get a post card ?

replace the bios chip ?

weedeewee wrote on 2024-12-20, 18:35:

get a post card ?

replace the bios chip ?

Did both. The POST card isn't of much help as it stops spitting any codes after 0B 0D, despite the board showing the POST sequence. It's the only mainboard to do so, as I tested a QDI P6I440BX and it went through the normal POST codes.

As for BIOS chip, I flashed the latest BIOS (JI4333) over both its original chip and a newer SST 39SF010 from a similar PA-2013, except rev2.1, that was dead. Same symptoms.

Hi, maybe you've already tried, but from what you say it could be a disk controller problem (?), if you can disable the on-bord one, you should see if it starts from a floppy, a DOS or Windows 9X boot disk is enough, if it starts in the sense that it doesn't crash, then maybe you could try a PCI controller, just to understand if it works, and if it goes there is something in the integrated one that doesn't work well.
If instead it crashes immediately after changing the BIOS screen, then it could be a CPU problem(?), try a simple one, like a Pentium 75-166 not MMX if you have it.

PC@LIVE wrote on 2024-12-20, 21:47:

Hi, maybe you've already tried, but from what you say it could be a disk controller problem (?), if you can disable the on-bord one, you should see if it starts from a floppy, a DOS or Windows 9X boot disk is enough, if it starts in the sense that it doesn't crash, then maybe you could try a PCI controller, just to understand if it works, and if it goes there is something in the integrated one that doesn't work well.
If instead it crashes immediately after changing the BIOS screen, then it could be a CPU problem(?), try a simple one, like a Pentium 75-166 not MMX if you have it.

Tried all kinds of CPUs (as per the original POST - P150, Cyrix 6x86 non-MMX, K6-2, P166MMX), as well as disabled all onboard storage controllers (both IDE channels and even floppy). Same result, hangs with a flashing cursor.

Offboard controllers result in, more-or-less, the same outcome. The only difference is that the hang occurs AFTER the offboard controllers (in my case, tested with AHA-2940UW SCSI controller as well as a PCI Promise Ultra100 TX2) finish loading their own BIOSes.

Have you checked over every inch of the board and every solder joint? A chipped SMD resistor\capacitor, bent leg on a chip, a bent-over solder joint on the back of the board or even an arrant bit of some debris that happens to be bridging something could be a possibility. I have found things like this on boards that I have looked at (and worked on) multiple times. They can really be hard to spot.

Sounds like you have been really thorough though...

Ozzuneoj wrote on 2024-12-20, 22:32:

Have you checked over every inch of the board and every solder joint? A chipped SMD resistor\capacitor, bent leg on a chip, a bent-over solder joint on the back of the board or even an arrant bit of some debris that happens to be bridging something could be a possibility. I have found things like this on boards that I have looked at (and worked on) multiple times. They can really be hard to spot.

Sounds like you have been really thorough though...

Everything. I've even ultrasound cleaned the board and to no avail.

PcBytes wrote on 2024-12-20, 22:33:

Ozzuneoj wrote on 2024-12-20, 22:32:

Have you checked over every inch of the board and every solder joint? A chipped SMD resistor\capacitor, bent leg on a chip, a bent-over solder joint on the back of the board or even an arrant bit of some debris that happens to be bridging something could be a possibility. I have found things like this on boards that I have looked at (and worked on) multiple times. They can really be hard to spot.

Sounds like you have been really thorough though...

Everything. I've even ultrasound cleaned the board and to no avail.

Puzzling indeed...

Just throwing this out there, but are you positive that you are using the correct BIOS for your board revision? These boards can have some confusing markings, as mentioned in this thread:
Re: FIC PA-2013 Ver 2.1 and Voodoo3 3000 AGP

Quote from me almost 8 years ago:

On my board, it says 2.1 right where your yellow arrow is pointing, next to the back of ISA slot 2. However, in between PCI slots 3 and 4 the following is printed on the PCB in white: 2.0 E-0036 . So, here it says 2.0, but on the bottom left corner it says 2.1. I believe the printed number by the ISA slot is what people should look for.

And some more information about board revisions here:
Re: FIC PA-2013 Ver 2.1 and Voodoo3 3000 AGP

I'm more than confident I have checked everything out.

I have a dead Rev2.1 board (which the newer SST 39SF010 chip actually came from) and was able to pinpoint my POST-ing board to being the earlier 2.0 revision.
I got my BIOS off Jan's page, however the issue has been present with the original BIOSes as well (1.15JI31S for my "working" board, 1.15JI37S for the dead one).

The only thing I have not tried is downgrading the BIOS core version (4.60PGA) to the older 4.51PG that seems to be on TRW.

As for my revision, it's definitely a Rev2.0 board. There's no 2.0 E-0036 sticker anywhere.

PcBytes wrote on 2024-12-20, 23:12:

I'm more than confident I have checked everything out. […]
Show full quote

I'm more than confident I have checked everything out.

I have a dead Rev2.1 board (which the newer SST 39SF010 chip actually came from) and was able to pinpoint my POST-ing board to being the earlier 2.0 revision.
I got my BIOS off Jan's page, however the issue has been present with the original BIOSes as well (1.15JI31S for my "working" board, 1.15JI37S for the dead one).

The only thing I have not tried is downgrading the BIOS core version (4.60PGA) to the older 4.51PG that seems to be on TRW.

As for my revision, it's definitely a Rev2.0 board. There's no 2.0 E-0036 sticker anywhere.

It was worth a shot. 😀

These are nice boards. I hope you can get it sorted out.

Since it is able to get through POST, that probably rules out a lot of things... but what is left is certainly puzzling. I'm wondering if one small component of supporting logic on the board has failed.

A bit of quick googling on this issue seems to turn up some posts about cache-related problems, but I know you have disabled the onboard cache already. Keyboard or keyboard controller issues may also be another possibility.

Has it always been this way as long as you have owned it, and you recapped it to try to fix the problem? Or did the problem start at some point while you had it? Either way it may be worth looking at anything that could be affected by externally-introduced static discharge... like, components near the keyboard\mouse\USB ports for example. I've seen damaged components near these ports (possibly from a previous owner plugging in a PS/2 device while the PC is on) create all sorts of strange problems.

Though it is more likely to be helpful for a board that won't even POST, if you have access to a thermal camera (even a cheap one) it can sometimes help to locate components that are either unusually warm or unusually cold compared to others.

Also... you've probably checked them a dozen times, but check the jumper configuration and be sure that none of the jumpers are loose. I have seen loose jumpers also cause strange problems.

Keyboard works fine - I can access the BIOS and no ghost-stuck-keys happen.
As for heating up - the only thing heating up is the chipset itself, however I found out that it's normal.
I also checked the jumpers - a dozen times, to be sure I installed them correctly for the CPU I use.

And for owning it - it's about 4 or 5 days since I've owned it. I didn't power it up pre-recap as the original caps were in very bad shape.

PcBytes wrote on 2024-12-21, 01:59:

Keyboard works fine - I can access the BIOS and no ghost-stuck-keys happen. As for heating up - the only thing heating up is the […]
Show full quote

Keyboard works fine - I can access the BIOS and no ghost-stuck-keys happen.
As for heating up - the only thing heating up is the chipset itself, however I found out that it's normal.
I also checked the jumpers - a dozen times, to be sure I installed them correctly for the CPU I use.

And for owning it - it's about 4 or 5 days since I've owned it. I didn't power it up pre-recap as the original caps were in very bad shape.

Forgive me for saying lots of obvious stuff, but have you tried it with no keyboard or other peripherals? Of course, to test this you might need to set the BIOS to not pause boot on keyboard errors.

No clue but the history of survival of those FIC 2013 boards is not a good one, not compared to their 503+ which seem to rarely have an issue.
Sounds like most everything you got in that "lot" is defective in one form or other....hope you got them extremely inexpensive 😀
Added: early ATX boards or those combo AT/ATX PSU boards are known for issues in my tiny local circle of geeks.

Horun wrote on 2024-12-21, 05:16:

No clue but the history of survival of those FIC 2013 boards is not a good one, not compared to their 503+ which seem to rarely have an issue.
Sounds like most everything you got in that "lot" is defective in one form or other....hope you got them extremely inexpensive 😀
Added: early ATX boards or those combo AT/ATX PSU boards are known for issues in my tiny local circle of geeks.

Now you've got me thinking about mine...

It was in my first really awesome intentional "retro" PC, which I built for DOS and 98SE back in 2005 or so. It had a Rendition Verite 1000 and 2x Voodoo 2 12MB in SLI, with a K6-2 500 and a SB16 CT2230, and I had it in one of those weird stubby AT-looking ATX towers with the side mounted power supply. If I pulled the cover off of the case it would expose all of the CPU speed jumpers on the PA-2013 (since they are near the top right corner), so I could easily change the speed for sensitive games. Also had a DVD drive in it so I could burn scads of files and old games to DVDs on my main PC and read them in 98SE, including disc images for games with FMV. 20 years ago this was much MUCH cheaper than having dedicated flash drives for this stuff.

Anyway, I migrated the board and the hard drive to a nicer and less irritating ATX mid-tower (now with just a TNT2 Pro and Yamaha YMF-719E) probably 8 years ago after learning a bunch of stuff here on Vogons. Now, it has been at least a few years since I have messed with it. I just opened it up and the caps all still look good, so I think mine is plague free at least. I will test it out tomorrow to see if it still functions or if it has become another statistic. 😀

EDIT: Just checked it and it seems to be totally fine. I remember now, the hard drive (DiamondMaxPlus 9) died several years ago and I was so bummed that the whole setup was lost that I never bothered setting it back up. Hearing the dreaded clunk of death when I powered it on today jogged my memory... bleh...

Ozzuneoj wrote on 2024-12-21, 02:42:

PcBytes wrote on 2024-12-21, 01:59:

Keyboard works fine - I can access the BIOS and no ghost-stuck-keys happen. As for heating up - the only thing heating up is the […]
Show full quote

Keyboard works fine - I can access the BIOS and no ghost-stuck-keys happen.
As for heating up - the only thing heating up is the chipset itself, however I found out that it's normal.
I also checked the jumpers - a dozen times, to be sure I installed them correctly for the CPU I use.

And for owning it - it's about 4 or 5 days since I've owned it. I didn't power it up pre-recap as the original caps were in very bad shape.

Forgive me for saying lots of obvious stuff, but have you tried it with no keyboard or other peripherals? Of course, to test this you might need to set the BIOS to not pause boot on keyboard errors.

Just tried that right now and same result. It'll count RAM, pass the IDE detection, then hang with just a flashing cursor right before it should display the screen below (not mine, from amibay) :

EDIT: Nevermind, I somehow managed to get it unstuck. Setting the boot process to SCSI,C,A brought the screen above.

EDIT2: It still won't boot from any drive. The problem worsens though - IT AFFECTS OFFBOARD SCSI/IDE DRIVES TOO!

PcBytes wrote on 2024-12-21, 11:00:

Just tried that right now and same result. It'll count RAM, pass the IDE detection, then hang with just a flashing cursor right […]
Show full quote

Ozzuneoj wrote on 2024-12-21, 02:42:

PcBytes wrote on 2024-12-21, 01:59:

Keyboard works fine - I can access the BIOS and no ghost-stuck-keys happen. As for heating up - the only thing heating up is the […]
Show full quote

Keyboard works fine - I can access the BIOS and no ghost-stuck-keys happen.
As for heating up - the only thing heating up is the chipset itself, however I found out that it's normal.
I also checked the jumpers - a dozen times, to be sure I installed them correctly for the CPU I use.

And for owning it - it's about 4 or 5 days since I've owned it. I didn't power it up pre-recap as the original caps were in very bad shape.

Forgive me for saying lots of obvious stuff, but have you tried it with no keyboard or other peripherals? Of course, to test this you might need to set the BIOS to not pause boot on keyboard errors.

Just tried that right now and same result. It'll count RAM, pass the IDE detection, then hang with just a flashing cursor right before it should display the screen below (not mine, from amibay) :

EDIT: Nevermind, I somehow managed to get it unstuck. Setting the boot process to SCSI,C,A brought the screen above.

EDIT2: It still won't boot from any drive. The problem worsens though - IT AFFECTS OFFBOARD SCSI/IDE DRIVES TOO!

Seeing this image you have to understand exactly what he is doing (or is trying to do) right after the writing:
Verifying DMI Pool Data….
Usually it should access the drives, but there is something that is not right as it seems, maybe it could be a bad contact (soldering), if you have the possibility to start it, pressing for example on the CPU or other chips, you could find with some luck 🍀, which can be defective, maybe I would hypothesize that it is the I/O chip to have some problem, having something compatible on scrap boards, maybe it could be replaced, if that is the problem, and the BIOS chip and the content is correct, for this last thing, you would need the same card, or someone with the same card to tell you which one uses.

I pressed on every chip to no avail. It either does the same as in the original post or stops with a similar blinking cursor after Verifying DMI Pool Data.

PcBytes wrote on 2024-12-21, 15:31:

I pressed on every chip to no avail. It either does the same as in the original post or stops with a similar blinking cursor after Verifying DMI Pool Data.

I imagined that you had tried various things, well I would say to look at the I/O chip welds, and if you do well, to renew them all, this however guarantees the connection between chip and MB, but if the chip is damaged obviously it doesn't solve anything. You could also check if there is any interruption in the lines of that chip, a track could be defective, in short it could be a long job, but the easiest solution, would be to replace the I/O chip, at least so you would be sure that it works, and even if it is not said that this solves the problem.

Redid all the joints on the cache and SIO, same issue.

PcBytes wrote on 2024-12-22, 08:58:

Redid all the joints on the cache and SIO, same issue.

Ok well then I would exclude the welds, it remains to be understood if the problem is hw, that is a chip or something like SMD broken, or it is sw, here we need to understand if the BIOS is 100% working, possibly what problem is there? In the past it happened to me a few times, that the problem manifested itself despite the fact that the Hardware had no failure.