[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gmp-devel
Subject:    Code Submition ( Improvement for mpn_mul_fft )
From:       tege () swox ! com (Torbjorn Granlund)
Date:       2004-10-27 14:58:53
Message-ID: 86mzy81fn7.fsf () king ! swox ! se
[Download RAW message or body]

"Jeunder Yu" <gis91542@cis.nctu.edu.tw> writes:

  > How does the speed compare between the old and new code?  In the
  > GMP world, speed is all that counts, of course.  :-)

  CPU: Pentium 4 2.6GHz
  RAM: 512MB

  Two operands are 2^n limbs with random value.

  size    old     new
  ===================  n   0.013   0.011
  n   0.031   0.027
  n   0.072   0.064
  n   0.160   0.145
  n   0.374   0.348
  n   0.824   0.767
  n   1.682   1.441
  n    3.398   2.845
  n!   7.252   6.053
  n"  16.625  13.957
  n#  34.203  29.497
  n$ 408.061 350.596
  (virtual memory swap in/out for n>$)

  Is this comparison enough?

Quite nice improvements!

Paul Zimmermann improved our present code a while ago, but I
think your code is faster than even the improved code.

--
Torbj?rn

[prev in list] [next in list] [prev in thread] [next in thread]