System keeps crashing

[Yannick]

Hi all,

I have the following system to mine for bitcoins:

core i7 920
asus p6t-se
3x2gb corsair xms3 1333mhz 9-9-9-24
thermaltake element s case
hdd wd re3 500gb sata300 hdd
coolermaster v8 cpu cooler
2x brand new sapphire 5870 gpu's
Corsair CMPSU-750AX (750 Watt)
W7 ultimate

I'm using MSI Afterburner to set core clock 950 MHz and memory clock 350 MHz. (Mining at ~ 420 Mhash/s)

But my system restarts sometimes for no reason while mining. I can't figure out how to fix this problem. I keep my cards temperature well below 80°c.

Anyone an idea to fix this annoying problem?

Thank you!

[whopper]

what is your power supply?

[Yannick]

what is your power supply?

Oh yes, I forgot to mention:

Corsair CMPSU-750AX (750 Watt)

[Agozyen]

Can be one of three things, to much heat, unstable power, or overclocking too much.

 It doesn't have to be your video cards that over heat to force your system to shut down.  What are your CPU temps?

 Does your system crash with just one card in?

 Lastly, does your system crash when cards are not overclocked?


 One thing about overclocking, just because the slider goes to 950 doesn't mean it can actually do 950.  The first step to determine what's wrong here is to set the cards back to factory clocks and see what happens.

[joulesbeef]

you might want to check your event logs.. most of the time they will mention what caused the restart.

go in cntrl panel, admin tools and event viewer.. i would start with system.. looking for red icons.. and then I would check application. most likely something that caused a reboot will be listed in system

you got some nice ram and cpu there.. I take it this was your main system before mining?

what did you use for video before?

Did you uninstall that video cards drivers before shuttng down and taking it out?

did you use a driver sweeper?(get a good one from guru3d)

also might want to make totally sure the cards are seated all the way in.. and yeah power supply voltage swings can cause it as well.. might want to monitor voltages and post them here, so we can see if any are out of spec.

[Yannick]

Can be one of three things, to much heat, unstable power, or overclocking too much.

 It doesn't have to be your video cards that over heat to force your system to shut down.  What are your CPU temps?

 Does your system crash with just one card in?

 Lastly, does your system crash when cards are not overclocked?


 One thing about overclocking, just because the slider goes to 950 doesn't mean it can actually do 950.  The first step to determine what's wrong here is to set the cards back to factory clocks and see what happens.

My CPU temps are fine, I guess, ~49°C.

When the cards are not overclocked, the system doesn't crash. I suppose I have to test and see what my cards overclock limit is.

[Yannick]

you might want to check your event logs.. most of the time they will mention what caused the restart.

go in cntrl panel, admin tools and event viewer.. i would start with system.. looking for red icons.. and then I would check application. most likely something that caused a reboot will be listed in system

you got some nice ram and cpu there.. I take it this was your main system before mining?

what did you use for video before?

Did you uninstall that video cards drivers before shuttng down and taking it out?

did you use a driver sweeper?(get a good one from guru3d)

also might want to make totally sure the cards are seated all the way in.. and yeah power supply voltage swings can cause it as well.. might want to monitor voltages and post them here, so we can see if any are out of spec.

I checked the event logs, and they don't mention what caused the restart. It only says that my system has been rebooted without a proper shutdown and can cause serious damage.

I bought the case, mobo, chipset, ram and hdd from a friend. Then I bought 2 brand new 5870's to start mining. I also bought a brand new 750W gold cert PSU. The hdd has been formatted and I freshly installed W7, gpu's (crossfire) and used the ati 10.7 drivers (the newest drivers cause a slight lower Mhash/sec).

How can I monitor the voltages? I guess the voltage swings are causing the reboots. But it seems that if I don't lock my system (windowskey+L) the system doesn't reboot, but if it's locked for a while, it seems to reboot from itself. Strange!

[joulesbeef]

you can get something like SIW.. system information for windows and click on sensors.

[hmblm1245]

You may want to try turning off the W7 Aero themes. i have a quad 5830 miner with random freezing on stock speeds with my AC blowing right on the cards (gpu1 58c gpu2 60c gpu3 60c gpu4 48c). after turning off aero it became stable under full gpu load and i could over clock them to 1000mhz. just a thought.

[Agozyen]

Can be one of three things, to much heat, unstable power, or overclocking too much.

 It doesn't have to be your video cards that over heat to force your system to shut down.  What are your CPU temps?

 Does your system crash with just one card in?

 Lastly, does your system crash when cards are not overclocked?


 One thing about overclocking, just because the slider goes to 950 doesn't mean it can actually do 950.  The first step to determine what's wrong here is to set the cards back to factory clocks and see what happens.

My CPU temps are fine, I guess, ~49°C.

When the cards are not overclocked, the system doesn't crash. I suppose I have to test and see what my cards overclock limit is.

Overclock in small increments and test each increase until you are sure it's stable.  Once instability sets in just go back to last good overclock and you are set.