GCC AMD Znver3 Compiler Optimization Levels

Benchmarks for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2012160-PTS-GCCZNVER86&export=pdf&grr&sor&rro.

GCC AMD Znver3 Compiler Optimization LevelsProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen Resolution-O2 -march=x86-64-O3 -march=x86-64-O3 -march=znver2-O2 -march=znver3-O3 -march=znver3-O2 -march=znver3 -flto-O3 -maech=znver3 -flto-Ofast -march=znver3 -fltoAMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3003 BIOS)AMD Starship/Matisse16GB2000GB Corsair Force MP600AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (2100/875MHz)AMD Navi 10 HDMI AudioASUS MG28URealtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 20.045.10.0-051000rc6daily20201205-generic (x86_64) 20201204GNOME Shell 3.36.4X Server 1.20.8modesetting 1.20.84.6 Mesa 21.0.0-devel (git-1a53572 2020-12-09 focal-oibaf-ppa) (LLVM 11.0.0)1.2.145GCC 11.0.0 20201213ext43840x2160OpenBenchmarking.orgEnvironment Details- -O2 -march=x86-64: CXXFLAGS="-O2 -march=x86-64" CFLAGS="-O2 -march=x86-64"- -O3 -march=x86-64: CXXFLAGS="-O3 -march=x86-64" CFLAGS="-O3 -march=x86-64"- -O3 -march=znver2: CXXFLAGS="-O3 -march=znver2" CFLAGS="-O3 -march=znver2"- -O2 -march=znver3: CXXFLAGS="-O2 -march=znver3" CFLAGS="-O2 -march=znver3"- -O3 -march=znver3: CXXFLAGS="-O3 -march=znver3" CFLAGS="-O3 -march=znver3"- -O2 -march=znver3 -flto: CXXFLAGS="-O2 -march=znver3 -flto" CFLAGS="-O2 -march=znver3 -flto"- -O3 -maech=znver3 -flto: CXXFLAGS="-O3 -march=znver3 -flto" CFLAGS="-O3 -march=znver3 -flto"- -Ofast -march=znver3 -flto: CXXFLAGS="-Ofast -march=znver3 -flto" CFLAGS="-Ofast -march=znver3 -flto"Compiler Details- --disable-multilib --enable-checking=releaseProcessor Details- Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected Disk Details- -O3 -march=znver3: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Python Details- -O3 -march=znver3: Python 2.7.18 + Python 3.8.5

GCC AMD Znver3 Compiler Optimization Levelsfftw: Stock - 2D FFT Size 4096mrbayes: Primate Phylogeny Analysishmmer: Pfam Database Searchgraphics-magick: Noise-Gaussianaobench: 2048 x 2048 - Total Timegraphics-magick: Enhancedgraphics-magick: Sharpengraphics-magick: Resizingscimark2: Compositemt-dgemm: Sustained Floating-Point Ratebuild-imagemagick: Time To Compilekvazaar: Bosphorus 4K - Slowkvazaar: Bosphorus 4K - Mediumc-ray: Total Time - 4K, 16 Rays Per Pixelcoremark: CoreMark Size 666 - Iterations Per Secondlibraw: Post-Processing Benchmarkdaphne: OpenMP - NDT Mappingkvazaar: Bosphorus 4K - Very Fastaom-av1: Speed 8 Realtimekvazaar: Bosphorus 1080p - Slowkvazaar: Bosphorus 1080p - Mediumkvazaar: Bosphorus 4K - Ultra Fastfftw: Stock - 1D FFT Size 4096kvazaar: Bosphorus 1080p - Very Fastsmallpt: Global Illumination Renderer; 128 Sampleskvazaar: Bosphorus 1080p - Ultra Fast-O2 -march=x86-64-O3 -march=x86-64-O3 -march=znver2-O2 -march=znver3-O3 -march=znver3-O2 -march=znver3 -flto-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto8352.879.34185.78842935.03541922416052883.787.52656716.39811.4811.6860.896700520.67320151.45921.7129.5560.8845.0946.3253.1312892108.724.829194.528169.277.07485.59642629.99138222517213660.746.48455719.00011.6911.8631.364717480.23761363.90931.0330.4860.0745.7546.7454.4312436111.775.068198.848927.175.69784.79143425.01644437719074407.327.51218919.37011.5911.7225.251721620.06751075.26964.7230.3462.2844.9746.3854.7514096111.544.677199.358823.779.54385.12645230.71143035921502625.597.39174215.80411.5211.7346.720669924.92186150.95966.9929.8762.1044.8946.3153.8313771109.424.567195.139314.176.31785.02044125.06543236121353958.177.30709318.85911.5811.7425.254716026.62027173.83971.3130.3162.3344.9146.4254.7313702111.504.673200.129718.279.51384.92944527.36347637718542924.237.34563850.40711.6811.8446.419681281.70624249.32970.4630.7662.0845.6847.0454.9115232113.414.607201.851012573.91984.11943025.84844537019063855.307.69608963.47611.8812.0325.480736850.43363973.73971.4531.4462.5246.3047.3256.1215584116.074.656203.898755.945.27476.94350424.50544932819014101.897.75895266.84312.1212.2727.342728279.79221175.23969.9132.4063.7647.2548.5458.6912149.0121.744.602220.99OpenBenchmarking.org

FFTW

Build: Stock - Size: 2D FFT Size 4096

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 4096-O3 -march=x86-64-O2 -march=x86-64-Ofast -march=znver3 -flto-O2 -march=znver3-O3 -march=znver2-O3 -march=znver3-O2 -march=znver3 -flto-O3 -maech=znver3 -flto2K4K6K8K10KSE +/- 312.57, N = 12SE +/- 88.61, N = 3SE +/- 86.92, N = 3SE +/- 39.49, N = 3SE +/- 15.97, N = 3SE +/- 18.57, N = 3SE +/- 96.35, N = 3SE +/- 25.21, N = 38169.28352.88755.98823.78927.19314.19718.210125.0-Ofast -march=znver3 -flto-O2 -march=znver3-O3 -march=znver2-O3 -march=znver3-O2 -march=znver3 -flto-O3 -march=znver3 -flto1. (CC) gcc options: -pthread -lm

Timed MrBayes Analysis

Primate Phylogeny Analysis

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny Analysis-O2 -march=znver3-O2 -march=znver3 -flto-O2 -march=x86-64-O3 -march=x86-64-O3 -march=znver3-O3 -march=znver2-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto20406080100SE +/- 0.10, N = 3SE +/- 0.19, N = 3SE +/- 0.78, N = 9SE +/- 0.07, N = 3SE +/- 0.72, N = 9SE +/- 0.64, N = 15SE +/- 0.06, N = 3SE +/- 0.29, N = 379.5479.5179.3477.0776.3275.7073.9245.27-O2 -march=znver3-O2 -march=znver3 -flto-O2 -march=x86-64-march=x86-64-march=znver3-march=znver2-march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm

Timed HMMer Search

Pfam Database Search

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database Search-O2 -march=x86-64-O3 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O2 -march=znver3 -flto-O3 -march=znver2-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto20406080100SE +/- 0.15, N = 3SE +/- 0.15, N = 3SE +/- 0.13, N = 3SE +/- 0.18, N = 3SE +/- 0.09, N = 3SE +/- 0.11, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 385.7985.6085.1385.0284.9384.7984.1276.94-O2 -march=x86-64-O3 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O2 -march=znver3 -flto-O3 -march=znver2-O3 -march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -pthread -lhmmer -leasel -lm

GraphicsMagick

Operation: Noise-Gaussian

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Noise-Gaussian-O3 -march=x86-64-O2 -march=x86-64-O3 -maech=znver3 -flto-O3 -march=znver2-O3 -march=znver3-O2 -march=znver3 -flto-O2 -march=znver3-Ofast -march=znver3 -flto110220330440550SE +/- 4.91, N = 12SE +/- 3.51, N = 3SE +/- 1.53, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 1.45, N = 3SE +/- 1.53, N = 3426429430434441445452504-O3 -march=x86-64-O2 -march=x86-64-O3 -march=znver3 -flto-O3 -march=znver2-O3 -march=znver3-O2 -march=znver3 -flto-O2 -march=znver3-Ofast -march=znver3 -flto1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

AOBench

Size: 2048 x 2048 - Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterAOBenchSize: 2048 x 2048 - Total Time-O2 -march=x86-64-O2 -march=znver3-O3 -march=x86-64-O2 -march=znver3 -flto-O3 -maech=znver3 -flto-O3 -march=znver3-O3 -march=znver2-Ofast -march=znver3 -flto816243240SE +/- 2.22, N = 15SE +/- 1.92, N = 15SE +/- 1.25, N = 15SE +/- 0.23, N = 3SE +/- 0.08, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.13, N = 335.0430.7129.9927.3625.8525.0725.0224.51-O2 -march=x86-64-O2 -march=znver3-march=x86-64-O2 -march=znver3 -flto-march=znver3 -flto-march=znver3-march=znver2-Ofast -march=znver3 -flto1. (CC) gcc options: -lm -O3

GraphicsMagick

Operation: Enhanced

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Enhanced-O3 -march=x86-64-O2 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto-O2 -march=znver3 -flto100200300400500SE +/- 1.53, N = 3SE +/- 5.29, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.88, N = 3SE +/- 0.33, N = 3382419430432444445449476-O3 -march=x86-64-O2 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O3 -march=znver3 -flto-Ofast -march=znver3 -flto-O2 -march=znver3 -flto1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Sharpen-O2 -march=x86-64-O3 -march=x86-64-Ofast -march=znver3 -flto-O2 -march=znver3-O3 -march=znver3-O3 -maech=znver3 -flto-O3 -march=znver2-O2 -march=znver3 -flto80160240320400SE +/- 0.58, N = 3SE +/- 0.67, N = 3SE +/- 0.88, N = 3SE +/- 4.33, N = 3SE +/- 2.33, N = 3SE +/- 0.67, N = 3SE +/- 1.20, N = 3SE +/- 1.20, N = 3224225328359361370377377-O2 -march=x86-64-O3 -march=x86-64-Ofast -march=znver3 -flto-O2 -march=znver3-O3 -march=znver3-O3 -march=znver3 -flto-O3 -march=znver2-O2 -march=znver3 -flto1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Resizing

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Resizing-O2 -march=x86-64-O3 -march=x86-64-O2 -march=znver3 -flto-Ofast -march=znver3 -flto-O3 -maech=znver3 -flto-O3 -march=znver2-O3 -march=znver3-O2 -march=znver35001000150020002500SE +/- 1.53, N = 3SE +/- 14.34, N = 3SE +/- 1.15, N = 3SE +/- 1.76, N = 3SE +/- 1.20, N = 3SE +/- 8.37, N = 3SE +/- 4.91, N = 316051721185419011906190721352150-O2 -march=x86-64-O3 -march=x86-64-O2 -march=znver3 -flto-Ofast -march=znver3 -flto-O3 -march=znver3 -flto-O3 -march=znver2-O3 -march=znver3-O2 -march=znver31. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

SciMark

Computational Test: Composite

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Composite-O2 -march=znver3-O2 -march=x86-64-O2 -march=znver3 -flto-O3 -march=x86-64-O3 -maech=znver3 -flto-O3 -march=znver3-Ofast -march=znver3 -flto-O3 -march=znver29001800270036004500SE +/- 34.75, N = 3SE +/- 9.37, N = 3SE +/- 25.20, N = 3SE +/- 12.53, N = 3SE +/- 130.68, N = 12SE +/- 35.50, N = 15SE +/- 33.94, N = 3SE +/- 11.01, N = 32625.592883.782924.233660.743855.303958.174101.894407.32-O2 -march=znver3-O2 -march=x86-64-O2 -march=znver3 -flto-O3 -march=x86-64-O3 -march=znver3 -flto-O3 -march=znver3-Ofast -march=znver3 -flto-O3 -march=znver21. (CC) gcc options: -lm

ACES DGEMM

Sustained Floating-Point Rate

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point Rate-O3 -march=x86-64-O3 -march=znver3-O2 -march=znver3 -flto-O2 -march=znver3-O3 -march=znver2-O2 -march=x86-64-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto246810SE +/- 0.050043, N = 15SE +/- 0.086169, N = 3SE +/- 0.071987, N = 3SE +/- 0.055721, N = 3SE +/- 0.069362, N = 3SE +/- 0.110286, N = 3SE +/- 0.089342, N = 6SE +/- 0.087025, N = 36.4845577.3070937.3456387.3917427.5121897.5265677.6960897.758952-march=x86-64-march=znver3-O2 -march=znver3 -flto-O2 -march=znver3-march=znver2-O2 -march=x86-64-march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -O3 -march=native -fopenmp

Timed ImageMagick Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed ImageMagick Compilation 6.9.0Time To Compile-Ofast -march=znver3 -flto-O3 -maech=znver3 -flto-O2 -march=znver3 -flto-O3 -march=znver2-O3 -march=x86-64-O3 -march=znver3-O2 -march=x86-64-O2 -march=znver31530456075SE +/- 0.30, N = 3SE +/- 0.32, N = 3SE +/- 0.34, N = 3SE +/- 0.18, N = 15SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.22, N = 15SE +/- 0.02, N = 366.8463.4850.4119.3719.0018.8616.4015.80

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Slow

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Slow-O2 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O2 -march=znver3 -flto-O3 -march=x86-64-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto3691215SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 311.4811.5211.5811.5911.6811.6911.8812.12-O2 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O2 -march=znver3 -flto-O3 -march=x86-64-O3 -march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Medium

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Medium-O2 -march=x86-64-O3 -march=znver2-O2 -march=znver3-O3 -march=znver3-O2 -march=znver3 -flto-O3 -march=x86-64-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto3691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 311.6811.7211.7311.7411.8411.8612.0312.27-O2 -march=x86-64-O3 -march=znver2-O2 -march=znver3-O3 -march=znver3-O2 -march=znver3 -flto-O3 -march=x86-64-O3 -march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

C-Ray

Total Time - 4K, 16 Rays Per Pixel

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per Pixel-O2 -march=x86-64-O2 -march=znver3-O2 -march=znver3 -flto-O3 -march=x86-64-Ofast -march=znver3 -flto-O3 -maech=znver3 -flto-O3 -march=znver3-O3 -march=znver21428425670SE +/- 0.16, N = 3SE +/- 0.11, N = 3SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.11, N = 3SE +/- 0.09, N = 360.9046.7246.4231.3627.3425.4825.2525.25-O2 -march=x86-64-O2 -march=znver3-O2 -march=znver3 -flto-march=x86-64-Ofast -march=znver3 -flto-march=znver3 -flto-march=znver3-march=znver21. (CC) gcc options: -lm -lpthread -O3

Coremark

CoreMark Size 666 - Iterations Per Second

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per Second-O2 -march=znver3-O2 -march=znver3 -flto-O2 -march=x86-64-O3 -march=znver3-O3 -march=x86-64-O3 -march=znver2-Ofast -march=znver3 -flto-O3 -maech=znver3 -flto160K320K480K640K800KSE +/- 4015.82, N = 3SE +/- 5589.70, N = 3SE +/- 2080.13, N = 3SE +/- 449.85, N = 3SE +/- 371.91, N = 3SE +/- 1896.73, N = 3SE +/- 1316.80, N = 3SE +/- 1257.49, N = 3669924.92681281.71700520.67716026.62717480.24721620.07728279.79736850.43-march=znver3-march=znver3 -flto-march=x86-64-O3 -march=znver3-O3 -march=x86-64-O3 -march=znver2-Ofast -march=znver3 -flto-O3 -march=znver3 -flto1. (CC) gcc options: -O2 -lrt" -lrt

LibRaw

Post-Processing Benchmark

OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing Benchmark-O2 -march=znver3 -flto-O2 -march=znver3-O2 -march=x86-64-O3 -march=x86-64-O3 -maech=znver3 -flto-O3 -march=znver3-Ofast -march=znver3 -flto-O3 -march=znver220406080100SE +/- 0.36, N = 3SE +/- 0.46, N = 3SE +/- 0.41, N = 3SE +/- 0.24, N = 3SE +/- 0.89, N = 3SE +/- 1.48, N = 15SE +/- 0.56, N = 3SE +/- 0.59, N = 349.3250.9551.4563.9073.7373.8375.2375.26-O2 -march=znver3 -flto-O2 -march=znver3-O2 -march=x86-64-O3 -march=x86-64-O3 -march=znver3 -flto-O3 -march=znver3-Ofast -march=znver3 -flto-O3 -march=znver21. (CXX) g++ options: -fopenmp -ljpeg -lz -lm

Darmstadt Automotive Parallel Heterogeneous Suite

Backend: OpenMP - Kernel: NDT Mapping

OpenBenchmarking.orgTest Cases Per Minute, More Is BetterDarmstadt Automotive Parallel Heterogeneous SuiteBackend: OpenMP - Kernel: NDT Mapping-O2 -march=x86-64-O3 -march=x86-64-O3 -march=znver2-O2 -march=znver3-Ofast -march=znver3 -flto-O2 -march=znver3 -flto-O3 -march=znver3-O3 -maech=znver3 -flto2004006008001000SE +/- 13.06, N = 3SE +/- 3.82, N = 3SE +/- 0.72, N = 3SE +/- 3.99, N = 3SE +/- 3.18, N = 3SE +/- 2.71, N = 3SE +/- 2.43, N = 3SE +/- 2.69, N = 3921.71931.03964.72966.99969.91970.46971.31971.451. (CXX) g++ options: -O3 -std=c++11 -fopenmp

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Very Fast

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Very Fast-O2 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O3 -march=x86-64-O2 -march=znver3 -flto-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto816243240SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 329.5529.8730.3130.3430.4830.7631.4432.40-O2 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O3 -march=x86-64-O2 -march=znver3 -flto-O3 -march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

AOM AV1

Encoder Mode: Speed 8 Realtime

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 2.0Encoder Mode: Speed 8 Realtime-O3 -march=x86-64-O2 -march=x86-64-O2 -march=znver3 -flto-O2 -march=znver3-O3 -march=znver2-O3 -march=znver3-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto1428425670SE +/- 0.99, N = 15SE +/- 0.73, N = 6SE +/- 0.79, N = 3SE +/- 0.37, N = 3SE +/- 0.23, N = 3SE +/- 0.65, N = 3SE +/- 0.59, N = 9SE +/- 0.25, N = 360.0760.8862.0862.1062.2862.3362.5263.76-march=x86-64-O2 -march=x86-64-O2 -march=znver3 -flto-O2 -march=znver3-march=znver2-march=znver3-march=znver3 -flto-Ofast -march=znver3 -flto1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Kvazaar

Video Input: Bosphorus 1080p - Video Preset: Slow

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Slow-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O2 -march=x86-64-O2 -march=znver3 -flto-O3 -march=x86-64-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto1122334455SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 344.8944.9144.9745.0945.6845.7546.3047.25-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O2 -march=x86-64-O2 -march=znver3 -flto-O3 -march=x86-64-O3 -march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

Kvazaar

Video Input: Bosphorus 1080p - Video Preset: Medium

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Medium-O2 -march=znver3-O2 -march=x86-64-O3 -march=znver2-O3 -march=znver3-O3 -march=x86-64-O2 -march=znver3 -flto-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto1122334455SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 346.3146.3246.3846.4246.7447.0447.3248.54-O2 -march=znver3-O2 -march=x86-64-O3 -march=znver2-O3 -march=znver3-O3 -march=x86-64-O2 -march=znver3 -flto-O3 -march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Ultra Fast

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Ultra Fast-O2 -march=x86-64-O2 -march=znver3-O3 -march=x86-64-O3 -march=znver3-O3 -march=znver2-O2 -march=znver3 -flto-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto1326395265SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.31, N = 3SE +/- 0.13, N = 3SE +/- 0.10, N = 353.1353.8354.4354.7354.7554.9156.1258.69-O2 -march=x86-64-O2 -march=znver3-O3 -march=x86-64-O3 -march=znver3-O3 -march=znver2-O2 -march=znver3 -flto-O3 -march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

FFTW

Build: Stock - Size: 1D FFT Size 4096

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 4096-Ofast -march=znver3 -flto-O3 -march=x86-64-O2 -march=x86-64-O3 -march=znver3-O2 -march=znver3-O3 -march=znver2-O2 -march=znver3 -flto-O3 -maech=znver3 -flto3K6K9K12K15KSE +/- 564.61, N = 15SE +/- 80.35, N = 3SE +/- 135.67, N = 3SE +/- 54.77, N = 3SE +/- 161.07, N = 3SE +/- 75.72, N = 3SE +/- 16.50, N = 3SE +/- 105.64, N = 312149.012436.012892.013702.013771.014096.015232.015584.0-Ofast -march=znver3 -flto-O3 -march=znver3-O2 -march=znver3-O3 -march=znver2-O2 -march=znver3 -flto-O3 -march=znver3 -flto1. (CC) gcc options: -pthread -lm

Kvazaar

Video Input: Bosphorus 1080p - Video Preset: Very Fast

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Very Fast-O2 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O3 -march=x86-64-O2 -march=znver3 -flto-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto306090120150SE +/- 0.51, N = 3SE +/- 0.20, N = 3SE +/- 0.17, N = 3SE +/- 0.15, N = 3SE +/- 0.30, N = 3SE +/- 0.14, N = 3SE +/- 0.16, N = 3SE +/- 0.24, N = 3108.72109.42111.50111.54111.77113.41116.07121.74-O2 -march=x86-64-O2 -march=znver3-O3 -march=znver3-O3 -march=znver2-O3 -march=x86-64-O2 -march=znver3 -flto-O3 -march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

Smallpt

Global Illumination Renderer; 128 Samples

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 128 Samples-O3 -march=x86-64-O2 -march=x86-64-O3 -march=znver2-O3 -march=znver3-O3 -maech=znver3 -flto-O2 -march=znver3 -flto-Ofast -march=znver3 -flto-O2 -march=znver31.14032.28063.42094.56125.7015SE +/- 0.018, N = 3SE +/- 0.004, N = 3SE +/- 0.016, N = 3SE +/- 0.025, N = 3SE +/- 0.012, N = 3SE +/- 0.009, N = 3SE +/- 0.005, N = 3SE +/- 0.003, N = 35.0684.8294.6774.6734.6564.6074.6024.567-march=x86-64-O2 -march=x86-64-march=znver2-march=znver3-march=znver3 -flto-O2 -march=znver3 -flto-Ofast -march=znver3 -flto-O2 -march=znver31. (CXX) g++ options: -fopenmp -O3

Kvazaar

Video Input: Bosphorus 1080p - Video Preset: Ultra Fast

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Ultra Fast-O2 -march=x86-64-O2 -march=znver3-O3 -march=x86-64-O3 -march=znver2-O3 -march=znver3-O2 -march=znver3 -flto-O3 -maech=znver3 -flto-Ofast -march=znver3 -flto50100150200250SE +/- 0.75, N = 3SE +/- 0.09, N = 3SE +/- 0.89, N = 3SE +/- 0.35, N = 3SE +/- 0.25, N = 3SE +/- 0.55, N = 3SE +/- 0.63, N = 3SE +/- 0.20, N = 3194.52195.13198.84199.35200.12201.85203.89220.99-O2 -march=x86-64-O2 -march=znver3-O3 -march=x86-64-O3 -march=znver2-O3 -march=znver3-O2 -march=znver3 -flto-O3 -march=znver3 -flto-Ofast -march=znver3 -flto1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

Phoronix Test Suite v10.8.5