Hard Crash

[peterb]

Don't know about the temp monitoring tools. Any configuration settings that allows you to select which card?


As for other things - memtest86+ is an obvious (and easy) one to do. It could be a PSU issue. Have you tried running some of the stress testing applications to see if occurs ((and so ruling out the game itself)

[Phage]

memtest I can download, but what are the stress testers you were thinking of ? Furmark ?

[Phage]

Hmmm. Looking around, this seems to be a fairly known issue http://www.google.co.uk/search?hl=en...crash&gs_rfai=
Let's see what the other games do.

On GPUZ there are three GPU temp readings. Does anyone know what they are measuring ?

EDIT: Have found the answer. At the bottom of the sensors tab there is drop down box. You can only select one card at a time, so select one card and then open another instance of GPUZ and select the other.

[jim]

Can't see the image

[Phage]

Can't see the image

Hmmmm. It was there last night. (Imageshack)
I can't do anything from here though as my work firewall prevents access to any of the image sites I'm afraid.

[Phage]

Interesting - seems to be software.

Have a look at the file below. Whilst playing BC2 temps are excellent, and both cards are behaving normally. When the game is closed, the GPUs both switch to 2D mode and frequencies drop to 175. BUT, after a few seconds it appears that the second GPU zooms back up max output. The odd thing is that temps are not increasing on that GPU as you would expect if it was constantly running at 100%.




This leads me to assume that both GPUZ and MSI Afterburner are reporting the second card at 875 *incorrectly*, (some sort of default reading ?). Further, shouldn't ATI overdrive show both cards in the drop down menu at the top ? Mine only shows one GPU.



I think that the ATI driver may be completely shutting the 2nd GPU down to the point where is doesn't register at all. <sigh> This means that I really have no idea what's going on. I think I have ruled out temps (Short of an all-night session)..What do you think ?

[peterb]

I think I'd be tempted to remove GPUZ and see if it happens again - if it does, I'd remove the ATI application (does it do anything else other than report temperatures) and test again.

[Donni10]

MSI Afterburner is a good replacement to CCC

[Phage]

I think that I don't *actually* have an issue. I think that both afterburner and GPUZ are failing to read the second card becuase CCC has shut it down.

I found this quote from someone who had the same problem with two Vapor-X cards.

"Now I see that GPU1 in the sidebar gadget drops to 157/300 MHz when I'm on the desktop and GPU2 drops to 850/1200 MHz. However, GPU2's temperature is within 3 to 5 C of GPU1's which is running at 2D speeds so it's clear to me that the second card *is* idle, it's just that Afterburner/RivaTuner cannot read the values because PowerPlay has disabled it so uses the BIOS defaults (or something). Oddly, the temperature of the second card can still be read by the sidebar gadget even though the clock speeds can't. Don't know why that is."
http://forums.guru3d.com/showthread.php?t=323006

[Phage]

Found the problem. It was heat - but not due to the clocks on the 2nd card, but becuase the fans didn't accelerate with load. I don't know why this would be the case, but with a manual fan profile on MSI afterburner all is fine. Any ideas on why the CCC auto fans wouldn't work ? I`m running 10.10 WHQL drivers.