Quote from: Claymore on June 23, 2017, 12:16:30 PM

Quote from: Aquazi on June 22, 2017, 10:39:44 PM

Something I really dislike about your miner Claymore is the fact that if I'm dual mining ETC, when DevFee starts I lose another 30-40 seconds because it needs to swap to a different DAG and then after DevFee ends it changes again to a different DAG (132 on ETC 130 on ETH).
I really, really dislike it about your miner.
Just set an ETC pool for your devfee if mining ETC for christ sake, I understand you do a good job with your miners (you also make an awful lot of money, but it's cool anyway, it's deserved), but that's a long way from 1-2%, it's much closer to around 5% of lost hashrate each our with this dag swapping.

You can use pools that accept wallet address directly and remove "-allpools 1" option.

Quote from: simon66 on June 23, 2017, 04:46:07 AM

Hi Claymore,
Because of the AMD init bug (where cards are not initiated sequentially) the current miner version (9.5) is unable to know which card is which.
After looking at HWiNFO I noticed that each card has a DeviceInstanceId which doesn't change when I reboot my machine. Why can't I use that to target a OC on that card? Right now every time my machine reboot I have to manually OC each card (Kinda annoying).
Doesn't sound like an easy implementation but it doesn't hurt to ask

Miner cannot read DeviceInstanceId via OpenCL. But recently I finally found a way to match the OpenCL list of GPUs and the list of GPUs that is used in ADL for temperature/fans/OC management, it must help, I will try to put it in next update.
Also now I can enumerate GPUs as AfterBurner does.

Which is the way for?