Hello all. So I am using some 5700's and I can hit 56MH/s (can hit 60MH/s but way more incorrect shares) with them but I still get a considerable amount of incorrect shares. Has anyone found an optimal setting for a better accepted/incorrect ratio? On Claymore I only get 52Mh/s but 0 rejected/incorrect shares. Maybe it is just pushed too far and I might be better off going back to Claymore..
What do you guys think?
Thank you @Binary100100 and @joblo for your input. I was using a couple switches -openclLocalWork 128 -openclGlobalMultiplier 4096 which gave much higher speeds but led to a lot of incorrect shares. I re-adjusted a little bit ago and am only getting ~52MH/s per card @450W from the wall (3 cards). I think I might tinker a bit more to see if I can hit 53MH/s for each.
I was hoping that I can replace my 1070 rigs that have 6 X 1070's with just 3 X 5700's but it doesn't quite look like that can be the case. Maybe Phoenix will have some more optimizations up their sleeve to help achieve this!! ;-)