Windows randomly shutting down every 1-2 hours. BSOD Dump Report within.

Looks like a different issue if you ask me, plus that is from 2010 and the nVidia drivers have changed a lot from back then.

I'd say it is highly likely to be a hardware issue, the likely culprits being the GPU or the Power Supply. As mentioned, install MSI afterburner and either reduce the factory overclocks down to standard GTX560 Ti 448 Edition speeds or up the GPU vcore slightly, make sure MSI Afterburner is set to change settings at boot and see if it cures the issue.
 
QUick update before it crashes again, installed windows to HDD, same problem. I can eliminate the hard drives.

Installed MSI Afterburner, voltage slider is locked, so I've slid the Core Clock and Memory Clock down as far as they will go and applied those settings (now at 405Mhz and 950Mhz respectively). Apparently that's all I can change for that. If it crashes again I think it's safe to assume it's not the GPU? (Although it shouldn't be anyway, because FurMark doesn't cause it to lock up or crash, and given that it's been stable for 9 months of BF3, Minecraft and more I figure it'd be something other than this).

Anyway I can increase the core voltage in afterburner? Doesn't let me at the moment.
 
Guess what, it conked out again, even with those super-low clock settings.

I've moved to my laptop for the minute, it's now shutting down within about 5 minutes of booting up to desktop, if it boots at all. I've just taken out two of the RAM sticks to see if I can eliminate any of those. The other two are insanely hard to get to, means I have to take the cooler off, so I'll see about those two if it crashes again.

BTW, according to HWMonitor, my GPU's core voltage is 0.91V. Trying to unlock the MSI adjustment at the moment but not having much luck
 
Last edited:
Okay, maybe you guys are right, although I can't understand why.

I downloaded the MSI program, set all values I could alter to minimum. and ran a Furmark Benchmark test for 15 minutes. No crashes, slow but no crashes. Reset my GPU settings (to their defaults), ran the test, it was fine for a few seconds then bam, everything goes dead again, all peripherals dead apart from the PC itself.

So what gives? It's been stable for 9+ months now, and all of a sudden it does this? How is that even possible? Underclocked it by about 100Mhz as well, and it still occurs. Something just isn't right here. Time to RMA the GPU maybe? I'm going to wander up to my computer shop tomorrow and get them to test the PSU if they can, to try an eliminate that.

EDIT: FFS. Just booted it and before I even got to the desktop it shut off again. This is taking the biscuit now.
 
Last edited:
Well, if the GPU was out of warranty I'd edit the BIOS to force it to run slower (had to do that with a 8800GT with the infamous Nvidia defect). But if the card is still under warranty that seems drastic.

Pity you don't have either a spare card/onboard or a spare PSU to try it out.

Sounds more likely to be the graphic card though. What you can do if you can get it to boot again, is to try and underclock it using Afterburner and then try a CPU stress tester like IntelBurn. Then if it crashes its also likely to be the PSU.

At the end of the day, the only way to test hardware reliably is to swap stuff around and eliminate items.
 
TheJamsh said:
Okay, maybe you guys are right, although I can't understand why.

I downloaded the MSI program, set all values I could alter to minimum. and ran a Furmark Benchmark test for 15 minutes. No crashes, slow but no crashes. Reset my GPU settings (to their defaults), ran the test, it was fine for a few seconds then bam, everything goes dead again, all peripherals dead apart from the PC itself.

So what gives? It's been stable for 9+ months now, and all of a sudden it does this? How is that even possible? Underclocked it by about 100Mhz as well, and it still occurs. Something just isn't right here. Time to RMA the GPU maybe? I'm going to wander up to my computer shop tomorrow and get them to test the PSU if they can, to try an eliminate that.

EDIT: FFS. Just booted it and before I even got to the desktop it shut off again. This is taking the biscuit now.
Click to expand...

Of course it is possible, hardware can fail under stress just like anything, just a fact of life when it comes to computers.

If you have a credit card or some spare cash, my suggestion would be to buy a new PSU and see if that solves the issues outright. If it does RMA the old unit and sell the replacement when it arrives, that way you don't have much downtime.


KompuKare said:
Well, if the GPU was out of warranty I'd edit the BIOS to force it to run slower (had to do that with a 8800GT with the infamous Nvidia defect). But if the card is still under warranty that seems drastic.

Pity you don't have either a spare card/onboard or a spare PSU to try it out.

Sounds more likely to be the graphic card though. What you can do if you can get it to boot again, is to try and underclock it using Afterburner and then try a CPU stress tester like IntelBurn. Then if it crashes its also likely to be the PSU.

At the end of the day, the only way to test hardware reliably is to swap stuff around and eliminate items.
Click to expand...

It won't be out of warranty, GTX560Ti's 448 Core Versions haven't even been out a year.
 
Well I'll try the Intel Stress test idea, and see if it fails (I'm expecting that it will), and I'll also give OCUK a ring tomorrow and see what they come back with. I might also go up to my local shop tomorrow and see if they can come up with any ideas.

There is a BIOS update for my Motherboard available, I might try flashing it just for good measure to be sure I've really tried everything. At least I've eliminated the CD-Drive, the RAM, and the HDD & SDD now. Just the four really costly parts :(

Everything is less than a year old, I think February was when I bought it all. I had to RMA the Mobo and CPU right away, and that was a nightmare. I'll try these last few resorts and see where I get. Might even try plugging it in somewhere else in the house to see if I'm overloading my sockets or something.

EDIT: Running the Intel Burn tests, so far so good. It's completed 10 passes at both standard and high without any complaints. Going for the highest available now, will let you know how I get on. Leans more and more towards the GPU, especially given that my CPU voltage has gone up by about .15 V.

EDIT 2: God damn... I really thought it was stable this time. Crashes out again on the third IntelBurn test on the highest setting.
 
Last edited:
Well, the GPU has gone back to OCUK under RMA, possible warranty replacement.

I managed to get the crash to replicate itself 99% of the time using a GPU stress test, while the IntelBurn and various other tests reported no faults. Apparently I won't be charged if there's nothing wrong with it, so can't complain.

Ace customer support from OCUK as always ;)
 
After nearly a month of being away, I'm being issued a replacement. About bloody time Zotac...

EDIT: ARGH. Brand new graphics card, same damn problem! It has to be the PSU then right!?
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom