Neoverse-V1 Compiler Tests

amazon testing on Ubuntu 22.04 via the Phoronix Test Suite by michael larabel 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 2206087-PTS-SVE2607864
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
Allow Limiting Results To Certain Suite(s)

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
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
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
armv8.4-a
December 31 1969
  5 Hours, 3 Minutes
armv8.4-a+sve
December 31 1969
  4 Hours, 57 Minutes
Invert Behavior (Only Show Selected Data)
  5 Hours
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):

armv8.4-a

Kernel Notes: Transparent Huge Pages: madvise
Environment Notes: CXXFLAGS="-O3 -march=armv8.4-a" CFLAGS="-O3 -march=armv8.4-a"
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-nls --disable-werror --enable-checking=yes,extra,rtl --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-objc-gc=auto --enable-plugin --enable-shared --host=aarch64-linux-gnu --program-prefix= --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Python Notes: Python 3.10.4
Security Notes: 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: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

armv8.4-a+sve

Processor: ARMv8 Neoverse-V1 (32 Cores), Motherboard: Amazon EC2 c7g.8xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 62GB, Disk: 301GB Amazon Elastic Block Store, Network: Amazon Elastic

OS: Ubuntu 22.04, Kernel: 5.15.0-1004-aws (aarch64), Compiler: GCC 12.0.0 20220117, File-System: ext4, System Layer: amazon

Kernel Notes: Transparent Huge Pages: madvise
Environment Notes: CXXFLAGS="-O3 -march=armv8.4-a+sve" CFLAGS="-O3 -march=armv8.4-a+sve"
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-nls --disable-werror --enable-checking=yes,extra,rtl --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-objc-gc=auto --enable-plugin --enable-shared --host=aarch64-linux-gnu --program-prefix= --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Python Notes: Python 3.10.4
Security Notes: 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: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Neoverse-V1 Compiler TestsOpenBenchmarking.orgPhoronix Test SuiteARMv8 Neoverse-V1 (32 Cores)Amazon EC2 c7g.8xlarge (1.0 BIOS)Amazon Device 020062GB301GB Amazon Elastic Block StoreAmazon ElasticUbuntu 22.045.15.0-1004-aws (aarch64)GCC 12.0.0 20220117ext4amazonProcessorMotherboardChipsetMemoryDiskNetworkOSKernelCompilerFile-SystemSystem LayerNeoverse-V1 Compiler Tests PerformanceSystem Logs- Transparent Huge Pages: madvise- armv8.4-a: CXXFLAGS="-O3 -march=armv8.4-a" CFLAGS="-O3 -march=armv8.4-a" - armv8.4-a+sve: CXXFLAGS="-O3 -march=armv8.4-a+sve" CFLAGS="-O3 -march=armv8.4-a+sve" - --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-nls --disable-werror --enable-checking=yes,extra,rtl --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-objc-gc=auto --enable-plugin --enable-shared --host=aarch64-linux-gnu --program-prefix= --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - 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: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

armv8.4-a vs. armv8.4-a+sve ComparisonPhoronix Test SuiteBaseline+6.8%+6.8%+13.6%+13.6%+20.4%+20.4%27.2%25.2%22%16.4%9.1%8.7%8.3%6%5.9%4.9%4.9%4.9%4.3%4%3.8%3.8%3.6%3.2%3.2%2.8%2.6%2.3%2.1%2%2%WAV To Opus EncodeCPU - SqueezeNet v1.1T.T.S.SCPU - DenseNetHWB Color SpaceJPEG - 7WAV To MP3CPU - MobileNet v27.5%F.F.T7.4%C75527.4%CPU - SqueezeNet v27.3%poly1305-aes6.3%Speed 10 Realtime - Bosphorus 4K5.9%Rotate32 - 256 - 575.5%16 - 256 - 575.2%8 - 256 - 575.1%D.L.M.FS.F.P.RTwofish - DecryptNoise-GaussianC26704.2%JPEG - 8SwirlTwofishN.C.P.MCoreMark Size 666 - I.P.S3.6%ResizingSpeed 8 Realtime - Bosphorus 1080pTotal Time3%Speed 9 Realtime - Bosphorus 1080p44100 - 1024Unkeyed Algorithms2.4%44100 - 512Trace Time2.1%Composite480000 - 1024480000 - 512Opus Codec EncodingTNNeSpeak-NG Speech EngineTNNGraphicsMagickJPEG XL libjxlLAME MP3 EncodingTNNLuaJITNgspiceTNNNettleAOM AV1KripkeGraphicsMagickLiquid-DSPLiquid-DSPLiquid-DSPLuaJITACES DGEMMBotanGraphicsMagickNgspiceJPEG XL libjxlGraphicsMagickBotanOpenJPEGCoremarkGraphicsMagickAOM AV1StockfishAOM AV1Stargate Digital Audio WorkstationCrypto++Stargate Digital Audio WorkstationPOV-RayLuaJITStargate Digital Audio WorkstationStargate Digital Audio Workstationarmv8.4-aarmv8.4-a+sve

Neoverse-V1 Compiler Testsencode-opus: WAV To Opus Encodetnn: CPU - SqueezeNet v1.1espeak: Text-To-Speech Synthesistnn: CPU - DenseNetgraphics-magick: HWB Color Spacejpegxl: JPEG - 7encode-mp3: WAV To MP3tnn: CPU - MobileNet v2luajit: Fast Fourier Transformngspice: C7552tnn: CPU - SqueezeNet v2nettle: poly1305-aesaom-av1: Speed 10 Realtime - Bosphorus 4Kkripke: graphics-magick: Rotateliquid-dsp: 32 - 256 - 57liquid-dsp: 16 - 256 - 57liquid-dsp: 8 - 256 - 57luajit: Dense LU Matrix Factorizationmt-dgemm: Sustained Floating-Point Ratebotan: Twofish - Decryptgraphics-magick: Noise-Gaussianngspice: C2670jpegxl: JPEG - 8graphics-magick: Swirlbotan: Twofishopenjpeg: NASA Curiosity Panorama M34coremark: CoreMark Size 666 - Iterations Per Secondgraphics-magick: Resizingaom-av1: Speed 8 Realtime - Bosphorus 1080pstockfish: Total Timeaom-av1: Speed 9 Realtime - Bosphorus 1080pstargate: 44100 - 1024cryptopp: Unkeyed Algorithmsstargate: 44100 - 512povray: Trace Timeluajit: Compositestargate: 480000 - 1024stargate: 480000 - 512graphics-magick: Enhancedwebp: Quality 100, Losslessaom-av1: Speed 10 Realtime - Bosphorus 1080plczero: Eigenstargate: 96000 - 1024astcenc: Mediumcompress-zstd: 19 - Compression Speedastcenc: Thoroughmrbayes: Primate Phylogeny Analysisstargate: 96000 - 512rnnoise: compress-zstd: 3 - Compression Speedlczero: BLASdraco: Church Facadeprimesieve: 1e12 Prime Number Generationnettle: sha512caffe: GoogleNet - CPU - 200himeno: Poisson Pressure Solverbotan: AES-256luajit: Sparse Matrix Multiplynettle: chachalammps: Rhodopsin Proteindraco: Lioncompress-zstd: 19, Long Mode - Compression Speedcaffe: AlexNet - CPU - 200openssl: SHA256botan: Blowfishxmrig: Wownero - 1Mencode-flac: WAV To FLACastcenc: Exhaustivejpegxl: PNG - 7redis: GETcompress-zstd: 19, Long Mode - Decompression Speedx264: Bosphorus 1080ponnx: GPT-2 - CPU - Standardbotan: ChaCha20Poly1305 - Decryptcompress-zstd: 19 - Decompression Speedonnx: ArcFace ResNet-100 - CPU - Standardxmrig: Monero - 1Mnettle: aes256botan: ChaCha20Poly1305redis: SETbotan: Blowfish - Decryptluajit: Monte Carlox264: Bosphorus 4Kencode-wavpack: WAV To WavPackonnx: bertsquad-12 - CPU - Standardluajit: Jacobi Successive Over-Relaxationwebp: Quality 100, Highest Compressioncompress-zstd: 3, Long Mode - Compression Speedcompress-zstd: 3, Long Mode - Decompression Speedaom-av1: Speed 8 Realtime - Bosphorus 4Kgromacs: MPI CPU - water_GMX50_baregmpbench: Total Timesysbench: CPUbotan: AES-256 - Decryptopenssl: RSA4096aobench: 2048 x 2048 - Total Timeonnx: super-resolution-10 - CPU - Standardbotan: CAST-256botan: KASUMIbotan: KASUMI - Decryptsmallpt: Global Illumination Renderer; 128 Samplesbotan: CAST-256 - Decryptc-ray: Total Time - 4K, 16 Rays Per Pixelopenssl: RSA4096onnx: fcn-resnet101-11 - CPU - Standardjpegxl: PNG - 8armv8.4-aarmv8.4-a+sve18.320257.70436.5872730.4097873.218.054260.778661.55103.93371.130871.9061.882041431675777052333333527000001763633333355.5312.813927246.155494102.55826.301225239.70357205789646.9248002339120.1357485680152.466.370035459.8701636.07291619.8471282.596.3226846.00538673223.852190.2713114.7298484.883372.99.1435237.5424.41405517.6226937.8128179358.438498.831238075561.5592875494.3131151.57740.2521.330535440.04363427603943570278.87411811.238.31135.36478.322523377.923250.9168.9212364382.5143083.49388645.44435.91389.3751865840.13288.505343.2748.4320.488773901.028.6301241.33820.862.132.2774152.396726.405477.5715090.533.4815413108.78662.01762.2773.895108.59919.296356359.6730.6714.402205.79929.9822346.322106779.587.440280.243615.71111.64476.301820.5165.621927092336116686366673354233331677333333521.1313.442233258.148515106.91027.341272248.88755196762066.5011632414123.9555823340156.716.538163449.0242086.21350020.2631309.036.4501826.12445571823.403193.6813334.8077974.809274.09.0131234.2554.47484817.3857027.2129778438.533504.331251255508.2824355442.6491162.33733.5921.149530940.34393127428176880280.57011877.838.51535.19268.362513289.23263.7169.5812317383.9453094.89358669.84447.04390.3111861924.13289.032343.8548.5120.515772902.168.6401242.73824.862.192.2754155.696666.765474.3215088.133.4945411108.75462.0062.2613.896108.62319.299356407.8730.67OpenBenchmarking.org

Opus Codec Encoding

Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus Encodearmv8.4-a+svearmv8.4-a510152025SE +/- 0.02, N = 5SE +/- 0.00, N = 514.4018.32-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.1armv8.4-a+svearmv8.4-a60120180240300SE +/- 0.14, N = 3SE +/- 0.08, N = 3205.80257.70-march=armv8.4-a+sve - MIN: 205.46 / MAX: 206.28-march=armv8.4-a - MIN: 256.95 / MAX: 258.391. (CXX) g++ options: -O3 -fopenmp -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl

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 Synthesisarmv8.4-a+svearmv8.4-a816243240SE +/- 0.30, N = 20SE +/- 0.31, N = 1629.9836.59-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -std=c99

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: DenseNetarmv8.4-a+svearmv8.4-a6001200180024003000SE +/- 22.16, N = 3SE +/- 12.63, N = 32346.322730.40-march=armv8.4-a+sve - MIN: 2268.4 / MAX: 2446.6-march=armv8.4-a - MIN: 2665.42 / MAX: 2834.731. (CXX) g++ options: -O3 -fopenmp -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: HWB Color Spacearmv8.4-a+svearmv8.4-a2004006008001000SE +/- 0.33, N = 3SE +/- 1.00, N = 31067978-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -fopenmp -O3 -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: JPEG - Encode Speed: 7armv8.4-a+svearmv8.4-a20406080100SE +/- 0.12, N = 3SE +/- 0.13, N = 379.5873.21-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -funwind-tables -O2 -fPIE -pie

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 MP3armv8.4-a+svearmv8.4-a246810SE +/- 0.002, N = 3SE +/- 0.004, N = 37.4408.054-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lm

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v2armv8.4-a+svearmv8.4-a60120180240300SE +/- 0.69, N = 3SE +/- 0.10, N = 3280.24260.78-march=armv8.4-a+sve - MIN: 277.9 / MAX: 282.31-march=armv8.4-a - MIN: 259.13 / MAX: 262.381. (CXX) g++ options: -O3 -fopenmp -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl

LuaJIT

This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Fast Fourier Transformarmv8.4-a+svearmv8.4-a140280420560700SE +/- 10.69, N = 3SE +/- 0.39, N = 3615.71661.55-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -O3 -U_FORTIFY_SOURCE -fno-stack-protector

Ngspice

Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552armv8.4-a+svearmv8.4-a20406080100SE +/- 1.04, N = 3SE +/- 0.51, N = 3111.64103.93-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v2armv8.4-a+svearmv8.4-a20406080100SE +/- 0.09, N = 3SE +/- 0.07, N = 376.3071.13-march=armv8.4-a+sve - MIN: 76.07 / MAX: 76.53-march=armv8.4-a - MIN: 70.76 / MAX: 71.581. (CXX) g++ options: -O3 -fopenmp -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl

Nettle

GNU Nettle is a low-level cryptographic library used by GnuTLS and other software. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.8Test: poly1305-aesarmv8.4-a+svearmv8.4-a2004006008001000SE +/- 5.37, N = 3SE +/- 1.52, N = 3820.51871.90-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -ggdb3 -lnettle -lgmp -lm -lcrypto

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4Karmv8.4-a+svearmv8.4-a1530456075SE +/- 0.22, N = 3SE +/- 0.47, N = 365.6261.88-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Kripke

Kripke is a simple, scalable, 3D Sn deterministic particle transport code. Its primary purpose is to research how data layout, programming paradigms and architectures effect the implementation and performance of Sn transport. Kripke is developed by LLNL. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.4armv8.4-a+svearmv8.4-a40M80M120M160M200MSE +/- 298633.53, N = 3SE +/- 226703.11, N = 3192709233204143167-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -fopenmp

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Rotatearmv8.4-a+svearmv8.4-a130260390520650SE +/- 1.20, N = 3SE +/- 0.00, N = 3611577-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -fopenmp -O3 -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

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: 32 - Buffer Length: 256 - Filter Length: 57armv8.4-a+svearmv8.4-a150M300M450M600M750MSE +/- 1978807.16, N = 3SE +/- 125476.87, N = 3668636667705233333-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57armv8.4-a+svearmv8.4-a80M160M240M320M400MSE +/- 20275.88, N = 3SE +/- 30550.50, N = 3335423333352700000-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 8 - Buffer Length: 256 - Filter Length: 57armv8.4-a+svearmv8.4-a40M80M120M160M200MSE +/- 26666.67, N = 3SE +/- 12018.50, N = 3167733333176363333-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

LuaJIT

This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Dense LU Matrix Factorizationarmv8.4-a+svearmv8.4-a8001600240032004000SE +/- 86.66, N = 3SE +/- 6.02, N = 33521.133355.53-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -O3 -U_FORTIFY_SOURCE -fno-stack-protector

ACES DGEMM

This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point Ratearmv8.4-a+svearmv8.4-a3691215SE +/- 0.05, N = 3SE +/- 0.04, N = 313.4412.81-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -march=native -fopenmp

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Twofish - Decryptarmv8.4-a+svearmv8.4-a60120180240300SE +/- 0.11, N = 3SE +/- 0.20, N = 3258.15246.161. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Noise-Gaussianarmv8.4-a+svearmv8.4-a110220330440550SE +/- 0.58, N = 3SE +/- 0.67, N = 3515494-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -fopenmp -O3 -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

Ngspice

Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670armv8.4-a+svearmv8.4-a20406080100SE +/- 0.18, N = 3SE +/- 0.06, N = 3106.91102.56-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: JPEG - Encode Speed: 8armv8.4-a+svearmv8.4-a612182430SE +/- 0.03, N = 3SE +/- 0.01, N = 327.3426.30-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -funwind-tables -O2 -fPIE -pie

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Swirlarmv8.4-a+svearmv8.4-a30060090012001500SE +/- 1.33, N = 3SE +/- 0.33, N = 312721225-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -fopenmp -O3 -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Twofisharmv8.4-a+svearmv8.4-a50100150200250SE +/- 0.26, N = 3SE +/- 0.12, N = 3248.89239.701. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

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 M34armv8.4-a+svearmv8.4-a12K24K36K48K60KSE +/- 89.48, N = 3SE +/- 19.06, N = 35519657205-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -rdynamic

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 Secondarmv8.4-a+svearmv8.4-a200K400K600K800K1000KSE +/- 416.47, N = 3SE +/- 169.09, N = 3762066.50789646.92-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O2 -O3 -lrt" -lrt

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Resizingarmv8.4-a+svearmv8.4-a5001000150020002500SE +/- 1.86, N = 3SE +/- 22.36, N = 324142339-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -fopenmp -O3 -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080parmv8.4-a+svearmv8.4-a306090120150SE +/- 0.09, N = 3SE +/- 0.03, N = 3123.95120.13-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Stockfish

This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 512 CPU threads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 13Total Timearmv8.4-a+svearmv8.4-a12M24M36M48M60MSE +/- 645132.01, N = 3SE +/- 721518.45, N = 145582334057485680-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -lgcov -lpthread -O3 -fno-exceptions -std=c++17 -pedantic -flto -fprofile-use -fno-peel-loops -fno-tracer -flto=jobserver

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080parmv8.4-a+svearmv8.4-a306090120150SE +/- 0.20, N = 3SE +/- 0.03, N = 3156.71152.46-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 44100 - Buffer Size: 1024armv8.4-a+svearmv8.4-a246810SE +/- 0.002411, N = 3SE +/- 0.002515, N = 36.5381636.3700351. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

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: Unkeyed Algorithmsarmv8.4-a+svearmv8.4-a100200300400500SE +/- 0.25, N = 3SE +/- 0.25, N = 3449.02459.87-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -fPIC -pthread -pipe

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 44100 - Buffer Size: 512armv8.4-a+svearmv8.4-a246810SE +/- 0.003428, N = 3SE +/- 0.002564, N = 36.2135006.0729161. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

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 Timearmv8.4-a+svearmv8.4-a510152025SE +/- 0.04, N = 3SE +/- 0.03, N = 320.2619.85-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -pipe -O3 -ffast-math -R/usr/lib -lXpm -lSM -lICE -lX11 -lIlmImf -lIlmImf-2_5 -lImath-2_5 -lHalf-2_5 -lIex-2_5 -lIexMath-2_5 -lIlmThread-2_5 -lIlmThread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

LuaJIT

This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Compositearmv8.4-a+svearmv8.4-a30060090012001500SE +/- 18.19, N = 3SE +/- 0.41, N = 31309.031282.59-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -O3 -U_FORTIFY_SOURCE -fno-stack-protector

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 480000 - Buffer Size: 1024armv8.4-a+svearmv8.4-a246810SE +/- 0.002250, N = 3SE +/- 0.002118, N = 36.4501826.3226841. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 480000 - Buffer Size: 512armv8.4-a+svearmv8.4-a246810SE +/- 0.002030, N = 3SE +/- 0.001956, N = 36.1244556.0053861. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Enhancedarmv8.4-a+svearmv8.4-a160320480640800SE +/- 0.33, N = 3SE +/- 0.33, N = 3718732-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -fopenmp -O3 -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Losslessarmv8.4-a+svearmv8.4-a612182430SE +/- 0.18, N = 3SE +/- 0.01, N = 323.4023.85-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -fvisibility=hidden -O3 -lm -ljpeg -lpng16 -ltiff

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080parmv8.4-a+svearmv8.4-a4080120160200SE +/- 0.20, N = 3SE +/- 0.28, N = 3193.68190.27-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

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: Eigenarmv8.4-a+svearmv8.4-a30060090012001500SE +/- 14.64, N = 5SE +/- 13.65, N = 313331311-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -flto -O3 -pthread

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 96000 - Buffer Size: 1024armv8.4-a+svearmv8.4-a1.08182.16363.24544.32725.409SE +/- 0.002826, N = 3SE +/- 0.000903, N = 34.8077974.7298481. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: Mediumarmv8.4-a+svearmv8.4-a1.09872.19743.29614.39485.4935SE +/- 0.0080, N = 3SE +/- 0.0125, N = 34.80924.8833-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -flto -pthread

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: 19 - Compression Speedarmv8.4-a+svearmv8.4-a1632486480SE +/- 0.03, N = 3SE +/- 0.23, N = 374.072.9-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lz -llzma

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: Thorougharmv8.4-a+svearmv8.4-a3691215SE +/- 0.0027, N = 3SE +/- 0.0046, N = 39.01319.1435-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -flto -pthread

Timed MrBayes Analysis

This test performs a bayesian analysis of a set of primate genome sequences in order to estimate their phylogeny. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny Analysisarmv8.4-a+svearmv8.4-a50100150200250SE +/- 0.18, N = 3SE +/- 0.07, N = 3234.26237.54-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -std=c99 -pedantic -lm

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 96000 - Buffer Size: 512armv8.4-a+svearmv8.4-a1.00682.01363.02044.02725.034SE +/- 0.002191, N = 3SE +/- 0.003502, N = 34.4748484.4140551. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

RNNoise

RNNoise is a recurrent neural network for audio noise reduction developed by Mozilla and Xiph.Org. This test profile is a single-threaded test measuring the time to denoise a sample 26 minute long 16-bit RAW audio file using this recurrent neural network noise suppression library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 2020-06-28armv8.4-a+svearmv8.4-a48121620SE +/- 0.05, N = 3SE +/- 0.05, N = 317.3917.62-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pedantic -fvisibility=hidden

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 Speedarmv8.4-a+svearmv8.4-a15003000450060007500SE +/- 15.42, N = 3SE +/- 14.45, N = 37027.26937.8-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lz -llzma

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: BLASarmv8.4-a+svearmv8.4-a30060090012001500SE +/- 6.96, N = 3SE +/- 13.99, N = 512971281-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -flto -O3 -pthread

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: Church Facadearmv8.4-a+svearmv8.4-a2K4K6K8K10KSE +/- 7.00, N = 3SE +/- 6.64, N = 378437935-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3

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 Generationarmv8.4-a+svearmv8.4-a246810SE +/- 0.043, N = 3SE +/- 0.022, N = 38.5338.438-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3

Nettle

GNU Nettle is a low-level cryptographic library used by GnuTLS and other software. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.8Test: sha512armv8.4-a+svearmv8.4-a110220330440550SE +/- 0.07, N = 3SE +/- 0.04, N = 3504.33498.83-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -ggdb3 -lnettle -lgmp -lm -lcrypto

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: GoogleNet - Acceleration: CPU - Iterations: 200armv8.4-a+svearmv8.4-a30K60K90K120K150KSE +/- 105.70, N = 3SE +/- 49.72, N = 3125125123807-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -fPIC -rdynamic -lglog -lgflags -lprotobuf -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

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 Solverarmv8.4-a+svearmv8.4-a12002400360048006000SE +/- 5.76, N = 3SE +/- 2.66, N = 35508.285561.56-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256armv8.4-a+svearmv8.4-a12002400360048006000SE +/- 9.30, N = 3SE +/- 14.75, N = 35442.655494.311. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

LuaJIT

This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Sparse Matrix Multiplyarmv8.4-a+svearmv8.4-a30060090012001500SE +/- 7.20, N = 3SE +/- 3.14, N = 31162.331151.57-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -O3 -U_FORTIFY_SOURCE -fno-stack-protector

Nettle

GNU Nettle is a low-level cryptographic library used by GnuTLS and other software. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.8Test: chachaarmv8.4-a+svearmv8.4-a160320480640800SE +/- 0.55, N = 3SE +/- 0.61, N = 3733.59740.25-march=armv8.4-a+sve - MIN: 442.26 / MAX: 956.22-march=armv8.4-a - MIN: 454.21 / MAX: 956.531. (CC) gcc options: -O3 -ggdb3 -lnettle -lgmp -lm -lcrypto

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin Proteinarmv8.4-a+svearmv8.4-a510152025SE +/- 0.01, N = 3SE +/- 0.09, N = 321.1521.33-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -lm

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: Lionarmv8.4-a+svearmv8.4-a11002200330044005500SE +/- 2.40, N = 3SE +/- 2.65, N = 353095354-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ 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: 19, Long Mode - Compression Speedarmv8.4-a+svearmv8.4-a918273645SE +/- 0.00, N = 3SE +/- 0.03, N = 340.340.0-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lz -llzma

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: AlexNet - Acceleration: CPU - Iterations: 200armv8.4-a+svearmv8.4-a9K18K27K36K45KSE +/- 12.55, N = 3SE +/- 31.22, N = 34393143634-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -fPIC -rdynamic -lglog -lgflags -lprotobuf -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

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: SHA256armv8.4-a+svearmv8.4-a6000M12000M18000M24000M30000MSE +/- 32102278.66, N = 3SE +/- 25639974.71, N = 32742817688027603943570-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Blowfisharmv8.4-a+svearmv8.4-a60120180240300SE +/- 0.10, N = 3SE +/- 0.29, N = 3280.57278.871. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

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: Wownero - Hash Count: 1Marmv8.4-a+svearmv8.4-a3K6K9K12K15KSE +/- 26.88, N = 3SE +/- 27.59, N = 311877.811811.2-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -fexceptions -fno-rtti -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

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 FLACarmv8.4-a+svearmv8.4-a918273645SE +/- 0.01, N = 5SE +/- 0.02, N = 538.5238.31-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: Exhaustivearmv8.4-a+svearmv8.4-a816243240SE +/- 0.01, N = 3SE +/- 0.02, N = 335.1935.36-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -flto -pthread

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: PNG - Encode Speed: 7armv8.4-a+svearmv8.4-a246810SE +/- 0.01, N = 3SE +/- 0.02, N = 38.368.32-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -funwind-tables -O2 -fPIE -pie

Redis

Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: GETarmv8.4-a+svearmv8.4-a500K1000K1500K2000K2500KSE +/- 9056.40, N = 3SE +/- 1605.36, N = 32513289.202523377.92-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -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: 19, Long Mode - Decompression Speedarmv8.4-a+svearmv8.4-a7001400210028003500SE +/- 0.59, N = 3SE +/- 7.62, N = 33263.73250.9-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lz -llzma

x264

This is a multi-threaded test of the x264 video encoder run on the CPU with a choice of 1080p or 4K video input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2022-02-22Video Input: Bosphorus 1080parmv8.4-a+svearmv8.4-a4080120160200SE +/- 0.10, N = 3SE +/- 0.07, N = 3169.58168.921. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -lm -lpthread -O3 -flto

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: Standardarmv8.4-a+svearmv8.4-a3K6K9K12K15KSE +/- 12.91, N = 3SE +/- 63.90, N = 31231712364-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -ffunction-sections -fdata-sections -march=native -mtune=native -flto -fno-fat-lto-objects -ldl -lrt

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305 - Decryptarmv8.4-a+svearmv8.4-a80160240320400SE +/- 0.13, N = 3SE +/- 0.02, N = 3383.95382.511. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

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: 19 - Decompression Speedarmv8.4-a+svearmv8.4-a7001400210028003500SE +/- 6.60, N = 3SE +/- 8.46, N = 33094.83083.4-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lz -llzma

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: Standardarmv8.4-a+svearmv8.4-a2004006008001000SE +/- 0.17, N = 3SE +/- 0.88, N = 3935938-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -ffunction-sections -fdata-sections -march=native -mtune=native -flto -fno-fat-lto-objects -ldl -lrt

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: 1Marmv8.4-a+svearmv8.4-a2K4K6K8K10KSE +/- 6.18, N = 3SE +/- 9.56, N = 38669.88645.4-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -fexceptions -fno-rtti -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Nettle

GNU Nettle is a low-level cryptographic library used by GnuTLS and other software. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.8Test: aes256armv8.4-a+svearmv8.4-a10002000300040005000SE +/- 0.39, N = 3SE +/- 3.11, N = 34447.044435.91-march=armv8.4-a+sve - MIN: 3925.11 / MAX: 5627.84-march=armv8.4-a - MIN: 3927.32 / MAX: 5628.861. (CC) gcc options: -O3 -ggdb3 -lnettle -lgmp -lm -lcrypto

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305armv8.4-a+svearmv8.4-a80160240320400SE +/- 0.07, N = 3SE +/- 0.04, N = 3390.31389.381. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

Redis

Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SETarmv8.4-a+svearmv8.4-a400K800K1200K1600K2000KSE +/- 7427.93, N = 3SE +/- 794.78, N = 31861924.131865840.13-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Blowfish - Decryptarmv8.4-a+svearmv8.4-a60120180240300SE +/- 0.03, N = 3SE +/- 0.07, N = 3289.03288.511. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

LuaJIT

This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Monte Carloarmv8.4-a+svearmv8.4-a70140210280350SE +/- 0.54, N = 3SE +/- 0.35, N = 3343.85343.27-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -O3 -U_FORTIFY_SOURCE -fno-stack-protector

x264

This is a multi-threaded test of the x264 video encoder run on the CPU with a choice of 1080p or 4K video input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2022-02-22Video Input: Bosphorus 4Karmv8.4-a+svearmv8.4-a1122334455SE +/- 0.01, N = 3SE +/- 0.08, N = 348.5148.431. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -lm -lpthread -O3 -flto

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 WavPackarmv8.4-a+svearmv8.4-a510152025SE +/- 0.03, N = 5SE +/- 0.00, N = 520.5220.49-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -rdynamic

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: Standardarmv8.4-a+svearmv8.4-a170340510680850SE +/- 0.50, N = 3SE +/- 0.44, N = 3772773-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -ffunction-sections -fdata-sections -march=native -mtune=native -flto -fno-fat-lto-objects -ldl -lrt

LuaJIT

This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Jacobi Successive Over-Relaxationarmv8.4-a+svearmv8.4-a2004006008001000SE +/- 1.00, N = 3SE +/- 0.68, N = 3902.16901.02-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -O3 -U_FORTIFY_SOURCE -fno-stack-protector

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest Compressionarmv8.4-a+svearmv8.4-a246810SE +/- 0.017, N = 3SE +/- 0.006, N = 38.6408.630-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -fvisibility=hidden -O3 -lm -ljpeg -lpng16 -ltiff

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, Long Mode - Compression Speedarmv8.4-a+svearmv8.4-a30060090012001500SE +/- 4.88, N = 3SE +/- 5.37, N = 31242.71241.3-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3, Long Mode - Decompression Speedarmv8.4-a+svearmv8.4-a8001600240032004000SE +/- 1.28, N = 3SE +/- 3.95, N = 33824.83820.8-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -pthread -lz -llzma

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4Karmv8.4-a+svearmv8.4-a1428425670SE +/- 0.61, N = 3SE +/- 0.43, N = 362.1962.13-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2022.1Implementation: MPI CPU - Input: water_GMX50_barearmv8.4-a+svearmv8.4-a0.51231.02461.53692.04922.5615SE +/- 0.002, N = 3SE +/- 0.001, N = 32.2752.277-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3

GNU GMP GMPbench

GMPbench is a test of the GNU Multiple Precision Arithmetic (GMP) Library. GMPbench is a single-threaded integer benchmark that leverages the GMP library to stress the CPU with widening integer multiplication. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGMPbench Score, More Is BetterGNU GMP GMPbench 6.2.1Total Timearmv8.4-a+svearmv8.4-a90018002700360045004155.64152.3-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O3 -lm

Sysbench

This is a benchmark of Sysbench with the built-in CPU and memory sub-tests. Sysbench is a scriptable multi-threaded benchmark tool based on LuaJIT. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPUarmv8.4-a+svearmv8.4-a20K40K60K80K100KSE +/- 2.72, N = 3SE +/- 8.50, N = 396666.7696726.40-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -O2 -funroll-loops -O3 -rdynamic -ldl -laio -lm

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256 - Decryptarmv8.4-a+svearmv8.4-a12002400360048006000SE +/- 8.64, N = 3SE +/- 5.58, N = 35474.325477.571. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

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.orgsign/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096armv8.4-a+svearmv8.4-a11002200330044005500SE +/- 0.53, N = 3SE +/- 0.78, N = 35088.15090.5-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

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 Timearmv8.4-a+svearmv8.4-a816243240SE +/- 0.01, N = 3SE +/- 0.00, N = 333.4933.48-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -lm -O3

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: Standardarmv8.4-a+svearmv8.4-a12002400360048006000SE +/- 2.17, N = 3SE +/- 0.93, N = 354115413-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -ffunction-sections -fdata-sections -march=native -mtune=native -flto -fno-fat-lto-objects -ldl -lrt

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256armv8.4-a+svearmv8.4-a20406080100SE +/- 0.01, N = 3SE +/- 0.02, N = 3108.75108.791. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMIarmv8.4-a+svearmv8.4-a1428425670SE +/- 0.00, N = 3SE +/- 0.00, N = 362.0062.021. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMI - Decryptarmv8.4-a+svearmv8.4-a1428425670SE +/- 0.00, N = 3SE +/- 0.00, N = 362.2662.281. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

Smallpt

Smallpt is a C++ global illumination renderer written in less than 100 lines of code. Global illumination is done via unbiased Monte Carlo path tracing and there is multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 128 Samplesarmv8.4-a+svearmv8.4-a0.87661.75322.62983.50644.383SE +/- 0.003, N = 3SE +/- 0.002, N = 33.8963.895-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -fopenmp -O3

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256 - Decryptarmv8.4-a+svearmv8.4-a20406080100SE +/- 0.01, N = 3SE +/- 0.02, N = 3108.62108.601. (CXX) g++ options: -fstack-protector -pthread -lbotan-2 -ldl -lrt

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 Pixelarmv8.4-a+svearmv8.4-a510152025SE +/- 0.00, N = 3SE +/- 0.03, N = 319.3019.30-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -lm -lpthread -O3

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.orgverify/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096armv8.4-a+svearmv8.4-a80K160K240K320K400KSE +/- 10.52, N = 3SE +/- 8.85, N = 3356407.8356359.6-march=armv8.4-a+sve-march=armv8.4-a1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: Standardarmv8.4-a+svearmv8.4-a1632486480SE +/- 0.00, N = 3SE +/- 0.00, N = 37373-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -ffunction-sections -fdata-sections -march=native -mtune=native -flto -fno-fat-lto-objects -ldl -lrt

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: PNG - Encode Speed: 8armv8.4-a+svearmv8.4-a0.15080.30160.45240.60320.754SE +/- 0.00, N = 3SE +/- 0.00, N = 30.670.67-march=armv8.4-a+sve-march=armv8.4-a1. (CXX) g++ options: -O3 -funwind-tables -O2 -fPIE -pie

armv8.4-a

Kernel Notes: Transparent Huge Pages: madvise
Environment Notes: CXXFLAGS="-O3 -march=armv8.4-a" CFLAGS="-O3 -march=armv8.4-a"
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-nls --disable-werror --enable-checking=yes,extra,rtl --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-objc-gc=auto --enable-plugin --enable-shared --host=aarch64-linux-gnu --program-prefix= --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Python Notes: Python 3.10.4
Security Notes: 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: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 31 December 1969 19:00 by user ubuntu.

armv8.4-a+sve

Processor: ARMv8 Neoverse-V1 (32 Cores), Motherboard: Amazon EC2 c7g.8xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 62GB, Disk: 301GB Amazon Elastic Block Store, Network: Amazon Elastic

OS: Ubuntu 22.04, Kernel: 5.15.0-1004-aws (aarch64), Compiler: GCC 12.0.0 20220117, File-System: ext4, System Layer: amazon

Kernel Notes: Transparent Huge Pages: madvise
Environment Notes: CXXFLAGS="-O3 -march=armv8.4-a+sve" CFLAGS="-O3 -march=armv8.4-a+sve"
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-nls --disable-werror --enable-checking=yes,extra,rtl --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-objc-gc=auto --enable-plugin --enable-shared --host=aarch64-linux-gnu --program-prefix= --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Python Notes: Python 3.10.4
Security Notes: 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: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 31 December 1969 19:00 by user ubuntu.