Quote from: chunkin on February 19, 2014, 08:58:50 PM

I can't seem to get my asus 280x 6card system stable on vertminer. All 3 of my rigs run Bamt 1.3. This is my only 6card configuration
This config achieves 375kh/s but only lasts for ~10mins then system freezes. I've tried default voltage for all six cards. Same outcome. Power is not the issue 1200w+850w for this system
Vertminer 0.5.3

Code:

"auto-fan" : true,
"gpu-fan" : "15-75",
"intensity" : "13",
"vectors" : "1",
"worksize" : "256",
"lookup-gap" : "2",
"thread-concurrency" : "8192",
"api-port" : "4028",
"expiry" : "30",
"gpu-dyninterval" : "7",
"gpu-platform" : "0",
"gpu-threads" : "2",
"gpu-engine" : "1050,1050,1050,1050,1050,1050",
"gpu-memclock" : "1500",
"gpu-powertune" : "20",
"gpu-vddc" : "1.2",
"temp-target" : "75",
"scrypt-vert" : true,
"log" : "5",
"no-pool-disable" : true,
"queue" : "0",
"scan-time" : "15",
"failover-only" : true,
"kernel-path" : "/usr/local/bin"
}

Updated app sdk to 2.9 and upped coreclock to 1100 pumps out 393kh/s but crashes same as before. My other 2 systems run 3 gigabyte 280x's and have no problem running 8191,8192,8193 TC with 1500mem. One 3card has 2GB's ram and the other has 4GB's ram. Both are very old dual core systems.

6card rig is brand new hardware 8GB's 6core. I've tried newer catalyst drivers, same outcome +lower hashrate. Also tried vertminer 0.5.2 and 0.5.4pre2 = lower hashrate+crash. Next on my list is to try 64bit xbuntu.
I'm pulling my hair out getting to the bottom of this one. I can drop the memory down to 1250Mhz but the wu is under 300 per card unacceptable. For now I'm stuck mining doge or whatever for the 6cards. Any ideas from anyone? Anybody with 6card rigs having issues with vertminer?

Edit: I rebuilt 0.5.3 from source and am 40mins in. No crash yet...

I believe, your problem is the video driver. I had a simillar trouble and I reinstalled the driver many times untill it stopped to freeze.
Another issue could be one of your video cards. Try the following : Start the miner with 1 videocard ( cgminer -d0 -c your.config.file) and wait to see if 1-st vGPU is ok. Repeat the procedure for the rest of your GPU's ( cgminer -d1 etc.) If all of them work stable launched as single GPU, try the following : cgminer -d0 -c your.conf and turn on from chminer control menu ( press letter "G", enable them one by one, like waiting few minutes to see how stable is your system after the next enabled GPU)
wish you good luck discovering the problem