Apple M1 Compilers

Apple M1 compiler testing for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2204107-NE-APPLEM1CO68
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Limit displaying results to tests within:

Audio Encoding 3 Tests
C/C++ Compiler Tests 9 Tests
Compression Tests 2 Tests
CPU Massive 11 Tests
Creator Workloads 11 Tests
Cryptography 3 Tests
Encoding 4 Tests
HPC - High Performance Computing 3 Tests
Imaging 3 Tests
Common Kernel Benchmarks 3 Tests
Machine Learning 2 Tests
Multi-Core 7 Tests
NVIDIA GPU Compute 2 Tests
Programmer / Developer System Benchmarks 2 Tests
Raytracing 2 Tests
Renderers 3 Tests
Server 2 Tests
Server CPU Tests 7 Tests
Single-Threaded 4 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
GCC 11.2.0
April 08 2022
  5 Hours, 11 Minutes
Clang 13.0.1
April 09 2022
  5 Hours, 35 Minutes
Invert Hiding All Results Option
  5 Hours, 23 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


Apple M1 CompilersOpenBenchmarking.orgPhoronix Test SuiteApple M1 @ 2.06GHz (4 Cores / 8 Threads)Apple Mac mini (M1 2020)8GB251GB APPLE SSD AP0256Q + 2 x 0GB APPLE SSD AP0256QllvmpipeBroadcom NetXtreme BCM57762 PCIe + Broadcom BRCM4378 + Broadcom Device 5f69Arch Linux ARM5.17.0-rc7-asahi-next-20220310-5-2-ARCH (aarch64)KDE Plasma 5.24.4X Server 1.21.1.34.5 Mesa 22.0.1 (LLVM 13.0.1 128 bits)GCC 11.2.0 + Clang 13.0.1Clang 13.0.1ext41920x1080ProcessorMotherboardMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerOpenGLCompilersFile-SystemScreen ResolutionApple M1 Compilers PerformanceSystem Logs- CXXFLAGS="-O3 -flto" CFLAGS="-O3 -flto"- GCC 11.2.0: --build=aarch64-unknown-linux-gnu --disable-libssp --disable-libstdcxx-pch --disable-multilib --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-fix-cortex-a53-835769 --enable-fix-cortex-a53-843419 --enable-gnu-indirect-function --enable-gnu-unique-object --enable-languages=c,c++,fortran,go,lto,objc,obj-c++,d --enable-lto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-unknown-linux-gnu --mandir=/usr/share/man --with-arch=armv8-a --with-isl --with-linker-hash-style=gnu - MQ-DEADLINE / relatime,rw / Block Size: 4096- Scaling Governor: apple-cpufreq schedutil- Python 3.10.4- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected

GCC 11.2.0 vs. Clang 13.0.1 ComparisonPhoronix Test SuiteBaseline+43.2%+43.2%+86.4%+86.4%+129.6%+129.6%74.9%35.4%31.8%31.7%31.7%30%28.3%19.6%15.4%6.1%5.7%5.1%4.8%4.3%3.9%3.6%3.3%3.2%3.2%3.1%3.1%3.1%2.7%2.2%CPU - resnet50172.7%CPU - alexnet168.1%CPU - resnet18148.7%CPU - vgg16138.1%CPU - mnasnet132.5%CPU - efficientnet-b0127.5%CPU-v2-v2 - mobilenet-v2123.8%CPU-v3-v3 - mobilenet-v3106%CPU - yolov4-tiny90.5%CPU - googlenet88.5%Vector MathCPU - shufflenet-v273.3%Unkeyed Algorithms59.4%CPU - mobilenet54.2%CPU - squeezenet_ssd41.3%CPU - regnety_400m37.4%Total Time - 4.1.R.P.P36.3%Keyed Algorithms35.7%Memory Copying4 - 256 - 572 - 256 - 571 - 256 - 578 - 256 - 57Matrix Math2048 x 2048 - Total Time21.6%CoreMark Size 666 - I.P.S21.3%WAV To FLACAll Algorithms16%Trace Time212.7%WAV To WavPack12.3%WAV To MP312.2%I.E.C.P.K.AP.P.S5.8%3, Long Mode - Compression Speed5.6%SHA256T.T.S.S5.1%6, Lossless6D.T4.2%19 - D.S9 - D.S3.6%3 - D.S3.6%N.C.P.M3 - D.S19, Long Mode - D.S3, Long Mode - D.S8 - Compression Speed3.1%10, Lossless8, Long Mode - D.S8 - D.STimed Time - Size 1,0003%Eigen19 - Compression SpeedNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNStress-NGNCNNCrypto++NCNNNCNNNCNNC-RayCrypto++Stress-NGLiquid-DSPLiquid-DSPLiquid-DSPLiquid-DSPStress-NGAOBenchCoremarkFLAC Audio EncodingCrypto++POV-Raylibavif avifencWavPack Audio EncodingLAME MP3 EncodingCrypto++Himeno BenchmarkZstd Compressionlibavif avifencOpenSSLeSpeak-NG Speech Enginelibavif avifenclibavif avifenclibjpeg-turbo tjbenchZstd CompressionLZ4 CompressionLZ4 CompressionOpenJPEGZstd CompressionZstd CompressionZstd CompressionZstd Compressionlibavif avifencZstd CompressionZstd CompressionSQLite SpeedtestLeelaChessZeroZstd CompressionGCC 11.2.0Clang 13.0.1

Apple M1 Compilerscryptopp: All Algorithmscryptopp: Keyed Algorithmscryptopp: Unkeyed Algorithmscryptopp: Integer + Elliptic Curve Public Key Algorithmslczero: Eigenxmrig: Monero - 1Mxmrig: Wownero - 1Mcompress-lz4: 1 - Compression Speedcompress-lz4: 1 - Decompression Speedcompress-lz4: 3 - Compression Speedcompress-lz4: 3 - Decompression Speedcompress-lz4: 9 - Compression Speedcompress-lz4: 9 - Decompression Speedcompress-zstd: 3 - Compression Speedcompress-zstd: 3 - Decompression Speedcompress-zstd: 8 - Compression Speedcompress-zstd: 8 - Decompression Speedcompress-zstd: 19 - Compression Speedcompress-zstd: 19 - Decompression Speedcompress-zstd: 3, Long Mode - Compression Speedcompress-zstd: 3, Long Mode - Decompression Speedcompress-zstd: 8, Long Mode - Compression Speedcompress-zstd: 8, Long Mode - Decompression Speedcompress-zstd: 19, Long Mode - Compression Speedcompress-zstd: 19, Long Mode - Decompression Speedcoremark: CoreMark Size 666 - Iterations Per Secondhimeno: Poisson Pressure Solveravifenc: 0avifenc: 2avifenc: 6avifenc: 6, Losslessavifenc: 10, Losslessc-ray: Total Time - 4K, 16 Rays Per Pixelpovray: Trace Timeprimesieve: 1e12 Prime Number Generationaobench: 2048 x 2048 - Total Timeencode-flac: WAV To FLACencode-mp3: WAV To MP3espeak: Text-To-Speech Synthesisopenjpeg: NASA Curiosity Panorama M34openssl: SHA256openssl: RSA4096openssl: RSA4096liquid-dsp: 1 - 256 - 57liquid-dsp: 2 - 256 - 57liquid-dsp: 4 - 256 - 57liquid-dsp: 8 - 256 - 57tjbench: Decompression Throughputsqlite-speedtest: Timed Time - Size 1,000draco: Liondraco: Church Facadestress-ng: Cryptostress-ng: IO_uringstress-ng: Matrix Mathstress-ng: Vector Mathstress-ng: Memory Copyingstress-ng: Socket Activityncnn: CPU - mobilenetncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - shufflenet-v2ncnn: CPU - mnasnetncnn: CPU - efficientnet-b0ncnn: CPU - googlenetncnn: CPU - vgg16ncnn: CPU - resnet18ncnn: CPU - alexnetncnn: CPU - resnet50ncnn: CPU - yolov4-tinyncnn: CPU - squeezenet_ssdncnn: CPU - regnety_400mencode-wavpack: WAV To WavPackGCC 11.2.0Clang 13.0.1954.956113508.836448539.2818271766.98588012632247.22798.221909.4527018.551.9917490.948.9417478.53341.23850.2721.54016.422.73546.2240.04221.1693.04416.318.83765.4179896.5994117577.316534287.397143.44214.09415.6536.07064.43772.01729.11827.45870.6487.23922.2895389080596910501408.599370.52877866757611000115230000151120000206.17735051.372374756491511.75144281.6723588.9623954.102763.254331.7114.402.612.342.172.524.1813.3233.787.3111.8117.1617.2014.265.8817.205823.153532374.896175338.4083691875.52352012972209.72804.821875.5426736.451.3216877.449.8916863.33301.13977.7699.64141.023.23684.3253.74356.6703.44553.418.83887.3148361.3624407158.970486303.550161.61213.51614.9295.88787.82462.41629.62633.40259.0748.12423.4295202484745273501391.499445.43789733375898667151820000196510000197.94522552.900377257221527.17147040.9830254.2141899.943741.174313.4822.215.844.823.765.869.5125.1180.4418.1831.6646.7932.7720.158.0819.320OpenBenchmarking.org

Crypto++

Crypto++ is a C++ class library of cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/second, More Is BetterCrypto++ 8.2Test: All AlgorithmsClang 13.0.1GCC 11.2.02004006008001000SE +/- 0.17, N = 3SE +/- 0.23, N = 3823.15954.961. (CXX) g++ options: -O3 -flto -fPIC -pthread -pipe

OpenBenchmarking.orgMiB/second, More Is BetterCrypto++ 8.2Test: Keyed AlgorithmsClang 13.0.1GCC 11.2.0110220330440550SE +/- 1.08, N = 3SE +/- 0.07, N = 3374.90508.841. (CXX) g++ options: -O3 -flto -fPIC -pthread -pipe

OpenBenchmarking.orgMiB/second, More Is BetterCrypto++ 8.2Test: Unkeyed AlgorithmsClang 13.0.1GCC 11.2.0120240360480600SE +/- 0.01, N = 3SE +/- 0.04, N = 3338.41539.281. (CXX) g++ options: -O3 -flto -fPIC -pthread -pipe

OpenBenchmarking.orgMiB/second, More Is BetterCrypto++ 8.2Test: Integer + Elliptic Curve Public Key AlgorithmsClang 13.0.1GCC 11.2.0400800120016002000SE +/- 1.78, N = 3SE +/- 0.67, N = 31875.521766.991. (CXX) g++ options: -O3 -flto -fPIC -pthread -pipe

LeelaChessZero

LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: EigenClang 13.0.1GCC 11.2.030060090012001500SE +/- 18.26, N = 3SE +/- 10.69, N = 3129712631. (CXX) g++ options: -flto -O3 -pthread

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.12.1Variant: Monero - Hash Count: 1MClang 13.0.1GCC 11.2.05001000150020002500SE +/- 7.70, N = 3SE +/- 9.05, N = 32209.72247.2-funroll-loops-static-libgcc -static-libstdc++1. (CXX) g++ options: -O3 -flto -fexceptions -fno-rtti -Ofast -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.12.1Variant: Wownero - Hash Count: 1MClang 13.0.1GCC 11.2.06001200180024003000SE +/- 1.95, N = 3SE +/- 1.83, N = 32804.82798.2-funroll-loops-static-libgcc -static-libstdc++1. (CXX) g++ options: -O3 -flto -fexceptions -fno-rtti -Ofast -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Compression SpeedClang 13.0.1GCC 11.2.05K10K15K20K25KSE +/- 3.05, N = 3SE +/- 5.50, N = 321875.5421909.451. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Decompression SpeedClang 13.0.1GCC 11.2.06K12K18K24K30KSE +/- 8.86, N = 3SE +/- 1.47, N = 326736.427018.51. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedClang 13.0.1GCC 11.2.01224364860SE +/- 0.09, N = 3SE +/- 0.05, N = 351.3251.991. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression SpeedClang 13.0.1GCC 11.2.04K8K12K16K20KSE +/- 3.46, N = 3SE +/- 0.40, N = 316877.417490.91. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedClang 13.0.1GCC 11.2.01122334455SE +/- 0.12, N = 3SE +/- 0.09, N = 349.8948.941. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedClang 13.0.1GCC 11.2.04K8K12K16K20KSE +/- 3.18, N = 3SE +/- 1.03, N = 316863.317478.51. (CC) gcc options: -O3

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Compression SpeedClang 13.0.1GCC 11.2.07001400210028003500SE +/- 39.46, N = 3SE +/- 6.19, N = 33301.13341.21. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Decompression SpeedClang 13.0.1GCC 11.2.09001800270036004500SE +/- 0.75, N = 3SE +/- 0.87, N = 33977.73850.21. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Compression SpeedClang 13.0.1GCC 11.2.0160320480640800SE +/- 4.97, N = 3SE +/- 3.70, N = 3699.6721.51. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Decompression SpeedClang 13.0.1GCC 11.2.09001800270036004500SE +/- 3.02, N = 3SE +/- 1.95, N = 34141.04016.41. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression SpeedClang 13.0.1GCC 11.2.0612182430SE +/- 0.17, N = 3SE +/- 0.07, N = 323.222.71. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression SpeedClang 13.0.1GCC 11.2.08001600240032004000SE +/- 1.62, N = 3SE +/- 0.15, N = 33684.33546.21. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3, Long Mode - Compression SpeedClang 13.0.1GCC 11.2.060120180240300SE +/- 3.51, N = 3SE +/- 2.00, N = 15253.7240.01. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3, Long Mode - Decompression SpeedClang 13.0.1GCC 11.2.09001800270036004500SE +/- 0.40, N = 3SE +/- 0.25, N = 154356.64221.11. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8, Long Mode - Compression SpeedClang 13.0.1GCC 11.2.0150300450600750SE +/- 2.38, N = 3SE +/- 2.35, N = 3703.4693.01. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8, Long Mode - Decompression SpeedClang 13.0.1GCC 11.2.010002000300040005000SE +/- 3.13, N = 3SE +/- 1.55, N = 34553.44416.31. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression SpeedClang 13.0.1GCC 11.2.0510152025SE +/- 0.21, N = 4SE +/- 0.13, N = 318.818.81. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Decompression SpeedClang 13.0.1GCC 11.2.08001600240032004000SE +/- 0.69, N = 4SE +/- 0.92, N = 33887.33765.41. (CC) gcc options: -O3 -flto -pthread -lz -llzma -llz4

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondClang 13.0.1GCC 11.2.040K80K120K160K200KSE +/- 146.27, N = 3SE +/- 59.94, N = 3148361.36179896.601. (CC) gcc options: -O2 -O3 -flto -lrt" -lrt

Himeno Benchmark

The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverClang 13.0.1GCC 11.2.016003200480064008000SE +/- 3.44, N = 3SE +/- 2.64, N = 37158.977577.321. (CC) gcc options: -O3 -flto

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 0Clang 13.0.1GCC 11.2.070140210280350SE +/- 0.16, N = 3SE +/- 0.92, N = 3303.55287.401. (CXX) g++ options: -O3 -fPIC -flto -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 2Clang 13.0.1GCC 11.2.04080120160200SE +/- 0.72, N = 3SE +/- 0.32, N = 3161.61143.441. (CXX) g++ options: -O3 -fPIC -flto -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 6Clang 13.0.1GCC 11.2.048121620SE +/- 0.15, N = 5SE +/- 0.14, N = 613.5214.091. (CXX) g++ options: -O3 -fPIC -flto -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 6, LosslessClang 13.0.1GCC 11.2.048121620SE +/- 0.21, N = 3SE +/- 0.18, N = 314.9315.651. (CXX) g++ options: -O3 -fPIC -flto -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 10, LosslessClang 13.0.1GCC 11.2.0246810SE +/- 0.047, N = 3SE +/- 0.049, N = 35.8876.0701. (CXX) g++ options: -O3 -fPIC -flto -lm

C-Ray

This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per PixelClang 13.0.1GCC 11.2.020406080100SE +/- 0.05, N = 3SE +/- 0.04, N = 387.8264.441. (CC) gcc options: -lm -lpthread -O3 -flto

POV-Ray

This is a test of POV-Ray, the Persistence of Vision Raytracer. POV-Ray is used to create 3D graphics using ray-tracing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace TimeClang 13.0.1GCC 11.2.01632486480SE +/- 0.64, N = 5SE +/- 0.85, N = 462.4272.02-R/usr/lib1. (CXX) g++ options: -pipe -O3 -ffast-math -flto -lSDL -lpthread -lXpm -lSM -lICE -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

Primesieve

Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 7.71e12 Prime Number GenerationClang 13.0.1GCC 11.2.0714212835SE +/- 0.03, N = 3SE +/- 0.01, N = 329.6329.121. (CXX) g++ options: -O3 -flto

AOBench

AOBench is a lightweight ambient occlusion renderer, written in C. The test profile is using a size of 2048 x 2048. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAOBenchSize: 2048 x 2048 - Total TimeClang 13.0.1GCC 11.2.0816243240SE +/- 0.01, N = 3SE +/- 0.03, N = 333.4027.461. (CC) gcc options: -lm -O3 -flto

FLAC Audio Encoding

This test times how long it takes to encode a sample WAV file to FLAC format ten times. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.3WAV To FLACClang 13.0.1GCC 11.2.01632486480SE +/- 0.01, N = 5SE +/- 0.01, N = 559.0770.65-fvisibility=hidden1. (CXX) g++ options: -O3 -flto -logg -lm

LAME MP3 Encoding

LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3Clang 13.0.1GCC 11.2.0246810SE +/- 0.005, N = 3SE +/- 0.011, N = 38.1247.239-pipe1. (CC) gcc options: -O3 -flto -lncurses -lm

eSpeak-NG Speech Engine

This test times how long it takes the eSpeak speech synthesizer to read Project Gutenberg's The Outline of Science and output to a WAV file. This test profile is now tracking the eSpeak-NG version of eSpeak. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech SynthesisClang 13.0.1GCC 11.2.0612182430SE +/- 0.03, N = 4SE +/- 0.03, N = 423.4322.291. (CC) gcc options: -O3 -flto -std=c99 -lpthread -lm

OpenJPEG

OpenJPEG is an open-source JPEG 2000 codec written in the C programming language. The default input for this test profile is the NASA/JPL-Caltech/MSSS Curiosity panorama 717MB TIFF image file converting to JPEG2000 format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenJPEG 2.4Encode: NASA Curiosity Panorama M34Clang 13.0.1GCC 11.2.012K24K36K48K60KSE +/- 161.48, N = 3SE +/- 92.73, N = 352024538901. (CXX) g++ options: -O3 -flto -rdynamic

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.0Algorithm: SHA256Clang 13.0.1GCC 11.2.02000M4000M6000M8000M10000MSE +/- 3887401.32, N = 3SE +/- 12283962.01, N = 384745273508059691050-Qunused-arguments1. (CC) gcc options: -pthread -O3 -flto -lssl -lcrypto -ldl

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096Clang 13.0.1GCC 11.2.030060090012001500SE +/- 0.15, N = 3SE +/- 0.78, N = 31391.41408.5-Qunused-arguments1. (CC) gcc options: -pthread -O3 -flto -lssl -lcrypto -ldl

OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096Clang 13.0.1GCC 11.2.020K40K60K80K100KSE +/- 16.80, N = 3SE +/- 18.59, N = 399445.499370.5-Qunused-arguments1. (CC) gcc options: -pthread -O3 -flto -lssl -lcrypto -ldl

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 1 - Buffer Length: 256 - Filter Length: 57Clang 13.0.1GCC 11.2.08M16M24M32M40MSE +/- 2905.93, N = 3SE +/- 3527.67, N = 337897333287786671. (CC) gcc options: -O3 -flto -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 2 - Buffer Length: 256 - Filter Length: 57Clang 13.0.1GCC 11.2.016M32M48M64M80MSE +/- 1763.83, N = 3SE +/- 2081.67, N = 375898667576110001. (CC) gcc options: -O3 -flto -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 4 - Buffer Length: 256 - Filter Length: 57Clang 13.0.1GCC 11.2.030M60M90M120M150MSE +/- 0.00, N = 3SE +/- 0.00, N = 31518200001152300001. (CC) gcc options: -O3 -flto -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 8 - Buffer Length: 256 - Filter Length: 57Clang 13.0.1GCC 11.2.040M80M120M160M200MSE +/- 0.00, N = 3SE +/- 0.00, N = 31965100001511200001. (CC) gcc options: -O3 -flto -pthread -lm -lc -lliquid

libjpeg-turbo tjbench

tjbench is a JPEG decompression/compression benchmark that is part of libjpeg-turbo, a JPEG image codec library optimized for SIMD instructions on modern CPU architectures. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.1.0Test: Decompression ThroughputClang 13.0.1GCC 11.2.050100150200250SE +/- 2.27, N = 3SE +/- 0.03, N = 3197.95206.18-lm1. (CC) gcc options: -O3 -flto -rdynamic

SQLite Speedtest

This is a benchmark of SQLite's speedtest1 benchmark program with an increased problem size of 1,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Clang 13.0.1GCC 11.2.01224364860SE +/- 0.07, N = 3SE +/- 0.04, N = 352.9051.371. (CC) gcc options: -O3 -flto -lz

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.0Model: LionClang 13.0.1GCC 11.2.08001600240032004000SE +/- 0.58, N = 3SE +/- 2.73, N = 3377237471. (CXX) g++ options: -O3 -flto

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.0Model: Church FacadeClang 13.0.1GCC 11.2.012002400360048006000SE +/- 3.79, N = 3SE +/- 7.21, N = 3572256491. (CXX) g++ options: -O3 -flto

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.13.02Test: CryptoClang 13.0.1GCC 11.2.030060090012001500SE +/- 0.94, N = 3SE +/- 16.47, N = 31527.171511.751. (CC) gcc options: -O3 -flto -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -pthread -lkmod -lc -latomic

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.13.02Test: IO_uringClang 13.0.1GCC 11.2.030K60K90K120K150KSE +/- 271.95, N = 3SE +/- 28.54, N = 3147040.98144281.671. (CC) gcc options: -O3 -flto -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -pthread -lkmod -lc -latomic

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.13.02Test: Matrix MathClang 13.0.1GCC 11.2.06K12K18K24K30KSE +/- 0.69, N = 3SE +/- 332.61, N = 330254.2123588.961. (CC) gcc options: -O3 -flto -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -pthread -lkmod -lc -latomic

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.13.02Test: Vector MathClang 13.0.1GCC 11.2.09K18K27K36K45KSE +/- 2.19, N = 3SE +/- 195.44, N = 1541899.9423954.101. (CC) gcc options: -O3 -flto -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -pthread -lkmod -lc -latomic

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.13.02Test: Memory CopyingClang 13.0.1GCC 11.2.08001600240032004000SE +/- 15.21, N = 3SE +/- 6.71, N = 33741.172763.251. (CC) gcc options: -O3 -flto -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -pthread -lkmod -lc -latomic

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.13.02Test: Socket ActivityClang 13.0.1GCC 11.2.09001800270036004500SE +/- 13.20, N = 3SE +/- 4.58, N = 34313.484331.711. (CC) gcc options: -O3 -flto -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -pthread -lkmod -lc -latomic

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mobilenetClang 13.0.1GCC 11.2.0510152025SE +/- 0.01, N = 3SE +/- 0.17, N = 322.2114.40MIN: 22.15 / MAX: 22.25-lgomp -lpthread - MIN: 9.21 / MAX: 25.21. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU-v2-v2 - Model: mobilenet-v2Clang 13.0.1GCC 11.2.01.3142.6283.9425.2566.57SE +/- 0.01, N = 3SE +/- 0.05, N = 35.842.61MIN: 5.81 / MAX: 5.87-lgomp -lpthread - MIN: 2.48 / MAX: 12.21. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU-v3-v3 - Model: mobilenet-v3Clang 13.0.1GCC 11.2.01.08452.1693.25354.3385.4225SE +/- 0.01, N = 3SE +/- 0.01, N = 34.822.34MIN: 4.8 / MAX: 4.85-lgomp -lpthread - MIN: 2.32 / MAX: 2.491. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: shufflenet-v2Clang 13.0.1GCC 11.2.00.8461.6922.5383.3844.23SE +/- 0.00, N = 3SE +/- 0.01, N = 33.762.17MAX: 3.85-lgomp -lpthread - MIN: 2.15 / MAX: 2.481. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mnasnetClang 13.0.1GCC 11.2.01.31852.6373.95555.2746.5925SE +/- 0.01, N = 2SE +/- 0.01, N = 35.862.52MIN: 5.84 / MAX: 5.87-lgomp -lpthread - MIN: 2.48 / MAX: 2.841. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: efficientnet-b0Clang 13.0.1GCC 11.2.03691215SE +/- 0.00, N = 3SE +/- 0.02, N = 39.514.18MIN: 9.47 / MAX: 9.67-lgomp -lpthread - MIN: 4.13 / MAX: 8.11. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: googlenetClang 13.0.1GCC 11.2.0612182430SE +/- 0.01, N = 3SE +/- 0.10, N = 325.1113.32MIN: 25.07 / MAX: 25.16-lgomp -lpthread - MIN: 9.14 / MAX: 21.971. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: vgg16Clang 13.0.1GCC 11.2.020406080100SE +/- 0.01, N = 3SE +/- 0.14, N = 380.4433.78MIN: 80.22 / MAX: 80.95-lgomp -lpthread - MIN: 30.68 / MAX: 45.721. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: resnet18Clang 13.0.1GCC 11.2.048121620SE +/- 0.01, N = 3SE +/- 0.04, N = 318.187.31MIN: 18.14 / MAX: 18.23-lgomp -lpthread - MIN: 6.17 / MAX: 16.921. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: alexnetClang 13.0.1GCC 11.2.0714212835SE +/- 0.00, N = 3SE +/- 0.10, N = 331.6611.81MIN: 31.62 / MAX: 33.42-lgomp -lpthread - MIN: 9.48 / MAX: 21.581. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: resnet50Clang 13.0.1GCC 11.2.01122334455SE +/- 0.01, N = 3SE +/- 0.08, N = 346.7917.16MIN: 46.7 / MAX: 46.9-lgomp -lpthread - MIN: 15.54 / MAX: 27.861. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: yolov4-tinyClang 13.0.1GCC 11.2.0816243240SE +/- 0.00, N = 3SE +/- 0.07, N = 332.7717.20MIN: 32.68 / MAX: 32.88-lgomp -lpthread - MIN: 14.01 / MAX: 27.331. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: squeezenet_ssdClang 13.0.1GCC 11.2.0510152025SE +/- 0.00, N = 3SE +/- 0.17, N = 320.1514.26MIN: 20.08 / MAX: 20.21-lgomp -lpthread - MIN: 9.6 / MAX: 28.571. (CXX) g++ options: -O3 -flto -rdynamic

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: regnety_400mClang 13.0.1GCC 11.2.0246810SE +/- 0.00, N = 3SE +/- 0.03, N = 38.085.88MIN: 8.05 / MAX: 8.15-lgomp -lpthread - MIN: 5.78 / MAX: 8.621. (CXX) g++ options: -O3 -flto -rdynamic

WavPack Audio Encoding

This test times how long it takes to encode a sample WAV file to WavPack format with very high quality settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPackClang 13.0.1GCC 11.2.0510152025SE +/- 0.00, N = 5SE +/- 0.00, N = 519.3217.211. (CXX) g++ options: -O3 -flto -rdynamic

72 Results Shown

Crypto++:
  All Algorithms
  Keyed Algorithms
  Unkeyed Algorithms
  Integer + Elliptic Curve Public Key Algorithms
LeelaChessZero
Xmrig:
  Monero - 1M
  Wownero - 1M
LZ4 Compression:
  1 - Compression Speed
  1 - Decompression Speed
  3 - Compression Speed
  3 - Decompression Speed
  9 - Compression Speed
  9 - Decompression Speed
Zstd Compression:
  3 - Compression Speed
  3 - Decompression Speed
  8 - Compression Speed
  8 - Decompression Speed
  19 - Compression Speed
  19 - Decompression Speed
  3, Long Mode - Compression Speed
  3, Long Mode - Decompression Speed
  8, Long Mode - Compression Speed
  8, Long Mode - Decompression Speed
  19, Long Mode - Compression Speed
  19, Long Mode - Decompression Speed
Coremark
Himeno Benchmark
libavif avifenc:
  0
  2
  6
  6, Lossless
  10, Lossless
C-Ray
POV-Ray
Primesieve
AOBench
FLAC Audio Encoding
LAME MP3 Encoding
eSpeak-NG Speech Engine
OpenJPEG
OpenSSL:
  SHA256
  RSA4096
  RSA4096
Liquid-DSP:
  1 - 256 - 57
  2 - 256 - 57
  4 - 256 - 57
  8 - 256 - 57
libjpeg-turbo tjbench
SQLite Speedtest
Google Draco:
  Lion
  Church Facade
Stress-NG:
  Crypto
  IO_uring
  Matrix Math
  Vector Math
  Memory Copying
  Socket Activity
NCNN:
  CPU - mobilenet
  CPU-v2-v2 - mobilenet-v2
  CPU-v3-v3 - mobilenet-v3
  CPU - shufflenet-v2
  CPU - mnasnet
  CPU - efficientnet-b0
  CPU - googlenet
  CPU - vgg16
  CPU - resnet18
  CPU - alexnet
  CPU - resnet50
  CPU - yolov4-tiny
  CPU - squeezenet_ssd
  CPU - regnety_400m
WavPack Audio Encoding