SPEC CPU result comparison

15th October 2026 - Normalized report

SPEC2017 FPrate: Standard -Ofast native - LLVM, AOCC and GCC

-Ofast -march=native

AMD EPYC 8635P (zen5 based Sorano)

Run-times

Benchmark AOCC GCC
503.bwaves_r 100.00% 77.87%
507.cactuBSSN_r 100.00% 127.72%
508.namd_r 100.00% 107.57%
510.parest_r 100.00% 87.91%
511.povray_r 100.00% 87.83%
519.lbm_r 100.00% 91.74%
521.wrf_r 100.00% 97.06%
526.blender_r 100.00% 131.37%
527.cam4_r 100.00% 85.51%
538.imagick_r 100.00% 53.03%
544.nab_r 100.00% 86.54%
549.fotonik3d_r 100.00% 100.67%
554.roms_r 100.00% 95.71%
LLVM clang version 23.1.3 (/labs/mjambor/git/llvm 0d261d1ca552c95a8f007e061c787ac7132fbcbc), flang version 23.1.3 (/labs/mjambor/git/llvm 0d261d1ca552c95a8f007e061c787ac7132fbcbc)
AOCC AMD clang version 17.0.6 (CLANG: AOCC_5.2.0-Build#2035 2026_04_10)
GCC GCC 17.0.0 20261015 (experimental) (revision 84348add236100e5f69a3e9ec4f7b1038bf2b83d)
In the case of LLVM -fno-finite-math-only was used to compile 511.povray_r (that benchmark only) to avoid a known issue.
Naturally, the smaller the run-times, the better.

Non-debug executable sizes

Benchmark AOCC GCC
503.bwaves_r 100.00% 72.08%
507.cactuBSSN_r 100.00% 79.90%
508.namd_r 100.00% 133.26%
510.parest_r 100.00% 122.70%
511.povray_r 100.00% 111.17%
519.lbm_r 100.00% 52.05%
521.wrf_r 100.00% 96.69%
526.blender_r 100.00% 86.72%
527.cam4_r 100.00% 122.63%
538.imagick_r 100.00% 133.04%
544.nab_r 100.00% 102.90%
549.fotonik3d_r 100.00% 98.93%
554.roms_r 100.00% 110.84%
Smaller binary sizes are always better.