Adding hardware monitor flag --HWMON causing exception/exit under Windows 10
I just upgraded my Win10 Nvidia drivers and ethminer (to cuda10/v17) and now I get an error using same start script.
Expected behavior
--HWMON INT flag should work. If I set it to 0 (off) or 1, I see same exit/error.
The actual setting doesn't matter. Just the presence of the flag itself.
And yes, I am running under an administrator cmd for everything. Same as before.
Screenshots (Optional)

Work-around
(If I drop the --HWMON flag, it starts mining fine.)
Environment (please complete the following information):
Win10
nVidia 1060/6gb
ethminer v17/Cuda10 (release)
i have had the same, but not clever enough to remove this setting. I thought the issue was due to a faulty driver update or similar.
Yeah, I just start eliminating flags from command line till i find the stinker. lol
Might be easy fix since the presence of flag itself and not necessarily it's setting is breaking it.
(In other words: You would think --HWMON 0 would be fine since that is 'OFF'. But, nope.
--HWMON if not set in the CLI arguments list is meant to be off
--HWMON 1 means to monitor fans and temperatures
--HWMON 2 same as 1 plus power drain
This said you must have some problems on your Nvidia CUDA driver setup as it's been tested multiple times and works properly.
@AndreaLanfranchi Understood & TY. Not sure what the issue could be - everything same. Driver upgraded as noted. ethminer updated as noted. Nothing else. Will report back if I find the problem.
Also, it appears I am not alone in this issue^^ - so your 'multiple tests' may not be complete? Have you tried it on win10 box with 1060/slightly older card w/ newest stable nvidia driver (Minus the 'nvidia experience' bloat - which i never install)? thanks
On my end I have the "Nvidia experience" installed and have the same issue.
I am able to reproduce it on my 1080 with latest drivers, clean install. if the problem is still present after reboot i will try to analyse and fix the issue.
Happens on precompiled cuda9 and 10 exes
Thanks for reproducing it @smurfy .. yeah I am busy w/ work at moment. Pretty sure I rebooted mine? (tho now you got me wondering.. lol)
@AndreaLanfranchi If those are the flags then need to update --help?
This is v17 release:
ethminer.exe --help
Ethminer - GPU Ethereum miner
Usage: ethminer.exe [OPTIONS] [pool...]
(..snip..)
--HWMON INT in [0 - 1] 0 - Displays gpu temp, fan percent. 1 - and power usage. Note for Linux: The program uses sysfs for power, which requires running with root privileges.
Assuming you still meant 0 or 1 (but understand that it is 'on' either way now.. sorry for confusion!)
--HWMON 1 means to monitor fans and temperatures
--HWMON 2 same as 1 plus power drain
@smurphy
I'm back. Just rebooted win10 miner box twice.
Same .exe exit popup when using HWMON switch, unfortunately.
@bmatthewshea yea, still happening after reboot, so i can have a look.. it crashes way early for me.. if i enable it. must be somewhat related to hwmon/driver, because it worked before updating driver. (had quite old cuda 9 only installed)
Basically same here.
Yeah, the only reason I updated the driver is because I wanted to try new cuda10 release/build. Had been using v14 ethminer iirc.
Driver seems to be fine other than the hwmon issue.
.. nVidia had to go mess with something apparently. lol
found the problem, but going to sleep now. will create a fix later.
Please try this build (when ready): https://github.com/ethereum-mining/ethminer/releases/tag/v0.17.1-rc.0.
@chfast All good now. Just had chance to test.
--HWMON works.
m 21:33:10 main Speed 23.83 Mh/s gpu0 23.83 53C 50% 78W A0 Time: 00:02
Thanks & thanks @smurfy