Computer Crashing - Possible Hardware Problem?

Quaponally

Honorable
Oct 9, 2012
9
0
10,510
Hi! Take a deep breath, this is going to be a long one. So about 9 months ago I built a gaming PC. Specs are at the bottom of this post. Then a couple of months ago I bought a second monitor and it was shortly after this that all my troubles began. The new monitor is a 22" samsung HD monitor and I also have a 19" SD monitor connected as well. I don't think this is the problem however, merely a coincidence. I thought I would let you know just in case.

The problem itself is that my computer spontaneously crashes. At first it was just while I was playing games but it didn't seem to crash at any one game in particular or even games that are particularly resource hungry. During these early days it even crashed while playing a flash game. The one thing it appeared to have in common was 3D graphics so I removed my graphics card and took it to a friend who has a very similar build to me and he tested it for a whole week without incident. During this time without the graphics card my computer continued to crash so that was one thing ticked off the list (my mobo has onboard graphics).

The next thing I tested was the RAM. I left memtest to run overnight first on both sticks of RAM together then on each stick of RAM individually and in various different DIMM slots. No runs produced any errors.

I then turned to software. In fact my initial bit of troubleshooting had me doing graphics driver sweeps and reinstalling before I looked at any hardware problems. I have experienced the crashing problem on two separate installs of Windows 7 with and without the service pack. I also tried reinstalling directx and the .NET framework as that was suggested somewhere. This however led to more frequent crashes which prompted a system restore that seemed to get it back to a less frequent level of crashing.

I have also updated the BIOS but it had no effect. I have carefully monitored the system temperatures during prolonged gameplay and the hottest I have seen anything is my GPU at 58C. I have also monitored the voltages for each rail from the PSU (using speedFan) and these remained within a 2% margin of their expected values. I realise testing these voltages with a voltmeter would be more exact but I don't have one!

This is where my knowledge of trouble shooting the problem ends so I'll elaborate a bit more on the problem itself. As I said before it started happening while playing games. The PC would just freeze up, the picture would freeze and it wouldn't respond to any commands. Sometimes it would continue making noise like the last fraction of a second of sound was playing over and over. Recently however it has started giving me the BSOD and happening more frequently when I'm not playing games. It seems the gaming part could have been a coincidence as I play games an awful lot and the graphics really have nothing to do with it after all. One time the BSOD gave me a code relating to directx which is what prompted me to try reinstalling it. All the other times however it gives me different error codes and I'm not sure what they relate to. Just this evening it gave me a new blue screen I haven't seen before, it just said that there is a hardware fault and that I should contact my hardware supplier. One final thing is that after it crashes and I turn off the machine by holding down the power button the backlight on my keyboard remains on. It will stay on indefinitely unless I flick the switch on the PSU. This has led me to believe that either the PSU or the way the mobo sends power to things is the problem.

All crashes have been experienced in windows 7. I have not yet experienced one while booted into my Ubuntu partition. I don't use Ubuntu as much and it could just be a coincidence. This machine is primarily for games, I have other linux machines that I use for work.

Any ideas or possible trouble shooting methods would be greatly appreciated as I'm all out of any myself.

The specs:
PSU: Thermaltake 700W TR2
MoBo: Asus M5A 88-M evo
CPU: AMD phenom II x4
GPU: Saphire 2GB Radeon HD6950
RAM: 2x4GB Corsair Vengence 1600MHz
optical drive: basic Asus DVD-RW drive
HDD: 1TB seagate barracuda
OS: Windows 7 64-bit ultimate and Ubuntu 12.04 64-bit
 

clutchc

Titan
Ambassador
Have you tried eliminating the Ubuntu and fresh installing Win7 on a single partition? At least for testing purposes.
The light on the keyboard may not be an issue. I've had this happen on keyboards too. And no crashes. Your KB isn't plugged into an always-on USB port is it... one designed for phone charging?
What are the resolutions of the 2 monitors? Why do you refer to one of them as Std Def?

I realize I haven't come up with any solutions, I'm still trying to get my head around it. But the hdwr fault my be a clue.
 

Quaponally

Honorable
Oct 9, 2012
9
0
10,510
Yeah the last time I reinstalled Windows I completely wiped everything and left it just with windows for a while but still had the crashing.

I don't know if it's an always on USB port. There's nothing about it in the mobo manual but it only does it after a crash and not if I shut down properly.

The 22" is 1080p and the other one is 1280x1024. I said it was SD because it's an old square monitor (5:4).
 

clutchc

Titan
Ambassador

Is the older monitor a CRT with an analog input? And are both connected to your HD 6950? I would disconnect the older monitor and uninstall the gfx driver completely. Run Driver Sweeper from safe mode and have it remove any remnants of ATI, AMD, and Nvidia graphics drivers it finds. Re-boot and re-install the latest version of the gfx driver for your card and O/S.
http://www.techspot.com/downloads/4266-driver-sweeper.html

If you have already tried the above test, we'll move on to something else. But it appears you've covered all the bases with your own troubleshooting. That would pretty much leave just the PSU and the Motherboard.
 

Quaponally

Honorable
Oct 9, 2012
9
0
10,510
The older monitor is still an LCD. It's not that old :kaola: It has a VGA input. I have done a driver sweep from safe mode before but I still kept the two monitors connected. I'll try it again now with just the new monitor and see if I can get it to crash afterwards. I'll keep you posted!
 

Quaponally

Honorable
Oct 9, 2012
9
0
10,510
I did a driver sweep and reinstalled everything with just the one monitor but it just crashed again so no change at all.
 

Quaponally

Honorable
Oct 9, 2012
9
0
10,510
Yes I did the fresh install with just the new monitor connected. And yes the BIOS has to be altered from default because my BIOS wants to use IDE but windows was installed with AHCI so it won't boot unless I change that. That's just a recent addition though since I updated the BIOS.

There was a new development today, I got it to crash while using the Ubuntu partition. It was the same kind of crash and happened while watching a youtube video.
 

Quaponally

Honorable
Oct 9, 2012
9
0
10,510
Also sorry for taking so long to reply. I was waiting for an email notification for the thread and it never arrived!

(it won't let me edit posts either!)
 

clutchc

Titan
Ambassador
If all you did was change BIOS from IDE mode to AHCI mode, that shouldn't be an issue. You have covered all bases when it come to troubleshooting. The memory checks out, the card proved OK on a different machine, PSU seems OK*, clean install of Win7, updated BIOS, driver sweeper, CCleaner... It is beginning to look more and more like a motherboard issue. Especially after the 'hrdwr fault' you said you got.

*The PSU could still be a culprit, but without a 2nd one to test with it is hard to say. Rather than rely on software for the PSU's voltage readings, it would be better to use a voltmeter or a PSU tester. Software can sometimes give erroneous results.
Example: http://www.amazon.co.uk/s/ref=nb_sb_noss?url=search-alias%3Daps&field-keywords=PSU+TESTER
I use the Antec model to test every new PSU I use in a build. It's faster than digging out the voltmemer when I'm not expecting there's trouble.
 

Quaponally

Honorable
Oct 9, 2012
9
0
10,510
Ok thanks a lot! I should be able to "borrow" a volt meter from the lab I work in and I'll try and see if I can find anything out about the PSU. I'd just be looking for strange drops in voltages on the different rails right? If that all checks out I guess I'll get on to Asus and see if I can get a replacement mobo!
 

Quaponally

Honorable
Oct 9, 2012
9
0
10,510
Ok I just finished testing the PSU and everything looks perfect! That link was real handy, thank you.

On a whim on saturday I plugged the PC into an extension lead instead of the wall and it hasn't crashed since. It obviously needs more time before I can say this fixed it but I'm pretty hopeful. If it does turn out to still be crashing then I guess it's the mother board and that's problem solved. Glad it's still under warranty!
 

clutchc

Titan
Ambassador
Good luck. If it runs without crashing, it will surprise me. I can't think of a scenerio where a different outlet would make a difference. But I've got my fingers crossed for you. Btw, be sure you have a decent size extension cable that isn't too long.