Computer White Screen Crashing

My computer has been exhibiting an odd issue that has recently become a problem in more than one game. My system will load and start playing games fine for while and at some point the screen will just got white. Sound will persist for a few seconds and either stop or stall on a small segment of a sound. Sometimes it crashes to a black screen, but maybe 1 in 10 times only. It usually take 30 seconds to a minute and then will write to the hard drive and then reboot.

At first it only happened in Driver: San Fransico some time late last year when I picked that game up. Then during the summer sale I was gifted a copy of Grid, and that game started, but would quickly crash. Got through one race, crashed and restarted and crashed while in the video settings menu.

Last week, the problem showed up in Firefall. Whenever I got into a high intensity situation, some after more than 15 minutes of other stuff, but sometimes after just a couple minutes, the system would crash. Always during the same event type and I'm almost certain it was when things started getting hectic on screen.

I've got the following:
i5 750 (Lynnfield)
Coolermaster Hyper 212+ cooler
Gigabyte P55M UD2
XFX Radeon 4870 1GB
8GB RAM (4x2GB)
Intel NIC
700 W Single Rail Rosewill PSU

No overclocking.

Throughout, I've been running ATI Tray Tools (to manage fan speed) and GPU temps via the OSD have remained below 90, and typically between 80 & 85.

CPU Temps are good, maybe getting into the high 40's or low 50's.

Some more history: Last year my computer was impacted by a storm. The damaged seemed to be limited to networking stuff. It took out the ethernet ports on the cable modem, router, and computers that were connected. The unconnected ethernet ports and wireless on the router still seemed to work. But my onboard ethernet was fried on this computer, so I replaced it with a dedicated card and disabled it in Windows and BIOS. I haven't seen too much else odd aside from the death of a 6 year old HDD in January of this year. Otherwise, the system has been stable.

The only thing I can find via google that worked is one person with an Nvidia card at one point had to roll back his drivers. I don't think I found much of a solution otherwise, just a LOT of speculation and no reports back of success.

I do not have extra components to test power, mobo or video card with. The only other desktop in the house uses an AGP vid card and was built in 2004.

Any help is appreciated. I guess I could try rolling back drivers to before they put the 4000 series on legacy support.

A few things:

1) GENERAL SYSTEM: first rule is to always have the latest drivers and make sure everything is seated correctly on the mobo. In addition, making sure anything with a fan is not obstructed / use air can to blow over the gpu/cpu and case fans that can intermittently get stuck and overheat.

2) GPU: does this happen ONLY in games or when any app is running? Or you could try running something other than a game that specifically taxes the GPU. e.g. download GIMP (free) and have it process several filters on a large RAW file or try a encoding video.

3) CPU: run a prime 95.

4) Can you clarify what you mean by, " Last year my computer was impacted by a storm. The damaged seemed to be limited to networking stuff." Did you suffer through a power surge? If that's the case, I'm sorry to say that it is possible that everything from your PSU and beyond could have been affected. Was your computer at the time connected to a surge protected/UPS? You could have suffered some damage that over time, degraded your system, capacitors, etc to what you are seeing now.

5) SOFTWARE: Install anything new? If so, try doing a system restore to a point where you believe you didn't have any problems or uninstalling the suspect program.

6) How old is your hard drive that your OS is on? It's very unlikely this is your problem, but if it's over 3 years old, it could be failing.'

Hope it helps
/ESA

ESA wrote:

A few things:

1.) I'll look at drivers, but my card is stuck in Legacy support mode now, being so old, so updates aren't too often, and it has been updated twice since I first saw the problem. I'll also try cleaning again. All heat monitoring programs are reporting temps as stated, and the vid card fan can very audibly be heard spinning up as the temps rise and both temp and fan speed are reported via the overlay in games via ATI Tray Tools.

2.) So far, only in games. Isn't GIMP the same as photoshop? Typically 2D programs don't really do anything much with vid cards, from what I've read of it.

3.) I think I tried that a while back, but can test again to get a better read on CPU temp, since that doesn't have an overlay display.

4.) What I meant is, there was probably an electrical surge while the computer was in sleep mode while I was not home. Again, the only things immediately apparent as damaged were the ethernet ports on the cable modem, router, the computer in question and another computer (the one I also mention in the OP). It was connected to a surge protector, but not the coax cable going into the cable modem along the path of the ethernet connections.

5.) Since the first time I saw the issue I have replaced the main system hard drive (see below/OP about HDD) and so, also done a fresh install of Windows 7 64-bit.

6.) The HDD mentioned int he OP was actually the OS drive, and it was actually December of 2012.

Thanks for the help so far.

I have the most up to date drivers for my card (Catalyst version 13.1).

Opened the case and cleared a minimal amount of dust (computer sits in sleep mode except for 2-3 hours a day) that was in there before the compressed air ran out. I did clean the GPU around the beginning of the year, or something like that. Then I put it back together after checking the vid card seating and just doing a wiggle / visual inspection on the 2 mobo power connectors. I should probably recheck/reseat the vid card power.

Intel Burn Test for 10 normal and 5 maximum (35 mins) runs was fine. Maximum CPU temps was 56 on one core, the others a couple degrees cooler, and that was with case fans at 1/2 speed. At full speed they dropped a couple degrees I can try prime 95, but just for temps, I don't know if it is going to give me anything new.

Looking at my order history, I ordered a new nic after the storm that hit the ethernet ports in June 2012. I am still willing to accept that it may be a PSU issue (and have been thinking that is the culprit all along), but want to check other (read: free) stuff first since I don't have spare money now.

EDIT: Just made it through a furmark burn in at 1920x1080. GPU did top out at 88.

EDIT 2: Was in Firefall for almost an hour before it went white, then black then BSOD and reboot (I guess it used to do that, but I had taken to just hitting reset). Critical Error says source: Kernel-Power

Log Name: System
Source: Microsoft-Windows-Kernel-Power
Date: 9/10/2013 1:03:54 AM
Event ID: 41
Task Category: (63)
Level: Critical
Keywords: (2)
User: SYSTEM
Computer: tomayto-PC
Description:
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
Event Xml:

41
2
1
63
0
0x8000000000000002

41489

System
tomayto-PC

278
0xfffffa800aa854e0
0xfffff88003a95efc
0x0
0x2
false
0

Ok, I can't get it to display all the text.... guess the markups are trying to take over. MAybe quote to see it. Does that help at all?

Well, I got a new vid card for Christmas from the wife. It seems that solved my problem.