Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.lang.pascal.misc > #869
| From | aminer <aminer@toto.net> |
|---|---|
| Newsgroups | comp.lang.pascal.misc |
| Subject | Here is the new scimark2 benchmarks: |
| Date | 2014-06-07 05:48 -0700 |
| Organization | albasani.net |
| Message-ID | <ln01ac$c2s$3@news.albasani.net> (permalink) |
Hello, Here is the new scimark2 benchmarks: For 64-bit fpc 2.7.1 we have: Composite Score MFlops: 567.02 FFT Mflops: 340.34 (N=1024) SOR Mflops: 717.01 (100 x 100) MonteCarlo: Mflops: 113.77 Sparse matmult Mflops: 811.47 (N=1000, nz=5000) LU Mflops: 852.50 (M=100, N=100) For 64-bit TDM GCC 4.8.1 we have: Composite Score: 750.46 FFT Mflops: 664.98 (N=1024) SOR Mflops: 723.33 (100 x 100) MonteCarlo: Mflops: 225.10 Sparse matmult Mflops: 978.88 (N=1000, nz=5000) LU Mflops: 1160.01 (M=100, N=100) As you have noticed the GCC is only around 20% faster than FPC 2.7.1 on the Sparse matmult and LU and it has the same speed as FPC on the SOR, that's cool, and i think FPC will get more optimized in the near future, so i am not worry about the performance of FreePascal cause i am sure that FreePascal will get much better. You can download fpc 2.7.1 from here: ftp://ftp.freepascal.org/pub/fpc/snapshot/trunk/ Thank you, Amine Moulay Ramdane.
Back to comp.lang.pascal.misc | Previous | Next | Find similar
Here is the new scimark2 benchmarks: aminer <aminer@toto.net> - 2014-06-07 05:48 -0700
csiph-web