I'm trying to contribute to this thread but I'm having issues just running the benchmark. My hardware:

AOpen AX59 Pro Super Socket 7 mainboard
1 stick 128MB SDRAM PC133
Asus graphics card (Ti4800SE)
Amd K6-2+ 550
Windows 98SE
VIA chipset drivers (latest from website)
Nvidia drivers (latest from website)
DX 8.1
3D Mark 2001 from the 3dmark website

What happens is when I run 3dmark 2001 after a fresh boot, I get to the main meny. After I press "benchmark" I get a crash to the desktop. When I run 3dmark2001 again I get an error:

Do I need to use an older version of 3DMark 2001? Or a special version of Direct X?

Please help me 😀

EDIT:

Seems to be a driver issue:

EDIT:

Ok so what happens is after a fresh boot I can run DXDIAG and it will show the Direct3D button. But when I press it I get the same crash to the desktop. It must be a driver issue. Maybe using the latest chipset and Nvidia driver is not the right approach...

Mau1wurf1977 wrote:

I'm trying to contribute to this thread but I'm having issues just running the benchmark. My hardware: […]
Show full quote

I'm trying to contribute to this thread but I'm having issues just running the benchmark. My hardware:

AOpen AX59 Pro Super Socket 7 mainboard
1 stick 128MB SDRAM PC133
Asus graphics card (Ti4800SE)
Amd K6-2+ 550
Windows 98SE
VIA chipset drivers (latest from website)
Nvidia drivers (latest from website)
DX 8.1
3D Mark 2001 from the 3dmark website

What happens is when I run 3dmark 2001 after a fresh boot, I get to the main meny. After I press "benchmark" I get a crash to the desktop. When I run 3dmark2001 again I get an error:

Do I need to use an older version of 3DMark 2001? Or a special version of Direct X?

Please help me 😀

EDIT:

Seems to be a driver issue:

EDIT:

Ok so what happens is after a fresh boot I can run DXDIAG and it will show the Direct3D button. But when I press it I get the same crash to the desktop. It must be a driver issue. Maybe using the latest chipset and Nvidia driver is not the right approach...

its probably drivers, when I tried running the latest Nvidia drivers on my 440BX motherboard, it cause weird issues, such as Unreal not working in D3D mode, inexplicable slowdown , or the characters in NOLF having no textures.

it could be something related to that, try different drivers.

interestingly I thought about the effect DX8 has on the 2001SE scores and decided to run 3DMark2000 on both the 700Mhz Coppermine/GF2TI machine, and the laptop mentioned previously.

Win98 machine - 5430 (I was mistaken beforehand)
2010 laptop - 5145

to recall the laptop has a 2Ghz Dual-Core CPU , 3GB of RAM and a GMA45000 chip.

even without RAM optimizations due to too much RAM , thats kind of funny.

Ok so I located chipset motherboard drivers on the AOpen FTP. Will re-install Windows and try these.

But what about Nvidia? I guess drivers from within a year of the card getting releases might be a good start.

Found their W98 driver archive: http://www.nvidia.com/object/win9x_archive.html

Mau1wurf1977 wrote:

I'm trying to contribute to this thread but I'm having issues just running the benchmark. My hardware: […]
Show full quote

I'm trying to contribute to this thread but I'm having issues just running the benchmark. My hardware:

AOpen AX59 Pro Super Socket 7 mainboard
1 stick 128MB SDRAM PC133
Asus graphics card (Ti4800SE)
Amd K6-2+ 550
Windows 98SE
VIA chipset drivers (latest from website)
Nvidia drivers (latest from website)
DX 8.1
3D Mark 2001 from the 3dmark website

What happens is when I run 3dmark 2001 after a fresh boot, I get to the main meny. After I press "benchmark" I get a crash to the desktop. When I run 3dmark2001 again I get an error:

Do I need to use an older version of 3DMark 2001? Or a special version of Direct X?

Please help me 😀

EDIT:

Seems to be a driver issue:

EDIT:

Ok so what happens is after a fresh boot I can run DXDIAG and it will show the Direct3D button. But when I press it I get the same crash to the desktop. It must be a driver issue. Maybe using the latest chipset and Nvidia driver is not the right approach...

suggested version:
via 4in1 driver 4.25 or 4.32
nvidia driver 30.82

I gave up on the AOpen board and using the Gigabyte GA-5AX with ALI chipset. The Direct3D test completed sucessfully. Might be simply a case of ALI chipset being a better / more stable for Windows? Got a few crashes on the AOpen board, even while installing Windows. Might be a hardware issue. But I'm getting another VIA chipset board soon, so I can investigate further.

Here are the result:

Gigabyte GA-5AX
128MB Apacer PC100 CL3 memory
Asus Geforce 4 Ti4800SE
AMD K6-III+ 400 MHz clocked at 100 x 4 = 400 MHz
Windows 98SE
43.45_win9x_english
DirectX9.0c-x86-x86_64-Dec2005

Test result:

Test bench:

Mau1wurf1977 wrote:

I gave up on the AOpen board and using the Gigabyte GA-5AX with ALI chipset. The Direct3D test completed sucessfully. Might be s […]
Show full quote

I gave up on the AOpen board and using the Gigabyte GA-5AX with ALI chipset. The Direct3D test completed sucessfully. Might be simply a case of ALI chipset being a better / more stable for Windows? Got a few crashes on the AOpen board, even while installing Windows. Might be a hardware issue. But I'm getting another VIA chipset board soon, so I can investigate further.

Here are the result:

Gigabyte GA-5AX
128MB Apacer PC100 CL3 memory
Asus Geforce 4 Ti4800SE
AMD K6-III+ 400 MHz clocked at 100 x 4 = 400 MHz
Windows 98SE
43.45_win9x_english
DirectX9.0c-x86-x86_64-Dec2005

Test result:

Test bench:

actually ali chipsets have worse agp support than the via ones. it was so bad that nvidia drivers would purposefully disable all agp features and force pci mode whenever it sees an ali chipset, to avoid all kinds of problems. you can use everest, sisoft or something else to confirm if your video card is running in pci mode.
you can force agp mode on in registry, but on my 5ax it simply gives a lot of BSOD in loading windows. i tried that in win2000, not sure if the win98 result would be the same.
my recent experience with mvp3 board is good, but not sure if its just my board.

noshutdown wrote:

actually ali chipsets have worse agp support than the via ones. it was so bad that nvidia drivers would purposefully disable all agp features and force pci mode whenever it sees an ali chipset, to avoid all kinds of problems. you can use everest, sisoft or something else to confirm if your video card is running in pci mode.
you can force agp mode on in registry, but on my 5ax it simply gives a lot of BSOD in loading windows. i tried that in win2000, not sure if the win98 result would be the same.
my recent experience with mvp3 board is good, but not sure if its just my board.

Is that right, fascinating 😀

Here the results with a K6-2+ 550:

I tried some utilities to look at this AGP issue:

At this point I haven't installed any chipset driver. I have the latest ULI driver from Nvidia and will try this next and see if the score improves.

Here the result after installing the ULI drivers from Nvidia (IntegratedDriver2.13). Version 2.20 gave me a crash however...

Pretty much the same score: 2561

Mau1wurf1977 wrote:

Here the result after installing the ULI drivers from Nvidia (IntegratedDriver2.13). Version 2.20 gave me a crash however...

Pretty much the same score: 2561

i reckon installing ali/uli's chipset driver won't do any good, they only help with ali's newer p6 and k7 chipsets. but well, maybe thats the difference between win98 and 2000.

Same graphics card and Ram in a BX440 system with a PIII 500. Quite a bit faster compared to the K6-2+

Ran it on a friend's system that I was repairing:
i7-3770, GTX-650 (vanilla; 384 core), 8GB RAM, and a freshly installed Windows 8.1

Mau1wurf1977 wrote:

Same graphics card and Ram in a BX440 system with a PIII 500. Quite a bit faster compared to the K6-2+

Is that a Katmai?

I think so. Slot 1, 100MHz FSB.

I think I may have set some sort of record here. I had to cheat a little by overclocking the 9800 Pro, and I also used an older driver (5.8 instead of my usual 7.3) but there it is. Over 12K on a P3!

Nice 😀

I just bought a new (old used) socket 939 motherboard with a single-core Athlon 64 CPU.
The same exact same model socket 939 motherboard as in the original post by Steve 😀
Your scores are in danger! Or would have been if I had any of those video cards. I have two 8800 GTX 512 though so I guess I will have to try those.
I also have two Geforce 7950 GX2 so if I get things going I might post some quad gpu scores... but those cards are not easy to work with.

Everything is pretty much on auto with stock voltage and silent air cooling. The 7900GTX is not overclocked...yet 😉
This is not a max clock I just tested first 2750 mhz and then 3000 mhz and it just works 😮
Not bad for 8-9 year old hardware. I like hazzard less overclocks!

A little bit faster memory and lower vcore. Everything else is the same.
The Asus A8N32-SLI overvolts a bit at default settings.

3dmark 2001 really loves memory speed!

I'm working on a Dual Tualatin 1.4 and came across this thread.. A few people were curious if a dual would help 3Dmark2001se and it doesn't
look like it does, or if it does it's very little. My numbers are in line with Single 1.4's.. However it looks like 3dmark 03 is multi-threaded, since a search on google has my score up with a lot of 2+ghz CPU's..

These are funny scores... I'm guessing the newer driver and card just aren't optimized for 99max.. Tried a couple different drivers
and the result was the same..

No overclock...

Dual P III-S 1.4ghz
1.5gig PC133 Cas 3 - ECC off
Geforce 6800 Ultra OC (BFG factory overlock 425mhz core)
Supermicro P3TDDE with VIA Apollo Pro 266T chipset...
Forceware 81.85's

Ran 3dmark 05, no pics cuz it's in an XML file..

4108 3Dmarks
1547 CPU marks

Quick compare on google suggests it is multi-threaded...

thats very impressive! getting 11200pts on a p3s... i thought it would be around 10000 at best since cpu is the obvious bottleneck.
i could only get 9300pts with p3s-1.4, 815et board and ti4600, even with the classic 30.82 driver. with newer drivers it degrades to 8900, and then 8600.