Choosing multiplication algorithm

Torbjorn Granlund tg at
Mon Sep 28 17:15:43 CEST 2009

Pedro Gimeno <parigalo at> writes:

  Torbjorn Granlund wrote:
  > I updated with some curious diagrams for
  > multiplication algorithm selection.
  > Now we just need to figure out how to best decide which algorithm to use
  > in GMP.  :-)
  One suggestion: maybe it's worth plotting also the "second best"
  algorithm and the "distance between the best and the second best", to
  graphically see if it's worth the effort to discern between two given
  algorithms within certain ranges.

It would be nice, but I am not sure how to fit that too into the

I suppose one may ask to show just Alg-A and Alg-B, and use a greyscale
betwen them to show the speed difference.

But that doesn't generalise to 15 algorithms...


More information about the gmp-devel mailing list