On Sun, Mar 1, 2009 at 10:57 AM, Cactus <rieman...@googlemail.com> wrote:
> For AMD64 GMP 4.2.4 with the MSVC stuff from my site and MPIR 0.9.x > with MSVC are essentially identical in performance since the MPIR and > GMP assembler code is identical. > > In MPIR 1.0.0 all the x64 assembler support is based on new code from > Jason M. With Jason's help I have learnt a lot I did not know about > AMD64 in the last week! Is your updated x64 assembler code in SVN trunk right now? I just pulled down the latest SVN trunk code but the speed doesn't seem to have changed much from 0.9.0. This is using your "Core2" project in MSVC for 64bit. The 32bit version I'm picking "Pentium4" since that seems to be the newest Intel architecture supported for 32bit. Intel Core2 Q9550 @ 3.4GHz (Vista 64bit) GMP 4.2.4 (MSVC 64bit) = 2m06.904s MPIR 0.9 (MSVC 64bit) = 2m07.032s MPIR SVN (MSVC 64bit) = 2m09.020s MPIR SVN (MSVC 32bit) = 6m43.741s GMP 4.2.4 (MSVC 32bit) = 6m45.073s MPIR 0.9 (MSVC 32bit) = 6m50.230s Jeff. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "mpir-devel" group. To post to this group, send email to mpir-devel@googlegroups.com To unsubscribe from this group, send email to mpir-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/mpir-devel?hl=en -~----------~----~----~----~------~----~------~--~---
