Quote from: shgth on August 12, 2017, 02:16:43 PM

Getting this kind of error every now and then "[17:01:03][GPU0]:Invalid nonce" on the right side of the miner. 4x past 1½h. Everything works and all but just figured to post about it.

EDIT: Using this kind of command-line "cwigm -c sigt -u shgth.xxx -p xxx --lodiff --max-temp=70 --pstate=0 -d 0"(replaced private stuff with xxx). Max-temp was there by default and it's ok value so didn't remove it but added --pstate and -d.

well ...

the --max-temp parameter is an experimental parameter - and does work but will pause your miner for milliseconds at a time to 'cool' the gpu ... this causes seesaw power consumption ( ie - high then low then high ) among other things ... this is great for those in a very high temp environment and dont mind losing a small bit hash - but not if you dont need to ...

if you want the miner to run constantly - remove that setting ...

-d 0 is a setting that will only force gpu device number 2 to run ... and only that card ... if you have more cards and want to run them also - then do so ...

--pstate is a ccminer parameter - and even though we adopted a few parameters from ccminer and sgminer for compatibility reasons - we created a number of our own parameters as well ... under normal circumstances - this will not force it to stay in pstate 0 mode ... but please double check that - as some windows users do run in administrator profiles ...

if you are overclocking - lower it ...

see what happens when you mess around with the settings and tweak them so the miner can run more consistently ... and hopefully less / or none invalid nonces ...

we are working on this also ...

#crysx