Sign in with
Sign up | Sign in
Your question

Critical Failure beginning graphically intensive applications

Last response: in Components
Share
January 21, 2013 10:22:55 AM

&&&&&&****** (Hopefully) Final word: It was the power supply, try swapping your power supply with a healthy one (possibly a beefier one, unless you're sure... and you've actually checked... that you're over the power you need) if you experience problems like I did.


Good morning, I wanted to rule out my graphics card as my current point of failure and was hoping you guys could help me solidify my theory.

----What I have:
Currently I'm running an intel 3570k, and an ASUS HD 7850 with a corsair HX750 PSU. I have a 180 GB Intel 330 SSD as my main drive with a caviar blue 500 gb secondary. I built this system about seven months ago and it performed gloriously in so many aspects (including very stressful games.)

---My problems, symptoms
The PSU should be overkill for this system but over the last 7 days I've had 10 crashes. Each time it's when I started to do something that stressed the system..

Example of the problem: Start World of warcraft, change settings to absolute lowest, cap fps at 25, run around for a minute... stop, check temps, 30 degrees for CPU and GPU, then go back. run around for a few more seconds, change the settings to just medium, hit apply and INSTANTLY black screen. not even a BSOD (I disabled the immediate restart option to try to see the blue screen)

---What I've done
Blew out the case with compressed air, cleaned the filter on the front. It needed this, without a doubt. Resolved nothing, hopefully wasn't the reason for my persistent problem.

Read a lot of forums, tried a lot of things. I just replaced my pci-e cable on the modular PSU in hopes that I'd either discover a loose connection or replace a faulty cable. no go, WoW Crashed before I even made it past the log in screen.

I downloaded Heaven DX11 benchmark, computer crashed about 3 seconds after it started playing.

I checked my temperatures (with Speedfan), my CPU is idling around 30C and my GPU at 27C.

I removed my RAM and reseated it, no fix, and then I did the built in Windows memory test as well, but it returned with no error. (I kept trying to run memtest86 but am too dumb to burn the iso to a cd somehow, every time I try to boot from the cd it just starts Windows)

In eventviewer it says there have been ~9 41 kernel error in the last week, with 7 today, but it doesn't have enough time to record anything. I turned off the "immediate restart" thing in admin tools, in hopes of seeing a BSOD message, but no change. Instant black, then instant restart.

I deleted a lot of excess files, defragged my SSD, uninstalled excess programs, updated windows and the drivers for my graphics card.

Based on my limited researching capabilities, it's either the PSU or the GPU, and based on the suddenness of all these problems I'd imagine it's a power problem.

What should I do? Corsair's customer service solution is just send it back but I don't want to pay shipping unless I can be sure it's the problem
a b ) Power supply
January 21, 2013 1:23:14 PM

1. Never ever defrag a SSD, totally screws with it.
2. Latest drivers for motherboard and GPU.
3. Could be PSU, Corsair have had issue with their PSU's, not sur eof this model.
m
0
l
January 21, 2013 5:52:43 PM

1. just looked that up, thanks for the tip
2. did GPU before, motherboard is up to date
3. hoping for a little more confidence before I pay shipping

Any one else have any advice? I just tried playing WoW again hoping maybe one element was not compatible, but multiple graphics settings (setting one at a time failure) caused a few instant crashes
m
0
l
Related resources
a b ) Power supply
January 21, 2013 6:03:22 PM

Yeah, it sounds like either a GPU error or a driver error.

Are the issues JUST with WoW, or with all other games?
m
0
l
January 21, 2013 7:32:01 PM

It was happening before I installed WoW, but since then I have uninstalled every other game. When I play the Heaven DX11 benchmark program it crashes very quickly.

I just ran memtest86 and got through one pass, left my computer, and came back to an messed up screen. going to do some scouting real quick
m
0
l
January 22, 2013 2:53:42 AM

Well I guess it's problem solved, for now. It was the power supply.

I went to best buy and picked up a few things. I wanted another 8gb of ram anyway so i tried switching the new ram with the old, but it still crashed. Guess I get to have 16 gb now

Next I tried the power supply. Got an 800w corsair, (still overkill, for sure, but I didn't want to risk a 500w and this was the next step up in their store). I replaced the modular PSU with a non-modular, so messy =(, and it's working fine now. I had previously replaced just the pcie cable going to the GPU on the modular PSU, but to no avail. It seems weird to me that the PSU would fail in just one element (it could run forever as long as I kept the load light)

I picked up an nvidia 650 in case the PSU didn't fix it (was really counting on the PSU lol) but i'll be able to return that easily.

Going to RMA my modular PSU with Corsair (and check to see if they'll pay for shipping) and if I can get it soon enough I should be able to return this power supply to best buy as well. Glad Corsair has a super easy RMA process.

I learned a lot about my computer through this process, so I'm not totally upset that the PSU failed, sort of inbetween.
m
0
l
!