Your board looks very similar to this one, it might be a rebrand https://www.ultimateretro.net/en/motherboards/2045

Moreover it has a bios for the chipset you're looking for so even if that's not the correct board, it should be similar enough to be working with your board.

Can you tell me what's the POST string for the bios you currently have ? (Or even dump it here), maybe we can figure out which board needs that bios and potentially add it to our database 😀

Edit: nevermind, we don't have that bios, I thought we had one but nope 🙁

Feel free to use our bios search function. Maybe you can find a bios for a different board that uses the same chipset

Edit 2: nope that's the only record of a bios for that chipset we have and it's empty 😢

You got me excited there for a minute - but unfortunately I found that listing about a month ago and that is as close as I have ever gotten.
I’ll dump a few of the BIOS chips I have and help with populating the database though if it helps someone in the same predicament as me 😀

Sorry for the false hope 🙁

Hopefully someday someone will upload a bios

No problems - thanks for trying though 😀

MrBuzzy wrote on 2022-01-03, 02:58:

Should we take this to a new thread? […]
Show full quote

Phoenix1975 wrote on 2022-01-03, 00:27:

Confirmed that “UNIFLASH -SAVE filename” works a treat and result in the same file that my programmer gets off the bios. I used […]
Show full quote

Phoenix1975 wrote on 2021-12-31, 12:28:

I use an EPROM programmer, but I believe it can be don in DOS using UNIFLASH.

Can get a copy and try it out on one of my boards first if you need instruction. If the chip is a xxx512 it should give you a 64k file, or 128k for a xxx010 or xxx011

Confirmed that “UNIFLASH -SAVE filename” works a treat and result in the same file that my programmer gets off the bios.
I used v1.40 which was easy to find. Just run it under dos with no memory managers etc running.
Good luck and thanks for your help

Should we take this to a new thread?

Here's a BIOS dump to test out. I wasn't able to use uniflash it requires PCI. This one was made with NSSI.
If you can confirm it works I'll provide it to UltimateRetro.

The sticker on the chip says ISA bus. I thought it might say ISA/VLB. Not even sure VLB is active.
And my board has fake cache confirmed with cachechk.
I'm interested to know how you go.

I tried the image - no go. I also tried running from 08000h and it posts up to 06h - which is where I think it tries to decompress the BIOS and fails there.
It looks like other areas of the image are missing as well - though looking through it it's certainly the image I have been chasing

I will fire up a similar 486 board and try out NSSI and also UNIFLASH with the "-BASE" option (F0000h for System BIOS??) which looks like can run without a PCI bus if you can identify the base address of the ROM.
Glad I bought a programmer - remembering all the 'fun' I had back in the 90s navigating AMIFLASH etc trying to find software that worked....

What is a CS8006 chipset? Who makes it?

Phoenix1975 wrote on 2022-01-03, 23:36:

I will fire up a similar 486 board and try out NSSI and also UNIFLASH with the "-BASE" option (F0000h for System BIOS??) which looks like can run without a PCI bus if you can identify the base address of the ROM.
Glad I bought a programmer - remembering all the 'fun' I had back in the 90s navigating AMIFLASH etc trying to find software that worked....

I just tried "UNIFLASH -BASE F0000" on another 486 without PCI and it worked - I got a full BIOS image.
It seems to want PCI for flashing, not reading thankfully.
Can you please give this a try?
Thanks 😀

Anonymous Coward wrote on 2022-01-04, 01:16:

What is a CS8006 chipset? Who makes it?

From google - HiNT Chipset (consists of HMC HM82C206 and HiNT CS8006)

Continuing from Re: 486 Retro Gaming Build

Here's another try using `UNIFLASH -BASE F0000`.

Note Uniflash reported it could not detect the type.

Weird. You usually only see HiNT used on those mini-EISA boards.
This one was made really late too.

Sorry, but no go. Twice the size, but contains exactly the same data.
All FF from 0h to 7500h, and the rest starts looking like the BIOS image from there.
Comparing with other AMI BIOS images I have, it seems that only the second half (8000h to FFFFh) is the BIOS, but something is blocking the first half from coming through - and it confirms that the start address of the BIOS image you are taking is correct...
Maybe the chipset does not allow direct access to this address rang, or there is some BIOS shadowing or something going on? Maybe someone here can shed light on this - it beats me...

Not quite sure what I’m looking at with those screenshots. Can see ‘92 was changed to ‘94 etc. I’ll try again with no shadowing on.

The screenshots were your BIOS capture at the top, and another UNICHIP board below. Was just showing we are on the right track.

I have been reading up on shadowing, and tried this which seemed to work on my board - UNIFLASH -BASE FFFF0000

It seems shadowing will relocate F0000 to FFFF0000 in RAM. Worth a try if you can

Found this here: Re: Preserving your BIOS EPROMS and other firmware.

Also, dumping mobo BIOS via NSSI will only work up to 64k images (so many 486+ boards will not be dumped correctly), and only if the ROM shadow is OFF. And even then the chipset might be internally mapping some SRAM onto it and allow writes to store some BIOS config values. So the dumped image will have wrong checksum and fail to boot

So, if the mobo supports disabling shadowing it might help. Or ffff0000 if we are lucky 😀

I didn't see an explicit option to disable rom shadowing yet. Only options to shadow select addresses(?)
I've initially loaded the default safe settings, and then saved both f0000 and ffff0000.

No go I'm afraid. The F0000 looks the same as previous, and the FFFF0000 is BIOS data, but does not look like a dump and shares no similarity to the other dumps. Thinking the PC may be saving it in RAM in a decompressed format?
Starting to think this may be the reason there are no dumps online - the chipset may be restricting access somehow.
Really just guessing now - but maybe if you can try at F0000 with cache turned off?
Also, make sure you are running pure dos without memory managers (himem etc). I disable config.sys/autoexec on startup.
Thanks for helping 😀

Well I'm not giving up, will get an EEPROM reader!

MrBuzzy wrote on 2022-01-05, 00:11:

Well I'm not giving up, will get an EEPROM reader!

Now we’re getting serious!

The programmer - you will not regret it - I got a TL866-II (from the xgecu official store) with a bunch of adaptors and have used it to fix a number of motherboards very successfully. A lot of faulty boards I purchase have corrupted EEPROMS so they can be quite an easy fix.