Verthash miner mines on some of the selected GPUs, puts the rate in the wrong GPU display and eventually the interface crashes and verthash stops mining.
Running the latest HiveOS on my rigs.
I have 4 8gb GPUs that I mine Ethash and 4 2gb GPUs that I mine Vert/Doge. I have mined each type on separate HiveOS rigs just fine, but combining them has interesting side effects.
The GPUs are assigned from 0-7: 8gb, 2gb, 2gb, 2gb, 2gb, 8gb, 8gb, 8gb
I can select the 4 8gb GPUs in either phoenixminer and nbminer and mine just fine and the rates show up on the correct GPU and the worker display show the miner using only those 4.
When I use just verthashminer, selecting various numbers of 2gb GPUs, the worker display shows all GPUs selected.
The first GPU in the selection list does not start (I have reversed selection order to verify) and eventually times out and stops verthashminer. I can see the other selected GPUs running in the shell-in-a-box miner view and I can verify by GPU temp in the Hive worker view that the correct GPUs were selected by verthashminer. The hash rate displayed is always shifted to the GPU one to the left. After a while the worker view becomes disconnect from rig.
I get similar results when combining the miners in one flight sheet.
I suspect that if I can reorder the GPU locations in HiveOS the problem might go away, I don’t know how.