GCC 9 compiler tuning benchmarks by Michael Larabel for a future article on Phoronix.com.
-O0 Environment Notes: CXXFLAGS=-O0 CFLAGS=-O0Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-Og Environment Notes: CXXFLAGS=-Og CFLAGS=-OgCompiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O1 Environment Notes: CXXFLAGS=-O1 CFLAGS=-O1Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O2 Environment Notes: CXXFLAGS=-O2 CFLAGS=-O2Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O2 -ftree-vectorize -ftree-slp-vectorize Environment Notes: CXXFLAGS=-O2-ftree-vectorize-ftree-slp-vectorize CFLAGS=-O2-ftree-vectorize-ftree-slp-vectorizeCompiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O2 -march=znver1 Environment Notes: CXXFLAGS=-O2-march=znver1 CFLAGS=-O2-march=znver1Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O2 -flto Environment Notes: CXXFLAGS=-O2-flto CFLAGS=-O2-fltoCompiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O3 Environment Notes: CXXFLAGS=-O3 CFLAGS=-O3Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O3 -march=znver1 Environment Notes: CXXFLAGS=-O3-march=znver1 CFLAGS=-O3-march=znver1Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O3 -march=znver1 -flto Environment Notes: CXXFLAGS=-O3 march=znver1-flto CFLAGS=-O3-march=znver1-fltoCompiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-Ofast -march=znver1 Processor: 2 x AMD EPYC 7601 32-Core (64 Cores / 128 Threads), Motherboard: Dell 02MJ3T (1.2.5 BIOS), Chipset: AMD Family 17h, Memory: 16 x 32 GB DDR4-2400MT/s 36ASF4G72PZ-2G6D2, Disk: 120GB SSDSCKJB120G7R + 20 x 500GB Samsung SSD 860, Graphics: Matrox G200eW3, Monitor: VE228, Network: 2 x Broadcom BCM57416 NetXtreme-E 10GBase-T RDMA + 2 x Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 18.04, Kernel: 5.0.0-050000rc6-generic (x86_64) 20190210, Desktop: GNOME Shell 3.28.3, Display Server: X Server, Compiler: GCC 9.0.1 20190210, File-System: ext4, Screen Resolution: 1600x1200
Environment Notes: CXXFLAGS=-Ofast-march=znver1 CFLAGS=-Ofast-march=znver1Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
FLAC Audio Encoding This test times how long it takes to encode a sample WAV file to FLAC format five times. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 20 40 60 80 100 SE +/- 0.12, N = 5 SE +/- 0.11, N = 5 SE +/- 0.09, N = 5 SE +/- 0.08, N = 5 SE +/- 0.14, N = 5 SE +/- 0.09, N = 5 SE +/- 0.11, N = 5 SE +/- 0.12, N = 5 SE +/- 0.10, N = 5 SE +/- 0.10, N = 5 SE +/- 0.08, N = 5 96.77 15.58 15.01 13.65 13.70 13.89 13.64 13.61 13.85 14.21 13.95 -O0 1. (CXX) g++ options: -fvisibility=hidden -logg -lm
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 3K 6K 9K 12K 15K SE +/- 1.00, N = 3 SE +/- 165.58, N = 3 SE +/- 70.29, N = 3 SE +/- 160.95, N = 3 SE +/- 72.78, N = 3 SE +/- 15.71, N = 3 SE +/- 134.39, N = 3 SE +/- 115.66, N = 3 SE +/- 78.62, N = 3 SE +/- 49.65, N = 3 SE +/- 160.71, N = 8 2193 12642 13468 13391 13285 13346 13214 13555 12752 13110 13166 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -pthread -lm
Timed PHP Compilation This test times how long it takes to build PHP 5 with the Zend engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 7.1.9 Time To Compile -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O3 -O3 -march=znver1 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 SE +/- 0.18, N = 3 SE +/- 0.26, N = 3 SE +/- 0.25, N = 3 SE +/- 0.22, N = 3 SE +/- 0.33, N = 3 15.19 21.42 29.05 52.17 52.58 51.96 78.19 78.13 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O3 -O3 -march=znver1 1. (CC) gcc options: -pedantic -ldl -lz -lm
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 600 1200 1800 2400 3000 SE +/- 5.25, N = 3 SE +/- 3.66, N = 3 SE +/- 59.22, N = 3 SE +/- 11.61, N = 3 SE +/- 12.19, N = 3 SE +/- 14.53, N = 3 SE +/- 5.41, N = 3 SE +/- 3.37, N = 3 SE +/- 10.26, N = 3 SE +/- 2.13, N = 3 SE +/- 12.59, N = 3 516 2188 2411 2527 2515 2584 2299 2475 2482 2052 2579 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 400 800 1200 1600 2000 SE +/- 5.30, N = 3 SE +/- 18.65, N = 5 SE +/- 6.18, N = 3 SE +/- 12.17, N = 3 SE +/- 8.94, N = 3 SE +/- 23.45, N = 5 SE +/- 24.31, N = 3 SE +/- 7.96, N = 3 SE +/- 11.89, N = 3 SE +/- 35.09, N = 3 SE +/- 20.59, N = 3 434 1205 1519 1369 1724 1501 1307 1800 1961 1747 1825 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm
C-Ray This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 10 20 30 40 50 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 44.92 28.64 28.74 25.84 25.77 21.58 25.96 12.60 11.35 11.31 10.40 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm -lpthread -O3
LAME MP3 Encoding LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 10 20 30 40 50 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 41.79 16.78 14.32 14.07 10.96 14.00 14.14 10.84 10.57 10.38 9.80 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1200 2400 3600 4800 6000 SE +/- 5.17, N = 3 SE +/- 11.16, N = 3 SE +/- 6.37, N = 3 SE +/- 2.28, N = 3 SE +/- 26.82, N = 3 SE +/- 13.33, N = 3 SE +/- 5.87, N = 3 SE +/- 41.07, N = 3 SE +/- 47.88, N = 3 SE +/- 26.34, N = 3 SE +/- 10.24, N = 3 1708 4366 4632 4625 4805 5074 5091 4751 5006 5571 4885 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -pthread -lm
Timed ImageMagick Compilation This test times how long it takes to build ImageMagick. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 30 60 90 120 150 SE +/- 0.10, N = 3 SE +/- 0.04, N = 3 SE +/- 0.21, N = 8 SE +/- 0.10, N = 3 SE +/- 0.20, N = 3 SE +/- 0.45, N = 3 SE +/- 0.98, N = 3 SE +/- 0.44, N = 3 SE +/- 0.30, N = 3 SE +/- 0.45, N = 3 SE +/- 0.34, N = 3 5.23 7.89 18.42 23.63 23.91 23.78 98.67 25.06 24.88 118.48 25.21
Himeno Benchmark The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 200 400 600 800 1000 SE +/- 0.11, N = 3 SE +/- 4.02, N = 3 SE +/- 5.78, N = 3 SE +/- 5.25, N = 3 SE +/- 7.21, N = 3 SE +/- 2.90, N = 3 SE +/- 2.58, N = 3 SE +/- 6.27, N = 3 SE +/- 0.08, N = 3 SE +/- 2.81, N = 3 SE +/- 8.54, N = 3 383 772 785 1017 1007 1001 1022 1008 1011 1000 1022 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -O3 -mavx2
Timed Apache Compilation This test times how long it takes to build the Apache HTTP Server. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed Apache Compilation 2.4.7 Time To Compile -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 7 14 21 28 35 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 11.43 14.59 17.51 23.82 24.03 23.82 26.50 26.08 25.94 28.62 26.11
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Sharpen -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 40 80 120 160 200 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 82 156 180 181 180 183 183 174 183 183 182 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Enhanced -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 40 80 120 160 200 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 3.18, N = 3 90 173 187 189 188 191 190 181 191 186 193 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: HWB Color Space -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 50 100 150 200 250 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 SE +/- 0.33, N = 3 102 195 210 211 212 211 214 203 210 209 209 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Swirl -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 40 80 120 160 200 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.67, N = 3 96 181 194 195 196 196 196 189 195 194 196 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Noise-Gaussian -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 40 80 120 160 200 SE +/- 1.20, N = 3 SE +/- 0.88, N = 3 SE +/- 0.88, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 92 168 179 180 178 180 180 172 180 178 187 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 400 800 1200 1600 2000 SE +/- 0.35, N = 3 SE +/- 0.31, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 SE +/- 0.15, N = 3 SE +/- 0.21, N = 3 SE +/- 0.10, N = 3 SE +/- 0.12, N = 3 832 919 919 919 919 1016 918 1427 1689 1675 1676 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 300 600 900 1200 1500 SE +/- 0.03, N = 3 SE +/- 0.28, N = 3 SE +/- 0.27, N = 3 SE +/- 0.03, N = 3 SE +/- 0.22, N = 3 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 SE +/- 0.33, N = 3 SE +/- 0.06, N = 3 108 210 576 560 560 557 568 560 557 1480 561 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Rotate -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 40 80 120 160 200 SE +/- 0.33, N = 3 98 181 191 191 190 191 191 183 190 188 189 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
AOBench AOBench is a lightweight ambient occlusion renderer, written in C. The test profile is using a size of 2048 x 2048. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 20 40 60 80 100 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 92.50 77.41 56.61 55.54 55.53 54.35 55.52 53.53 51.49 52.08 51.73 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm -O3
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Resizing -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 30 60 90 120 150 SE +/- 1.40, N = 8 SE +/- 1.43, N = 12 SE +/- 2.52, N = 3 SE +/- 1.53, N = 3 SE +/- 1.50, N = 12 SE +/- 1.20, N = 3 SE +/- 1.94, N = 5 SE +/- 1.50, N = 8 SE +/- 1.32, N = 10 74 120 126 131 128 127 128 118 127 125 124 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
PostgreSQL pgbench This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Only -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 3K 6K 9K 12K 15K SE +/- 69.77, N = 3 SE +/- 149.12, N = 3 SE +/- 119.98, N = 3 SE +/- 48.94, N = 3 SE +/- 295.94, N = 3 SE +/- 122.77, N = 3 SE +/- 172.87, N = 3 SE +/- 32.93, N = 3 SE +/- 101.16, N = 3 SE +/- 224.15, N = 6 SE +/- 125.79, N = 3 9063 13333 13303 14931 15353 15111 14851 15099 15188 16012 15352 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
Timed HMMer Search This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 2.3.2 Pfam Database Search -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 3 6 9 12 15 SE +/- 0.14, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 9.02 7.39 6.93 6.62 6.82 6.54 6.56 6.57 6.29 6.16 6.00 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -pthread -lhmmer -lsquid -lm
x264 This is a simple test of the x264 encoder run on the CPU (OpenCL support disabled) with a sample video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better x264 2018-09-25 H.264 Video Encoding -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O3 -O3 -march=znver1 -Ofast -march=znver1 30 60 90 120 150 SE +/- 0.12, N = 3 SE +/- 1.09, N = 3 SE +/- 1.49, N = 3 SE +/- 0.97, N = 3 SE +/- 0.47, N = 3 SE +/- 1.40, N = 3 SE +/- 0.81, N = 3 SE +/- 1.78, N = 3 SE +/- 0.52, N = 3 102 142 145 144 144 144 147 144 144 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -march=znver1 -Ofast -march=znver1 1. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
PostgreSQL pgbench This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1100 2200 3300 4400 5500 SE +/- 29.67, N = 3 SE +/- 47.02, N = 3 SE +/- 71.59, N = 9 SE +/- 49.50, N = 9 SE +/- 55.61, N = 6 SE +/- 12.79, N = 3 SE +/- 26.99, N = 3 SE +/- 29.58, N = 3 SE +/- 50.49, N = 3 SE +/- 66.20, N = 5 SE +/- 66.04, N = 9 4585 3767 4301 4167 4239 4272 4095 4262 5068 4319 4102 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
libjpeg-turbo tjbench tjbench is a JPEG decompression/compression benchmark part of libjpeg-turbo. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 1.5.3 Test: Decompression Throughput -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 30 60 90 120 150 SE +/- 0.76, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.67, N = 3 SE +/- 0.06, N = 3 SE +/- 0.91, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.07, N = 3 111 141 139 140 139 142 140 141 144 144 144 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm
PostgreSQL pgbench This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Write -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 200 400 600 800 1000 SE +/- 6.16, N = 3 SE +/- 4.15, N = 3 SE +/- 8.54, N = 3 SE +/- 17.89, N = 3 SE +/- 1.22, N = 3 SE +/- 12.77, N = 9 SE +/- 3.41, N = 3 SE +/- 12.77, N = 8 SE +/- 19.20, N = 3 SE +/- 9.80, N = 3 SE +/- 19.14, N = 3 886 1080 1065 1037 1125 1060 1127 1079 1145 1074 1125 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 60 120 180 240 300 SE +/- 1.00, N = 3 SE +/- 2.03, N = 3 SE +/- 0.44, N = 3 SE +/- 0.61, N = 3 SE +/- 0.71, N = 3 SE +/- 1.88, N = 3 SE +/- 2.57, N = 3 SE +/- 0.51, N = 3 SE +/- 0.03, N = 3 SE +/- 1.03, N = 3 SE +/- 1.04, N = 3 201 257 226 230 231 229 232 232 227 230 221 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm
PostgreSQL pgbench This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 110K 220K 330K 440K 550K SE +/- 4794.85, N = 3 SE +/- 3395.99, N = 3 SE +/- 6061.06, N = 3 SE +/- 2768.55, N = 3 SE +/- 3875.50, N = 3 SE +/- 5952.34, N = 3 SE +/- 4765.38, N = 3 SE +/- 8068.95, N = 9 SE +/- 3819.62, N = 3 SE +/- 7546.41, N = 4 SE +/- 1629.04, N = 3 419700 507203 515102 515340 529699 510425 520570 490551 505031 454256 508384 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.8.0-jumbo-1 Test: Traditional DES -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 60M 120M 180M 240M 300M SE +/- 2756947.41, N = 3 SE +/- 2445677.03, N = 3 SE +/- 2774527.11, N = 10 SE +/- 2178423.16, N = 3 SE +/- 2839112.24, N = 3 SE +/- 2041895.52, N = 3 SE +/- 642920.77, N = 3 SE +/- 3859011.69, N = 12 SE +/- 2346357.84, N = 3 SE +/- 1374420.40, N = 3 SE +/- 1656338.97, N = 3 218232000 239289333 257067200 257407667 257058000 255957000 260736667 253868583 260019667 254777333 258770667 1. (CC) gcc options: -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Stack -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -Ofast -march=znver1 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 6.00 6.02 6.00 6.01 5.98 5.80 6.32 6.05 5.80 5.80 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -Ofast -march=znver1 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Hierarchical INTegration This test runs the U.S. Department of Energy's Ames Laboratory Hierarchical INTegration (HINT) benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: DOUBLE -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 130M 260M 390M 520M 650M SE +/- 8177784.85, N = 3 SE +/- 9099115.17, N = 9 SE +/- 1617493.63, N = 3 SE +/- 6546514.72, N = 3 SE +/- 2338894.48, N = 3 SE +/- 9814749.06, N = 4 SE +/- 10585229.89, N = 3 SE +/- 1832535.45, N = 3 SE +/- 7042504.19, N = 3 SE +/- 7234705.24, N = 9 SE +/- 7419115.18, N = 3 598545342 597234266 585060029 599481605 602535297 617516626 626640400 595428047 589289926 618644101 605331833 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -O3 -march=native -lm
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 136 Ragdolls -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -Ofast -march=znver1 0.729 1.458 2.187 2.916 3.645 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 3.14 3.14 3.15 3.14 3.15 3.05 3.24 3.15 3.06 3.06 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -Ofast -march=znver1 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 2019-02-17 1080p 8-bit YUV To VP9 Video Encode -Og -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -march=znver1 -flto -Ofast -march=znver1 20 40 60 80 100 SE +/- 1.17, N = 3 SE +/- 1.09, N = 3 SE +/- 0.18, N = 3 SE +/- 0.75, N = 3 SE +/- 0.50, N = 3 SE +/- 0.30, N = 3 92.68 94.82 95.91 95.79 97.26 97.80 -Og -ftree-vectorize -ftree-slp-vectorize -march=native -O3 -march=znver1 -Ofast -march=znver1 1. (CC) gcc options: -fPIE -fPIC -O2 -flto -fvisibility=hidden -mavx -pie -rdynamic -lpthread -lrt -lm
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Convex -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -Ofast -march=znver1 1.215 2.43 3.645 4.86 6.075 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 5.36 5.37 5.37 5.37 5.37 5.19 5.40 5.38 5.18 5.18 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -Ofast -march=znver1 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
VP9 libvpx Encoding This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9/WebM format using a sample 1080p video. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.0 vpxenc VP9 1080p Video Encode -Og -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 20.39 20.34 20.05 20.86 20.13 -Og -std=c++11 -O2 -ftree-vectorize -ftree-slp-vectorize -std=c++11 -O2 -march=native -std=c++11 -march=znver1 -flto -Ofast -march=znver1 -std=c++11 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE
SVT-AV1 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2019-02-03 1080p 8-bit YUV To AV1 Video Encode -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 0.3893 0.7786 1.1679 1.5572 1.9465 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 9 SE +/- 0.02, N = 8 SE +/- 0.01, N = 3 SE +/- 0.02, N = 6 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 1.69 1.73 1.70 1.69 1.67 1.70 1.69 1.73 1.68 1.71 1.70 1. (CC) gcc options: -mavx2 -fPIE -fPIC -O2 -pie -lpthread -lm
VP9 libvpx Encoding This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9/WebM format using a sample 1080p video. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.0 vpxenc VP9 1080p Video Encode -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.18, N = 5 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 12.50 12.52 12.53 12.54 12.56 12.42 12.31 12.41 12.75 12.37 -O0 -std=c++11 -Og -std=c++11 -O1 -std=c++11 -O2 -std=c++11 -O2 -ftree-vectorize -ftree-slp-vectorize -std=c++11 -O2 -march=znver1 -std=c++11 -std=c++11 -march=znver1 -std=c++11 -march=znver1 -flto -Ofast -march=znver1 -std=c++11 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE
x265 This is a simple test of the x265 encoder run on the CPU with a sample 1080p video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better x265 3.0 H.265 1080p Video Encoding -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -Ofast -march=znver1 8 16 24 32 40 SE +/- 0.67, N = 3 SE +/- 0.58, N = 3 SE +/- 0.37, N = 11 SE +/- 0.24, N = 3 SE +/- 0.46, N = 3 SE +/- 0.09, N = 3 SE +/- 0.38, N = 3 SE +/- 0.41, N = 3 SE +/- 0.18, N = 3 SE +/- 0.01, N = 3 35.00 34.76 35.62 34.55 35.41 34.80 35.07 35.21 35.57 34.91 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -Ofast -march=znver1 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 3000 Fall -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -Ofast -march=znver1 1.1723 2.3446 3.5169 4.6892 5.8615 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 5.14 5.15 5.16 5.14 5.14 5.08 5.21 5.16 5.07 5.09 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -Ofast -march=znver1 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Raytests -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -Ofast -march=znver1 0.702 1.404 2.106 2.808 3.51 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 3.11 3.11 3.12 3.12 3.11 3.10 3.05 3.11 3.09 3.09 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -Ofast -march=znver1 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Stockfish This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -Ofast -march=znver1 20M 40M 60M 80M 100M SE +/- 1595135.10, N = 3 SE +/- 823190.91, N = 3 SE +/- 549524.55, N = 3 SE +/- 1016511.53, N = 3 SE +/- 468403.16, N = 3 SE +/- 324013.38, N = 3 SE +/- 579693.09, N = 3 SE +/- 673773.08, N = 3 SE +/- 402849.09, N = 3 SE +/- 460638.54, N = 3 105868175 105709690 105698092 104480422 104197865 106084276 104536605 104121840 106497994 106507244 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -march=znver1 -Ofast -march=znver1 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -msse -msse3 -mpopcnt -flto
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Convex Trimesh -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -Ofast -march=znver1 0.3038 0.6076 0.9114 1.2152 1.519 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.35 1.35 1.35 1.35 1.35 1.32 1.33 1.35 1.32 1.32 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -Ofast -march=znver1 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Prim Trimesh -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -Ofast -march=znver1 0.2498 0.4996 0.7494 0.9992 1.249 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.11 1.11 1.11 1.11 1.11 1.11 1.09 1.11 1.11 1.11 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -Ofast -march=znver1 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
SVT-AV1 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2019-02-15 1080p 8-bit YUV To AV1 Video Encode -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1.3298 2.6596 3.9894 5.3192 6.649 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 5.88 5.86 5.87 5.81 5.89 5.84 5.90 5.90 5.89 5.84 5.91 1. (CC) gcc options: -mavx -fPIE -fPIC -O2 -pie -lpthread -lm
Hierarchical INTegration This test runs the U.S. Department of Energy's Ames Laboratory Hierarchical INTegration (HINT) benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 60M 120M 180M 240M 300M SE +/- 321625.93, N = 3 SE +/- 144731.32, N = 3 SE +/- 1109543.66, N = 3 SE +/- 232284.24, N = 3 SE +/- 54052.05, N = 3 SE +/- 211192.91, N = 3 SE +/- 1057235.60, N = 3 SE +/- 219028.41, N = 3 SE +/- 1208988.13, N = 3 SE +/- 67545.97, N = 3 SE +/- 193963.83, N = 3 267404445 267368671 268455578 267311970 267172145 267268023 268173400 267315647 268506472 267239405 267055407 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -O3 -march=native -lm
TSCP This is a performance test of TSCP, Tom Kerrigan's Simple Chess Program, which has a built-in performance benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 200K 400K 600K 800K 1000K SE +/- 333.13, N = 5 SE +/- 333.13, N = 5 SE +/- 542.88, N = 5 SE +/- 507.80, N = 5 SE +/- 508.06, N = 5 SE +/- 272.00, N = 5 SE +/- 331.91, N = 5 SE +/- 272.00, N = 5 SE +/- 667.00, N = 5 SE +/- 270.20, N = 5 SE +/- 507.80, N = 5 865459 865187 864102 864373 864916 864915 864101 864915 865732 863018 864373 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -march=znver1 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -O3 -march=native
ctx_clock Ctx_clock is a simple test program to measure the context switch time in clock cycles. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Clocks, Fewer Is Better ctx_clock Context Switch Time -Og -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -flto -O3 -march=znver1 -flto 30 60 90 120 150 132 132 132 132 -Og -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -flto -O3 -march=znver1 -flto 1. (CC) gcc options:
Zstd Compression This test measures the time needed to compress a sample file (an Ubuntu file-system image) using Zstd compression. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 6 12 18 24 30 SE +/- 0.38, N = 4 SE +/- 0.29, N = 12 SE +/- 0.25, N = 12 SE +/- 0.49, N = 9 SE +/- 0.33, N = 12 SE +/- 0.35, N = 12 SE +/- 0.31, N = 12 SE +/- 0.38, N = 12 SE +/- 0.44, N = 12 SE +/- 0.24, N = 11 SE +/- 0.21, N = 12 23.12 14.39 14.11 14.48 13.67 14.71 14.08 13.66 14.37 13.16 13.77 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -pthread -lz -llzma
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.8.0-jumbo-1 Test: Blowfish -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 14K 28K 42K 56K 70K SE +/- 215.16, N = 12 SE +/- 1339.16, N = 9 SE +/- 1098.57, N = 4 SE +/- 1387.83, N = 12 SE +/- 1953.49, N = 12 SE +/- 1967.27, N = 12 SE +/- 1395.50, N = 12 SE +/- 1049.43, N = 3 SE +/- 1082.96, N = 12 SE +/- 1598.30, N = 12 SE +/- 1454.31, N = 11 15179 56453 65995 62718 63586 61309 65117 65806 66823 58764 62841 1. (CC) gcc options: -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1000 2000 3000 4000 5000 SE +/- 30.37, N = 3 SE +/- 139.26, N = 3 SE +/- 32.13, N = 3 SE +/- 54.41, N = 3 SE +/- 57.07, N = 3 SE +/- 173.68, N = 3 SE +/- 129.41, N = 3 SE +/- 42.05, N = 3 SE +/- 65.35, N = 3 SE +/- 178.79, N = 3 SE +/- 107.53, N = 3 512 2539 3466 2609 4396 3231 2515 4307 4851 3300 4089 -O0 -Og -O1 -O2 -O2 -ftree-vectorize -ftree-slp-vectorize -O2 -march=znver1 -O2 -flto -O3 -O3 -march=znver1 -O3 -march=znver1 -flto -Ofast -march=znver1 1. (CC) gcc options: -lm
-O0 Environment Notes: CXXFLAGS=-O0 CFLAGS=-O0Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 15 February 2019 20:24 by user root.
-Og Environment Notes: CXXFLAGS=-Og CFLAGS=-OgCompiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 19 February 2019 05:54 by user root.
-O1 Environment Notes: CXXFLAGS=-O1 CFLAGS=-O1Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 16 February 2019 07:27 by user root.
-O2 Environment Notes: CXXFLAGS=-O2 CFLAGS=-O2Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 16 February 2019 15:25 by user root.
-O2 -ftree-vectorize -ftree-slp-vectorize Environment Notes: CXXFLAGS=-O2-ftree-vectorize-ftree-slp-vectorize CFLAGS=-O2-ftree-vectorize-ftree-slp-vectorizeCompiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 18 February 2019 05:50 by user root.
-O2 -march=znver1 Environment Notes: CXXFLAGS=-O2-march=znver1 CFLAGS=-O2-march=znver1Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 17 February 2019 07:16 by user root.
-O2 -flto Environment Notes: CXXFLAGS=-O2-flto CFLAGS=-O2-fltoCompiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 18 February 2019 20:32 by user root.
-O3 Environment Notes: CXXFLAGS=-O3 CFLAGS=-O3Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 16 February 2019 21:58 by user root.
-O3 -march=znver1 Environment Notes: CXXFLAGS=-O3-march=znver1 CFLAGS=-O3-march=znver1Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 15 February 2019 13:07 by user root.
-O3 -march=znver1 -flto Environment Notes: CXXFLAGS=-O3 march=znver1-flto CFLAGS=-O3-march=znver1-fltoCompiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 18 February 2019 12:40 by user root.
-Ofast -march=znver1 Processor: 2 x AMD EPYC 7601 32-Core (64 Cores / 128 Threads), Motherboard: Dell 02MJ3T (1.2.5 BIOS), Chipset: AMD Family 17h, Memory: 16 x 32 GB DDR4-2400MT/s 36ASF4G72PZ-2G6D2, Disk: 120GB SSDSCKJB120G7R + 20 x 500GB Samsung SSD 860, Graphics: Matrox G200eW3, Monitor: VE228, Network: 2 x Broadcom BCM57416 NetXtreme-E 10GBase-T RDMA + 2 x Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 18.04, Kernel: 5.0.0-050000rc6-generic (x86_64) 20190210, Desktop: GNOME Shell 3.28.3, Display Server: X Server, Compiler: GCC 9.0.1 20190210, File-System: ext4, Screen Resolution: 1600x1200
Environment Notes: CXXFLAGS=-Ofast-march=znver1 CFLAGS=-Ofast-march=znver1Compiler Notes: --disable-multilib --enable-checking=releaseSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 17 February 2019 14:18 by user root.