Troubleshooting DISASTER! Can you figure this out?

[chron]

Hello everyone, thanks for taking the time to view my problem. I'll keep things as brief as possible.

My system:
P5W DH Deluxe, E6600, G.Skill DDR2-800, X1950XT 512MB, OCZ 700W GamerXStream, 1x80GB SATA, 1x320GB SATA

My problem:
Computer freezes (locks up) after detection of IDE drives then reboots.
If the computer makes it past the IDE detection then the windows load screen freezes (locks up) and reboots.
If I make it into windows, the computer might work fine for a while, then lock up and reboot.
Please keep in mind; I see no pattern to the crashes.

What I have done so far:
I reset CMOS.
I removed all cards I didn't need to boot (sound, TV tuner)
I have disconnected both hard drives in order to boot from CD and reflash bios.
I have tried flashing the bios (system locks up before bios flash can start)
I have replaced the PSU from a 500W to a 700W.
I have downclocked the CPU from 266MHz fsb to 150MHz fsb.
I set RAM timings to SPD, set voltage to 2.2, set speed to 200MHz (50% decrease.)
Put in two sticks of ballistix DDR2-667 but computer did not post (incompatible?)
No noticable difference in symptoms after any of these actions!

What may have caused it:
I recently updated the P5W's bios to 1901. There were no issues at first.
While downloading a call of juarez patch in the background via a torrent, I opened up a large video file (episode of futurama.) The computer automatically rebooted itself the moment I opened the video file. Symptoms started immediately upon restart.

My personal suspicions:
I think the bios may be corrupt or messing up the system, or perhaps the motherboard is just faulty in some way.

The hard drive is making no concernable noise, and has always operated at a safe temperature, but when I connect the hard drives to different SATA connections, it seems to help it get past the "IDE Detection" step (but again, keep in mind that diconnecting both drives resulted in the same sort of lockups.)

I don't suspect CPU or RAM as the same symptoms occur no matter what voltage/speed they are at. (Plus, the cpu is fairly new. About a month.)

I've heard of "hot flashing" where you take out the bios chip while the computer is on, so is it safe to assume that once the computer gets past post and starts to load windows that it is no longer a major effect on the system? I mean, once it gets to windows boot screen, could the bios chip STILL cause a lockup/reboot?
PLEASE HELP!

[xtrempinoy]

if your system boots up then its not a bios problem.. umm try to enable EVERYTHING on your motherboard..and if that doesn't work, try the opposite and try disabling everything you don't need... there might be a feature in the bios that messing up your computer

[chron]

Well I've disabled everything, unluding the legacy usb support, every onboard device, and every chipset/cpu feature. So far it has gotten past post twice in a row and made it into windows, but the problem still exists where windows reboots itself.

I'm starting to rule out bios issues and now my intuition is leaning toward a bad stick of ram, although I've always been under the impression that ram gives BSOD more than reboots/restarts.

Perhaps a componant of the motherboard its self is dead?

[DAK1640]

Try re-loading windows

[chron]

I would re-load windows if the system didn't lockup. I've got vital data on the hard drives anyways.

[Lood007]

Make a bootable memtest86+ cd, boot to it and run the test. Disconnect all hard drives when you do this.

[DAK1640]

Try 1 stick of Ram...Do you have a different GPU to try ?

[pcdoc1]

It's either a corrupt bios or a mem problem... Try one stick of ram and re-flash...

[chron]

It's either a corrupt bios or a mem problem... Try one stick of ram and re-flash...

Try 1 stick of Ram...Do you have a different GPU to try ?

Its DDR2 - doesn't it need to be ran with 2 sticks at a time?

Make a bootable memtest86+ cd, boot to it and run the test. Disconnect all hard drives when you do this.

I have one from my last set of Corsair sticks. When it starts to run the system locks up and reboots.

[Lood007]

Have you checked your CPU temps?
Oh, and DDR2 does NOT need to run in pairs.

[rmd2001]

Run Scandisk to check hard drive

[chron]

Well, I tried one stick of the ram with no improvement. I think I can rule out ram as I tested both sticks this way.

CPU temperatures are fine (25C idle, don't know about load as the system locks up.) If the cpu were overheating I think I'd feel it. The base of the tuniq tower 120 feels cool.

The only thing left is for me to test the cpu. I'll take the P4 out of my mom's dell and put it in. If the same symptoms occur is it safe to say the motherboard is corrupt?

Run Scandisk to check hard drive

With the hard drives disconnected, the same lockups occur in programs like memtest 86 and the ezflash utility.

[Tac0]

The only thing left is for me to test the cpu. I'll take the P4 out of my mom's dell and put it in.

I dunno cpu's either work or NOT...they don't flake out and die. I'd suggest loading Windows on a known good or new hdd.

[chron]

Replacing the CPU with a P4 resulted in the SAME SYMPTOMS! I havn't been able to test this P4 because it pisses my mom off when I take her computer apart and she thinks I'll break it (a computer I built for her lol.) My mom is at work tonight, what a perfect time to take this p4 out of it ;P

The only thing left is the motherboard. I have replaced the video card with a PCI Radeon 32MB 7200, tried single sticks of ram, disconnected CD drives, replaced the psu from 500W to 700W, removed all pci cards i didn't need, disconnected hard drives, reconnected hard drives to new SATA outlets on the MB, you name it. Finally I have to accept the motherboard simply is broken somehow.


UPDATE:

took 80gb drive with os and hooked it up to sata1 and then disconnected the 320gb... Computer downstairs boots fine; fault motherboard is still the correct diagnosis.


THANKS FOR ALL YOUR HELP EVERYONE

[DAK1640]

BIOS is corrupt

[Doom5]

What heatsink are you using? The backplate making electrical contact with the board after tension is applied can cause problems; I've seen this happen twice personally, and more times over on multiple message boards.

[K33B4]

I've had a problem similar to this due to IDE drives on my AMD system with DFI Lanparty UT Ultra-D. The only thing i did to fix the whole issue was to unplug the ide drives and boot the system to SATA drive. Then i plugged up the ide drives while the system was on and happen to fix the problem. Now be cautious bout pluggin up the IDE drives while the system is running because each system reacts to situations differently. Mine worked without a glitch but can't say this will be a 100% fix. You can't run IDE HDD's insync with SATA HDD's due to some sort of glitch in OS. I do not think that your MOBO is messing up.

[chron]

lol you plugged the ide drives in while the system was on and it actually FIXED a problem not cause one? Thats pretty cool... You must be kind of a risk taker like me when it comes to electronics and stuff. If the new motherboard does not fix the problem I think I'll just have to use the 320gb hd only instead of the 80gb plus the 320gb. Its all SATA btw.

The reply about the bios being corrupt might be correct, but it also might be an electrical componant with the motherboard. Either way RMA gets here thursday and I think I'll flash to 1503 and leave it alone forever. 1901 seems to have caused problems for a lot of people and although I can't say with 100% certainty that it 1901 was the culprit, I will say it seems odd to come accross the problems shortly after updating to that bios version.

[LazyBum]

What heatsink are you using? The backplate making electrical contact with the board after tension is applied can cause problems; I've seen this happen twice personally, and more times over on multiple message boards.

Yeah that can make some weird crap happen, I had some real frustrating problems with a SLI-DR until I put some electrical tape over any exposed metal.

[chron]

Yeah that can make some weird crap happen, I had some real frustrating problems with a SLI-DR until I put some electrical tape over any exposed metal.

I actually messed up an e6400 with a tuniq tower 120. Touched the motherboard with a screw and shorted it out. Hopefully intel will rma.

I lined the motherboard tray with duct tape, and even with the tuniq tower unscrewd and just lieing on it, i get the same issues.

Actually, my MB has that problem where if its too tight it won't boot, and usualy when i see this I can recognize if its a matter of a heat sink being too tight.

We will see on thursday I guess when the new MB gets here. I really don't see what else it could be.

[DAK1640]

I hope it all gets sorted out quickly for you. I know how frustrating that can be...

[chron]

I hope it all gets sorted out quickly for you. I know how frustrating that can be...

Thanks man. Hey, I noticed you can run 3.6GHz at 1.32v!? Has it always OC'd that well or does the 1503 bios provide the best performance? On the 1707 bios I needed 1.55v to hit 3.6Ghz on my e6600.

In fact... what are all your voltage settings? How can mine be more like yours!? lol

[chron]

new motherboard solved everything...

I'm using 1602 rite now with no problems.