Slower 3D performance in XP Pro?

I have recently changed to a dual-boot with WinME and XP Pro.

Using 3DMark 2001 se I have noticed lower scores in XP than in ME.

Have any others noticed this on their systems?

My ME 3DMark is 8223, and in XP the mark is 7868.
Not much of a concern, just curious.

Both OS's are using the same VIA chipset driver version and the same video card drivers. The video card is a Radeon 8500le.

TIA,

Dale

 

Moving this to hardware. Probably more answers there.

 

It is very close to impossible that an ME video driver would work in XP and vice versa. If you are using the same one, I'm surprized you have video at all on one of them!

Where did you get these drivers, and which one are you using for both, ME or XP? (Actually, it must be the XP one because XP would not allow you to install an ME driver).

I have a Radeon 7000 and I think it was slower on XP than ME until I updated the driver. Then it showed no delays in any games or other videos. I didn't benchmark it, though, so this is fairly subjective. Pauses were gone, however.

 

Well, not exactly the same driver, but the same version of the drivers needed for XP and for ME.
These are the Catalyst 3.2 Drivers from ATI's website at...

http://www.atitech.com/support/products/pc/radeon8500/radeon8500le128drivers.html

I should have mentioned that I am not having any issues, just curious as to why the numbers are lower in XP. Thought maybe I could learn something.

Dale

 

Oh, I see. The same installer for all OS's, but different drivers.

Just out of curiosity, try another benchmark, like pcpitstop.com and SiSoft Sandra if you have it.

Is the difference noticable to you or too close to call?

Same version of DirectX in both?

Were there separate values for OpenGL and DirectX?

 

Thanks for the reply;

Yes the same version of DirectX is installed on both (DX9)

I do have Sandra and the numbers from it in all benchmarks are a ~tiny~ bit higher in XP over ME. So the opposite is true with this benchmark. With XP being faster.

No separate values for Open GL or Direct 3D.

No noticable difference in the actual performance from a user's perspective. Just in that benchmark.

So no, I don't think there is a problem, it's more of a curiosity/learning thing.

Dale

 

OK. So it doesn't look like a problem. Thanks for the info. If anyone asks about this, I'll know . It depends what tests each benchmark performs, and what weight is assigned to each result.