Open appkisstronics opened 6 years ago
Probably....... but splitting the nonces into more threads should be implemented for cpu also....
you split the nonces into the number of threads that you want..... here I did it manually....
example for an 8 threads cpu :
4294967295 / 8 = 536870911.875
So : -n 0000000000 thread 1 -n 536870911 thread 2 -n 1073741823 thread 3 -n 1610612736 thread 4 etc..
each thread is 70000 hash / s , and when I do only one thread, it's 140 000 hash / s
I guess it's one thread per real cpu.
but 70000 * 8 = 560 000 hash / s instead of 140 000 hash / s
in other words, an old I7, if this was implemented, would do 4 times more hashes per second, reducing the total time needed by a factor of 4.
Please do let me know is there any way to execute this on GPU ?