Try a different power supply.

The 300W model is most likely underpowered and the PSU-shaped object with an Apevia label should not be connected to any AC outlets if I'm to believe a few reviews of that thing. Most likely some of the capacitors inside are already bulging.

The 300w PSU has 22A on the 3.3v rail and 30A on the 5v rail. The Apevia has 16A on the 3.3v and 5v rails, it is brand new and I don't see any issues with the caps... Given how minimalistic I have the system right now, I cant see either of them being maxed out. I would expect it to reboot if that was the case and not just freeze. I went digging through my parts bin and found another power supply that I might be able to test with. I don't know the history and it falls between the other two. It has 24A on the 3.3v rail and 17A on the 5v rail.

Now having said that, I just finished the clean install on the Western Digital drive. So far it appears as it might be the drive! I have been running the game for nearly 30min now... I had not installed the Audigy drivers, since I was testing without the card installed anyway, so will do that next and see if it continues to run as expected.

Still does not change my opinion on the Apevia box. There's enough reported to fail after a few months of regular use. I can't find any reviews on reputable sites.

evasive wrote on 2020-05-28, 18:53:

Still does not change my opinion on the Apevia box. There's enough reported to fail after a few months of regular use. I can't find any reviews on reputable sites.

No doubting it is cheap, not apposed to replacing it if need be... Just want to focus on if it is the issue at this point.

I am installing the Audigy now and will repeat my test to be sure installing the drivers doest cause an issue.

I would be surprised the hard drive was the sole cause. Which makes me wonder if there is something about how the bios was detecting the drive or some other bios setting that might have an impact on the primary IDE channel.

Installing the Audigy drivers appears to be causing the issue. I put the card in without installing the drivers, canceling through any devices Windows detected. Then ran the game again. Was able to validate it was stable. Then installed the drivers (VXD) and rebooted, validated I had sound and then ran the game. This time it locked up as it has done before.

So I uninstalled the Audigy drivers and rebooted, running the game without any freezing even though the card is still connected.

I am going to reinstall the drivers again, but this time pull out the card. This will replicate the steps I took with the setup on the Maxtor drive. If I get the same results, I think it would be safe to assume it has to be the drivers.

Maybe a problem with IRQ? While using my Soyo 440BX based motherboard the OS sometime freezes when I run a game that tries to inicialize D3D or OpenGL mode (using TNT2). After reloading BIOS default settings and restart everything works fine.

Interesting... I could try forcing IRQs in the BIOS. The stability mystery deepens. I reinstalled the Audigy drivers (still VXD) and rebooted, changing nothing else. And sure enough it froze again. So once again I uninstalled them leaving the card in the system so I could validate a stable system again. It was.

Reinstalled the drivers again so that I could pull the card with the drivers still installed but this time it has continued to be stable! Now I am really confused. Ran the game for a good 10 min then quit and rebooted. Went back in thinking maybe that was an exception. But on testing again it is still stable.

The only doubt I had about resource conflicts was that the problem persisted when I pulled the card on the previous setup with the Maxtor drive (I never uninstalled the drivers there). I am going to put the Maxtor drive back in and uninstall the Audigy drivers and see if that is also stable.

Okay, now I am *really* at a loss... Put the Maxtor drive back in without changing anything else and the system has been stable. Audigy drivers were installed from prior testing. Only thing that is similar is that the card was put back in with drivers already installed.

maybe it has taken a different IRQ and now is not sharing it with anything else. Some cards can be really fussy about that.

I just never encountered such an issue. Wonder if even having the drivers installed without the card in the system could cause such a conflict?

Will keep installing other games and testing. Might grab a graphical benchmark and have it just run on a loop...

It would not surprise me in the least if you add hardware windows decides to throw out the table, tell the bios to do its thing and end up with another IRQ sharing on your sound card: round 3...

I have first hand experienced the early incarnations of PnP in DOS. Let's say I'm not surprised yet.

Well with the SB16 installed again, still stable. I am going to lock in the IRQs for the PCI slots. I had IRQ7 excluded and PnP OS set to disabled before. The Audigy has been using IRQ11 while it has been stable, so I will lock it to that.

So now I have Slot 1 set to 5 since it shares with the AGP, slot 3 set to 11, and slots 2,4/5 set to NA... Hopefully that keeps them from changing... Serial and Parallel ports are disabled, will probably disable USB too since I am not using it.

well I am back to the freezes... Pulled the SB16 again and re-enabled IRQ for USB...

I would consider leaving it in a minimal state and running prime95 on it to tax the cpu and see if it holds for an hour or so .

If ok , I would loop 3dmark on it for an hour or so .

That way, you can establish a stability baseline .

If still OK after that, I would start adding non essential peripherals .

EDIT : P3B-F is normally very reliable and compatible . I ran an AGP TNT, a PCI network card and at least 2 sound cards on the one I had back in the day (GUS PnP and Live and I forget what else) under Windows 98 SE with zero issues . USB was enabled and I had parallel port scanner and then a USB one .

I don't think is a resource conflict. My guess is the increased stability with removed sound cards is a red herring , but I could be wrong

The P3B-F should have an i/o voltage jumper I think somewhere near the upper right corner. Maybe that'll do the trick?

Edit: vcore and multipliers are set correctly in BIOS I assume?

Probably capacitors acting up after all this years. Does it freeze with all listed CPUs?

The Serpent Rider wrote on 2020-05-28, 22:33:

Probably capacitors acting up after all this years. Does it freeze with all listed CPUs?

What brand are the caps on this one ?

Okay will start with Prime95...

I/O voltage jumper is still in the default position, set to 3.50v

Yeah, had it freeze at least once with all the CPUs I have...

Having trouble making out the brand of the caps, will have to pull the board from the case. There are at least 3 different types, purple, black, and green...

Pulled both sound cards, and started a Prime95 torture test... Small FFTs. Using the PII 350, two sticks of Mushkin memory, hd, rom, and CF adapter...