Is my new GPU bad?

[seanr7]

I got a new ASUS DirectCU II GTX 770 a few days ago. It installed fine and I really like the card, but, occasionally, my entire system crashes when I leave a game. Sometimes its when I've only been playing for like 20 minutes, sometimes when I've been playing for hours. In the game I've been playing, the card rarely reaches above 70% load and the temperature never gets beyond 70C. The card is running at stock clocks and voltages.

Is my card bad? I really love this card, but if I've got a defective one it's better if I get it taken care of before it's too late.

 

[pit_1209]

Your cpu temps are also good?

 

[seanr7]

Yeah. I have a stock clocked CPU and I am fanatical about cleaning the heatsinks, keeping dust out, and verifying fan functionality.

 

[lucasz]

What PSU do you have?

 

[seanr7]

Corsair GS600. I'm planning on upgrading it soon because I really want one that's modular and to give me more headroom on the wattage. But as far as all the research I did, 600w was enough for this GPU and the rest of the hardware I have without much trouble. I could be wrong as well, but wouldn't a PSU issue cause issues at times other than exactly the second I exit the game every time it happens?

 

[lucasz]

If the PSU is old it could be the problem, but I doubt it.

 

[seanr7]

I've only had it for 2 years, is that old in PSU time? I've never really owned a computer I've needed to replace one in, but I do know that they fail eventually.

 

[pit_1209]

It´s always the same game or it happens with every game?

 

[lucasz]

If it is only 2 years old it should be fine

 

[pit_1209]

I've only had it for 2 years, is that old in PSU time? I've never really owned a computer I've needed to replace one in, but I do know that they fail eventually.

It´s not old in fact is under the warranty period.

 

[seanr7]

It's happened after playing 3 different games.

 

[BlankInsanity]

By any chance are you using this on an old motherboard? the VBIOS of the GPU may not be fully supported by your motherboard's BIOS, however I highly doubt this is the problem and if it is, it should be last resort to think on.

I think it could be a problem with your PSU cutting out when the GPU goes under load. What games are you playing? 70% load seems like your play medium GPU strain games rather than the high GPU intensive games like battlefield, AC4, Barman Origins, AAA games of the sort.

also tell us your CPU's temperature. CPUs unlike GPUs can't handle the high temperatures that reference GPUs experience which is in the mid 80s low 90s. CPUs cut out at around the med 70s. So make sure you CPU isn't going above 70C, also is it overclocked? overclocking on a stock cooler isn't reccomended, try reverting to default clocks if it is overclocked

also like lucasz said, what PSU are you using?

 

[BlankInsanity]

The GS600 powers the GTX 770 just fine. 600W check. 42A requires on the +12V Rail, PSU supplies 49A so that's checked aswell. Like I said before try using a stress test program and try again

 

[seanr7]

My motherboard is about 2 years old, CPU is at stock clocks already. I haven't verified the CPU temps, but it's stock clocks and a functioning stock cooler, so I doubt it'll be high, but I definitely will check.

And the games I've tested it with are AC4 (happens the most because I'm playing that the most lately), Skyrim with the entire Steam Workshop wolftyler1997 Graphics collection installed, and Hitman: Absolution.

 

[seanr7]

And I'm gonna see how a little while of Prime-95 effects the CPU temps as far as stress testing goes. Is there any good GPU stress testing software? Like 3D Mark 11 on loop or something?

 

[pit_1209]

Try unigine heaven benchmark for the gpu with xtreme settings, very demanding may I say.

 

[seanr7]

P-95 seems stable at 95C under 100% load with small FFTs. It rises there really quickly then stays there. And I'll look into the Unigine Heaven bench.

 

[seanr7]

And the benchmark ran and never got beyond 70C and CPU sat around 50% load and 58C.

 

[pit_1209]

What´s your cpu again? and you´re telling us that in prime 95 it goes to 95c temp??

 

[seanr7]

Its an i5-2500. And prime gets it to 95C and it idles at around 40C. From the thermal document I read from Intel, that seems within spec.

 

[seanr7]

And just for some more info, the system is not going to BSOD. I have memory dumps enabled and there are none for me to look at. Also, Windows is not detecting a driver crash either. But when the issue does happen, both my monitors go to a random solid color and stay there, no matter what I do. The only thing that fixes it is a hard reboot.

 

[pit_1209]

Its an i5-2500. And prime gets it to 95C and it idles at around 40C. From the thermal document I read from Intel, that seems within spec.

95c it´s by no means within spec in fact is extremely high for that processor at stock clocks, that seems to be your problem, you should reapply thermal paste or get an aftermarket cooler.

72.6c it´s the max temp on the IHS, that´s what software see as cpu temps.
http://ark.intel.com/products/52209/Intel-Core-i5-2500-Processor-6M-Cache-up-to-3_70-GHz

 

[seanr7]

Alright. Thanks for the help. I'll get that taken care of.