Phoenix Miner detects one more gpu that what is installed?

[knofan]

Hey

This has been bugging me for some time now. I have tried three scenarios.

1: Use NH OS, no problems.
2: Use Win10 with NH miner, no problems
3: Use Win10 and Phoenix miner and point it to NH, it detects one more gpu than what is installed on the system.

I would like to use the miner stand alone while i tweek but ever time i start it up i get two gpus detected. The GPU is an PowerColor 5700xt.

I have disabled the onboard display in bios.

\bins\11.4\PhoenixMiner_5.1c_Windows>PhoenixMiner.exe -pool stratum+tcp://daggerhashimoto.eu.nicehash.com:3353 -wal 3KsX4rwbD....H92KQXN7P7SjK -worker rig1 -pass x -proto 4 -stales 0 -mt 2
Phoenix Miner 5.1c Windows/msvc - Release build
-----------------------------------------------

Unable to enum CUDA GPUs: unknown error
OpenCL driver version: 20.7.1
Available GPUs for mining:
GPU1: AMD Radeon RX 5700 XT (pcie 3), OpenCL 2.0, 8 GB VRAM, 40 CUs
GPU2: AMD Radeon RX 5700 XT (pcie 3), OpenCL 2.0, 8 GB VRAM, 40 CUs
Eth: the pool list contains 1 pool (1 from command-line)
Eth: primary pool: daggerhashimoto.eu.nicehash.com:3353
Starting GPU mining
GPU1: AMD driver 20.8.3
GPU2: AMD driver 20.8.3
Eth: Connecting to ethash pool daggerhashimoto.eu.nicehash.com:3353 (proto: Nicehash)
GPU1: 49C 0% 13W, GPU2: 49C 0% 13W
GPUs power: 26.0 W
Eth: Connected to ethash pool daggerhashimoto.eu.nicehash.com:3353 (172.65.200.133)
Eth: Subscribed to ethash pool

I have no idea what wrong. Can anyone help?

/H

[1715371938]

1715371938

[1715371938]

1715371938

[1715371938]

1715371938

[cornl]

But is it mining on both? I mean, is this your card hashrate divided by two? Or one is just like a ghost? Try comparing your results with others with the same chipset

[mkh90t]

add the -gpus command on your started to select only one of them

-gpus 1

That should solve your problem.  I just crashed my little windows rig this morning messing with some OC settings.  After the drivers reset my AMD rx 580s, I reloaded my normal OC settings and sarted right back up.  I noticed gpu 1 and gpu 2 were taking forever to make the DAG file, and even once complete they never started to mine before fatal error and miner restarted.

I switched to ETChash as it is slightly different, dags slow, but finished and started mining, but in the range of 3-4 mh/s.  Checked to make sure was still in compute, I was.  Then I tried Teamredminer and noticed at its start i saw the message that it was excluding a gpu in pci slot 1 as it was a duplicate and it loaded right on up.

So I looked at the logs and true enough, I went from 4 gpus to 5 gpus.  Once I added -gpus 1,3,4,5 it loaded up like it used to and no more issues.