Quote from: arulbero on October 24, 2018, 07:31:46 AM

Quote from: Remember remember the 5th of November on October 20, 2018, 09:06:57 AM

I was reading up on whether it was possible to get faster performance on key generation and came across this post https://crypto.stackexchange.com/questions/29885/can-a-billion-elliptic-curve-keys-be-generated-on-a-laptop-in-less-than-an-hour

The very last post claims pretty high numbers and wanted to know if it was bull or real. If it's real what would the math be for those kinds of speeds? And what is referred by "symmetrie" of seck256k1?

I wrote that answer.

The current speed of the public keys generation on my laptop is 185 MKeys/s (I compute only compressed keys, only the "x" coordinates). It's real.

If I computed "y" coordinates too I could exploit the fact that (x,y) and (x, p-y) (that is the "symmetrie") are 2 valid points and maybe I would get more speed.

I'm talking about public keys generation, not addresses generation. The performance of addresses generation is obviously lower, about 12.7 MAddresses/s (always on cpu)

If we compare vanitygen and my library on writing addresses speed (not only generation) the difference is even bigger:

--> https://bitcointalk.org/index.php?topic=25804.msg23710724#msg23710724

Secp256k1 library too is using optimizations to get more speed: https://bitcointalk.org/index.php?topic=2934774.msg30174356#msg30174356

Should I assume that you do not wish to make the math available? Or is it already implemented in libsecp256k1? Cause millions of combos per second is pretty good, on a CPU too.