i3930K benchmark result with libgmp-5.0.4

[j. charton]

Hi Torbjörn,

Yes, I've made the previous bench with Turbo mode on (default mode) which means CPU frequency up to 3.8 GHz (not locked at 3.2 GHz)

Following is the new bench result with Turbo mode disabled.
But what is strange in this case is that the bench for the core i5 2500 (gmp 5.0.1) reported on http://gmplib.org/gmpbench.html is a bit better.
Are you sure the bench reported for the i5 2500 is also with turbo mode disabled ?
I hardly imagine the i5 2500 beating the i7 3930k.


Using gmp version: 5.0.4
Compiling benchmarks
Running benchmarks (propagated score accuracy exceeds printed intermediates)
  Category base
    Program multiply (weight=1)
      GMPbench.base.multiply(128)                        7.814e+07
      GMPbench.base.multiply(512)                        1.622e+07
      GMPbench.base.multiply(8192)                        1.72e+05
      GMPbench.base.multiply(131072)                          2951
      GMPbench.base.multiply(2097152)                          108
      GMPbench.base.multiply(128,128)                     7.83e+07
      GMPbench.base.multiply(512,512)                    1.184e+07
      GMPbench.base.multiply(8192,8192)                  1.087e+05
      GMPbench.base.multiply(131072,131072)                   2049
      GMPbench.base.multiply(2097152,2097152)                 55.3
      GMPbench.base.multiply(15000,10000)                 5.78e+04
      GMPbench.base.multiply(20000,10000)                4.214e+04
      GMPbench.base.multiply(30000,10000)                2.693e+04
      GMPbench.base.multiply(16777216,512)                     393
      GMPbench.base.multiply(16777216,262144)                 16.8
    GMPbench.base.multiply                         37074
    Program divide (weight=1)
      GMPbench.base.divide(8192,32)                      1.627e+06
      GMPbench.base.divide(8192,64)                      1.655e+06
      GMPbench.base.divide(8192,128)                     9.813e+05
      GMPbench.base.divide(8192,4096)                    3.079e+05
      GMPbench.base.divide(131072,65536)                      2887
      GMPbench.base.divide(8388608,4194304)                   11.4
      GMPbench.base.divide(8192,8064)                    1.393e+07
      GMPbench.base.divide(16777216,262144)                   8.21
    GMPbench.base.divide                           36370
    Program gcd (weight=0.5)
      GMPbench.base.gcd(128,128)                         2.329e+06
      GMPbench.base.gcd(512,512)                         5.256e+05
      GMPbench.base.gcd(8192,8192)                            9080
      GMPbench.base.gcd(131072,131072)                         140
      GMPbench.base.gcd(1048576,1048576)                      7.12
    GMPbench.base.gcd                             6440.4
    Program gcdext (weight=0.5)
      GMPbench.base.gcdext(128,128)                      1.892e+06
      GMPbench.base.gcdext(512,512)                       3.82e+05
      GMPbench.base.gcdext(8192,8192)                         5514
      GMPbench.base.gcdext(131072,131072)                     86.2
      GMPbench.base.gcdext(1048576,1048576)                   4.68
    GMPbench.base.gcdext                          4377.6
  GMPbench.base                          19274
  Category app
    Program rsa (weight=1)
      GMPbench.app.rsa(512)                              2.935e+04
      GMPbench.app.rsa(1024)                                  4977
      GMPbench.app.rsa(2048)                                   790
    GMPbench.app.rsa                              4868.8
    Program pi (weight=1)
      GMPbench.app.pi(10000)                                   746
      GMPbench.app.pi(100000)                                 36.4
      GMPbench.app.pi(1000000)                                1.87
    GMPbench.app.pi                               37.031
  GMPbench.app                          424.61
GMPbench:         2860.7

Jean-Louis

 

----- Mail original -----
De : Torbjorn Granlund <tg at gmplib.org>
À : j. charton <j_chrtn at yahoo.com>
Cc : "gmp-discuss at gmplib.org" <gmp-discuss at gmplib.org>
Envoyé le : Mardi 13 mars 2012 18h36
Objet : Re: i3930K benchmark result with libgmp-5.0.4

"j. charton" <j_chrtn at yahoo.com> writes:

  I'm not really sure that this is the right email address for this
  purpose; but I would like to report the following bench result.

These numbers are higher than I would have expected.

Do you have "Turbo" enabled?  While that perhaps is useful in some
situations, it makes benchmarking give results that are somewhat less
useful.

-- 
Torbjörn