GPU or CPU crashing?

[BigChickenJim]

So since the fans, i have had no issues. does that still point to a bad card if 70c was causing lock up?

If the card is failing at 72C it's a bad card. However, the fact that the fans help does seem to point to a temperature in the system. What kind of CPU cooler do you use? What are your CPU load temps during gaming? What are your motherboard temps during heavy gaming?

 

[justin4pack]

When the cpu was overclocked it would run 60c on load and now that its not oc, it runs around 55c on load. While playing bf3 max temp was 60c on gpu. Played for about an hour with no failures. Not sure what to use to monitor mb temps. Also whats the diff from refrence or none refrence.

 

[Termie]

When the cpu was overclocked it would run 60c on load and now that its not oc, it runs around 55c on load. While playing bf3 max temp was 60c on gpu. Played for about an hour with no failures. Not sure what to use to monitor mb temps. Also whats the diff from refrence or none refrence.

You have a reference cooler on your card. That means it was designed by AMD. It should provide proper cooling to all components. Non-reference custom coolers often cool the GPU more efficiently but leave out cooling of other components.

Your CPU isn't overheating either, so it seems everything is in order other than your video card, which appears to have a very negative reaction to heat. If you can't tolerate the noise from 100% fans, you may need to either RMA it or return it.

 

[justin4pack]

Yea i bought it 2nd hand so i guess ill have to deal with the noise. Its only 60% tho so not to bad

 

[BigChickenJim]

When the cpu was overclocked it would run 60c on load and now that its not oc, it runs around 55c on load. While playing bf3 max temp was 60c on gpu. Played for about an hour with no failures. Not sure what to use to monitor mb temps. Also whats the diff from refrence or none refrence.

HWinfo is the best for full-system monitoring. Use that to check your motherboard temps.

60C is a little warm depending on the CPU; AMD FX chips are generally only rated to 60C under sustained stress. What kind of CPU do you have? Are you using an aftermarket cooler?

I'm glad the issue has abated, but if you have a problem with one of your components overheating it's best to know which one in order to avoid possibly serious issues later on.

 

[justin4pack]

I updated my Rig sig , anyhow i thought the i7-920 ran a little hotter then recent Gen

 

[BigChickenJim]

Apologies, but I often post from my phone and can't see signatures. It doesn't appear that 60C is in the danger zone for that chip, but I definitely wouldn't want my CPU running that hot consistently. Then again, I'm conditioned to think with AMD standards in mind.

Did you check the other temps on the mobo with HWinfo?

 

[Sweepr]

60c is not exactly hot for an overclocked i7 920.

 

[justin4pack]

So i want to update whats happened. First of all my testing methods are not the best. Anyway so when i started the thread, when my video card reached around 70c in bf3 my comp would hang and hard lock. Figured it was the video card and have since been trying to keep it under 70c full load. Well then i decided to download kombuster. I can run kombuster for 10min under 100% load with temps reaching 78c and the computer is fine....wth?? so what ever the cause its only happening when the computer is in full use, i can run prime for the same and no issues with current OC. So now i am thinking the card is fine it must be something else. I wa reading where the 7xxx series card had issues with x58 chipsets. It would cause a sporadic hang. sounds familiar. I installed the recommended build 13.1. But in bf3 for around 5 6 min issue returned.

 

[BigChickenJim]

So i want to update whats happened. First of all my testing methods are not the best. Anyway so when i started the thread, when my video card reached around 70c in bf3 my comp would hang and hard lock. Figured it was the video card and have since been trying to keep it under 70c full load. Well then i decided to download kombuster. I can run kombuster for 10min under 100% load with temps reaching 78c and the computer is fine....wth?? so what ever the cause its only happening when the computer is in full use, i can run prime for the same and no issues with current OC. So now i am thinking the card is fine it must be something else. I wa reading where the 7xxx series card had issues with x58 chipsets. It would cause a sporadic hang. sounds familiar. I installed the recommended build 13.1. But in bf3 for around 5 6 min issue returned.

Ok... So as we suspected, this is not a GPU or CPU temp issue. That leaves us a few possibilities:

1. Something else is overheating. This is my guess since the extra fans seemed to help the issue for a while (or are the issues now also occurring with the fans?). While Prime95 really needs to run for a minimum of 30 min. to give a somewhat valid stability/temperature indication, I'd say the fact that you didn't crash out quickly as you do in-game rules out VRM or CPU overheating. If I had to guess why you'd only see temp-related crashes during "full-system" use, I'd say your northbridge is getting too hot. The only way to know for sure is to test.

2. You have a driver problem. You say you installed the "recommended" 13.1 build, but I'm pretty sure that the latest build for the 7870 is 13.4 or 13.5 (not on my computer or I'd check). Delete the old drivers, skip the auto-downloader, and get the latest driver package directly from AMD's site. You can also try downloading the latest CAP since it contains improvements for single-card setups.

3. Something's fishy with your mobo. I did a little research about x58/7xxx incompatibility, but I quickly discovered that a driver update was released to fix the issue late last year (http://support.amd.com/us/kbarticles/Pages/AMDCatalyst1211betadriver.aspx). As far as I can tell, there shouldn't be anything inherently problematic about a x58 board and a 7870 at this point.

You now know the card is fine and that the CPU is fine, but it's possible that there's something funny going on with your board. I have no idea what that could be. Unfortunately, the only way to test this theory is to swap boards and retest.

4. You are having some kind of system memory corruption issue. Are you running Prime95 "blend"? Running Memtest for a few hours and watching for errors would be good.

Out of curiosity, do you see any error or critical error entries in the Event Viewer log that correspond to your crash times? If so, telling us the codes and descriptions could help narrow down the possibilities.

 

[2is]

Power supply? Try running both a GPU stress test like Kombuster and a CPU stress test like intel burn in at the same time.

I know what you have SHOULD be enough, but stranger things have happened.

 

[justin4pack]

All of a sudden I am getting "your usb device can perform faster" pop ups with various devices. Possible MB to blame?

 

[BigChickenJim]

All of a sudden I am getting "your usb device can perform faster" pop ups with various devices. Possible MB to blame?

If your headers are not working correctly it could indeed point to a mobo problem. However, the mention of "pop ups" made me realize that no one that I've seen has suggested malware as a potential cause of your issues. What AV software do you use?

Also, did you ever check those northbridge temps?

 

[justin4pack]

What I meant by pop ups was notifications. Also nb temp was fine. Right around 60 on load.

 

[BigChickenJim]

What I meant by pop ups was notifications. Also nb temp was fine. Right around 60 on load.

I know what you meant, but it just popped (yep, I punned it) the malware idea into my head. 60C is a little warm for a NB, but it wouldn't (shouldn't) cause system failure.

I guess that narrows things down to a PSU problem or a software issue.