VOGONS


First post, by GroundVale284

User metadata
Rank Newbie
Rank
Newbie

I had received a Gigabyte GA-5SMM motherboard from Ebay, in what appeared to be an original anti-static bag, similar to an OEM replacement part. Plugging in as much of the small front panel accommodations, and the RAM that the board came with, I got the motherboard to boot with a red COMPAQ Logo, after which it attempted to boot into my spare Windows 98 machine's drive. Obviously, it would not be able to boot it properly, but I just wanted to see if it could actually boot into something.

After that was confirmed, I was hoping to get into the BIOS to adjust some settings and ensure that it boots to CD first to install a new version of Win98. I checked the manual, cleared the CMOS battery, and attempted to power on the machine again. Afterwards, things had definitely gone wrong.

Pressing the Power Switch runs power through the machine, but the machine fails to boot. My constant speed fans spin, but the CPU fan isn't spinning and feels a bit warm when running. I've left a PCI board tester in both situations, and instead of the usual flashes of symbols that go on during boot, I receive the "--" "--" in the digit display.
The following lights are illuminated in the diagnostic PCI Card:
-12V, +12V, +5V, +3.3V: Illuminated
CLK, RESET: Illuminated
IRDY, FRAME: Not Illuminated.

When the PC had proceeded normally but not booted into an OS, the PCI Card shows the following:
Digit Display: "06""0A".
-12V, +12V, +5V, +3.3V: Illuminated
CLK: Illuminated
IRDY, FRAME, RESET: Not Illuminated.

I've tried a lot of things so far, and nothing had appeared so far to wake it out of its permanently resetting state. Removing the RAM from the machine and powering it on causes the same results in the PCI Diagnostic Tool.

I've checked the back end of the board and there doesn't appear to be any debris causing shorts, and all the Front panel pins appear to be un-bent and intact. No shorts there either.

Did I goof up and accidentally slaughter a board I just picked up?

Reply 1 of 9, by GroundVale284

User metadata
Rank Newbie
Rank
Newbie

OK, for some reason what appeared to solve the issue was plugging in a PCI Video Card. No clue why in the world an integrated graphics card would wish to default to a Non integrated display adapter, but compaq was a bit wild back then.
I will still need to diagnose IO issues to get started with getting a Windows 98 installation on this machine.

Reply 2 of 9, by rasz_pl

User metadata
Rank l33t
Rank
l33t

>default to a Non integrated display adapter

was definitely not the source of your problem. That determination happens Very far into BIOS initialization process.

>RESET held high

did you connect all the case buttons? you might have broken/sticking reset

https://github.com/raszpl/sigrok-disk FM/MFM/RLL decoder
https://github.com/raszpl/FIC-486-GAC-2-Cache-Module (AT&T Globalyst)
https://github.com/raszpl/386RC-16 ram board
https://github.com/raszpl/440BX Reference Design adapted to Kicad

Reply 3 of 9, by GroundVale284

User metadata
Rank Newbie
Rank
Newbie
rasz_pl wrote on Yesterday, 08:39:
>default to a Non integrated display adapter […]
Show full quote

>default to a Non integrated display adapter

was definitely not the source of your problem. That determination happens Very far into BIOS initialization process.

>RESET held high

did you connect all the case buttons? you might have broken/sticking reset

I've swapped the power and reset buttons and both function just fine. No shorts on the front panel either.

When the board was stuck in its "Reset" line on, starting the board without RAM did not change the results. Using a Screwdriver to power on the board without any front panel connections had the same results as well.

I'll have to confirm by capturing some in-BIOS screenshots later, but in BIOS there's an "Init Display First" setting that is set to PCI instead of AGP by default after reset.
My assumption is that It checks if there's a VGA card in the PC as one of its first checks or something, because not a single digit on the PCI tester displays, and the RESET line is constantly on without the card.

I'm quite distrustful of pressing the RESET button at all. When I shut it down, I just hold the power button instead of touching the reset button to speed up the reboot.

Currently the BIOS I'm dealing with is extremely limited, and has barely half of the features shown in the BIOS manual. For some reason, F10 enters the BIOS instead of "Delete" in the manual. Do you happen to know any "advanced BIOS" keyboard shortcuts that I could try to get more features?

Reply 4 of 9, by weedeewee

User metadata
Rank l33t
Rank
l33t
GroundVale284 wrote on 2026-05-18, 18:58:

I got the motherboard to boot with a red COMPAQ Logo,

GroundVale284 wrote on Yesterday, 17:31:

For some reason, F10 enters the BIOS instead of "Delete" in the manual.

F10 is pretty normal for a Compaq

Right to repair is fundamental. You own it, you're allowed to fix it.
How To Ask Questions The Smart Way
Do not ask Why !
https://www.vogonswiki.com/index.php/Serial_port

Reply 5 of 9, by EduBat

User metadata
Rank Member
Rank
Member

Maybe you had something metallic inside the PCI slot, connecting the reset pin to ground, which then moved away after you inserted a card... (Just a thought...)

Reply 6 of 9, by rasz_pl

User metadata
Rank l33t
Rank
l33t
GroundVale284 wrote on Yesterday, 17:31:

I've swapped the power and reset buttons and both function just fine. No shorts on the front panel either.

but was Reset from front panel connected when you had problems booting? They can get stuck just temporarily. That would completely explain what happened

GroundVale284 wrote on Yesterday, 17:31:

Using a Screwdriver to power on the board without any front panel connections had the same results as well.

well, there goes my theory

GroundVale284 wrote on Yesterday, 17:31:

in BIOS there's an "Init Display First" setting that is set to PCI instead of AGP by default after reset.
My assumption is that It checks if there's a VGA card in the PC as one of its first checks or something, because not a single digit on the PCI tester displays, and the RESET line is constantly on without the card.

CPU doesnt run _at all_ when reset is high, and checking which Video to initialize happens quite far into the Bios code.

GroundVale284 wrote on Yesterday, 17:31:

Currently the BIOS I'm dealing with is extremely limited

I would just reflash it to normal non branded bios

https://github.com/raszpl/sigrok-disk FM/MFM/RLL decoder
https://github.com/raszpl/FIC-486-GAC-2-Cache-Module (AT&T Globalyst)
https://github.com/raszpl/386RC-16 ram board
https://github.com/raszpl/440BX Reference Design adapted to Kicad

Reply 7 of 9, by bracecomputerlab

User metadata
Rank Newbie
Rank
Newbie

I think what happened to you is what exactly happened to me.
In fact, I also obtained my GIGABYTE GA-5SMM from a Compaq box back in 1999.
If my research is correct, the computer in question was Presario 5360.
It had AMD-K6-2 380 MHz, 64 MB of PC100 SDRAM, 4.3 GB Seagate hard drive, and Windows 98.
I upgraded the system with the last release (I think it was version F6) of GIGABYTE retail BIOS due to it having better large (i.e., > 32 GB) hard drive support.
I eventually switched to a 60 GB IBM hard drive and installed Windows 2000.
If you switch to GIGABYTE retail BIOS, Compaq's restore software will no longer work properly so be aware!

I think I got it free from an Internet based startup that required me to connect up via their dial-up (yes, people still used 56K dial-up modem until early 2000s) service, and had to run their built-in adware (spyware) at least 20 hours per month.
That was the requirement for getting this free system.
This was more than a quarter century ago event, so I do not remember everything 100% correctly.
The full package also included a 15 inch monitor.
The system just showed up one day in front of my house delivered free of charge.
I think the whole system cost about $600 to $700 retail back in 1999.
A year or so later, this startup's spyware technology was purportedly purchased by another firm, so the users were told that we can have the computer and even gave instructions on how to permanently remove this spyware from our computers.
I used this computer until 2007 or so when I switched to ASUS P2B + Pentium II 400 MHz, and was quite surprised to see my Xilinx FPGA design software ran much faster on Pentium II.
That was pretty much the end of front line use of Compaq Presario 5360.
After that, I still played around to experiment with Ubuntu, but did not use it too frequently anymore.

Although it is OT, this mainboard has an odd bug that prevents proper operation if 512 MB or more RAM is installed.
It gives intermittent errors in memtest86 if one installs 512 MB or more RAM.
Based on my testing, it was running reliably up to 448 MB of RAM (3 slots, 6 rows of SDRAM).
256 Mbit SDRAM is apparently supported by the mainboard, not that it matters.

Okay, here is what is relevant to you.
I have had the exactly same issue of this mainboard suddenly dying one day.
I thought about a bad power supply killing the mainboard, although the power supply I used with it was always a small sized Delta power supply that originally came with the system.
Can the NOR flash that came with the mainboard just suddenly lose its content?
This happened to me between 2018 to 2021 (I do not really remember which year exactly.), and because I have a lot of other boards, I still have not bothered to troubleshoot what went wrong with this system.
That's just my 2 cents, and I hope it is relevant to you.
Reflashing the BIOS should be considered, although aflash memory writers costs around $60.

Reply 8 of 9, by GroundVale284

User metadata
Rank Newbie
Rank
Newbie
bracecomputerlab wrote on Today, 05:39:
I think what happened to you is what exactly happened to me. In fact, I also obtained my GIGABYTE GA-5SMM from a Compaq box back […]
Show full quote

I think what happened to you is what exactly happened to me.
In fact, I also obtained my GIGABYTE GA-5SMM from a Compaq box back in 1999.
If my research is correct, the computer in question was Presario 5360.
It had AMD-K6-2 380 MHz, 64 MB of PC100 SDRAM, 4.3 GB Seagate hard drive, and Windows 98.
I upgraded the system with the last release (I think it was version F6) of GIGABYTE retail BIOS due to it having better large (i.e.,
If you switch to GIGABYTE retail BIOS, Compaq's restore software will no longer work properly so be aware!

That was pretty much the end of front line use of Compaq Presario 5360.

I thought about a bad power supply killing the mainboard, although the power supply I used with it was always a small sized Delta power supply that originally came with the system.
Can the NOR flash that came with the mainboard just suddenly lose its content?
This happened to me between 2018 to 2021 (I do not really remember which year exactly.), and because I have a lot of other boards, I still have not bothered to troubleshoot what went wrong with this system.
That's just my 2 cents, and I hope it is relevant to you.
Reflashing the BIOS should be considered, although aflash memory writers costs around $60.

Thank you very much for the additional information!
From my best estimation this motherboard is definitely some sort of replacement for a compaq computer in the 5300 Series. Before I reset the BIOS, the primary video feed was going through the integrated graphics, as far as I'm aware.
F10 brings me to a super limited BIOS, there's no BIOS settings as far as I could see for any degree of configuration for drives.

I'm planning on using an 80GB Maxtor IDE HDD for the project, where there's 30GB for the OS and the rest would store games and such. Does anyone else know how to access a more extensive BIOS for this machine? I'm hoping to restore the machine, minus the case, and plus potentially a socket 7 CPU Upgrade, to be a compaq machine at least on paper.
I think I'll have to continue my studies on why exactly the motherboard has been acting like that, but it gives me hope this little system is somewhat operational aside from figuring out why the heck it can't boot to anything in particular.

Reply 9 of 9, by GroundVale284

User metadata
Rank Newbie
Rank
Newbie
bracecomputerlab wrote on Today, 05:39:
I upgraded the system with the last release (I think it was version F6) of GIGABYTE retail BIOS due to it having better large (i […]
Show full quote

I upgraded the system with the last release (I think it was version F6) of GIGABYTE retail BIOS due to it having better large (i.e., > 32 GB) hard drive support.
I eventually switched to a 60 GB IBM hard drive and installed Windows 2000.
If you switch to GIGABYTE retail BIOS, Compaq's restore software will no longer work properly so be aware!

Although it is OT, this mainboard has an odd bug that prevents proper operation if 512 MB or more RAM is installed.
It gives intermittent errors in memtest86 if one installs 512 MB or more RAM.

The motherboard itself Allegedly comes with a Cyrix MII PR333 (haven't been able to 100% Confirm, but the jumpers are set to match it), and 64MB of SDRAM. I have 3x 128MB SDRAM to try and upgrade it with for a maximum of 380MB of RAM, but I'm keeping the build conservative if possible.

Could another reason why I can't actually see the drive I've hooked up into it be the fact that the drive I'm using is 60GB large? The best solution to this may be a CompactFlash Adapter and a 32GB CompactFlash card.
The Windows 98 installation I was using for testing is on a 512GB Drive with a 27GB partition I had whipped up on GParted using my Athalon XP build, cause I have no other IDE compatible motherboards and I am unsure what to trust to use an IDE adapter for USB.

Also, unless it's aboslutely needed, I'm thinking of keeping the BIOS if I can manage it.