CGMiner keeps on crashing?

[ragingazn628]

hey guys,

I'm currently using the latest CGminer and a couple of days ago CGminer has been turning itself off every 2 hours. I don't know the reason behind this. Other than the fact that I just did a fresh win7 install again.

[ragingazn628]

Same drivers

SDK 2.1
11.12 Catalyst

here's the error log:

Problem signature:
  Problem Event Name:   APPCRASH
  Application Name:   cgminer.exe
  Application Version:   0.0.0.0
  Application Timestamp:   4f47034c
  Fault Module Name:   aticaldd.dll
  Fault Module Version:   6.14.10.1646
  Fault Module Timestamp:   4ebb3717
  Exception Code:   c0000005
  Exception Offset:   0017e52c
  OS Version:   6.1.7601.2.1.0.256.1
  Locale ID:   1033
  Additional Information 1:   0a9e
  Additional Information 2:   0a9e372d3b4ad19135b953a78882e789
  Additional Information 3:   0a9e
  Additional Information 4:   0a9e372d3b4ad19135b953a78882e789

Read our privacy statement online:
  http://go.microsoft.com/fwlink/?linkid=104288&clcid=0x0409

If the online privacy statement is not available, please read our privacy statement offline:
  C:\Windows\system32\en-US\erofflps.txt

[ragingazn628]

I was getting "C05" exceptions when I pushed card too hard.
Trying to set Max VDDC outside of BIOS range, or pushing core clocks too far.

Try to lower your clocks, bring voltages to stock and gradually move them up/down to your optimal settings
until you get some stability.

cgminer does not catch exceptions (probably for the sake of efficiency) so if driver goes boo boo you get what you get.
It might be cgminer code that does not check its pointers before every access (again for efficiency) and expects driver to allocate
stuff properly, then it passes a null pointer or handle in opencl call and we get c05

they are NOT overclocked

they are under voltage tho.

5870s are at .9v

[Buckwheet]

I am getting the same problem on a brand new 3x 7970 install. I installed 12.3 drivers and they are XFX Black Editions running stock everything. Win 7 - 64 clean install, with just 12.3 installed and CGMiner. Runs for about 30 minutes to 2 hours and crashes out. At first I saw this thread and thought it might be CGMiner so then I tried Diablo, with the same result.

So there must be something I am missing during the install or setup.

[ragingazn628]

it's still happening... anyone know why?

same drivers / sdk
same windows as last install

same cgminer ;o

anything wrong with my settings?

"intensity" : "5,5,5,5,5",
"vectors" : "2,2,2,2,2",
"worksize" : "256,256,256,256,256",
"kernel" : "phatk,phatk,phatk,phatk,phatk",
"gpu-engine" : "800,850,850,775,775",
"gpu-fan" : "0-85,0-85,0-85,0-85,0-85",
"gpu-memclock" : "300,300,300,300,300",
"gpu-memdiff" : "0,0,0,0,0",
"gpu-powertune" : "0,0,0,0,0",
"gpu-vddc" : "1.000,1.000,1.000,1.000,1.000",
"temp-cutoff" : "95,95,95,95,95",
"temp-overheat" : "85,85,85,85,85",
"temp-target" : "65,65,65,65,65",
"api-port" : "4028",
"auto-fan" : true,
"expiry" : "120",
"gpu-dyninterval" : "7",
"gpu-platform" : "0",
"gpu-threads" : "4",
"log" : "5",
"queue" : "1",
"retry-pause" : "5",
"scan-time" : "60",
"temp-hysteresis" : "3",
"shares" : "0",
"kernel-path" : "/usr/local/bin"
}

[ragingazn628]

it's still happening... anyone know why?

same drivers / sdk
same windows as last install

same cgminer ;o

Are you running the same Windows SPs, updates?
Turn off unused Win services, updates, firewall, IIS etc.

From my limited experience, my 7970s do not like any memory clocks below 800, and core clocks above 1180.
1170/800 seems stable.

Voltage, well, it is kinda on its own.

I am running 3 5870s and 1 5970

yes all those are off as well

[Buckwheet]

I ended up doing a re-install and not doing any updates. 12.2 drivers and my 3x7970 became stable.