backing up the BIOS

...and upload it to vogons. I think there's a later revision available.

Received the board yesterday and spruced it up by swapping the battery with a CR2032 with a battery holder; removed D2 to prevent the charging circuit from being used. Plus, I repaired the slight trace damage the board had from the battery and it works. The board is supposed to work with a WB cache CPU, but it doesn't want to boot with it (I get trippy 1960's colors at the Windows 95 splash screen). Even in WT mode, the CPU runs like an i486 DX2-66, which is not supposed to happen.

Going to swap a BIOS chip with another version and see what happens.

Newer bios (from 1997). I don't have revision info though.

Well, the L2 cache jumpers and TAG chip were wrong. After changing that, it boots with the CPU set to WB mode. I'll test the other BIOS and see what happens.

I tested the 1997 BIOS, but my HDD refuses to boot or throws bad cluster errors with ScanDisk and won't continue. Going to try it again.

Hello,

ich have the same mainboard and was wondering, if someone could tell me the values ot the capacitators sitting on L3 and L4 (to buy them), on my board both are missing. I get the "Keyboard error or keyboard not present" error message.
Maybe i'm totally wrong these are not needed an i have look into a different direction.
The board is booting up in general until this keyboard error message.

I would appreciate any answer.

kimschi wrote on 2021-06-01, 18:41:

ich have the same mainboard and was wondering, if someone could tell me the values ot the capacitators sitting on L3 and L4 (to buy them), on my board both are missing. I get the "Keyboard error or keyboard not present" error message.
Maybe i'm totally wrong these are not needed an i have look into a different direction.
The board is booting up in general until this keyboard error message.

If you are talking about the components called L3 and L4: Those are not capacitors, those are inductors. Most likely they are somewhere in the 10µH range. They should read a very low value when you probe them with a resistance meter. They don't usually go bad.

On the other hand, the 4SAW has a peculiar layout: The keyboard connector is connected to its own ground which is not directly connected to system ground. You need to plug the jumper next to the mouse port header to provide proper ground to the keyboard connector (at least that is the case on my copy of the board that has a lot of battery-induced corrosion).

thank you very much for your answer, i will replace them with these inductors .

I'm very happy, this board is working again.
Now i have to update the Bios.

Hello,
I have exact board, do you have the motherboard's manual (for this exact model)?
Thank you

kimschi wrote on 2021-06-02, 16:58:

thank you very much for your answer, i will replace them with these inductors .

I'm very happy, this board is working again.
Now i have to update the Bios.

Hi, do this board accept EDO ram? Thanks

lhtan73 wrote on 2024-03-27, 01:15:

kimschi wrote on 2021-06-02, 16:58:

thank you very much for your answer, i will replace them with these inductors .

I'm very happy, this board is working again.
Now i have to update the Bios.

Hi, do this board accept EDO ram? Thanks

https://theretroweb.com/motherboards/s/soyo-sy-4saw

bjwil1991 wrote on 2020-05-24, 02:45:

Bought a good, functional (battery leakage present, will fix the issue by replacing the battery with a CR2032 (got the battery h […]
Show full quote

Bought a good, functional (battery leakage present, will fix the issue by replacing the battery with a CR2032 (got the battery holder, as well as the board manual, BIOS, and battery replacement guide on vogonsdrivers.com prior to) Soyo SY-4SAW2 Rev WA3 motherboard.

Odd portion is the keyboard port. It's been replaced with a PS/2 keyboard connector, which is odd, yet satisfying, plus, it has a header for a PS/2 mouse, which is good (might look and see if that actually works).

Board works accordingly, just needs a new clock battery, slight trace patches, and that'll do.

Gallery: https://imgur.com/gallery/seKiVQF

Front of the board:

BIOS:

Hardware, RAM, cache, and other info:

Cache (512KB to be exact):

PS/2 keyboard connector, PS/2 mouse header (4x2), and NiCd battery (slight leakage):

Closeup of the battery and slight trace corrosion (will be easy to fix):

Rear of the board:

The board will be swapped with the PCChips M912 v1.7 board that is installed at the moment as well. I have a PCI video card, network card, 176MB worth (128MB (2x 64MB) EDO 60ns and 48MB (3x 16MB) FPM 70ns) of RAM, sound cards, VLB video card, Evergreen 586 (will set to Write-Back mode to test the turbo functionality to see if the board works with both WT and WB cache for turbo), and everything else that's already installed (FDDs, HDD, CD drives, keyboard, and mouse) will be used instead of swapping those around.

The board itself supports a PS/2 mouse, which is unusual for an AT 486 motherboard to support it, but it's pretty cool.

Has the following:
4x PCI slots
4x ISA slots + 1 VLB slot
512KB L2 cache installed (maximum supported, sadly, but will do)
PS/2 keyboard connector in place of the AT keyboard connector, which is also interesting to see on a motherboard
PS/2 mouse header on the motherboard (will test that out with a PS/2 mouse connector header I have in my Socket 7 system)
Integrated I/O (FDD, Primary IDE, Secondary IDE, Parallel, and Serial port headers)
Leaking NiCd battery that'll be removed and a CR2032 battery will be installed for the CMOS settings and D2 will be removed to prevent the charging circuit from being used
The board itself is in working order and will be getting the treatment to make it more appealing.

This'll be mostly for benchmarkings and game playing, including running Windows 95.

The pictures above are the seller's photos in case you're wondering. Once I receive the board, I'll be doing the battery swap, trace inspection, cleaning, and repairs (to be safe), then post pictures of the board itself, and last but not least, backing up the BIOS and making another BIOS chip as a spare in case the one in the system stops working one day.

Hi, Saw that you uses EDO ram for this board, workable? Need to set anything in BIOS for the EDO ram to work? Thanks

lhtan73 wrote on 2024-03-27, 01:15:

kimschi wrote on 2021-06-02, 16:58:

thank you very much for your answer, i will replace them with these inductors .

I'm very happy, this board is working again.
Now i have to update the Bios.

Hi, do this board accept EDO ram? Thanks

The SY-4SA and SY-4SAW have the SiS496 / SiS497 chipset. Newer BIOS releases contain the necessary code to auto-detect EDO RAM and initialize the chipset accordingly. You don't need to configure anything to make use of EDO RAM, but there is one important twist: Only the very latest revision of the SiS496 chip supports EDO RAM, and a lot of SY-4SA(W) boards contain older revisions. You need to check the revision code printed on the SiS496 chip. If it is anything except "PR" (like for example "OR" or "NV"), the hardware does not implement the EDO support bits that get programmed by new BIOS versions.

mkarcher wrote on 2024-03-30, 13:33:

lhtan73 wrote on 2024-03-27, 01:15:

kimschi wrote on 2021-06-02, 16:58:

thank you very much for your answer, i will replace them with these inductors .

I'm very happy, this board is working again.
Now i have to update the Bios.

Hi, do this board accept EDO ram? Thanks

The SY-4SA and SY-4SAW have the SiS496 / SiS497 chipset. Newer BIOS releases contain the necessary code to auto-detect EDO RAM and initialize the chipset accordingly. You don't need to configure anything to make use of EDO RAM, but there is one important twist: Only the very latest revision of the SiS496 chip supports EDO RAM, and a lot of SY-4SA(W) boards contain older revisions. You need to check the revision code printed on the SiS496 chip. If it is anything except "PR" (like for example "OR" or "NV"), the hardware does not implement the EDO support bits that get programmed by new BIOS versions.

Hi, Thanks for the critical information. Indeed mine do not have the latest chip with "PR", too bad. But for 64M FPM, most are double sided modules. Saw in some post to add pin header to J19 and closed the pin1 & pin2 will enable all the memory banks to accept double sided modules. Have you tried?

Also with the new BIOS, the speaker pins on JP18 seems not working as my internal speaker is not working at all not even a beep when there is boot error.

lhtan73 wrote on 2024-04-05, 02:38:

Hi, Thanks for the critical information. Indeed mine do not have the latest chip with "PR", too bad. But for 64M FPM, most are double sided modules. Saw in some post to add pin header to J19 and closed the pin1 & pin2 will enable all the memory banks to accept double sided modules. Have you tried?

While many 64M FPM modules have chips on both sides, they are still electrically "single-sided" modules. They have 32 (without parity) or 36 (with parity) chips, each providing one bit, to get one "side" of 32 bits. So you don't need to add modifications to allow 4 of these modules to work. On the other hand, adding that many memory chips (128 / 144) might overload the RAM controller and result in unstable operation. If that is the case, no jumper will be able to solve the issue.

lhtan73 wrote on 2024-04-05, 02:38:

Also with the new BIOS, the speaker pins on JP18 seems not working as my internal speaker is not working at all not even a beep when there is boot error.

I was wrong about having the 4SAW, I have the 4SA. I did not yet notice any PC speaker problems, but I know about a different issue with the latest BIOS (for the 4SA, I don't remember if it shares BIOS with the 4SAW): It breaks VL support. My board is currently broken due to battery leakage at the moment, and didn't get around to fix it for good, so I can't right now whether PC speaker support is broken with the most recent BIOS.