Thanks, I did the change to app_info.

I have left things running for a few days to see how it's all working out but I have noticed today that the GPU is erroring out with computation errors a lot though it appear some might have gone through .

See here http://setiathome.berkeley.edu/results.php?hostid=5883516&offset=0&show_names=0&state=5&appid=

Could some knowledgeable person have a look and see if all is ok or what the erros mean so I can fix. I have tried searching these forums for the error (-1073741819 (0xffffffffc0000005)) number but did not find anything.

From what I can see is that nothing on the GPU has worked since 8th March.

Thanks

PS, my other identical PC with GTX460 arrived today and that has started so I should see how these 2 identical PC's work against each other.

---

Im no expert but I will try to help. After you first posted and were told you had the wrong apps, Did you uninstall boinc and lunatics? or just try and reinstall? Have you pulled the cover off and taken a look see if any hairballs are clogging up the fans and intakes or exhaust ports?

Have you updated your video drivers to be compatable with the version of lunatics you are using?

Im sure someone who knows a heck of a lot more will be around shortly.

---

[/quote]

Old James

Well one of them is a 0xc00000005 error which is normally associated with heat or hardware failure. You are having a lot of them fall back to host cpu processing, out of memory errors etc. But it is all related directly to the video card.

1. How hot are your gpus getting?
2. How much air flow do you have into your case?
3. Have you checked and removed any excess dust that may be in your case?
4. What's the ambient temp in your computer room?
5. Have you tested your memory?
6. Are you overclocking any?
7. Have you stress tested your gpus?

I'm willing to beat you have a dieing piece of hardware though. My 8800 was throwing out those same errors a month or so ago, then it mysteriously stopped doing it for about a day or so. Then the card completely died. Bad memory on the gpu itself. I had two cards and it would hang the drivers when the one card would die and reset the driver. When it did it would trash both work units on both cards, one with a memory error and the other with a time out one.

I would highly suggest you stress test each gpu separately before going any farther with anything else. Try using furmark etc. and look for artifacts, bsods, lockups, etc. etc. while it's running. It will also give you the temps make a note of those as well.

---

Traveling through space at ~67,000mph!

I have noticed today that the GPU is erroring out with computation errors a lot though it appear some might have gone through .

Did you finish off all existing Work Units before chaning the app_info file & then downloading new work?
If not, most of them will time out.

---

Grant
Darwin NT

I looked at the last bunch only - you are having GPU memory issues.
Several intances of 'Falling back to HOST CPU processing...' - which doesn't work in current x32f.
Try a reboot to clear GPU RAM.
Anything graphics intensive running on the side?

---

Carola
-------
I'm multilingual - I can misunderstand people in several languages!