Soyo 486 PCI/VESA GREEN MAINBOARD dead (changed: NOW ALIVE)

Hot-swapping and Boot-Block flash & Boot block flash and floppy support
Post Reply
viveka
New visitors - please read the rules.
Posts: 14
Joined: Sun Jan 26, 2003 7:54 pm
Location: Brazil

Hi all. This is my first message. I hope someone can help me...
I have a "Soyo 486 PCI/VESA GREEN MAINBOARD" motherboard,
and the original BIOS was a Award:
Award Modular BIOS v4.50 PG
1984-95 Award Software Inc
REV WA
09/11/95-SIS-496-497/A/B-2A4IBS29C-00
Award Plug & Play V.1.0
I know it's old but it worked really good with Linux!
It was running pretty good with a Intel 486 DX4 at 100 MHz.
So I have got a AMD 5x86 - P75 - 133 MHz.
I installed it on the mainboard and made the adjusts on the
jumpers, but it worked only at most at 120 MHz.
So I got a - use at your own risk - BIOS upgrade file at
Soyo site.
I did the flashing with awdflash, but at the end the machine
froze. It seemed the flashing was done completely.
I rebooted the machine and it worked fine at 133 Mhz
for some 10 minutes, and then it suddenly went dead and
it doesn't switch on anymore.
I am looking for the original BIOS file for this motherboard,
if anyone have and help me, I would find a way to
retribute. Other ideas are welcome.
Thanks all
Last edited by viveka on Sun Feb 02, 2003 7:34 pm, edited 1 time in total.
Rainbow
The UniFlasher
Posts: 3122
Joined: Wed Mar 20, 2002 4:16 pm
Location: Slovakia
Contact:

If it suddenly died, I doubt that it's BIOS problem. Have you set the CPU voltage properly?
Patched and tested BIOSes are at http://wims.rainbow-software.org
UniFlash - Flash anything anywhere
viveka
New visitors - please read the rules.
Posts: 14
Joined: Sun Jan 26, 2003 7:54 pm
Location: Brazil

Surely. Both CPUS, the 486 DX4 and the 586-P75 work with 3,45 volts, so that I didnt change this jumper on the motherboard.
NickS
BIOS Bodhisattva
Posts: 3145
Joined: Fri May 03, 2002 10:34 am
Location: Thames Valley, UK

The Intel Overdrive DX4 had the voltage regulator on the chip - I had a friend who could not operate at full speed on the lower voltage, the Overdrive wanted the higher voltage. Are you sure the jumper is set to 3.45V ?
viveka
New visitors - please read the rules.
Posts: 14
Joined: Sun Jan 26, 2003 7:54 pm
Location: Brazil

Surely it was set to 3,45 Volts. The DX4 worked for 8 years without any problem. It was in fact a very stable machine. The DX4 is not a Intel Overdrive chip.
NickS
BIOS Bodhisattva
Posts: 3145
Joined: Fri May 03, 2002 10:34 am
Location: Thames Valley, UK

Time to get a voltmeter out and find whether you still have 3.45V, or whether an on-board regulator has blown. This has happened to me with a 486 board where the 3.45V regulator was the size of a small transistor with no heatsink! I used the Overdrive version of DX4-100 instead.

I also have a 586-P75 which ran happily at 160 MHz until it was retired.
viveka
New visitors - please read the rules.
Posts: 14
Joined: Sun Jan 26, 2003 7:54 pm
Location: Brazil

Now I've got a new mainboard, a Shuttle HOT-433, and the processor Interl 486-DX4 is running OK on it, at 100 MHz. I guess the processor isnt hurt. The Soyo motherboard is still dead, be with the 486, be with the AMD-586-133MHz.
It worked 10 minutes after the flashing, so it suddenly stopped and never more worled :(
Rainbow
The UniFlasher
Posts: 3122
Joined: Wed Mar 20, 2002 4:16 pm
Location: Slovakia
Contact:

SHARP PQ30RV21: Variable Outpt Low Power-Loss Voltage Regulator.
Datasheet here http://sharp-world.com/products/device/ ... 0rv1_e.pdf
Pin 1 (DC Input) should have 5V
Pin 2 (DC Output) should have 3.45V
Pin 3 (GND) should be connected to ground
Patched and tested BIOSes are at http://wims.rainbow-software.org
UniFlash - Flash anything anywhere
NickS
BIOS Bodhisattva
Posts: 3145
Joined: Fri May 03, 2002 10:34 am
Location: Thames Valley, UK

I'm not sure whether we should lock this thread or whether Soupy can cut and paste the relevant bits of http://www.wimsbios.net/phpBB2/viewtopi ... highlight= as viveka seems to have posted twice on the same problem from different angles.
viveka
New visitors - please read the rules.
Posts: 14
Joined: Sun Jan 26, 2003 7:54 pm
Location: Brazil

How I did a successful hot flash and brought to life my motherboard

(You people helped me, now I shall help the others, so I will tell my story)

I am the happy owner of a Soyo motherboard – the 486 PCI/VESA Green Mainboard – 4SA W2/W5. Its chipset is SiS-85C496-497, very stable in my opinion.

It had been running with a processor Intel 80486 DX-4 3.45V at 100MHz since year 1996. In September, 2002, I got a processor AMD X5-P75 (which is also called 586) that runs at 133 MHz.

Well, the motherboard's user's guide, the original manual that came along with the hardware, said that the board supported the 586 processor. So I put the processor, adjusted the jumpers and turned it on. But the processor was recognized as a AMD 486DX Enhanced and was running only at 120 MHz.

In fact, in the user's guide, the jumper settings for the AMD 486DX Enhanced and the X5 were the same, except for the clock rate settings. Maybe it was a printing error.

I went to Soyo site and discovered that a second version of the manual was available, printed in 1996. I downloaded it and compared it entirely with my printed version from September 1995. There were basicaly two main differences, and one of them was exactly that of the X5 processor configuration.

I reprogrammed the mainboard according the new manual but it didn't make any good. The processor was still being run at 120 MHz.

Again at the Soyo site, I saw some BIOS files. There were two BIOS files available: the older 4SAW53, and the newer 4SAW0911. I downloaded 4SAW0911.BIN, the last version for my mainboard. The BIOS was from Award, so I got AWDFLASH version 8.16 too.

The first strange thing was that AWDFLASH identified the BIOS chip correctly: a Macronix MXIC 28F1000P/12V, but it didn't manage to read the current BIOS program in order to make a backup file. When ordered to do so, it just froze.

Well, if I couldn't save a backup version of the BIOS, I would not do the flashing.

Some days later I got the HWINFO-486 for DOS utility at the Internet. It came with an utilitary program called GETBIOS. So I run GETBIOS and it seemed to read the BIOS and save it to files E000.ROM and F000.ROM. Each file had 64 Kbytes, so I thought: "well, these are the 128 Kbytes of the BIOS, the fact that it is split into two files doesn't matter; after all, the file 4SAW0911.BIN has 128 Kbytes".

Happy for the backup done, I loaded AWDFLASH 8.16 and went on with the flashing.

The progress bar went till the end, and at this moment the machine froze.

"Well", I thought, "it went till the end, it worked".

I rebooted the machine and now... ufff, it had really worked. The processor now was being recognized as AMD Am5x86-P75 at 133 MHz. Good.

I entered the Setup, things were a little different there. I loaded DOS and HWINFO, it reconized the processor finely.

Some ten minutes after, I was still running HWINFO and it was saying that now the machine was at 75 MHz, something like that.

"Ooops", I thought, "I forgot to Load Setup Default Values at the BIOS Setup after the flashing", so I immediately turned the machine off.

I turned it on again and I stayed waiting, waiting, waitng...

I did it once more, and waiting, waiting, waiting... It didn't turn on anymore. No beeps.

No matter what I did, with Intel 486 or AMD 586, it didn't work.

I almost threw it away. I thought it had a relation with the bad-done BIOS flashing, then I begin to study the BIOS world.

People at the capital city fix motherboards and they say they can flash any BIOS. First I thought they had some special chip writer, but then I had the idea to use my Duron's mainboard. (Only later I found out that the chip sockets had to be compatible in order to hot flash.)

I asked people if it was possible to use a motherboard to flash a BIOS chip of another motherboard and they told me "yes, in fact many people do that, it's called hot swap [sic]".

Hey, that sounded good.

I opened the files E000.ROM and F000.ROM by curiosity at the Windows Notepad, and did the same with 4SAW0911.BIN. Man, they were very different. I did the same with other .BIN and conclued that it wouldn't be wise to flash E000.ROM+F000.ROM to the BIOS chip without asking first.

In the meantime I started to search for the original BIOS file. It was a 2A4IBS29C-00 REV WA dated 09/11/95. Bad luck: nobody had a motherboard like that, it seems I am its last user of the world. Pretty bad luck: Soyo provides BIOS upgrade files at their site, not the original BIOS files, the first ones which came on the board. What a ship!

I know that the original file would recognize the processor wrong, at 120 MHz again, but at least it would be able to put the motherboard back to work.

At Wimsbios web forum people told me that I couldn't flash those files [E/F]000.ROM obtained with GETBIOS. I would have to get a .BIN file!
(http://www.wimsbios.net/phpBB2/viewtopi ... 3327#13327)

At this moment I was tired of searching for the original BIOS. I would flash it with the 4SAW53.BIN this time. Modbin77 said it was a 2A4IBS29C-00 REV WA53.

At Wimsbios web forum people recommended using UniFlash. Now I was discovering the truth, AWDFLASH wasn't really any good. I got UniFlash 1.29.

In January 2003, I got my sencond 486 board, a Shuttle HOT-433 Version 4.0. It has chipset UMC UM8886BF/UM8881F.

Good luck: its BIOS chip socket was a 32-pin! It has BIOS chip Winbond W29EE011-15.

Bad luck: that chip BIOS has programming voltage of 5V.

Good luck: thank Godness, this motherboard has a jumper for setting the programming voltage! It is JP 8, 3-pins, and according to the manual, open jumpers is the same as pins 2 and 3 closed: 5 volts. JP8 was open, what means 5 Volts programming voltage.

I bought two chip sockets, and inserted them into the motherboard's chip socket, so that it became taller, and was more easy to remove and put the BIOS chips without touching other components.

I turned the system on with Winbond chip. Then DOS was loaded.
I removed Winbond chip. System is still working.
I closed jumper JP8 pins 1 and 2, now Vpp is 12V. System is still working.
I put the other chip, the MX, at the tall socket. System is still working.
I loaded UniFlash 1.29.
I "flashed with bootblock" the file 4SAW53.BIN.
UniFlash says: SUCESS.
I quit UniFlash and turn off the system.
I removed the contacts, opening jumper JP-8.
I removed MX and put Winbond chip back.
I turned on the system, Shuttle HOT-433, it is still working (good!).
I put the MX chip at the Soyo motherboard.
I turned it on.
Beep!

Man, my old 486 board is back to life. I hope it last some ten years more.
The system has 32 MB of FPM RAM, 1 MB Trident TGUI-9440, 2 HDs (1,6 Gbytes and 400 Mbytes) and network card. It was really worth saving it, wasn't it? :D Now I have to decide if it will be a Windows 95 or a Linux machine...

Thanks for everyone who helped me at Wimsbios Forum.
Rainbow
The UniFlasher
Posts: 3122
Joined: Wed Mar 20, 2002 4:16 pm
Location: Slovakia
Contact:

Nice story :) I just don't know how the BIOS became corrupted... If you have the possibility to hot-flash again, it might be worth flashing the latest BIOS again.
BTW.: You're lucky that I got a board with the UMC chipset and did support for it in UniFlash (hacked out of the Award BIOS with Turbo Debugger and some very old Awdflash) because there is no info about it anywhere.
Patched and tested BIOSes are at http://wims.rainbow-software.org
UniFlash - Flash anything anywhere
viveka
New visitors - please read the rules.
Posts: 14
Joined: Sun Jan 26, 2003 7:54 pm
Location: Brazil

Hi Rainbow.
I don't know either how the BIOS became corrupted, since it worked at first I thought everthing had to be fine. Perhaps AWDFLASH, which froze at the end of the flashing, didn't write the boot block correctly?
I don't think I have to hot-flash again, maybe using UniFlash I'll be able to correctly flash the latest 4SAW0911.BIN this time, directly on the Soyo motherboard (?).
Thanks for doing that about adding support for UMC chipset, it was extremely useful for me, it saved my board! :D Good hacking!
Rainbow
The UniFlasher
Posts: 3122
Joined: Wed Mar 20, 2002 4:16 pm
Location: Slovakia
Contact:

You can try it on the Soyo board - the SiS 496+497 chipset is supported too. Hope that the BIOS does not have some self-corrupting capability.
Patched and tested BIOSes are at http://wims.rainbow-software.org
UniFlash - Flash anything anywhere
NickS
BIOS Bodhisattva
Posts: 3145
Joined: Fri May 03, 2002 10:34 am
Location: Thames Valley, UK

My AMD 586-PR75 runs at 160MHz but the BIOS says 120 MHz. Speed check utilities show it is really 160 MHz, so I don't worry about it.
Post Reply