Random bluescreen crashes

  • Thread starter Thread starter Greycap
  • 6 comments
  • 898 views

Greycap

The Flying Finn
Premium
Messages
6,112
Finland
Finland
Hardware: Intel Core2Duo E6750, MSI GeForce 8800GTS 320MB factory OC, MSI P35 Neo motherboard, 4GB DDR2 800MHz RAM.

The thread title says most of it. Random, and I mean really random, BSODs.

As a background my PSU failed a couple of weeks ago and I kept using it until a new one arrived. The problems caused by the failing unit were simple, a power cut with no warnings and an immediate restart. I replaced the PSU, everything worked fine. Then I installed Windows 7 Pro 64 bit.

At first everything worked but pretty soon I began receiving those BSODs with error codes "50" and "1e". Both of which usually point to a memory problem. I thought the damaged PSU might have fried a RAM stick and tested every stick alone, no errors. Then I installed them all (four of them) and got an error at 1361.xx MB. OK, perhaps the memory is OK but it's the motherboard that is suffering from being electrocuted. At this point I took it to the shop for further investigation.

They tested with one stick, two sticks in single channel mode, two sticks in dual channel mode, in every possible configuration and positioning. In the end it seems that the motherboard is OK after all but one of the sticks might be defective. Oddly enough the error always occurred at the same spot, 1361.xx MB, regardless of the positioning of the troublemaker stick in the slots but it was thought to be the test's way of telling about the error. That stick was removed, no errors were received and I got the machine back. But this wasn't the end of the story.

The first boot went perfectly, the machine ran as it should for about half an hour until I shut it down. Turning it on again a few hours later brought a disaster, two bluescreens before the desktop even loaded and another a couple of seconds after the icons had appeared. And a couple more after that, the intervals being from a few seconds to approximately fifteen minutes. I also got a new error "7f" among the ones previously mentioned - they haven't gone anywhere. "1e" is still ruling its own little world.

It might be a hardware failure even though the tests don't show anything. It might be a software problem but I don't feel like reinstalling Windows just to be sure. It might even be a driver problem but getting rid of them would again necessitate reinstalling Windows.

What the hell shall I do next?
 
Have you checked the capacitors on the mobo to see if any are blown?
 
No. But I tried reinstalling Windows to make sure it's not a software problem - the weight on the word "tried" because it crashed twice already during the installation. When I finally managed to get it up and running it crashed in bare seconds after loading the desktop and the next try lasted for a few minutes. So it's most definitely not a software problem.

Took it back to the shop, the motherboard will be replaced under warranty. Fortunately it had a three year warranty. Now I'm looking forward to see what the thing comes up with next...
 
If that doesn't work, the next thing to check is the hard drive. They can be pretty sensitive to "power fluctuations".
 
Are you able to note what the error code is? I'd check all the ram (use Memtest86+, possibly with 1 stick at a time), followed by swapping out the hard drive. I've had similar issues with a drive becoming crippled in a RAID array. It kept working for a while with sporadic problems, then complete failure.
 
They've been something like 75% of "50" and "1e", both memory related, and the rest has been a mixed selection of "124", "7f" and nothing. Memtest86+ gave me no errors with one stick at a time but a few at 1361.xx MB with them all installed as noted in the opening post. The guys at the shop managed to replicate the error with two sticks at a time and then bracketed the problem to one single stick. What makes me wonder though is the fact that they mounted it in slot #1 and still got an error at that very same 1361.xx MB spot, so is it a stick problem after all? Sounds more like there's a problem in that slot in that spot to me but they managed to get a clean Memtest pass with the stick removed and the remaining three installed so they concluded that it's the root of the problem. At this point it could be mentioned that I once got a clean pass with all four of them but on the second pass red was flashing all over the screen.

I don't think it's the hard drive until proven otherwise, I ran a couple of thorough chkdsk checks on it when the crashing had already begun and there was nothing to be found, not even file system problems. Admittedly chkdsk may not find everything but if the drive was well on its way to total failure I have a feeling it might just notice at least something.
 
Back