Random shut downs

[wronggear]

I would like to start off by saying that I'm very new to the PC building scene and will probably make many newb statements and ask many newb questions, so please be patient while I take this all in. Now that that's out of the way, I've recently built what i hope will be my first dedicated gaming rig:

MOBO: ASUS M5A88-V EVO
CPU: AMD FX-8150
GPU: Radeon HD 6950
Memory: Kingston HyperX Blu 8GB (2 x 4GB) DDR3
OS: Windows 7 Home 64-bit
Power Supply: Corsair TX650 V2
HDD: Seagate Barracuda 2TB 7200 RPM

And because the room that it's going to be running in doesn't get the best air flow I'm going to be adding a Corsair H80 Liquid Cooler just because I'm paranoid about over heating (The part is in the mail as I type this).

I got everything hooked up and everything ran perfectly for the first day but now anytime I run a game it's prone to random shut downs. Not crashes to the desktop or reboots but full on shut downs and I can't seem to turn the system back on for several minutes afterwards. I'd read somewhere online that that could mean that the CPU is overheating so i downloaded both CPUID HWM and Core Temp and they both seem to agree that I idle around 16 degrees celsius and while playing Sniper Elite 2 it hovers around 51 degrees. I would find it odd that it's shutting down due to over heating though because according to AMD, 61 is the beginning of the "danger zone" for the 8150 so I'm not sure that that's whats causing the problem. Part of me wonders if I picked the wrong power supply. That maybe it's just not giving enough power to keep everything running while it's under load but I'm not quite experienced enough to know for sure. It's only when i play games though. I've been surfing reddit and just generally Fing around on the net for a couple hours now and it's been completely stable, so I'm at a loss. Any thoughts on what might be to blame? I'm kind of starting to feel like I just wasted a little over a grand so any help would be greatly appreciated.

 

[AdrianPerry]

Your PSU certainly shouldn't be the issue, unless its faulty. 650w is more than enough Power.

Overheating does sound like the culprit, either that or an unstable over-clock - have you tried to over-clock anything at all? If so - id recommend a CMOS reset.

Your temps don't sound all that accurate though, unless FX operates at considerably lower than SandyBridge? (not too hot on my AMD-CPU knowledge), but at idle SnadyBridge CPU's tend to be around the 30 degree's mark, 16 seems cold/inaccurate.

Have you checked the Event Viewer to see if there are any shutdown errors/warnings? (Event Viewer > System Logs)

EDIT: Have you made sure you have the motherboard mounted on stand-off screws in the case? Could possibly be motherboard shorting out if not - standard "newb" mistake hehe.

 

[wronggear]

I thought the temps seemed a little inaccurate as well as I've read that the 8150 runs hot and is prone to random heat spikes.

As far as the stand-off screws go, I guess I lucked out. I had no idea they were required before you mentioned them, however my case came with them pre-configured for my ATX MOBO so I just popped it in place and affixed the screws. Perfect fit.

I'm looking at my event log but to be honest, I have no idea WTF I'm looking at lol. It shows that I do, in fact, have 55 events: 43 Warning, 11 Errors and 1 Critical. Though they all just seem to be either a program delaying windows shutting off or a program starting up too slowly and making the start up process go a little slower.

I haven't tried overclocking at all yet and I don't plan to until I gain more knowledge in all this. the last thing I want to do is burn out my CPU or something lol. That would suck. Besides, for right now it does everything I want it to do... When it works, that is.

 

[FinneousPJ]

Get HWMonitor and run some tests with Prime95 and Unigine for graphics - most likely something is overheating and you need to find out what.

 

[AdrianPerry]

Yeah Prime95 and Unigine would be a good place to start troubleshooting. Run them one at a time, for about 30min-1hour and see if it throws up any errors.

Id start with Prime95, and do a mixed test that does both CPU and RAM.

That sounds like a lot of errors and warnings consider your PC is a new build!

 

[bobigwheels]

I would update your bios. I had basically the same problem as you do and tried that, now I haven't had a random crash since then.

 

[jsc]

... so i downloaded both CPUID HWM and Core Temp and they both seem to agree that I idle around 16 degrees celsius ...

Right there, that tells me that you are having problems with your temperature monitoring unless you are running chilled water cooling. To idle at 16 C on air or conventional water, your ambient air temp would need to be no higher than about 40 F.