Intel Core i7-4770K testing with a Intel DH87RL motherboard looking at the GCC 4.7, GCC 4.8, LLVM Clang 3.2, and LLVM Clang 3.3 compiler performance with core-avx2 Haswell optimizations. Intel Core i7 Haswell core-avx2 compiler benchmarks for a future article on Phoronix by Michael Larabel.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 1306206-SO-LLVMCLANG08
LLVM Clang 3.3 vs. GCC 4.8 - Intel Core-AVX2 Haswell,
"Timed HMMer Search 2.3.2 - Pfam Database Search",
Lower Results Are Better
"GCC 4.7.3",10.440405130386,10.37143611908,10.409691810608
"GCC 4.8.1",10.302880048752,10.29790687561,10.293302059174
"LLVM Clang 3.2",11.912589073181,11.929613113403,11.921445131302
"LLVM Clang 3.3",11.465604066849,10.738760948181,10.708938837051,10.863368034363
"Timed MAFFT Alignment 6.864 - Multiple Sequence Alignment",
Lower Results Are Better
"GCC 4.7.3",5.2620220184326,5.2378311157227,5.30047082901
"GCC 4.8.1",5.1383187770844,5.2347860336304,5.1375110149384
"LLVM Clang 3.2",5.8076939582825,6.1441459655762,5.9166049957275
"LLVM Clang 3.3",6.4378819465637,6.0353920459747,6.1004898548126
"BLAKE2 20121223 - Phoronix Test Suite v4.8.0m0",
Lower Results Are Better
"GCC 4.7.3",5.32,5.32,5.32
"GCC 4.8.1",5.29,5.31,5.3
"LLVM Clang 3.2",7.45,7.41,7.77
"LLVM Clang 3.3",7.79,7.42,7.19,7.38
"SciMark 2.0 - Computational Test: Monte Carlo",
Higher Results Are Better
"GCC 4.7.3",449.26,451.15,451.15,449.26
"GCC 4.8.1",615.32,615.32,615.32,615.32
"LLVM Clang 3.2",618.87,620.66,598.18,622.46
"LLVM Clang 3.3",620.66,617.09,620.66,620.66
"SciMark 2.0 - Computational Test: Fast Fourier Transform",
Higher Results Are Better
"GCC 4.7.3",252.4,250.19,250.92,253.15
"GCC 4.8.1",244.47,245.87,245.17,244.47
"LLVM Clang 3.2",245.17,243.08,249.46,249.46
"LLVM Clang 3.3",235.07,238.34,238.34,239.67
"SciMark 2.0 - Computational Test: Sparse Matrix Multiply",
Higher Results Are Better
"GCC 4.7.3",1180.4,1177.01,1177.01,1177.01
"GCC 4.8.1",1137.78,1122.19,1122.19,1113.04
"LLVM Clang 3.2",1144.13,1264.2,1268.11,1260.31
"LLVM Clang 3.3",1248.78,1272.05,1264.2,1268.11
"SciMark 2.0 - Computational Test: Dense LU Matrix Factorization",
Higher Results Are Better
"GCC 4.7.3",1864.8,1864.8,1858.3,1858.3
"GCC 4.8.1",1771.87,1777.78,1771.87,1771.87
"LLVM Clang 3.2",1698.51,1871.35,1754.39,1771.87
"LLVM Clang 3.3",1760.18,1845.44,1845.44,1858.3
"SciMark 2.0 - Computational Test: Jacobi Successive Over-Relaxation",
Higher Results Are Better
"GCC 4.7.3",1161.31,1161.31,1165.73,1165.73
"GCC 4.8.1",1161.31,1165.73,1165.73,1165.73
"LLVM Clang 3.2",1670.77,1670.77,1670.77,1670.77
"LLVM Clang 3.3",1670.77,1661.71,1666.23,1666.23
"TSCP 1.81 - AI Chess Performance",
Higher Results Are Better
"GCC 4.7.3",631626,631626,631626,631626,631626
"GCC 4.8.1",597373,599976,599976,599976,599976
"LLVM Clang 3.2",623757,624464,624464,624464,624464
"LLVM Clang 3.3",625173,624464,625884,625173,623052
"x264 2013-06-08 - H.264 Video Encoding",
Higher Results Are Better
"GCC 4.7.3",155.85,155.64,155.78,154.07,155.31
"GCC 4.8.1",156.18,155.86,155.39,157.74,156.54
"LLVM Clang 3.2",155.03,157.19,156.07,153.67,154.77
"LLVM Clang 3.3",151.67,153.91,154.39,153.97,151.79
"Himeno Benchmark 3.0 - Poisson Pressure Solver",
Higher Results Are Better
"GCC 4.7.3",1664.083187,1664.221912,1663.591591
"GCC 4.8.1",1552.322304,1616.320784,1610.635598
"LLVM Clang 3.2",1535.016107,1534.00483,1528.521124
"LLVM Clang 3.3",1394.778888,1442.362854,1422.548266
"Timed ImageMagick Compilation 6.8.1-10 - Time To Compile",
Lower Results Are Better
"GCC 4.7.3",93.280772924423,94.845242977142,92.897957086563
"GCC 4.8.1",78.569664955139,78.503214836121,78.743851184845
"LLVM Clang 3.2",31.805461883545,32.170036792755,31.842491865158
"LLVM Clang 3.3",33.714333057404,34.336052894592,34.986826896667
"Timed PHP Compilation 5.2.9 - Time To Compile",
Lower Results Are Better
"GCC 4.7.3",33.200973033905,32.719606161118,32.838427066803
"GCC 4.8.1",34.228738069534,32.764131069183,32.91902089119
"LLVM Clang 3.2",19.635462999344,19.543079137802,19.590476989746
"LLVM Clang 3.3",21.006337165833,20.687618970871,21.382838964462
"C-Ray 1.1 - Total Time",
Lower Results Are Better
"GCC 4.7.3",21.441,21.465,21.457
"GCC 4.8.1",17.008,17.029,17.022
"LLVM Clang 3.2",27.462,27.487,27.427
"LLVM Clang 3.3",27.022,27.031,27.038
"Primesieve 4.2 - 1e12 Prime Number Generation",
Lower Results Are Better
"GCC 4.7.3",79.469,79.302,79.305
"GCC 4.8.1",79.0811,79.3629,79.2619
"LLVM Clang 3.3",325.95,330.82,323.79
"Smallpt 1.0 - Global Illumination Renderer; 100 Samples",
Lower Results Are Better
"GCC 4.7.3",25,25,25
"GCC 4.8.1",24,25,25
"LLVM Clang 3.2",141,140,140
"LLVM Clang 3.3",141,145,141
"LAME MP3 Encoding 3.99.3 - WAV To MP3",
Lower Results Are Better
"LLVM Clang 3.2",12.734658002853,12.750730037689,12.722330093384,12.73823595047,12.757515192032
"LLVM Clang 3.3",14.457759141922,14.482392072678,14.441653966904,14.443991184235,14.438816070557
"FFmpeg 1.1 - H.264 HD To NTSC DV",
Lower Results Are Better
"GCC 4.7.3",12.873476028442,12.992598056793,12.711709022522
"GCC 4.8.1",12.74954199791,12.928920030594,12.74736905098
"LLVM Clang 3.2",12.69673705101,12.799638986588,12.804874181747
"LLVM Clang 3.3",13.416246891022,13.131498098373,12.998423099518
"Tachyon 0.98.9 - Total Time",
Lower Results Are Better
"LLVM Clang 3.2",10.3621,10.562,10.4007
"LLVM Clang 3.3",12.0577,10.7334,11.2167,10.4765,10.7153,10.7059
"Apache Benchmark 2.4.3 - Static Web Page Serving",
Higher Results Are Better
"GCC 4.7.3",25804.39,25641.04,25786.54
"GCC 4.8.1",25724.78,25850.49,25783.17
"LLVM Clang 3.2",25886.14,25834.28,25946.44
"LLVM Clang 3.3",25852.31,25082.74,24952.41