GCC Znver3 First Cut Benchmarks

Tests for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2012046-HA-GCCZNVER333&rdt&grt.

GCC Znver3 First Cut BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen Resolutionznver3x86-64znver1znver2haswellskylakeAMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VIII HERO (WI-FI) (2311 BIOS)AMD Starship/Matisse16GB2000GB Corsair Force MP600 + 2000GBNVIDIA GeForce RTX 3060 Ti 8GB (1665/7000MHz)NVIDIA Device 228bASUS MG28URealtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 20.045.4.0-54-generic (x86_64)GNOME Shell 3.36.4X Server 1.20.8NVIDIA 455.45.014.6.0OpenCL 1.2 CUDA 11.1.1141.2.142GCC 11.0.0 20201203ext43840x2160OpenBenchmarking.orgEnvironment Details- znver3: CXXFLAGS="-O3 -march=znver3" CFLAGS="-O3 -march=znver3"- x86-64: CXXFLAGS="-O3 -march=x86-64" CFLAGS="-O3 -march=x86-64"- znver1: CXXFLAGS="-O3 -march=znver1" CFLAGS="-O3 -march=znver1"- znver2: CXXFLAGS="-O3 -march=znver2" CFLAGS="-O3 -march=znver2"- haswell: CXXFLAGS="-O3 -march=haswell" CFLAGS="-O3 -march=haswell"- skylake: CXXFLAGS="-O3 -march=skylake" CFLAGS="-O3 -march=skylake"Compiler Details- --disable-multilib --enable-checking=releaseDisk Details- znver3: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Details- Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009 OpenCL Details- GPU Compute Cores: 4864Python Details- znver3: Python 2.7.18 + Python 3.8.5Security 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

GCC Znver3 First Cut Benchmarksmt-dgemm: Sustained Floating-Point Rateaobench: 2048 x 2048 - Total Timec-ray: Total Time - 4K, 16 Rays Per Pixelcoremark: CoreMark Size 666 - Iterations Per Secondcrafty: Elapsed Timedaphne: OpenMP - NDT Mappingdaphne: OpenMP - Points2Imagedav1d: Chimera 1080p 10-bitgraphics-magick: Sharpengraphics-magick: Enhancedgraphics-magick: Resizinghint: FLOATkvazaar: Bosphorus 4K - Ultra Fastkvazaar: Bosphorus 1080p - Ultra Fastencode-mp3: WAV To MP3libraw: Post-Processing Benchmarkrnnoise: scimark2: Compositescimark2: Sparse Matrix Multiplyscimark2: Jacobi Successive Over-Relaxationsmallpt: Global Illumination Renderer; 128 Samplessqlite-speedtest: Timed Time - Size 1,000vpxenc: Speed 0vpxenc: Speed 5webp: Quality 100znver3x86-64znver1znver2haswellskylake8.97234725.25825.202837788.08423412060994969.3829625.957802710263.973744472025539663428.1078555.82203.085.46876.0413.9894463.425091.123072.044.62741.90510.7639.131.6708.05418029.52131.212783951.86725011828707932.2929502.084524288189.622253821809521237587.6705355.07202.435.82265.4814.6333662.754977.772440.105.05942.18610.1736.031.7178.77831625.02826.073835605.87397511865359950.3029918.699671060251.113674462084523840766.0695456.22205.715.54674.5414.0953304.214392.612913.834.79042.25010.8238.951.6928.92213325.25425.321834342.44018911910299963.5329824.340387406265.163734462050529343739.0960655.85201.845.47876.0113.8544310.954918.892931.784.66442.08610.7738.971.6897.82931925.32825.455776752.21752811935854969.6029785.613813708175.362995242023519167343.7785657.53209.195.52074.1315.5124332.885062.072934.464.90642.22910.8139.181.6908.11148625.36925.085823512.88640611943425962.7330279.771134316262.403774362023520771613.7886255.75203.695.56575.0415.5754314.574955.382936.044.87242.78310.7738.871.684OpenBenchmarking.org

ACES DGEMM

Sustained Floating-Point Rate

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point Rateznver3x86-64znver1znver2haswellskylake3691215SE +/- 0.135072, N = 3SE +/- 0.072954, N = 3SE +/- 0.028561, N = 3SE +/- 0.094255, N = 3SE +/- 0.062576, N = 15SE +/- 0.117999, N = 48.9723478.0541808.7783168.9221337.8293198.111486-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -O3 -march=native -fopenmp

AOBench

Size: 2048 x 2048 - Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterAOBenchSize: 2048 x 2048 - Total Timeznver3x86-64znver1znver2haswellskylake714212835SE +/- 0.06, N = 3SE +/- 0.68, N = 15SE +/- 0.24, N = 3SE +/- 0.01, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 325.2629.5225.0325.2525.3325.37-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -lm -O3

C-Ray

Total Time - 4K, 16 Rays Per Pixel

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per Pixelznver3x86-64znver1znver2haswellskylake714212835SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 325.2031.2126.0725.3225.4625.09-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (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 Secondznver3x86-64znver1znver2haswellskylake200K400K600K800K1000KSE +/- 5296.50, N = 3SE +/- 651.07, N = 3SE +/- 2226.29, N = 3SE +/- 7614.04, N = 3SE +/- 563.60, N = 3SE +/- 1894.03, N = 3837788.08783951.87835605.87834342.44776752.22823512.89-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -O2 -O3 -lrt" -lrt

Crafty

Elapsed Time

OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed Timeznver3x86-64znver1znver2haswellskylake3M6M9M12M15MSE +/- 121191.73, N = 3SE +/- 18592.10, N = 3SE +/- 33532.81, N = 3SE +/- 44223.61, N = 3SE +/- 22104.24, N = 3SE +/- 69563.68, N = 31206099411828707118653591191029911935854119434251. (CC) gcc options: -pthread -lstdc++ -fprofile-use -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 Mappingznver3x86-64znver1znver2haswellskylake2004006008001000SE +/- 3.90, N = 3SE +/- 2.69, N = 3SE +/- 0.66, N = 3SE +/- 2.52, N = 3SE +/- 1.90, N = 3SE +/- 1.88, N = 3969.38932.29950.30963.53969.60962.731. (CXX) g++ options: -O3 -std=c++11 -fopenmp

Darmstadt Automotive Parallel Heterogeneous Suite

Backend: OpenMP - Kernel: Points2Image

OpenBenchmarking.orgTest Cases Per Minute, More Is BetterDarmstadt Automotive Parallel Heterogeneous SuiteBackend: OpenMP - Kernel: Points2Imageznver3x86-64znver1znver2haswellskylake6K12K18K24K30KSE +/- 109.65, N = 3SE +/- 108.90, N = 3SE +/- 159.95, N = 3SE +/- 240.50, N = 3SE +/- 289.82, N = 3SE +/- 113.76, N = 329625.9629502.0829918.7029824.3429785.6130279.771. (CXX) g++ options: -O3 -std=c++11 -fopenmp

dav1d

Video Input: Chimera 1080p 10-bit

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.7.0Video Input: Chimera 1080p 10-bitznver3x86-64znver1znver2haswellskylake60120180240300SE +/- 0.85, N = 3SE +/- 0.23, N = 3SE +/- 0.45, N = 3SE +/- 0.73, N = 3SE +/- 0.22, N = 3SE +/- 0.56, N = 3263.97189.62251.11265.16175.36262.40-march=znver3 - MIN: 175.34 / MAX: 478.58-march=x86-64 - MIN: 124.1 / MAX: 372.85-march=znver1 - MIN: 167.21 / MAX: 471.36-march=znver2 - MIN: 175.81 / MAX: 502.41-march=haswell - MIN: 93.55 / MAX: 459.71-march=skylake - MIN: 173.7 / MAX: 495.131. (CC) gcc options: -O3 -pthread

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Sharpenznver3x86-64znver1znver2haswellskylake80160240320400SE +/- 1.00, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3374225367373299377-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Enhanced

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Enhancedznver3x86-64znver1znver2haswellskylake110220330440550SE +/- 0.58, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3447382446446524436-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -fopenmp -O3 -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: Resizingznver3x86-64znver1znver2haswellskylake400800120016002000SE +/- 4.33, N = 3SE +/- 3.79, N = 3SE +/- 0.88, N = 3SE +/- 4.91, N = 3SE +/- 1.00, N = 3SE +/- 4.10, N = 3202518092084205020232023-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

Hierarchical INTegration

Test: FLOAT

OpenBenchmarking.orgQUIPs, More Is BetterHierarchical INTegration 1.0Test: FLOATznver3x86-64znver1znver2haswellskylake120M240M360M480M600MSE +/- 963540.42, N = 3SE +/- 1444160.80, N = 3SE +/- 114637.70, N = 3SE +/- 5908890.92, N = 3SE +/- 2024395.89, N = 3SE +/- 4152069.13, N = 3539663428.11521237587.67523840766.07529343739.10519167343.78520771613.79-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -O3 -march=native -lm

Kvazaar

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

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Ultra Fastznver3x86-64znver1znver2haswellskylake1326395265SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.20, N = 3SE +/- 0.10, N = 3SE +/- 0.01, N = 3SE +/- 0.22, N = 355.8255.0756.2255.8557.5355.75-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -lpthread -lm -lrt

Kvazaar

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

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Ultra Fastznver3x86-64znver1znver2haswellskylake50100150200250SE +/- 0.81, N = 3SE +/- 0.29, N = 3SE +/- 0.58, N = 3SE +/- 1.10, N = 3SE +/- 1.14, N = 3SE +/- 0.72, N = 3203.08202.43205.71201.84209.19203.69-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -lpthread -lm -lrt

LAME MP3 Encoding

WAV To MP3

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3znver3x86-64znver1znver2haswellskylake1.312.623.935.246.55SE +/- 0.006, N = 3SE +/- 0.015, N = 3SE +/- 0.009, N = 3SE +/- 0.010, N = 3SE +/- 0.020, N = 3SE +/- 0.007, N = 35.4685.8225.5465.4785.5205.565-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lncurses -lm

LibRaw

Post-Processing Benchmark

OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing Benchmarkznver3x86-64znver1znver2haswellskylake20406080100SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.16, N = 3SE +/- 0.20, N = 3SE +/- 0.11, N = 3SE +/- 0.39, N = 376.0465.4874.5476.0174.1375.04-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CXX) g++ options: -O3 -fopenmp -ljpeg -lz -lm

RNNoise

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 2020-06-28znver3x86-64znver1znver2haswellskylake48121620SE +/- 0.13, N = 3SE +/- 0.22, N = 3SE +/- 0.10, N = 3SE +/- 0.19, N = 3SE +/- 0.16, N = 3SE +/- 0.05, N = 313.9914.6314.1013.8515.5115.58-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -O3 -pedantic -fvisibility=hidden

SciMark

Computational Test: Composite

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Compositeznver3x86-64znver1znver2haswellskylake10002000300040005000SE +/- 0.19, N = 3SE +/- 5.93, N = 3SE +/- 13.80, N = 3SE +/- 1.02, N = 3SE +/- 11.48, N = 3SE +/- 4.37, N = 34463.423662.753304.214310.954332.884314.57-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -O3 -lm

SciMark

Computational Test: Sparse Matrix Multiply

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Sparse Matrix Multiplyznver3x86-64znver1znver2haswellskylake11002200330044005500SE +/- 2.08, N = 3SE +/- 10.14, N = 3SE +/- 1.53, N = 3SE +/- 1.75, N = 3SE +/- 44.33, N = 3SE +/- 10.56, N = 35091.124977.774392.614918.895062.074955.38-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -O3 -lm

SciMark

Computational Test: Jacobi Successive Over-Relaxation

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Jacobi Successive Over-Relaxationznver3x86-64znver1znver2haswellskylake7001400210028003500SE +/- 0.78, N = 3SE +/- 0.05, N = 3SE +/- 3.35, N = 3SE +/- 3.83, N = 3SE +/- 2.02, N = 3SE +/- 0.45, N = 33072.042440.102913.832931.782934.462936.04-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -O3 -lm

Smallpt

Global Illumination Renderer; 128 Samples

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 128 Samplesznver3x86-64znver1znver2haswellskylake1.13832.27663.41494.55325.6915SE +/- 0.005, N = 3SE +/- 0.010, N = 3SE +/- 0.007, N = 3SE +/- 0.007, N = 3SE +/- 0.014, N = 3SE +/- 0.005, N = 34.6275.0594.7904.6644.9064.872-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CXX) g++ options: -fopenmp -O3

SQLite Speedtest

Timed Time - Size 1,000

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000znver3x86-64znver1znver2haswellskylake1020304050SE +/- 0.25, N = 3SE +/- 0.18, N = 3SE +/- 0.19, N = 3SE +/- 0.61, N = 3SE +/- 0.23, N = 3SE +/- 0.15, N = 341.9142.1942.2542.0942.2342.78-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -O3 -ldl -lz -lpthread

VP9 libvpx Encoding

Speed: Speed 0

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.8.2Speed: Speed 0znver3x86-64znver1znver2haswellskylake3691215SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 310.7610.1710.8210.7710.8110.77-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11

VP9 libvpx Encoding

Speed: Speed 5

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.8.2Speed: Speed 5znver3x86-64znver1znver2haswellskylake918273645SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.33, N = 3SE +/- 0.15, N = 3SE +/- 0.02, N = 3SE +/- 0.27, N = 339.1336.0338.9538.9739.1838.87-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11

WebP Image Encode

Encode Settings: Quality 100

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100znver3x86-64znver1znver2haswellskylake0.38630.77261.15891.54521.9315SE +/- 0.020, N = 3SE +/- 0.004, N = 3SE +/- 0.004, N = 3SE +/- 0.003, N = 3SE +/- 0.004, N = 3SE +/- 0.001, N = 31.6701.7171.6921.6891.6901.684-march=znver3-march=x86-64-march=znver1-march=znver2-march=haswell-march=skylake1. (CC) gcc options: -fvisibility=hidden -O3 -pthread -lm -ljpeg -lpng16 -ltiff

Phoronix Test Suite v10.8.4