Quote from: danielyap on November 04, 2017, 09:05:41 PM

Quote from: greatauror28 on October 31, 2017, 05:52:35 PM

Quote from: leonix007 on October 31, 2017, 05:33:52 PM

Quote from: joboy84 on October 31, 2017, 04:38:38 PM

Hello,

I have the same error, but I cannot figure out why.

I had 2 Sapphire RX580 8Gb cards running since September.

They ran without any problems (Claymore settings 1260/2230/910)

PSU : EVGA 1600 T2

Since I added a 3rd card I have approx every 3-4 hours the message :

WATCHDOG: GPU 1 hangs in OpenCL call

Sometimes it's GPU 3, sometimes GPU 2...

I put specific values for GPU 3 (115/2170/910) and always the same problem.

I changed the Rizer, reinstalled the drivers but nothing seems to work. Maybe a Win10 issue ?

Thx

Try a clean uninstall of drivers using DDU

http://www.guru3d.com/files-details/display-driver-uninstaller-download.html

then install blockchain drivers for windows 10

If this doesn't work, I suggest updating to Win10 Fall Creator's Update and to AMD's 17.10.3.

It could very well be a windows update sneaked on you and wrecked your GPU drivers.

I'm running on RX580 Sapphire Nitro+. Was running on Win10-64Bit-Crimson-ReLive-Beta-Blockchain-Workloads-Aug23 getting 30mhs. Kept getting the OpenCL GPU Hang errors. I did the AMD 17.10.3 update and tried to run again but my hashrate greatly reduced to 23mhs. Any idea how to run on the new driver and get 30mhs?

Be sure to set the AMD Radeon settings to "Compute" instead of "Graphics".