Work in progress. Might release when I am happy with the results. In the next version I will include the ptx assembly code so it might work on compute 8.0 cards(rtx 3080, rtx 3060ti etc..) . sp-mod #3 only works for compressed keys, so #4 will include kernels for non compressed and both as well. Cuda 11.2 doesn't work on my dev machine. (windows 8.1) so I will try a bin with cuda 10.2 with ptx code included.