Man, that sucks, sounds similar to my old laptop, it restarted randomly into a BSOD error that looped. The only way i fixed it was to reinstall XP. So you routed out that it is not a HD error. I do not think it is a problem with the power supply, if it was it should just restart/shutdown your computer, not cause problems. Try booting into safe mode and see if that does anything. Do you remember if anything you did lately might have related to it? Install new software? Might want to give system restore a try. Also for Internet/Network problems winsockxpfix works wonders, it has fixed all my disconnection problems so far. http://www.snapfiles.com/get/winsockxpfix.html

 

Might be a heating problem. I'm assuming you have enough fans/circulation going through your system? Computers tend to get frustrated with dust as well. Lots of people (including myself) don't clean their computers very often.

Do you have any theories of why its doing this?

 

Make sure you're CPU is seated correctly, fans plugged in, CPU 4 pin power plugged in correctly etc. Try and rule out as much hardware as you can.

 

I've suffered identical problems when overclocking too far and thus overheating the CPU. I know you've got three fans in there, but those symptoms perfectly match an overheated CPU. Monitor your CPU temps and see what it climbs up to before it crashes.

It's possible that something (eg. CPU) is already part-fried and is just really temperature sensitive now, or that the thermal paste between the CPU and HSF has dried out and turned to powder.

nb. If you remove the CPU to check the paste, then you'll have to replace the thermal paste anyway. Once you peek at it, it's destroyed. If your CPU temps aren't sky rocketing then the paste is doing it's job and I'd leave it alone.

Check with Middy, but if CPU temp is the problem then underclocking might be the way to go as a temporary fix.

A second possible reason for having to wait a minute or two before restarting is that there's something stored in a temporary memory (not just RAM, but any chip on the mobo) that takes time to disipate once the machine is turned off.

If it's not CPU temperature related:

Have you tried swapping out memory modules? Just run on half-RAM, if you don't crash then you know which one is the problem.

Remove the video card immediately after the next crash, see if the machine will now restart. It'll probably beep at you for not having a video card, but just ignore that.

Make sure all of the cables (power, data, etc) inside your case are firmly seated. I've tripped over this problem once or twice myself. -_-

Your PSU might be fubar. A 650Watt Corsair is definitely good enough for your setup, but it doesn't mean it's not fucked anyway. A monitoring program will be able to tell you what voltages you're getting on your rails. If there's any unusual sags then you may be up for a new PSU.

How to Test for CPU Temperature and Power Issues

Get some motherboard monitoring software. I run an Asus Mobo, so I use Asus PC Probe II. It tells me that my CPU's Vcore is a steady 1.16V, my +3.3V rail is at 3.31V, +5V @ 5.02, and +12V @ 11.93. My CPU temp is at 36C, Mobo Temp @ 41C and fan speeds are all 1.5k - 2.3k RPM.

Asus PC Probe won't keep an auto-saved log unfortunately, so I have to keep my eyeballs on the screen.

Now, I'd stress test my PC with Prime95. If you have one core, then run it just once. If you have four cores, then you could run up to four different instances of it, just force each one to run on it's own core. If you have one running on every core then it'll heat up faster and test the whole CPU instead of just one quarter of it.

Watch your temps especially, but also your rails. If the temp climbs and plateaus without a crash, then CPU overheating isn't your problem. If it crashes without any sagging in the power rails then the CPU or mobo is the likely culprit. If the power sags noticeably (ask Middy how much is too much) just before the crash then the PSU is suspect.

 

How long had the system been running in steady state before the problem started? If the answer is months, my gut feeling is that it is the start of a motherboard problem. How old is the board? Try upgrading the firmware of your motherboard as a place to start (especially if you bought the board early after it was released).

If it was the hard drive, the new one would have fixed it, but you also have random failures of other peripherals (NICs). That's why I think motherboard. Are you currently overclocking your CPUs? If so, try backing off to the rated speed and see if that helps.

Like Yak said in his post try removing some components and see how this affects your situation.

Have you had any power outages at your house lately? Is the system protected with a surge protector? Silly question, but it has to be asked.

There's no reason a 650 watt power supply shouldn't be able to handle your system from the sounds of it. If you really need to rule it out drop $20 on a power supply tester at Tiger Direct.

http://www.tigerdirect.com/applications/searchtools/item-details.asp?EdpNo=1647108


To recover your files off the old hard drive, get a USB enclosure (sata or eide depending on your type) and slap it in there.

What and when was the last thing that you upgraded on the system? Did you happen to jot down the Stop message on the BSOD? It's the first paragraph right before the hex dump. A clue why it is happening could be found in there.

 

I Don't know if you've opened the case or not... But it could be the capacitors starting to burst. I can't tell you how many frustrating hours i spent trying to figure out what was wrong with my parents computer until i noticed the acid >.<

 

Hari Seldon brings up a good point. Bad capacitors on your motherboard or in your PSU could account for why you have to wait several minutes to reboot. The capacitors may be taking a little while to discharge.

Your caps may not be leaking, but if the top of them is in any way bulged out, then they've gone bad. They should be flat or slightly concaved on top. Go here for more info and pics: http://www.badcaps.net/pages.php?vid=5

According to that page, random reboots, VCore problems and high CPU temps can all be caused by bad caps. They strongly recommend that you test with a known good PSU though.

Inside your PSU - includes pics of caps in good condition.
What to Expect from Cheap PSU Testers

 

Does anyone think it might be the mobo battery getting weak?

 

No, that's completely different.

 

Ah, that's right. My bad...

Cool deal

 

Ya, most heat the CPU should be at is 38-40 something Celsius. 60's and 70's is very high. You're welcome.

 

LOLFAN

Hardware fault diagnosed and fixed within 24hrs. W00t!

blanked text doesn't work anymore -_-

 

I was actually wondering about this the other day. My system's CPU runs at 32C (at least that is what NVIDIA MonitorView says), I wasn't sure if this was normal or not. Guess I got my answer

 

Give me some credit nigger. I was the one who fixed the problem. Knew straight away when you told me your temps were near 70C. >:|

 

I know very little with regards to these issues, but I have experienced the blue screen thing similar to yours. Random shut downs or the computer freezing.

I had to have a new motherboard. That might be the problem.