For some reason I can’t seem to prevent VertHashminer from running on both AMD & Nvidia cards.
If I specify in miner setup “–all-cl-devices” Hiveos overrides that and adds “–all-cu-devices” no matter what settings I try to enter.
Even tried to edit /hive/miners/verthasminer/0.7.2/miner.conf to prevent using Nvidia cards, still no dice.
Changed owner to root and write protected the file, no matter, still same result.
Haven’t figured out a solution to this yet, running HiveOS 0.6-203@210415.
Any help would be greatly appreciated!
I’m experiencing the same issue. I only have a single AMD card next to 5 Nvidia cards, and specifying the cl device index also fails to solve the issue. No matter what, every card in the machine starts mining when I launch VertHashminer.
Hope someone gets it figured out soon! Until then, back to RVN…
(I just figured I’d try that last one since that format works with T-Rex and I was getting desperate). If I remember right, all of them threw some type of “incorrect parameter” errors except for the first one, which proceeded to mine with all 6 cards (5 nvidia, 1 AMD). I did run it once with --device-list in the parameters to confirm the index, which worked as expected. Cuda devices had indices 0 through 4, the single OpenCL device (also) had index 0.
So if we’re assuming “–all-cu-devices” gets added by HiveOS automatically, I guess it makes sense that it was filtering my cards down to all of my Nvidia cards, and also my only AMD card at index 0.
I did not try adding a “–cu-devices 0” in there as well to see if it would use 1 AMD card and 1 Nvidia card, but maybe I could try that later just to see what happens. My AMD card is an old 4gb that can still turn a profit on RVN and VTC, while the Nvidia cards are all mining ETH (which I don’t want to change).
My situation is very similar, have 2 Nvidia cards with 8GB+ vram and all AMD cards are 4GB, kind of annoying as I haven’t found a way to circumvent hiveos implementation of how the write the config for the miner.
Update:
Support got back to me already, they identified problem and will rectify in next update. Great work from their part with the response time, gj hiveos team!
