16*16*16 Its too hard for GPU . The current situation is not very good for HPP to move forward,And now the rules of output are more fixed, what do you feel?The choice of Lyra Matrix with 16 Cols, 16 Rows and 16 TimeCost was made with concertation of all Dev Team, we have done a lot of test using multiple parameters combinations, finally we come up with 16*16 and 16 for TimeCost.
The Goal is to have Egalitarian Mining 1:1 CPU:GPU.
Best Regards
Ramlin
HPP Team
I like lyra2 too, I test it in the 270x 280x 290 and 750ti 1050ti . I get it Best Regards.