VOGONS


First post, by SETBLASTER

User metadata
Rank Member
Rank
Member

someone offered me this motherboard
battery is in great condition, but before selling.. he wanted to test it , when he turned it on, the got a "bios ROM checksum error" "detecting floppy drive A media" "insert system disk and press enter"
does this error mean the bios should be reflashed?

this is the picture of the board. but there are no markings as of what model this board is, how do you know if its a 4SAW or a 4SAW2 or a 4SAW5 ?

also on vogons drivers there is a pack of bios for this board, but its not clear since

-one bios says : wa3 for 4saw2 ?

-another bios says : 53 for 4saw2 ?

-there is no bios for 4saw and no bios for 4saw5.

image-2021-05-06-222608.png

4sa1.jpg

Reply 1 of 18, by Horun

User metadata
Rank l33t++
Rank
l33t++

It looks same as some pictures of a 4SAW2 but that probably means nothing. here is one picture:
https://thumbs.worthpoint.com/zoom/images1/1/ … fdeca777158.jpg

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 2 of 18, by SETBLASTER

User metadata
Rank Member
Rank
Member
Horun wrote on 2021-05-07, 02:34:

It looks same as some pictures of a 4SAW2 but that probably means nothing. here is one picture:
https://thumbs.worthpoint.com/zoom/images1/1/ … fdeca777158.jpg

thanks, do you know why there are 2 bios? one named 4saw-wa3 and anotherone called 4saw-53 ?

Reply 3 of 18, by PC Hoarder Patrol

User metadata
Rank l33t
Rank
l33t

53 looks like a fix (from Sep 96) to deal with issues relating to some newer CD-ROM drives

EDIT: Don't see board differences mentioned in the manual either

https://web.archive.org/web/19970331114743/ht … al/4sw-mu20.pdf

Reply 4 of 18, by Horun

User metadata
Rank l33t++
Rank
l33t++

Am wondering if the variants are due to slightly different Winbond I/O chips, in factory they could alter suffix to signify a variant. 4SAW for 486, Sis, unknown on A, W for winbond i/o and it's rev. Just a guess in my feeble little mind 😁

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 18, by PC Hoarder Patrol

User metadata
Rank l33t
Rank
l33t

This page seems to indicate WA3 was for AMD X5 support & 53 was for a move to AWARD 4.51PG

https://web.archive.org/web/19970204221623/ht … bios/index.html

Also, here's a thread re. a 4SAW5 board, but it seems to have a UV EPROM What kind of new bios chip to use?

Reply 6 of 18, by mkarcher

User metadata
Rank l33t
Rank
l33t

The digit 2 or 5 at the end of the model number of the board just tells you what amount of L2 cache came preinstalled. The 2 indicates 256KB of cache in 4 chips of 64K x 8 (and a 16Kx8 tag ram), whereas the 5 indicates 512KB of cache in 4 chips of 128K x 8 and a 32K x 8 tag RAM. I'm unsure whether the name without the digit at the end is meant for a cacheless board, or meant as generic name for both the 4SAW2 and 4SAW5.

Reply 7 of 18, by Horun

User metadata
Rank l33t++
Rank
l33t++
mkarcher wrote on 2021-05-07, 19:04:

The digit 2 or 5 at the end of the model number of the board just tells you what amount of L2 cache came preinstalled. The 2 indicates 256KB of cache in 4 chips of 64K x 8 (and a 16Kx8 tag ram), whereas the 5 indicates 512KB of cache in 4 chips of 128K x 8 and a 32K x 8 tag RAM. I'm unsure whether the name without the digit at the end is meant for a cacheless board, or meant as generic name for both the 4SAW2 and 4SAW5.

Thanks and interesting. I did not see any ref to that in the manual PC Hoarder linked but take you word for it. Curious how you figured that out, the actual board images gave me no clue.. but am often cluleess ;p

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 9 of 18, by PC Hoarder Patrol

User metadata
Rank l33t
Rank
l33t

The cache (naming) option makes perfect sense, and even the only page which refers specifically to the 4SAW - a download page for the 0911 BIOS (improved ls-120 support) - has a link to the v2.0 4SAW2 / 4SAW5 manual. Even the product page fails to provide a hint

https://web.archive.org/web/19970204222456/ht … kt/sy4saw2.html

Reply 10 of 18, by mkarcher

User metadata
Rank l33t
Rank
l33t
Horun wrote on 2021-05-08, 00:49:

Curious how you figured that out, the actual board images gave me no clue.. but am often cluleess ;p

The last digit of the model number being the cache size is a common theme for Soyo boards at that time, e.g. the 4SA (dual-bank cache), 4SAW (cost-reduced version of the 4SA with single-bank cache) or the 5TF (a Pentium board with the 430TX chipset). I don't remember for what board I got to know about this numbering scheme, and where I read it, though. For the 5TF, there officially exists a 5TF0 version without any on-board cache (it still has a COAST slot for cache upgrade, though). I could not find references to a 4SA0 or 4SAW0, though.

Reply 11 of 18, by Horun

User metadata
Rank l33t++
Rank
l33t++

Thanks mkarcher ! am saving your text in my soyo archives.
Found this page that list one newer bios for the 4saw: 4SAW0911.BIN 1997-09-11
https://web.archive.org/web/20040611185659/ht … s/dlmissing.php
link to file: https://web.archive.org/web/20040919160441/ht … 86/4SAW0911.BIN or at https://www.elhvb.com/mobokive/Archive/Soyo/

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 13 of 18, by evasive

User metadata
Rank Oldbie
Rank
Oldbie
SETBLASTER wrote on 2021-05-07, 01:41:

does this error mean the bios should be reflashed?

No, it means the battery has gone flat after all this time of not being charged and because of that the board lost its settings. If it is kept under power for a while, the settings are done properly, most likely it will work. However, the battery by design is a chemical time-bomb, you better replace it wil a lithium coin cell solution or an external battery pack.

Reply 14 of 18, by chrismeyer6

User metadata
Rank l33t
Rank
l33t
evasive wrote on 2021-05-11, 07:25:
SETBLASTER wrote on 2021-05-07, 01:41:

does this error mean the bios should be reflashed?

No, it means the battery has gone flat after all this time of not being charged and because of that the board lost its settings. If it is kept under power for a while, the settings are done properly, most likely it will work. However, the battery by design is a chemical time-bomb, you better replace it wil a lithium coin cell solution or an external battery pack.

I second this statement. Reflashing your bios won't fix your dead battery that's why your system can't save it's time or bios settings.

Reply 15 of 18, by mkarcher

User metadata
Rank l33t
Rank
l33t
Horun wrote on 2021-05-08, 15:04:
Thanks mkarcher ! am saving your text in my soyo archives. Found this page that list one newer bios for the 4saw: 4SAW0911.BIN 1 […]
Show full quote

Thanks mkarcher ! am saving your text in my soyo archives.
Found this page that list one newer bios for the 4saw: 4SAW0911.BIN 1997-09-11
https://web.archive.org/web/20040611185659/ht … s/dlmissing.php
link to file: https://web.archive.org/web/20040919160441/ht … 86/4SAW0911.BIN or at https://www.elhvb.com/mobokive/Archive/Soyo/

Be careful with BIOS updates on that board! The most recent versions of the BIOS initialize the chipset in a wrong way (IIRC 4*PCI instead of 3*PCI + VL), so that the performance of VL graphics cards is extremely low. It's a single bit you need to flip, and you may be able to do that with modbin (IIRC it's award), but I didn't get around to do a professional fix yet, especially as my copy of this board also needs some trace rework near that "chemical time bomb".

Reply 17 of 18, by mkarcher

User metadata
Rank l33t
Rank
l33t
evasive wrote on 2021-05-17, 09:34:

If you can elaborate on that fix, we might be able to put the proper bioses on the UH19 page with a remark in the Notes about this issue.

It's about bit 3 in PCI config space register 0x57. It chooses between LBD#/LRDY# or REQ2#/GNT2#. This bit is set the wrong way in recent 4SAW BIOS images.