I've been having some weird problems lately. My computer randomly locks up. It doesn't matter WHAT I'm doing, it just locks up randomly. I believe I've narrowed it down to one or more things, but first, my specs:

Tyan Trinity KT-A (KT133-A chipset)
Athlon 1.4GHz (T-Bird)
768MB RAM (PC133)
GeForce2 MX 400
80GB 7200 RPM Western Digital
8GB Maxtor
CD-RW
DVD-ROM
300-350W PSU (forget exactly, and its a pain to check it out on my PC)
Windows XP Professional
Linux (various distributions, currently Mandrake)

The PC is approaching the age of three years. The heat was a problem, but not anymore, since I have a massive fan. The old hard drive (another 80GB Maxtor) died a while ago, too. Now, the exact description of my problem:

Like I said, random lockups. They happen in either OS. When I shut down the PC after locking up, the 3D card is freakishly hot, and on boot, I usually check the CPU temp, and its a very respectable temp (nowhere near anything that could cause lockups). I suspect power is the problem, but not 100% sure. The reason: It locked up, and it rebooted spontaneously seconds later, and when Windows returned, no BSOD alert (its a fresh install, not really changed or anything). I don't think its the power itself, but I'm going to pick up a UPS this weekend to test that. But, at a LAN party, it worked fine, for some strange reason, and it was running for several hours. However, I'm looking for a bit more input on this problem, and it may actually be heat affecting non-CPU components (its beginning to heat up here, and it gets pretty hot during the summer).

I also suspect power since I've gotten the following voltages from MBM and the BIOS, but I'm not a guru, so just check these out:


3.3 LOW 3.31 HIGH 3.33
5.0 LOW 4.51 HIGH 4.86
12.0 LOW 12.07 HIGH 12.54

Voltages are within 10% of rated value....

If the video card is overheating it can cause these kind of problems. I would look into adding a fan to it.

But how can the 3D card cause it to spontaneously reboot? Its also not caused me any problems in three years...

I don't know how it works exactly, but it is true that an overheating video card can shut down a system. If it is "freakishly hot" as you say, it is the most likely suspect and it could even be dying if it didn't run that hot in the past.

And Welcome to http://www.pcguide.com/ubb/pcgubb.gif

Could be that the fan is about to go on the card. I've lost 2. (I keep using old fans as replacements.) The heat sink could be full of dust.

Well that ain't healthy.

I was taking out the 3D card soon after locking up, and, behold, its a normal heat. Touching the motherboard around it, however, is extremely hot. Hmm...Thats not a really good sign for the motherboard...

Any tips for finally figuring out the root cause?

<counts money>

Make sure the slot for the card is clean, it could be that you have a dust caused short that is heating things up. Any type of short could be the reason for the heat. You could try running with a fan blowing on the video card (case open) and see if the problem happens again.

I was trying something else...I removed one of my RAM sticks (512MB, only about 6 months old, ordered from Crucial)...

IT WORKS.

And I'm angered by that.

You might want to try cleaning the slot out real well and clean the contacts on the RAM with alcohol (really pure alcohol) then try it again. Or try it in another slot. Or try it and run a RAM tester like the one HERE (http://www.bootdisk.com) to make sure the RAM is really bad. It may simply be that it wasn't seated properly or contacts dirty or.....

Checked the slot, and I currently have one of the two 128MB sticks in there.

Its running fine, and I'm going to run the memory tester I generally use (memtest86) when I get back (I'm leaving the country for about a month).

Thanks for all the help.