The only thing that comes to my mind is that this miner is modifying ntime. We don't allow modified ntimes, because we have no methods to evaluate whether modified ntime would be accepted by remote pool or not. There should be a way to force miner not to modify ntime; in fact, there is plenty of extranonce2 size so the ntime doesn't have to be modified. Guess that is the change compared to older miner.
ntime is modifed. As is valid on all good pools.
kano ... have you incorporated the "idle bug" fix in your builds that is mentioned here?
Bummer. There are still some strange issues going on with my AntMiner S1s working against the pool with "price" thresholds.
I just came home and found my two test AntMiner S1s hung ... they *said* the NiceHash pool was alive ... but they were NOT mining it ... and they were NOT mining my backup pools.
Both were mining earlier in the day, and then when the 0.06 contracts expired, they fell back to the backup pools. That is where they were when I left the house. I come home and realized they were in some strange state ... not mining NiceHash nor the backup pools.
I had to restart the miners, and then they immediately jumped back to the backup pools as there are no 0.06 contracts.
Something is broken ...
ckolivas informed me, that idlebug is fixed in latest cgminer. Do you use latest cgminer?