EPYC 7763 LLVM Clang Compiler Tests

AMD EPYC 7763 64-Core testing with a Supermicro H12SSL-i v1.01 (2.0 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2104123-IB-EPYC7763L12&grs&sor.

EPYC 7763 LLVM Clang Compiler TestsProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionClang 12.0Clang 11.0AMD EPYC 7763 64-Core @ 2.45GHz (64 Cores / 128 Threads)Supermicro H12SSL-i v1.01 (2.0 BIOS)AMD Starship/Matisse126GB3841GB Micron_9300_MTFDHAL3T8TDPASPEED2 x Broadcom NetXtreme BCM5720 2-port PCIeUbuntu 20.045.12.0-051200rc6daily20210408-generic (x86_64) 20210407GNOME Shell 3.36.4X Server 1.20.8Clang 12.0.0-++20210409092622+fa0971b87fb2-1~exp1~20210409193326.73ext41024x768Clang 11.0.0-2~ubuntu20.04.1OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseEnvironment Details- CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"Processor Details- Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa001119Python Details- Python 3.8.2Security 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

EPYC 7763 LLVM Clang Compiler Testsviennacl: CPU BLAS - dGEMM-NNviennacl: CPU BLAS - dGEMM-TNdav1d: Chimera 1080p 10-bitonednn: Convolution Batch Shapes Auto - f32 - CPUetcpak: DXT1etcpak: ETC1onednn: Convolution Batch Shapes Auto - u8s8f32 - CPUviennacl: CPU BLAS - dGEMM-NTetcpak: ETC2onednn: IP Shapes 3D - u8s8f32 - CPUbotan: Blowfishviennacl: CPU BLAS - dGEMM-TTngspice: C2670toybrot: C++ Threadswebp2: Quality 100, Compression Effort 5scimark2: Fast Fourier Transformtoybrot: C++ Taskstoybrot: TBBviennacl: CPU BLAS - dGEMV-Tlibraw: Post-Processing Benchmarkonednn: IP Shapes 3D - f32 - CPUfftw: Float + SSE - 1D FFT Size 32scimark2: Sparse Matrix Multiplygraphics-magick: Rotatetoybrot: OpenMPbotan: Twofish - Decryptonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUngspice: C7552jpegxl: PNG - 5onednn: Recurrent Neural Network Inference - f32 - CPUbotan: Twofishbotan: AES-256webp2: Quality 100, Lossless Compressionbotan: KASUMI - Decryptonednn: Recurrent Neural Network Inference - u8s8f32 - CPUsimdjson: DistinctUserIDfftw: Stock - 1D FFT Size 2048botan: AES-256 - Decryptfftw: Stock - 1D FFT Size 4096botan: KASUMIsimdjson: PartialTweetstscp: AI Chess Performancebotan: CAST-256 - Decryptgraphics-magick: Swirlscimark2: Compositepgbench: 100 - 250 - Read Writeonnx: yolov4 - OpenMP CPUpgbench: 100 - 250 - Read Write - Average Latencyonnx: fcn-resnet101-11 - OpenMP CPUsimdjson: LargeRandscimark2: Dense LU Matrix Factorizationbotan: CAST-256jpegxl: JPEG - 8avifenc: 6, Losslessfftw: Stock - 2D FFT Size 1024aom-av1: Speed 8 Realtime - Bosphorus 1080psvt-av1: Enc Mode 4 - 1080pfftw: Float + SSE - 1D FFT Size 4096simdjson: Kostyaonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUaom-av1: Speed 9 Realtime - Bosphorus 1080ppgbench: 100 - 1 - Read Onlypgbench: 100 - 1 - Read Only - Average Latencyjpegxl: PNG - 8webp: Quality 100, Losslessencode-opus: WAV To Opus Encodeonednn: Recurrent Neural Network Training - u8s8f32 - CPUfftw: Float + SSE - 1D FFT Size 2048avifenc: 10, Losslessfftw: Stock - 1D FFT Size 1024tachyon: Total Timeaom-av1: Speed 9 Realtime - Bosphorus 4Konednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUavifenc: 10securemark: SecureMark-TLSwebp: Quality 100, Lossless, Highest Compressionwebp: Quality 100onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUgraphics-magick: HWB Color Spacec-ray: Total Time - 4K, 16 Rays Per Pixelaom-av1: Speed 6 Two-Pass - Bosphorus 4Kwebp2: Quality 95, Compression Effort 7jpegxl: JPEG - 5aom-av1: Speed 4 Two-Pass - Bosphorus 4Kencode-flac: WAV To FLACjpegxl: JPEG - 7svt-hevc: 10 - Bosphorus 1080paom-av1: Speed 4 Two-Pass - Bosphorus 1080pgcrypt: svt-vp9: VMAF Optimized - Bosphorus 1080pliquid-dsp: 128 - 256 - 57graphics-magick: Noise-Gaussiansvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080px265: Bosphorus 4Kpovray: Trace Timepgbench: 100 - 100 - Read Write - Average Latencyfftw: Stock - 2D FFT Size 4096avifenc: 2jpegxl: PNG - 7liquid-dsp: 1 - 256 - 57pgbench: 100 - 100 - Read Writefftw: Stock - 2D FFT Size 2048svt-av1: Enc Mode 0 - 1080ponnx: shufflenet-v2-10 - OpenMP CPUwebp: Quality 100, Highest Compressioncompress-lz4: 9 - Compression Speedpgbench: 100 - 1 - Read Write - Average Latencypgbench: 100 - 1 - Read Writeonednn: Deconvolution Batch shapes_1d - f32 - CPUaom-av1: Speed 6 Realtime - Bosphorus 1080px265: Bosphorus 1080pliquid-dsp: 32 - 256 - 57aom-av1: Speed 8 Realtime - Bosphorus 4Kgraphics-magick: Enhanceddav1d: Chimera 1080pliquid-dsp: 64 - 256 - 57fftw: Float + SSE - 2D FFT Size 2048aom-av1: Speed 6 Two-Pass - Bosphorus 1080pfinancebench: Bonds OpenMPonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUsvt-av1: Enc Mode 8 - 1080pdav1d: Summer Nature 1080pmrbayes: Primate Phylogeny Analysisastcenc: Mediumonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUcompress-lz4: 3 - Compression Speedpgbench: 100 - 250 - Read Onlyaom-av1: Speed 6 Realtime - Bosphorus 4Kquantlib: compress-lz4: 3 - Decompression Speedfftw: Float + SSE - 2D FFT Size 4096svt-hevc: 7 - Bosphorus 1080ppgbench: 100 - 250 - Read Only - Average Latencysvt-vp9: Visual Quality Optimized - Bosphorus 1080pwebp: Defaultdav1d: Summer Nature 4Kbotan: ChaCha20Poly1305 - Decryptcoremark: CoreMark Size 666 - Iterations Per Secondonednn: IP Shapes 1D - f32 - CPUavifenc: 6botan: ChaCha20Poly1305financebench: Repo OpenMPsvt-hevc: 1 - Bosphorus 1080pastcenc: Exhaustivegraphics-magick: Sharpenfftw: Float + SSE - 2D FFT Size 1024webp2: Defaultwebp2: Quality 75, Compression Effort 7onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUencode-mp3: WAV To MP3fftw: Stock - 1D FFT Size 32onednn: IP Shapes 1D - u8s8f32 - CPUbotan: Blowfish - Decryptscimark2: Monte Carloastcenc: Thoroughpgbench: 100 - 100 - Read Onlyavifenc: 0compress-lz4: 9 - Decompression Speedscimark2: Jacobi Successive Over-Relaxationpgbench: 100 - 100 - Read Only - Average Latencyaom-av1: Speed 0 Two-Pass - Bosphorus 1080paom-av1: Speed 0 Two-Pass - Bosphorus 4Konnx: super-resolution-10 - OpenMP CPUonnx: bertsquad-10 - OpenMP CPUviennacl: CPU BLAS - dGEMV-Nviennacl: CPU BLAS - dDOTviennacl: CPU BLAS - dAXPYviennacl: CPU BLAS - dCOPYviennacl: CPU BLAS - sDOTviennacl: CPU BLAS - sAXPYviennacl: CPU BLAS - sCOPYgraphics-magick: Resizingfftw: Float + SSE - 1D FFT Size 1024Clang 12.0Clang 11.048.651.9308.321.221322718.525284.6422.0360665.7202.0850.710124380.05473.0118.87072206.690363.857437678062641.783.28507156494280.227127507321.190597.48195.95674.27593.972315.4094659.338374.03584.229590.1824.62104674682.4559862.082.6444.601570966133.04819933190.62566843334.4311120.848848.40132.82028.1325.2209088.388.7811.474454282.751305.10103.17243100.0410.8219.0167.5671307.49512545.7461080516.046838.112.367971302.703.36126520438.4492.1991.1725860515.8708.99207.00866.664.877.85466.38643.587.10236.924487.433643766667457488.2330.329.2961.6076744.125.17512.1555663000623197789.90.18399046.30948.500.30532811.4442526.8574.00156483333333.3910761198.2230706333333193522.1351596.8671870.313689118.0671244.1189.1164.00580.49194052.07107120917.222653.813911.522797345.300.234372.491.331541.56843.4041785466.2839691.077019.510850.49633246.83723941.0918.9936614362392.739109.5250.7797768.256133331.07507351.284675.136.7647106902247.88413926.51785.500.0940.530.21445649869.181987860443435747121365035083.688.3184.190.8411691872.759205.0651.6054079.3168.8190.594729319.23484.0103.82663957.366399.166836624767738.713.52787145904590.376657029302.405563.24790.52778.41563.200299.2144901.127392.84980.221562.9704.4110004.24895.5589438.679.1494.411638265127.74019153319.34544883464.6031080.819146.88128.58627.2426.0348809.686.0911.821466762.681271.91100.55249430.0400.818.5737.3921277.62500845.8791056416.409937.282.318591276.043.42926011937.7272.2401.1514061615.5999.14203.63465.584.957.97965.43652.747.20240.205481.053596533333463482.0229.949.4081.6266823.825.47212.0156307000616167878.50.18197976.24349.010.30233121.4575726.6173.36157840000033.1410681190.4130513666673174122.0051900.4348960.315522117.3921251.2588.6203.98370.48927852.35106550617.132640.213840.322913346.890.235373.991.336543.43840.6371790837.0100001.080119.536848.23633178.49869841.0119.0255613361812.743109.6360.7791018.250133241.07577351.075674.866.7674106936747.89413927.91785.420.0940.530.21452347151.293310431877462412495203450740OpenBenchmarking.org

ViennaCL

Test: CPU BLAS - dGEMM-NN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NNClang 11.0Clang 12.020406080100SE +/- 0.06, N = 15SE +/- 0.05, N = 1283.648.61. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMM-TN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TNClang 11.0Clang 12.020406080100SE +/- 0.02, N = 15SE +/- 0.09, N = 1288.351.91. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

dav1d

Video Input: Chimera 1080p 10-bit

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Chimera 1080p 10-bitClang 12.0Clang 11.070140210280350SE +/- 0.93, N = 3SE +/- 0.48, N = 3308.32184.19MIN: 220.53 / MAX: 490.51-lm - MIN: 114.52 / MAX: 310.51. (CC) gcc options: -O3 -march=native -pthread

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUClang 11.0Clang 12.00.27480.54960.82441.09921.374SE +/- 0.000480, N = 3SE +/- 0.018279, N = 40.8411691.221320MIN: 0.82MIN: 1.131. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

Etcpak

Configuration: DXT1

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: DXT1Clang 12.0Clang 11.06001200180024003000SE +/- 2.64, N = 3SE +/- 1.69, N = 32718.531872.761. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

Etcpak

Configuration: ETC1

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC1Clang 12.0Clang 11.060120180240300SE +/- 0.11, N = 3SE +/- 0.03, N = 3284.64205.071. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUClang 11.0Clang 12.00.45810.91621.37431.83242.2905SE +/- 0.00118, N = 3SE +/- 0.01922, N = 121.605402.03606MIN: 1.55MIN: 1.811. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

ViennaCL

Test: CPU BLAS - dGEMM-NT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NTClang 11.0Clang 12.020406080100SE +/- 0.03, N = 15SE +/- 0.56, N = 1279.365.71. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

Etcpak

Configuration: ETC2

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC2Clang 12.0Clang 11.04080120160200SE +/- 0.02, N = 3SE +/- 0.02, N = 3202.09168.821. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

oneDNN

Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUClang 11.0Clang 12.00.15980.31960.47940.63920.799SE +/- 0.008914, N = 3SE +/- 0.011383, N = 30.5947290.710124MIN: 0.53MIN: 0.641. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

Botan

Test: Blowfish

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: BlowfishClang 12.0Clang 11.080160240320400SE +/- 0.05, N = 3SE +/- 1.73, N = 3380.05319.231. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

ViennaCL

Test: CPU BLAS - dGEMM-TT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TTClang 11.0Clang 12.020406080100SE +/- 0.02, N = 14SE +/- 0.07, N = 1284.073.01. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

Ngspice

Circuit: C2670

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670Clang 11.0Clang 12.0306090120150SE +/- 0.06, N = 3SE +/- 0.53, N = 3103.83118.871. (CC) gcc options: -O3 -march=native -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

toyBrot Fractal Generator

Implementation: C++ Threads

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ ThreadsClang 11.0Clang 12.015003000450060007500SE +/- 25.04, N = 3SE +/- 30.90, N = 3639572201. (CXX) g++ options: -O3 -march=native -lpthread -lm -lgcc -lgcc_s -lc

WebP2 Image Encode

Encode Settings: Quality 100, Compression Effort 5

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Compression Effort 5Clang 12.0Clang 11.0246810SE +/- 0.006, N = 3SE +/- 0.022, N = 36.6907.3661. (CXX) g++ options: -O3 -march=native -fno-rtti -rdynamic -lpthread -ljpeg -lgif -lwebp -lwebpdemux

SciMark

Computational Test: Fast Fourier Transform

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Fast Fourier TransformClang 11.0Clang 12.090180270360450SE +/- 0.67, N = 3SE +/- 0.46, N = 3399.16363.851. (CC) gcc options: -O3 -march=native -lm

toyBrot Fractal Generator

Implementation: C++ Tasks

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ TasksClang 11.0Clang 12.016003200480064008000SE +/- 7.31, N = 3SE +/- 33.67, N = 3683674371. (CXX) g++ options: -O3 -march=native -lpthread -lm -lgcc -lgcc_s -lc

toyBrot Fractal Generator

Implementation: TBB

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: TBBClang 11.0Clang 12.015003000450060007500SE +/- 67.11, N = 7SE +/- 87.21, N = 3624767801. (CXX) g++ options: -O3 -march=native -lpthread -lm -lgcc -lgcc_s -lc

ViennaCL

Test: CPU BLAS - dGEMV-T

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-TClang 11.0Clang 12.0150300450600750SE +/- 1.41, N = 14SE +/- 4.04, N = 126776261. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

LibRaw

Post-Processing Benchmark

OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing BenchmarkClang 12.0Clang 11.01020304050SE +/- 0.12, N = 3SE +/- 0.33, N = 341.7838.711. (CXX) g++ options: -O3 -march=native -fopenmp -ljpeg -lz -lm

oneDNN

Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUClang 12.0Clang 11.00.79381.58762.38143.17523.969SE +/- 0.01639, N = 3SE +/- 0.04735, N = 33.285073.52787MIN: 3.15MIN: 3.291. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

FFTW

Build: Float + SSE - Size: 1D FFT Size 32

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 32Clang 12.0Clang 11.03K6K9K12K15KSE +/- 48.79, N = 3SE +/- 129.55, N = 315649145901. (CC) gcc options: -pthread -O3 -march=native -lm

SciMark

Computational Test: Sparse Matrix Multiply

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Sparse Matrix MultiplyClang 11.0Clang 12.010002000300040005000SE +/- 3.87, N = 3SE +/- 10.41, N = 34590.374280.221. (CC) gcc options: -O3 -march=native -lm

GraphicsMagick

Operation: Rotate

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: RotateClang 12.0Clang 11.0150300450600750SE +/- 2.60, N = 3SE +/- 1.33, N = 37126651. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

toyBrot Fractal Generator

Implementation: OpenMP

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: OpenMPClang 11.0Clang 12.016003200480064008000SE +/- 20.42, N = 3SE +/- 14.89, N = 3702975071. (CXX) g++ options: -O3 -march=native -lpthread -lm -lgcc -lgcc_s -lc

Botan

Test: Twofish - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Twofish - DecryptClang 12.0Clang 11.070140210280350SE +/- 0.16, N = 3SE +/- 0.15, N = 3321.19302.411. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUClang 11.0Clang 12.0130260390520650SE +/- 0.10, N = 3SE +/- 3.02, N = 3563.25597.48MIN: 551.31MIN: 580.81. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

Ngspice

Circuit: C7552

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552Clang 11.0Clang 12.020406080100SE +/- 1.37, N = 3SE +/- 1.11, N = 690.5395.961. (CC) gcc options: -O3 -march=native -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

JPEG XL

Input: PNG - Encode Speed: 5

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: PNG - Encode Speed: 5Clang 11.0Clang 12.020406080100SE +/- 0.24, N = 3SE +/- 0.17, N = 378.4174.271. (CXX) g++ options: -O3 -march=native -funwind-tables -Xclang -mrelax-all -O2 -fPIE -pie -pthread -ldl

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUClang 11.0Clang 12.0130260390520650SE +/- 0.83, N = 3SE +/- 9.50, N = 3563.20593.97MIN: 550.23MIN: 570.441. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

Botan

Test: Twofish

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: TwofishClang 12.0Clang 11.070140210280350SE +/- 0.13, N = 3SE +/- 0.09, N = 3315.41299.211. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: AES-256

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256Clang 11.0Clang 12.011002200330044005500SE +/- 2.16, N = 3SE +/- 2.14, N = 34901.134659.341. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

WebP2 Image Encode

Encode Settings: Quality 100, Lossless Compression

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Lossless CompressionClang 12.0Clang 11.090180270360450SE +/- 0.49, N = 3SE +/- 0.17, N = 3374.04392.851. (CXX) g++ options: -O3 -march=native -fno-rtti -rdynamic -lpthread -ljpeg -lgif -lwebp -lwebpdemux

Botan

Test: KASUMI - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMI - DecryptClang 12.0Clang 11.020406080100SE +/- 0.06, N = 3SE +/- 0.04, N = 384.2380.221. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUClang 11.0Clang 12.0130260390520650SE +/- 0.25, N = 3SE +/- 1.89, N = 3562.97590.18MIN: 551.49MIN: 575.411. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

simdjson

Throughput Test: DistinctUserID

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.8.2Throughput Test: DistinctUserIDClang 12.0Clang 11.01.03952.0793.11854.1585.1975SE +/- 0.00, N = 3SE +/- 0.00, N = 34.624.411. (CXX) g++ options: -O3 -march=native -pthread

FFTW

Build: Stock - Size: 1D FFT Size 2048

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 2048Clang 12.0Clang 11.02K4K6K8K10KSE +/- 7.75, N = 3SE +/- 28.76, N = 310467.010004.21. (CC) gcc options: -pthread -O3 -march=native -lm

Botan

Test: AES-256 - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256 - DecryptClang 11.0Clang 12.010002000300040005000SE +/- 1.35, N = 3SE +/- 4.78, N = 34895.564682.461. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

FFTW

Build: Stock - Size: 1D FFT Size 4096

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 4096Clang 12.0Clang 11.02K4K6K8K10KSE +/- 101.36, N = 3SE +/- 15.16, N = 39862.09438.61. (CC) gcc options: -pthread -O3 -march=native -lm

Botan

Test: KASUMI

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMIClang 12.0Clang 11.020406080100SE +/- 0.01, N = 3SE +/- 0.06, N = 382.6479.151. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

simdjson

Throughput Test: PartialTweets

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.8.2Throughput Test: PartialTweetsClang 12.0Clang 11.01.0352.073.1054.145.175SE +/- 0.01, N = 3SE +/- 0.01, N = 34.604.411. (CXX) g++ options: -O3 -march=native -pthread

TSCP

AI Chess Performance

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess PerformanceClang 11.0Clang 12.0400K800K1200K1600K2000KSE +/- 2852.59, N = 5SE +/- 1798.40, N = 5163826515709661. (CC) gcc options: -O3 -march=native

Botan

Test: CAST-256 - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256 - DecryptClang 12.0Clang 11.0306090120150SE +/- 0.01, N = 3SE +/- 0.01, N = 3133.05127.741. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

GraphicsMagick

Operation: Swirl

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SwirlClang 12.0Clang 11.0400800120016002000SE +/- 6.57, N = 3SE +/- 12.41, N = 3199319151. (CC) gcc options: -fopenmp -O3 -march=native -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: CompositeClang 11.0Clang 12.07001400210028003500SE +/- 15.12, N = 3SE +/- 1.11, N = 33319.343190.621. (CC) gcc options: -O3 -march=native -lm

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 250 - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read WriteClang 12.0Clang 11.012K24K36K48K60KSE +/- 702.52, N = 15SE +/- 883.12, N = 356684544881. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

ONNX Runtime

Model: yolov4 - Device: OpenMP CPU

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPUClang 11.0Clang 12.080160240320400SE +/- 1.42, N = 3SE +/- 4.15, N = 43463331. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -ffunction-sections -fdata-sections -ldl -lrt

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average LatencyClang 12.0Clang 11.01.03572.07143.10714.14285.1785SE +/- 0.054, N = 15SE +/- 0.074, N = 34.4314.6031. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

ONNX Runtime

Model: fcn-resnet101-11 - Device: OpenMP CPU

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: fcn-resnet101-11 - Device: OpenMP CPUClang 12.0Clang 11.0306090120150SE +/- 0.50, N = 3SE +/- 0.29, N = 31121081. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -ffunction-sections -fdata-sections -ldl -lrt

simdjson

Throughput Test: LargeRandom

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.8.2Throughput Test: LargeRandomClang 12.0Clang 11.00.1890.3780.5670.7560.945SE +/- 0.00, N = 3SE +/- 0.00, N = 30.840.811. (CXX) g++ options: -O3 -march=native -pthread

SciMark

Computational Test: Dense LU Matrix Factorization

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Dense LU Matrix FactorizationClang 11.0Clang 12.02K4K6K8K10KSE +/- 77.81, N = 3SE +/- 7.16, N = 39146.888848.401. (CC) gcc options: -O3 -march=native -lm

Botan

Test: CAST-256

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256Clang 12.0Clang 11.0306090120150SE +/- 0.02, N = 3SE +/- 0.02, N = 3132.82128.591. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

JPEG XL

Input: JPEG - Encode Speed: 8

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: JPEG - Encode Speed: 8Clang 12.0Clang 11.0714212835SE +/- 0.03, N = 3SE +/- 0.01, N = 328.1327.241. (CXX) g++ options: -O3 -march=native -funwind-tables -Xclang -mrelax-all -O2 -fPIE -pie -pthread -ldl

libavif avifenc

Encoder Speed: 6, Lossless

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, LosslessClang 12.0Clang 11.0612182430SE +/- 0.04, N = 3SE +/- 0.22, N = 325.2226.031. (CXX) g++ options: -O3 -fPIC -lm

FFTW

Build: Stock - Size: 2D FFT Size 1024

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 1024Clang 12.0Clang 11.02K4K6K8K10KSE +/- 48.25, N = 3SE +/- 45.95, N = 39088.38809.61. (CC) gcc options: -pthread -O3 -march=native -lm

AOM AV1

Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080pClang 12.0Clang 11.020406080100SE +/- 1.07, N = 3SE +/- 0.51, N = 388.7886.091. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

SVT-AV1

Encoder Mode: Enc Mode 4 - Input: 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 4 - Input: 1080pClang 11.0Clang 12.03691215SE +/- 0.16, N = 4SE +/- 0.17, N = 311.8211.471. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

FFTW

Build: Float + SSE - Size: 1D FFT Size 4096

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 4096Clang 11.0Clang 12.010K20K30K40K50KSE +/- 413.24, N = 15SE +/- 671.66, N = 1546676454281. (CC) gcc options: -pthread -O3 -march=native -lm

simdjson

Throughput Test: Kostya

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.8.2Throughput Test: KostyaClang 12.0Clang 11.00.61881.23761.85642.47523.094SE +/- 0.01, N = 3SE +/- 0.00, N = 32.752.681. (CXX) g++ options: -O3 -march=native -pthread

oneDNN

Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUClang 11.0Clang 12.030060090012001500SE +/- 9.75, N = 3SE +/- 1.78, N = 31271.911305.10MIN: 1252.33MIN: 1294.761. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

AOM AV1

Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080pClang 12.0Clang 11.020406080100SE +/- 0.31, N = 3SE +/- 0.53, N = 3103.17100.551. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 1 - Mode: Read Only

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read OnlyClang 11.0Clang 12.05K10K15K20K25KSE +/- 289.16, N = 3SE +/- 303.43, N = 324943243101. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average LatencyClang 11.0Clang 12.00.00920.01840.02760.03680.046SE +/- 0.001, N = 3SE +/- 0.001, N = 30.0400.0411. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

JPEG XL

Input: PNG - Encode Speed: 8

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: PNG - Encode Speed: 8Clang 12.0Clang 11.00.18450.3690.55350.7380.9225SE +/- 0.00, N = 3SE +/- 0.00, N = 30.820.801. (CXX) g++ options: -O3 -march=native -funwind-tables -Xclang -mrelax-all -O2 -fPIE -pie -pthread -ldl

WebP Image Encode

Encode Settings: Quality 100, Lossless

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessClang 11.0Clang 12.0510152025SE +/- 0.13, N = 3SE +/- 0.02, N = 318.5719.021. (CC) gcc options: -fvisibility=hidden -O3 -march=native -pthread -lm -lpng16 -ljpeg

Opus Codec Encoding

WAV To Opus Encode

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus EncodeClang 11.0Clang 12.0246810SE +/- 0.002, N = 5SE +/- 0.013, N = 57.3927.5671. (CXX) g++ options: -O3 -march=native -logg -lm

oneDNN

Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUClang 11.0Clang 12.030060090012001500SE +/- 7.11, N = 3SE +/- 3.61, N = 31277.621307.49MIN: 1252.39MIN: 1293.381. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

FFTW

Build: Float + SSE - Size: 1D FFT Size 2048

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 2048Clang 12.0Clang 11.011K22K33K44K55KSE +/- 439.50, N = 3SE +/- 582.34, N = 351254500841. (CC) gcc options: -pthread -O3 -march=native -lm

libavif avifenc

Encoder Speed: 10, Lossless

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10, LosslessClang 12.0Clang 11.01.32282.64563.96845.29126.614SE +/- 0.013, N = 3SE +/- 0.011, N = 35.7465.8791. (CXX) g++ options: -O3 -fPIC -lm

FFTW

Build: Stock - Size: 1D FFT Size 1024

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 1024Clang 12.0Clang 11.02K4K6K8K10KSE +/- 27.10, N = 3SE +/- 35.53, N = 310805105641. (CC) gcc options: -pthread -O3 -march=native -lm

Tachyon

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.99b6Total TimeClang 12.0Clang 11.048121620SE +/- 0.06, N = 3SE +/- 0.02, N = 316.0516.411. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread

AOM AV1

Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KClang 12.0Clang 11.0918273645SE +/- 0.43, N = 3SE +/- 0.31, N = 338.1137.281. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUClang 11.0Clang 12.00.53281.06561.59842.13122.664SE +/- 0.02389, N = 3SE +/- 0.02100, N = 32.318592.36797MIN: 1.92MIN: 2.011. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUClang 11.0Clang 12.030060090012001500SE +/- 9.46, N = 3SE +/- 3.92, N = 31276.041302.70MIN: 1249.65MIN: 1289.861. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

libavif avifenc

Encoder Speed: 10

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10Clang 12.0Clang 11.00.77151.5432.31453.0863.8575SE +/- 0.014, N = 3SE +/- 0.010, N = 33.3613.4291. (CXX) g++ options: -O3 -fPIC -lm

SecureMark

Benchmark: SecureMark-TLS

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSClang 12.0Clang 11.060K120K180K240K300KSE +/- 1778.47, N = 3SE +/- 407.86, N = 32652042601191. (CC) gcc options: -pedantic -O3

WebP Image Encode

Encode Settings: Quality 100, Lossless, Highest Compression

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest CompressionClang 11.0Clang 12.0918273645SE +/- 0.08, N = 3SE +/- 0.07, N = 337.7338.451. (CC) gcc options: -fvisibility=hidden -O3 -march=native -pthread -lm -lpng16 -ljpeg

WebP Image Encode

Encode Settings: Quality 100

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100Clang 12.0Clang 11.00.5041.0081.5122.0162.52SE +/- 0.001, N = 3SE +/- 0.000, N = 32.1992.2401. (CC) gcc options: -fvisibility=hidden -O3 -march=native -pthread -lm -lpng16 -ljpeg

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUClang 11.0Clang 12.00.26380.52760.79141.05521.319SE +/- 0.00653, N = 3SE +/- 0.00458, N = 31.151401.17258MIN: 1.09MIN: 1.121. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

GraphicsMagick

Operation: HWB Color Space

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: HWB Color SpaceClang 11.0Clang 12.0130260390520650SE +/- 0.88, N = 3SE +/- 0.67, N = 36166051. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

C-Ray

Total Time - 4K, 16 Rays Per Pixel

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per PixelClang 11.0Clang 12.048121620SE +/- 0.01, N = 3SE +/- 0.02, N = 315.6015.871. (CC) gcc options: -lm -lpthread -O3 -march=native

AOM AV1

Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4KClang 11.0Clang 12.03691215SE +/- 0.03, N = 3SE +/- 0.10, N = 39.148.991. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

WebP2 Image Encode

Encode Settings: Quality 95, Compression Effort 7

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 95, Compression Effort 7Clang 11.0Clang 12.050100150200250SE +/- 0.66, N = 3SE +/- 0.07, N = 3203.63207.011. (CXX) g++ options: -O3 -march=native -fno-rtti -rdynamic -lpthread -ljpeg -lgif -lwebp -lwebpdemux

JPEG XL

Input: JPEG - Encode Speed: 5

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: JPEG - Encode Speed: 5Clang 12.0Clang 11.01530456075SE +/- 0.14, N = 3SE +/- 0.20, N = 366.6665.581. (CXX) g++ options: -O3 -march=native -funwind-tables -Xclang -mrelax-all -O2 -fPIE -pie -pthread -ldl

AOM AV1

Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4KClang 11.0Clang 12.01.11382.22763.34144.45525.569SE +/- 0.07, N = 3SE +/- 0.04, N = 34.954.871. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

FLAC Audio Encoding

WAV To FLAC

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.2WAV To FLACClang 12.0Clang 11.0246810SE +/- 0.007, N = 5SE +/- 0.006, N = 57.8547.9791. (CXX) g++ options: -O3 -march=native -logg -lm

JPEG XL

Input: JPEG - Encode Speed: 7

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: JPEG - Encode Speed: 7Clang 12.0Clang 11.01530456075SE +/- 0.16, N = 3SE +/- 0.08, N = 366.3865.431. (CXX) g++ options: -O3 -march=native -funwind-tables -Xclang -mrelax-all -O2 -fPIE -pie -pthread -ldl

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pClang 11.0Clang 12.0140280420560700SE +/- 5.55, N = 3SE +/- 3.01, N = 3652.74643.581. (CC) gcc options: -O3 -march=native -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

AOM AV1

Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080pClang 11.0Clang 12.0246810SE +/- 0.01, N = 3SE +/- 0.04, N = 37.207.101. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Gcrypt Library

OpenBenchmarking.orgSeconds, Fewer Is BetterGcrypt Library 1.9Clang 12.0Clang 11.050100150200250SE +/- 0.44, N = 3SE +/- 0.28, N = 3236.92240.211. (CC) gcc options: -O3 -march=native -fvisibility=hidden

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080pClang 12.0Clang 11.0110220330440550SE +/- 1.37, N = 3SE +/- 0.23, N = 3487.43481.051. (CC) gcc options: -O3 -fcommon -march=native -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

Liquid-DSP

Threads: 128 - Buffer Length: 256 - Filter Length: 57

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 57Clang 12.0Clang 11.0800M1600M2400M3200M4000MSE +/- 883804.91, N = 3SE +/- 1559202.08, N = 3364376666735965333331. (CC) gcc options: -O3 -march=native -pthread -lm -lc -lliquid

GraphicsMagick

Operation: Noise-Gaussian

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Noise-GaussianClang 11.0Clang 12.0100200300400500SE +/- 1.00, N = 34634571. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pClang 12.0Clang 11.0110220330440550SE +/- 0.73, N = 3SE +/- 1.76, N = 3488.23482.021. (CC) gcc options: -O3 -fcommon -march=native -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

x265

Video Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KClang 12.0Clang 11.0714212835SE +/- 0.23, N = 3SE +/- 0.25, N = 330.3229.941. (CXX) g++ options: -O3 -march=native -rdynamic -lpthread -lrt -ldl -lnuma

POV-Ray

Trace Time

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace TimeClang 12.0Clang 11.03691215SE +/- 0.041, N = 3SE +/- 0.032, N = 39.2969.4081. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSDL -lXpm -lSM -lICE -lX11 -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lpthread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average LatencyClang 12.0Clang 11.00.36590.73181.09771.46361.8295SE +/- 0.004, N = 3SE +/- 0.011, N = 31.6071.6261. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

FFTW

Build: Stock - Size: 2D FFT Size 4096

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 4096Clang 11.0Clang 12.015003000450060007500SE +/- 60.67, N = 3SE +/- 35.20, N = 36823.86744.11. (CC) gcc options: -pthread -O3 -march=native -lm

libavif avifenc

Encoder Speed: 2

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 2Clang 12.0Clang 11.0612182430SE +/- 0.06, N = 3SE +/- 0.06, N = 325.1825.471. (CXX) g++ options: -O3 -fPIC -lm

JPEG XL

Input: PNG - Encode Speed: 7

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: PNG - Encode Speed: 7Clang 12.0Clang 11.03691215SE +/- 0.05, N = 3SE +/- 0.02, N = 312.1512.011. (CXX) g++ options: -O3 -march=native -funwind-tables -Xclang -mrelax-all -O2 -fPIE -pie -pthread -ldl

Liquid-DSP

Threads: 1 - Buffer Length: 256 - Filter Length: 57

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 1 - Buffer Length: 256 - Filter Length: 57Clang 11.0Clang 12.012M24M36M48M60MSE +/- 40360.87, N = 3SE +/- 790005.27, N = 356307000556630001. (CC) gcc options: -O3 -march=native -pthread -lm -lc -lliquid

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 100 - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read WriteClang 12.0Clang 11.013K26K39K52K65KSE +/- 162.92, N = 3SE +/- 400.92, N = 362319616161. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

FFTW

Build: Stock - Size: 2D FFT Size 2048

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 2048Clang 11.0Clang 12.02K4K6K8K10KSE +/- 27.38, N = 3SE +/- 65.76, N = 37878.57789.91. (CC) gcc options: -pthread -O3 -march=native -lm

SVT-AV1

Encoder Mode: Enc Mode 0 - Input: 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 0 - Input: 1080pClang 12.0Clang 11.00.04120.08240.12360.16480.206SE +/- 0.000, N = 3SE +/- 0.000, N = 30.1830.1811. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

ONNX Runtime

Model: shufflenet-v2-10 - Device: OpenMP CPU

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: shufflenet-v2-10 - Device: OpenMP CPUClang 12.0Clang 11.02K4K6K8K10KSE +/- 88.25, N = 12SE +/- 102.76, N = 8990497971. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -ffunction-sections -fdata-sections -ldl -lrt

WebP Image Encode

Encode Settings: Quality 100, Highest Compression

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest CompressionClang 11.0Clang 12.0246810SE +/- 0.018, N = 3SE +/- 0.004, N = 36.2436.3091. (CC) gcc options: -fvisibility=hidden -O3 -march=native -pthread -lm -lpng16 -ljpeg

LZ4 Compression

Compression Level: 9 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedClang 11.0Clang 12.01122334455SE +/- 0.46, N = 3SE +/- 0.42, N = 349.0148.501. (CC) gcc options: -O3

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 1 - Mode: Read Write - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Write - Average LatencyClang 11.0Clang 12.00.06860.13720.20580.27440.343SE +/- 0.002, N = 3SE +/- 0.000, N = 30.3020.3051. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 1 - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read WriteClang 11.0Clang 12.07001400210028003500SE +/- 14.62, N = 3SE +/- 3.48, N = 3331232811. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUClang 12.0Clang 11.00.3280.6560.9841.3121.64SE +/- 0.00123, N = 3SE +/- 0.00568, N = 31.444251.45757MIN: 1.34MIN: 1.351. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

AOM AV1

Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080pClang 12.0Clang 11.0612182430SE +/- 0.27, N = 3SE +/- 0.13, N = 326.8526.611. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

x265

Video Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pClang 12.0Clang 11.01632486480SE +/- 0.49, N = 3SE +/- 0.49, N = 374.0073.361. (CXX) g++ options: -O3 -march=native -rdynamic -lpthread -lrt -ldl -lnuma

Liquid-DSP

Threads: 32 - Buffer Length: 256 - Filter Length: 57

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 32 - Buffer Length: 256 - Filter Length: 57Clang 11.0Clang 12.0300M600M900M1200M1500MSE +/- 1331665.62, N = 3SE +/- 2255610.29, N = 3157840000015648333331. (CC) gcc options: -O3 -march=native -pthread -lm -lc -lliquid

AOM AV1

Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4KClang 12.0Clang 11.0816243240SE +/- 0.48, N = 3SE +/- 0.22, N = 333.3933.141. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

GraphicsMagick

Operation: Enhanced

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: EnhancedClang 12.0Clang 11.02004006008001000SE +/- 1.86, N = 3107610681. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

dav1d

Video Input: Chimera 1080p

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Chimera 1080pClang 12.0Clang 11.030060090012001500SE +/- 2.95, N = 3SE +/- 6.69, N = 31198.221190.41MIN: 700.24 / MAX: 1494.16-lm - MIN: 685.16 / MAX: 1496.361. (CC) gcc options: -O3 -march=native -pthread

Liquid-DSP

Threads: 64 - Buffer Length: 256 - Filter Length: 57

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 64 - Buffer Length: 256 - Filter Length: 57Clang 12.0Clang 11.0700M1400M2100M2800M3500MSE +/- 6045475.81, N = 3SE +/- 2452436.43, N = 3307063333330513666671. (CC) gcc options: -O3 -march=native -pthread -lm -lc -lliquid

FFTW

Build: Float + SSE - Size: 2D FFT Size 2048

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 2048Clang 12.0Clang 11.07K14K21K28K35KSE +/- 77.17, N = 3SE +/- 146.10, N = 331935317411. (CC) gcc options: -pthread -O3 -march=native -lm

AOM AV1

Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080pClang 12.0Clang 11.0510152025SE +/- 0.05, N = 3SE +/- 0.15, N = 322.1322.001. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

FinanceBench

Benchmark: Bonds OpenMP

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Bonds OpenMPClang 12.0Clang 11.011K22K33K44K55KSE +/- 10.95, N = 3SE +/- 4.51, N = 351596.8751900.431. (CXX) g++ options: -O3 -march=native -fopenmp

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUClang 12.0Clang 11.00.0710.1420.2130.2840.355SE +/- 0.000321, N = 3SE +/- 0.000247, N = 30.3136890.315522MIN: 0.3MIN: 0.31. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

SVT-AV1

Encoder Mode: Enc Mode 8 - Input: 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 8 - Input: 1080pClang 12.0Clang 11.0306090120150SE +/- 0.10, N = 3SE +/- 0.46, N = 3118.07117.391. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

dav1d

Video Input: Summer Nature 1080p

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 1080pClang 11.0Clang 12.030060090012001500SE +/- 2.13, N = 3SE +/- 7.87, N = 31251.251244.11-lm - MIN: 556.46 / MAX: 1394.06MIN: 549.81 / MAX: 1390.031. (CC) gcc options: -O3 -march=native -pthread

Timed MrBayes Analysis

Primate Phylogeny Analysis

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisClang 11.0Clang 12.020406080100SE +/- 0.98, N = 3SE +/- 0.98, N = 388.6289.121. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -O3 -std=c99 -pedantic -march=native -lm

ASTC Encoder

Preset: Medium

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: MediumClang 11.0Clang 12.00.90131.80262.70393.60524.5065SE +/- 0.0013, N = 3SE +/- 0.0116, N = 33.98374.00581. (CXX) g++ options: -O3 -march=native -flto -pthread

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUClang 11.0Clang 12.00.11070.22140.33210.44280.5535SE +/- 0.001652, N = 3SE +/- 0.002843, N = 30.4892780.491940MIN: 0.46MIN: 0.471. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

LZ4 Compression

Compression Level: 3 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedClang 11.0Clang 12.01224364860SE +/- 0.33, N = 3SE +/- 0.80, N = 352.3552.071. (CC) gcc options: -O3

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 250 - Mode: Read Only

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read OnlyClang 12.0Clang 11.0200K400K600K800K1000KSE +/- 6289.60, N = 3SE +/- 13844.42, N = 3107120910655061. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

AOM AV1

Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KClang 12.0Clang 11.048121620SE +/- 0.11, N = 3SE +/- 0.11, N = 317.2217.131. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

QuantLib

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21Clang 12.0Clang 11.06001200180024003000SE +/- 1.92, N = 3SE +/- 1.01, N = 32653.82640.21. (CXX) g++ options: -O3 -march=native -rdynamic

LZ4 Compression

Compression Level: 3 - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression SpeedClang 12.0Clang 11.03K6K9K12K15KSE +/- 71.01, N = 3SE +/- 15.91, N = 313911.513840.31. (CC) gcc options: -O3

FFTW

Build: Float + SSE - Size: 2D FFT Size 4096

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096Clang 11.0Clang 12.05K10K15K20K25KSE +/- 220.77, N = 3SE +/- 348.10, N = 922913227971. (CC) gcc options: -pthread -O3 -march=native -lm

SVT-HEVC

Tuning: 7 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pClang 11.0Clang 12.080160240320400SE +/- 3.43, N = 3SE +/- 1.56, N = 3346.89345.301. (CC) gcc options: -O3 -march=native -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average LatencyClang 12.0Clang 11.00.05290.10580.15870.21160.2645SE +/- 0.001, N = 3SE +/- 0.003, N = 30.2340.2351. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080pClang 11.0Clang 12.080160240320400SE +/- 1.91, N = 3SE +/- 1.11, N = 3373.99372.491. (CC) gcc options: -O3 -fcommon -march=native -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

WebP Image Encode

Encode Settings: Default

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: DefaultClang 12.0Clang 11.00.30060.60120.90181.20241.503SE +/- 0.001, N = 3SE +/- 0.001, N = 31.3311.3361. (CC) gcc options: -fvisibility=hidden -O3 -march=native -pthread -lm -lpng16 -ljpeg

dav1d

Video Input: Summer Nature 4K

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 4KClang 11.0Clang 12.0120240360480600SE +/- 1.43, N = 3SE +/- 1.79, N = 3543.43541.56-lm - MIN: 256.75 / MAX: 593.99MIN: 252.01 / MAX: 587.531. (CC) gcc options: -O3 -march=native -pthread

Botan

Test: ChaCha20Poly1305 - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305 - DecryptClang 12.0Clang 11.02004006008001000SE +/- 4.64, N = 3SE +/- 0.16, N = 3843.40840.641. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Coremark

CoreMark Size 666 - Iterations Per Second

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondClang 11.0Clang 12.0400K800K1200K1600K2000KSE +/- 971.31, N = 3SE +/- 984.68, N = 31790837.011785466.281. (CC) gcc options: -O2 -O3 -march=native -lrt" -lrt

oneDNN

Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUClang 12.0Clang 11.00.2430.4860.7290.9721.215SE +/- 0.00199, N = 3SE +/- 0.00127, N = 31.077011.08011MIN: 1.04MIN: 1.031. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

libavif avifenc

Encoder Speed: 6

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6Clang 12.0Clang 11.03691215SE +/- 0.014, N = 3SE +/- 0.022, N = 39.5109.5361. (CXX) g++ options: -O3 -fPIC -lm

Botan

Test: ChaCha20Poly1305

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305Clang 12.0Clang 11.02004006008001000SE +/- 4.85, N = 3SE +/- 0.62, N = 3850.50848.241. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

FinanceBench

Benchmark: Repo OpenMP

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Repo OpenMPClang 11.0Clang 12.07K14K21K28K35KSE +/- 0.81, N = 3SE +/- 64.93, N = 333178.5033246.841. (CXX) g++ options: -O3 -march=native -fopenmp

SVT-HEVC

Tuning: 1 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pClang 12.0Clang 11.0918273645SE +/- 0.17, N = 3SE +/- 0.09, N = 341.0941.011. (CC) gcc options: -O3 -march=native -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

ASTC Encoder

Preset: Exhaustive

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: ExhaustiveClang 12.0Clang 11.0510152025SE +/- 0.01, N = 3SE +/- 0.01, N = 318.9919.031. (CXX) g++ options: -O3 -march=native -flto -pthread

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SharpenClang 12.0Clang 11.01302603905206506146131. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

FFTW

Build: Float + SSE - Size: 2D FFT Size 1024

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 1024Clang 12.0Clang 11.08K16K24K32K40KSE +/- 165.99, N = 3SE +/- 530.09, N = 436239361811. (CC) gcc options: -pthread -O3 -march=native -lm

WebP2 Image Encode

Encode Settings: Default

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: DefaultClang 12.0Clang 11.00.61721.23441.85162.46883.086SE +/- 0.027, N = 3SE +/- 0.031, N = 32.7392.7431. (CXX) g++ options: -O3 -march=native -fno-rtti -rdynamic -lpthread -ljpeg -lgif -lwebp -lwebpdemux

WebP2 Image Encode

Encode Settings: Quality 75, Compression Effort 7

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 75, Compression Effort 7Clang 12.0Clang 11.020406080100SE +/- 0.10, N = 3SE +/- 0.10, N = 3109.53109.641. (CXX) g++ options: -O3 -march=native -fno-rtti -rdynamic -lpthread -ljpeg -lgif -lwebp -lwebpdemux

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUClang 11.0Clang 12.00.17540.35080.52620.70160.877SE +/- 0.001200, N = 3SE +/- 0.004246, N = 30.7791010.779776MIN: 0.73MIN: 0.731. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

LAME MP3 Encoding

WAV To MP3

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3Clang 11.0Clang 12.0246810SE +/- 0.021, N = 3SE +/- 0.003, N = 38.2508.2561. (CC) gcc options: -O3 -pipe -march=native -lncurses -lm

FFTW

Build: Stock - Size: 1D FFT Size 32

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 32Clang 12.0Clang 11.03K6K9K12K15KSE +/- 24.25, N = 3SE +/- 20.33, N = 313333133241. (CC) gcc options: -pthread -O3 -march=native -lm

oneDNN

Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUClang 12.0Clang 11.00.2420.4840.7260.9681.21SE +/- 0.00286, N = 3SE +/- 0.00395, N = 31.075071.07577MIN: 0.87MIN: 0.861. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

Botan

Test: Blowfish - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Blowfish - DecryptClang 12.0Clang 11.080160240320400SE +/- 0.04, N = 3SE +/- 2.03, N = 3351.28351.081. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

SciMark

Computational Test: Monte Carlo

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Monte CarloClang 12.0Clang 11.0150300450600750SE +/- 0.40, N = 3SE +/- 0.40, N = 3675.13674.861. (CC) gcc options: -O3 -march=native -lm

ASTC Encoder

Preset: Thorough

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: ThoroughClang 12.0Clang 11.0246810SE +/- 0.0028, N = 3SE +/- 0.0026, N = 36.76476.76741. (CXX) g++ options: -O3 -march=native -flto -pthread

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 100 - Mode: Read Only

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read OnlyClang 11.0Clang 12.0200K400K600K800K1000KSE +/- 1740.88, N = 3SE +/- 720.87, N = 3106936710690221. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

libavif avifenc

Encoder Speed: 0

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 0Clang 12.0Clang 11.01122334455SE +/- 0.04, N = 3SE +/- 0.07, N = 347.8847.891. (CXX) g++ options: -O3 -fPIC -lm

LZ4 Compression

Compression Level: 9 - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedClang 11.0Clang 12.03K6K9K12K15KSE +/- 23.21, N = 3SE +/- 65.90, N = 313927.913926.51. (CC) gcc options: -O3

SciMark

Computational Test: Jacobi Successive Over-Relaxation

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Jacobi Successive Over-RelaxationClang 12.0Clang 11.0400800120016002000SE +/- 0.08, N = 3SE +/- 0.12, N = 31785.501785.421. (CC) gcc options: -O3 -march=native -lm

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average LatencyClang 12.0Clang 11.00.02120.04240.06360.08480.106SE +/- 0.000, N = 3SE +/- 0.000, N = 30.0940.0941. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

AOM AV1

Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080pClang 11.0Clang 12.00.11930.23860.35790.47720.5965SE +/- 0.00, N = 3SE +/- 0.00, N = 30.530.531. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

AOM AV1

Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4KClang 11.0Clang 12.00.04730.09460.14190.18920.2365SE +/- 0.00, N = 3SE +/- 0.00, N = 30.210.211. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

ONNX Runtime

Model: super-resolution-10 - Device: OpenMP CPU

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: super-resolution-10 - Device: OpenMP CPUClang 11.0Clang 12.010002000300040005000SE +/- 169.87, N = 9SE +/- 126.29, N = 12452344561. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -ffunction-sections -fdata-sections -ldl -lrt

ONNX Runtime

Model: bertsquad-10 - Device: OpenMP CPU

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: bertsquad-10 - Device: OpenMP CPUClang 12.0Clang 11.0110220330440550SE +/- 10.30, N = 12SE +/- 5.55, N = 34984711. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -ffunction-sections -fdata-sections -ldl -lrt

ViennaCL

Test: CPU BLAS - dGEMV-N

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-NClang 12.0Clang 11.01530456075SE +/- 2.22, N = 12SE +/- 3.65, N = 1569.151.21. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOTClang 11.0Clang 12.02004006008001000SE +/- 1.49, N = 15SE +/- 17.06, N = 129338191. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPYClang 11.0Clang 12.02004006008001000SE +/- 1.59, N = 15SE +/- 20.06, N = 1210438781. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPYClang 11.0Clang 12.0400800120016002000SE +/- 8.32, N = 15SE +/- 15.32, N = 1118776041. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - sDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOTClang 11.0Clang 12.0100200300400500SE +/- 38.96, N = 15SE +/- 35.24, N = 124624341. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - sAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPYClang 11.0Clang 12.090180270360450SE +/- 34.43, N = 15SE +/- 15.69, N = 124123571. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - sCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPYClang 11.0Clang 12.0110220330440550SE +/- 36.50, N = 15SE +/- 15.30, N = 124954711. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -rdynamic -lOpenCL

GraphicsMagick

Operation: Resizing

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: ResizingClang 12.0Clang 11.05001000150020002500SE +/- 41.63, N = 12SE +/- 27.29, N = 3213620341. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

FFTW

Build: Float + SSE - Size: 1D FFT Size 1024

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 1024Clang 11.0Clang 12.011K22K33K44K55KSE +/- 585.78, N = 3SE +/- 952.64, N = 1250740503501. (CC) gcc options: -pthread -O3 -march=native -lm

Phoronix Test Suite v10.8.5