Attachment seems missing.
Attachment now seems added

Tbh though, Speedsys is a very artificial benchmark. If Doom runs fine - Doom runs fine, and if everything else you actually intend to use runs fine, then whatever is causing a single low benchmark is not impacting in actual use and not really a performance bottleneck.

But maybe the table shows more than appears without it. I'll look.

Edit:
Only other board in Phil's database with a Via VP2 (=AMD640) and there are no Speedsys results there anyway. As for Quake and Doom, the correlation with CPU speed or motherboard chipset looks weak, in other words it's all apples and oranges. Your overview does correlate nicely (i.e. you keep all other factors the same when swapping CPU), and in no way looks out of the ordinary. I'd still say it's a non-issue.

Btw, it's "SDRAM", or if you insist on a dash, "S-DRAM" or indeed "SDR-SRAM", not "SD"- that's something for memory cards.

Is there even a BIOS for this board that properly supports the K6-2/III+ processors?

Doornkaat wrote on 2021-03-04, 12:39:

Is there even a BIOS for this board that properly supports the K6-2/III+ processors?

Not really. The CD14 BIOS introduced support for the K6-2, CD114 Beta states experimental support for the K6-3 with the usual disclaimer. There is even a CN versions for boards with the 586B SB.
Anyway, it got the job done good enough.
There is another BIOS CD115 with documentation whatsoever. By the look and feel of it its no improvement, rather to the contrary.

K6-3(+) was rather for the fun of it. My revision of the board can only supply as little as 2.8V Core which is too much for comfort for the long run.

K6-2 was quite stable; K6 300 - rated for 2.2V - wasn‘t at all. That is why I decided to settle the proper processors...

Couldn't the lacking support for the K6-III+ be the problem then?

Doornkaat wrote on 2021-03-04, 19:12:

Couldn't the lacking support for the K6-III+ be the problem then?

I am not so much worried about the K6-3+ as it was just a proof of concept if you will - and it worked quite admirably: It was the only processor to run with a 75MHz FSB beyond 300MHz core speed.

I am rather trying to find out the reason for a low Speedsys score across the board.

Other issue in relation to this board: Does anyone have a hint as to which VIA drivers would be best suited for the VP2?

Since it is not an AGP chipset I am rather trying to get DMA mode working for the HDD in Windows 95 OSR 2.1.

I have tried 4.25(1)a and 4.35 but to no avail. Win 95 dosen't load anymore as soon as I enable DMA mode.

Again, Anandtech hinted at performance issues until he got hold of the correct drivers and he states "VIA busmaster IDE drivers v2.03". Unfortunately I was unable to dig them up...

Don't know about 2.03, but 2.3.18 should do the trick.
http://www.vogonsdrivers.com/getfile.php?file … &menustate=37,0

Despite the filename, the archive contains drivers for all named OSs, including Win95.

dionb wrote on 2021-03-07, 00:43:

Don't know about 2.03, but 2.3.18 should do the trick.

Muchas gracias, will give it a try and report back.

Will hopefully brighten my day. It got really dark since the middle notch on the CPU socket snapped off this morning when trying to mount the cooler on one of those beautiful but "obese" Cyrix MIIs...