• [^] # reprennons à 0

    Posté par . En réponse à la dépêche Comparatif Intel C++ 7.0 / Gcc 3.2.1. Évalué à 3. Dernière modification le 05 décembre 2021 à 17:57.

    Bon autant pour moi, j'avais pas fait recompiler les autres .o avec les bonnes options

    Maintenant, avec les options données par Coin.

    Sur le P4 1.7

    Donc, avec gcc 2.94.5 

    moa@dionysos:~/tmprm$ scimark2 
** ** 
** SciMark2 Numeric Benchmark, see [http://math.nist.gov/scimark(...)](http://math.nist.gov/scimark) ** 
** for details. (Results can be submitted to pozo@nist.gov) ** 
** ** 
Using 2.00 seconds min time per kenel. 
Composite Score: 311.14 
FFT Mflops: 204.72 (N=1024) 
SOR Mflops: 246.41 (100 x 100) 
MonteCarlo: Mflops: 42.34 
Sparse matmult Mflops: 405.80 (N=1000, nz=5000) 
LU Mflops: 656.41 (M=100, N=100)

    Avec gcc-3.2 

    moa@dionysos:~/tmprm$ ./scimark2 
** ** 
** SciMark2 Numeric Benchmark, see [http://math.nist.gov/scimark(...)](http://math.nist.gov/scimark) ** 
** for details. (Results can be submitted to pozo@nist.gov) ** 
** ** 
Using 2.00 seconds min time per kenel. 
Composite Score: 319.72 
FFT Mflops: 197.31 (N=1024) 
SOR Mflops: 238.48 (100 x 100) 
MonteCarlo: Mflops: 44.89 
Sparse matmult Mflops: 461.52 (N=1000, nz=5000) 
LU Mflops: 656.41 (M=100, N=100)

    Sur le P4 2Gh 

    <pseudo>@atlobk01:~/tmp$ ./scimark2 
** ** 
** SciMark2 Numeric Benchmark, see [http://math.nist.gov/scimark(...)](http://math.nist.gov/scimark) ** 
** for details. (Results can be submitted to pozo@nist.gov) ** 
** ** 
Using 2.00 seconds min time per kenel. 
Composite Score: 388.10 
FFT Mflops: 249.18 (N=1024) 
SOR Mflops: 302.98 (100 x 100) 
MonteCarlo: Mflops: 51.82 
Sparse matmult Mflops: 514.01 (N=1000, nz=5000) 
LU Mflops: 822.49 (M=100, N=100)

    et avec gcc 3.0 

    <pseudo>@atlobk01:~/tmp$ ./scimark2 
** ** 
** SciMark2 Numeric Benchmark, see [http://math.nist.gov/scimark(...)](http://math.nist.gov/scimark) ** 
** for details. (Results can be submitted to pozo@nist.gov) ** 
** ** 
Using 2.00 seconds min time per kenel. 
Composite Score: 417.00 
FFT Mflops: 242.93 (N=1024) 
SOR Mflops: 372.00 (100 x 100) 
MonteCarlo: Mflops: 51.82 
Sparse matmult Mflops: 595.78 (N=1000, nz=5000) 
LU Mflops: 822.49 (M=100, N=100)

    sur un biproc GenuineIntel Pentium III (Coppermine) 796.553

    avec gcc 2.95.4 

    ** ** 
** SciMark2 Numeric Benchmark, see [http://math.nist.gov/scimark(...)](http://math.nist.gov/scimark) ** 
** for details. (Results can be submitted to pozo@nist.gov) ** 
** ** 
Using 2.00 seconds min time per kenel. 
Composite Score: 174.09 
FFT Mflops: 118.82 (N=1024) 
SOR Mflops: 239.67 (100 x 100) 
MonteCarlo: Mflops: 33.22 
Sparse matmult Mflops: 206.74 (N=1000, nz=5000) 
LU Mflops: 271.98 (M=100, N=100)

    et avec gcc 3.0 

    <pseudo>@subversions:~/tmp&gt; ./scimark2 
** ** 
** SciMark2 Numeric Benchmark, see [http://math.nist.gov/scimark(...)](http://math.nist.gov/scimark) ** 
** for details. (Results can be submitted to pozo@nist.gov) ** 
** ** 
Using 2.00 seconds min time per kenel. 
Composite Score: 182.24 
FFT Mflops: 127.88 (N=1024) 
SOR Mflops: 267.63 (100 x 100) 
MonteCarlo: Mflops: 36.57 
Sparse matmult Mflops: 213.47 (N=1000, nz=5000) 
LU Mflops: 265.63 (M=100, N=100)

    finalement, sur un PII (Deschutes) 350 avec gcc 3.2

    (en fait gcc-3.2-base 1:3.2.1-0pre3) 

    Using 2.00 seconds min time per kenel. 
Composite Score: 54.91 
FFT Mflops: 47.60 (N=1024) 
SOR Mflops: 90.21 (100 x 100) 
MonteCarlo: Mflops: 13.75 
Sparse matmult Mflops: 60.24 (N=1000, nz=5000) 
LU Mflops: 62.75 (M=100, N=100)

    Beuh...