I've got a 3 year old Gateway with:
P4-1.8Ghz processor on an Intel Monterrey mobo, (no builtin video, LAN or anything).
512Meg Samsung RDRAM

The problem: intermittent freezeups even when entering BIOS. This started happening out of the blue with no mods done to the PC within 2 months before.

What I've Tried:
Switched power supply and Hard Drive (which was getting noisy).
Pulled each PCI card out one at a time till none left.
Switched AGP video cards.
Reseated Memory.

Took all components out of the case and cleaned everything up. System ran fine for 2 weeks straight (it was on 24/7) then problem came back.

Right now I'm running it barebones, (mobo on a piece of cardboard, just video card and one HD hooked up). And still freezes up.

Don't know much about heavy-duty mobo troubleshooting so I'm looking for help.

Check the MB for bulging, leaking, or ruptured capacitors.

If they look ok, then run Memtest86 to check the RAM.

no visually bad capacitors. and i also ran the memory test. still no answer.

Is it an LCD monitor? Or have you tried another monitor? How about a pci video card instead of an AGP? Maybe a port is bad, even the connector.

Check the mobo battery voltage and if the BIOS settings are what they should be.

Still sounds like the PS though... maybe the mobo connector where the PS lead goes.

Lastly, has the os been repaired or reinstalled? Boot sector virus a possibility?

RD RAM huh? To replace, you'd be better off getting another mobo with different RAM

have had it hooked up to multiple CRT monitors. I have 3 diff. AGP video cards but no PCI. Damn thing even locks up in the BIOS with no drives (optical or hard drives) even connected. I ran memtest against the RDRAM with no problems. Oh well, please let me know if anyone has some ideas.

How are the CPU and system temperatures?

can't find a utility that supports the mobo and the BIOS has no builtin temp viewer. I just reseated the processor, cleaned and applied new thermal on heatsink and swapped out cpu fan just in case it was getting buggy. Still intermittent lockups though. The lockups don't seem to be heat-related (as in system freezing faster each time). I was actually able to play a round of Dawn of War, giving the PC a workout, with no lockup (until about 1 minute after i finished a game. {:) This box freezes up whether idle or doing heavy processing.
This PC is not my main one, friends of mine will use it for when we have a LAN party. Is there anything else to test? I figure after all the testing i've done (switching out or removing parts) that it's either the mobo itself or the processor. And if its the mobo, then I'm clueless about down and dirty troubleshooting on that particular part.

See if this MB monitor works:
http://www.pcworld.com/downloads/file_description/0,fid,7309,00.asp

If not, then remove the computer's cover, and run an external fan.
If that does not help, then overheating is not the cause of the problem.

But it appears that hardware is definitely the cause if the system ran ok for 2 weeks after you took all of the components out of the case, and cleaned things up.
Since you've swapped out everything except the MB, CPU, and RAM, and ran it barebones with just the PS, MB, CPU, boot drive, and video card installed, then remove the RAM, clean its contacts with 91% pure(or higher alcohol) and a lint free cloth, and reseat it.
If no luck and multiple RAM modules are installed, then insert just one module at a time. If one of the modules causes the system to freeze, then you've found the culprit.
If they all cause freezing, then insert the RAM in different slots(one at a time).
If no good, then clean the RAM slots on the MB with a pencil eraser.

If that does not work, then either the MB or CPU may be the cause, and swapping them out is the usual method for determining which one is bad.

thanks for the advice. I'm pretty sure its not heat-related as I forgot to mention that I installed a third fan. I have two sticks of memory and 2 banks of 2 slots (4 total). I haven't tried one stick at a time though, duh just moving them in tandem from one bank to the other, duh. I'll let you know if thats the culprit.

Someone advised above to check for bad capacitors. Info and pictures Here (http://www.pcguide.com/vb/showthread.php?t=25482) to show you what to look for.

I agree this sounds like a hardware problem, you might end up swapping everything in it one item at a time, but if that turns out to be necessary it should isolate the problem. Sure sounds like something is getting hot...Definitely try a fan blowing inside.

tried one stick at a time. first trial locked up in 10 minutes. second trial lasted at least a few hours. The only guess left that i haven't explored is the AGP slot being bad. I'll have to see if i can borrow a pci video card from a bud cuz i don't have any.

If the 2 trials were done individually with 2 different RAM modules, then install just the one module that allowed the system to run for a few hours without freezing, and see how things go.

tried pci video card and still the same issue.

I have x2 128 Meg modules (non-ECC) and the mobo has two banks of 2 slots (X4 slots total). I 've also got x2 CRIMM dummy modules.
All the modules have to be inserted as a rule. I've read that the memory has to be installed in pairs. So for the first test I switched banks with the CRIMMs, no luck.


Then I paired up one module with one dummy in each bank and tried switching around each bank. The bios recognized both sticks but only totaled my memory to be 128 Meg overall during these scenarios with the mismatched pairs. My question is: Am I effectively testing one module at a time during this configuration in order to weed out a bad stick? The bios recognizes all the modules correctly but the total memory listed is wrong by half. Is that a speed issue from not pairing them up?

If your MB uses dual memory architecture, then a minimum of 2 identical RAM modules must be inserted. It's also possible that the 2 modules must populate specific slots. Your computer's or MB's manual should have info on this.

But if you said that your system ran ok for a few hours with just one of the modules installed, then it's possible that its mate is bad.

yes but read my second paragraph on my last post. Even though i didn't pair it up correctly and the bios only included one stick in its total does that mean the second stick while not detected could've caused a freeze?

I'd hate to go buy that rare RDRAM thats unusable by most mobos and not have it be the problem.