VOGONS


First post, by PDXTony

User metadata
Rank Newbie
Rank
Newbie

Hello awesome community!

I am running this exact motherboard (even contributed a BIOS dump and image!):
https://theretroweb.com/motherboards/s/bek-tronic-bek-v409

I have been enjoying a 486 DX-2 66mhz on it for some time and decided to get myself a Pentium Overdrive 83mhz CPU to drop in.
According to the manual, there are no jumper settings to change to go from a 486 to the Pentium Overdrive, so I didn't change anything.

The problem statement is - if I have external cache enabled in the BIOS, the memory count at bootup is slower than usual and it doesn't count up to the correct number of memory. It also doesn't continue after the memory is counted - it will hang during floppy drive seeking.

If I disable external cache in the BIOS, it counts all the memory normally, but there seems to be some stability issues at that point where Windows 95 either says there's not enough memory or MIMEM.SYS has detected unreliable XMS memory. The BIOS also allows me to enable/disable internal cache - if I disable that as well, there's no improvement in booting into Windows 95 without the same errors.

With the 486 CPU, enabling external cache works without any issues, and there have been zero stability issues in using the computer.

The question is, is this expected behavior?
If not, might you have a recommendation for what I should do to enable external cache and not have it do this?

Also, here's my actual motherboard in case you wanted to see exactly what's populated in each cache socket:
https://theretroweb.com/motherboard/image/bek … b5843158603.jpg

Thank you!
Tony

Last edited by PDXTony on 2023-07-15, 22:19. Edited 2 times in total.

Reply 1 of 5, by PDXTony

User metadata
Rank Newbie
Rank
Newbie

I'd be really interested if there's a different BIOS I could try with this motherboard. I have ROM programmer and some spare 256k and 512k EEPROMs!

Should I try 15ns cache instead of the current 20ns? BTW, the speeds for the currently populated are set to be 3-1-1-1 since it's 20ns at 33fsb.

Reply 2 of 5, by TheMobRules

User metadata
Rank Oldbie
Rank
Oldbie

From the date codes on the chips of your board it seems to be from mid to late 1993. Keep in mind that Intel changed the electrical specs of the Pentium Overdrive shortly before its release, so motherboards from before 1995 usually do not work properly with that CPU. In some cases you can make it work but at the expense of crippling performance by disabling L1 WB cache for example.

Reply 3 of 5, by PDXTony

User metadata
Rank Newbie
Rank
Newbie
TheMobRules wrote on 2023-07-15, 17:57:

From the date codes on the chips of your board it seems to be from mid to late 1993. Keep in mind that Intel changed the electrical specs of the Pentium Overdrive shortly before its release, so motherboards from before 1995 usually do not work properly with that CPU. In some cases you can make it work but at the expense of crippling performance by disabling L1 WB cache for example.

Well, crud, that must be it! Oh well. Thank you for your reply 😀

Curious, what would the chances of the 486 Overdrive DX40DPR100 100MHz V1.1 working on this motherboard?

Reply 4 of 5, by Horun

User metadata
Rank l33t++
Rank
l33t++

That ODPR100 will probably work but most likely be reported as a DX2 due to your BIOS. I am running one on a similar VLB board w/o issues.

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 5 of 5, by PDXTony

User metadata
Rank Newbie
Rank
Newbie

Well I decided to experiment by flashing the latest version of Mr. BIOS for this specific chipset.

Guess what? The Pentium Overdrive CPU now works perfectly.

So for those searching and having similar issues, it could be a BIOS issue and an update is in store.