4in1 drivers were installed before dx8 and matrox drivers on sys2. On system 1 they were installed afterwards. Also, system 1 wasn't a clean install, system 2 was. So it would be logical to think system 1 would score less because it ain't a clean install, but it's the other way around

Otherwise, all Win settings and BIOS settings are identical. It could be that my BIOS is a newer version, since it's a later revision; I haven't checked that yet. My friend's gonna get another 128 megs so I'll check again when he has the same amount of RAM than I have.

There's no significant disk-acces during the bechmark..... I still think it's weird

EDIT:
Both harddrives are ATA66 but they're not the same. We do have different CD-players and soundcards, but I doubt that has anything to do with a videocard benchmark

[Werdna]

I'd probably look at the typical things like IRQ settings etc.

how do they both go in real world situations?

[chucky2]

What kind of soundcards?

[IcedEarth]

Originally posted by chucky2
What kind of soundcards?

Mine is a SB128, his was a Vortex >something<. It appears that his soundcard conflicted with his network card. We've replaced his Vortex with a SB128 and now he has 1014 3D marks. It's a bit more but I still have more than he does....

Oh well, it's not MY problem

[[GDI]Raptor]

Uh I'm guessing sys 2 gets 960 Nvidiot marks?

What is Nvidiot marks?????

[chucky2]

Are the hard drives different>

Maybe thats that G400's max out at. No two pieces of hardware are exactly that same, so maybe on video card runs a little faster than the other?

Chuck