1260L June

Intel Xeon E3-1260L v5 testing with a ASRock E3V5 WS (P7.10 BIOS) and llvmpipe on Ubuntu 20.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2106100-IB-1260LJUNE71
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

Limit displaying results to tests within:

AV1 5 Tests
Bioinformatics 2 Tests
C++ Boost Tests 2 Tests
CAD 2 Tests
Timed Code Compilation 8 Tests
C/C++ Compiler Tests 16 Tests
CPU Massive 15 Tests
Creator Workloads 20 Tests
Cryptocurrency Benchmarks, CPU Mining Tests 2 Tests
Cryptography 4 Tests
Encoding 8 Tests
Game Development 5 Tests
HPC - High Performance Computing 6 Tests
Machine Learning 2 Tests
Molecular Dynamics 2 Tests
MPI Benchmarks 3 Tests
Multi-Core 23 Tests
NVIDIA GPU Compute 3 Tests
Intel oneAPI 3 Tests
OpenMPI Tests 4 Tests
Programmer / Developer System Benchmarks 9 Tests
Python Tests 6 Tests
Scientific Computing 4 Tests
Software Defined Radio 4 Tests
Server CPU Tests 9 Tests
Single-Threaded 2 Tests
Telephony 2 Tests
Texture Compression 4 Tests
Video Encoding 8 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
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
1
June 07 2021
  1 Day, 1 Hour, 5 Minutes
2
June 08 2021
  22 Hours, 43 Minutes
3
June 09 2021
  21 Hours, 41 Minutes
Invert Hiding All Results Option
  23 Hours, 10 Minutes

Only show results where is faster than
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):


1260L JuneProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen Resolution123Intel Xeon E3-1260L v5 @ 3.90GHz (4 Cores / 8 Threads)ASRock E3V5 WS (P7.10 BIOS)Intel Xeon E3-1200 v5/E3-15008GB120GB INTEL SSDSC2BW12llvmpipeRealtek ALC892Intel I219-LMUbuntu 20.105.8.0-41-generic (x86_64)GNOME Shell 3.38.1X Server 1.20.94.5 Mesa 20.2.1 (LLVM 11.0.0 256 bits)GCC 10.2.0ext41024x768OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Scaling Governor: intel_pstate powersave - CPU Microcode: 0xe2 - Thermald 2.3 Python Details- Python 3.8.6Security Details- itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable

123Result OverviewPhoronix Test Suite100%101%101%102%oneDNNTimed LLVM CompilationStockfishXmrigTimed Wasmer CompilationEmbreeMobile Neural NetworkGROMACSSecureMarkGNU RadioASTC EncoderAOM AV1Timed MrBayes AnalysisLuaRadiosrsLTEChia Blockchain VDFKTX-Software toktxSVT-HEVCZstd CompressionXcompact3d Incompact3dBRL-CADChaos Group V-RAYVP9 libvpx EncodingViennaCLdav1dHelsinglibavif avifencSVT-VP9Timed Mesa CompilationTimed Erlang/OTP CompilationLiquid-DSPGNU GMP GMPbenchBasis UniversalVOSK Speech Recognition ToolkitOpenSCADNgspicePJSIPTimed FFmpeg CompilationTimed HMMer SearchTimed GDB GNU Debugger CompilationSVT-AV1Botanlibgav1Google DracoTimed Node.js CompilationtoyBrot Fractal GeneratorTimed Linux Kernel CompilationIntel Open Image Denoise

1260L Junetoktx: Zstd Compression 9luaradio: Complex Phasecompress-zstd: 3 - Compression Speedmnn: MobileNetV2_224viennacl: CPU BLAS - sDOTastcenc: Mediumavifenc: 10, Losslessembree: Pathtracer - Asian Dragonembree: Pathtracer ISPC - Asian Dragon Objaom-av1: Speed 6 Realtime - Bosphorus 4Kxmrig: Wownero - 1Mstockfish: Total Timemnn: inception-v3mnn: SqueezeNetV1.0build-wasmer: Time To Compilevpxenc: Speed 5 - Bosphorus 1080pcompress-zstd: 8, Long Mode - Compression Speedcompress-zstd: 19, Long Mode - Compression Speedaom-av1: Speed 8 Realtime - Bosphorus 4Kcompress-zstd: 3, Long Mode - Compression Speedonednn: IP Shapes 3D - u8s8f32 - CPUdraco: Liononednn: IP Shapes 3D - f32 - CPUembree: Pathtracer ISPC - Asian Dragonviennacl: CPU BLAS - dGEMV-Txmrig: Monero - 1Mcompress-zstd: 8 - Compression Speedembree: Pathtracer - Asian Dragon Objembree: Pathtracer ISPC - Crowndav1d: Summer Nature 4Kdav1d: Chimera 1080pgromacs: MPI CPU - water_GMX50_barecompress-zstd: 19 - Decompression Speedaom-av1: Speed 9 Realtime - Bosphorus 1080psvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080popenscad: Retro Cargnuradio: Signal Source (Cosine)compress-zstd: 8 - Decompression Speedonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUavifenc: 2chia-vdf: Square Assembly Optimizedbotan: Blowfish - Decryptsrslte: PHY_DL_Testmrbayes: Primate Phylogeny Analysissecuremark: SecureMark-TLSonednn: Convolution Batch Shapes Auto - f32 - CPUviennacl: CPU BLAS - dGEMM-TTviennacl: CPU BLAS - dGEMM-TNliquid-dsp: 4 - 256 - 57viennacl: CPU BLAS - dGEMM-NTsrslte: OFDM_Testluaradio: FM Deemphasis Filterbasis: UASTC Level 0luaradio: Five Back to Back FIR Filtersgnuradio: FIR Filterdav1d: Summer Nature 1080ptoktx: UASTC 4 + Zstd Compression 19compress-zstd: 3 - Decompression Speeddav1d: Chimera 1080p 10-bitviennacl: CPU BLAS - dCOPYcompress-zstd: 19 - Compression Speedgnuradio: Hilbert Transformopenscad: Mini-ITX Caseaom-av1: Speed 4 Two-Pass - Bosphorus 1080pavifenc: 6, Losslesshelsing: 12 digitsvt-hevc: 7 - Bosphorus 1080ponednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUsvt-hevc: 10 - Bosphorus 1080popenscad: Leonardo Phone Case Slimonednn: Recurrent Neural Network Training - f32 - CPUtoktx: UASTC 3 + Zstd Compression 19gnuradio: IIR Filtersvt-av1: Preset 8 - Bosphorus 1080pavifenc: 10onednn: IP Shapes 1D - u8s8f32 - CPUaom-av1: Speed 8 Realtime - Bosphorus 1080pdraco: Church Facadeincompact3d: input.i3d 129 Cells Per Directionpjsip: OPTIONS, Statelessmnn: mobilenet-v1-1.0brl-cad: VGR Performance Metricv-ray: CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUsrslte: PHY_DL_Testbuild-nodejs: Time To Compileopenscad: Projector Mount Swivelviennacl: CPU BLAS - dAXPYvpxenc: Speed 0 - Bosphorus 4Ksvt-hevc: 1 - Bosphorus 1080ptoybrot: OpenMPbotan: Blowfishviennacl: CPU BLAS - sAXPYonednn: Recurrent Neural Network Inference - f32 - CPUbotan: Twofish - Decryptaom-av1: Speed 6 Realtime - Bosphorus 1080pavifenc: 6viennacl: CPU BLAS - dDOTviennacl: CPU BLAS - dGEMV-Nmnn: resnet-v2-50liquid-dsp: 2 - 256 - 57aom-av1: Speed 6 Two-Pass - Bosphorus 4Kvpxenc: Speed 0 - Bosphorus 1080psvt-av1: Preset 4 - Bosphorus 4Klibgav1: Summer Nature 1080paom-av1: Speed 6 Two-Pass - Bosphorus 1080psvt-av1: Preset 4 - Bosphorus 1080ponednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUluaradio: Hilbert Transformonednn: Deconvolution Batch shapes_1d - f32 - CPUliquid-dsp: 1 - 256 - 57onednn: Deconvolution Batch shapes_3d - f32 - CPUlibgav1: Chimera 1080p 10-bitngspice: C7552basis: UASTC Level 3basis: ETC1Stoktx: Zstd Compression 19gnuradio: FM Deemphasis Filtervpxenc: Speed 5 - Bosphorus 4Konednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUopenscad: Pistolbuild-mesa: Time To Compilebuild-llvm: Ninjasvt-vp9: Visual Quality Optimized - Bosphorus 1080pbotan: Twofishcompress-zstd: 19, Long Mode - Decompression Speedlibgav1: Summer Nature 4Kembree: Pathtracer - Crowncompress-zstd: 8, Long Mode - Decompression Speedchia-vdf: Square Plain C++gmpbench: Total Timebuild-erlang: Time To Compilebotan: KASUMI - Decryptlibgav1: Chimera 1080pvosk: onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUhelsing: 14 digitbuild-linux-kernel: Time To Compilebasis: UASTC Level 2toybrot: C++ Threadsbuild-gdb: Time To Compileavifenc: 0toybrot: C++ Tasksbuild-ffmpeg: Time To Compileastcenc: Exhaustivecompress-zstd: 3, Long Mode - Decompression Speedonednn: Recurrent Neural Network Inference - u8s8f32 - CPUbotan: CAST-256 - Decryptsvt-av1: Preset 8 - Bosphorus 4Kbuild-llvm: Unix Makefilesastcenc: Thoroughbotan: AES-256 - Decryptbotan: CAST-256ngspice: C2670hmmer: Pfam Database Searchbotan: KASUMIsvt-vp9: VMAF Optimized - Bosphorus 1080ppjsip: OPTIONS, Statefulbotan: ChaCha20Poly1305botan: AES-256onednn: Convolution Batch Shapes Auto - u8s8f32 - CPUtoktx: UASTC 3liquid-dsp: 8 - 256 - 57botan: ChaCha20Poly1305 - Decryptonednn: Recurrent Neural Network Training - u8s8f32 - CPUpjsip: INVITEviennacl: CPU BLAS - dGEMM-NNviennacl: CPU BLAS - sCOPYoidn: RTLightmap.hdr.4096x4096oidn: RT.ldr_alb_nrm.3840x2160oidn: RT.hdr_alb_nrm.3840x2160aom-av1: Speed 0 Two-Pass - Bosphorus 1080paom-av1: Speed 4 Two-Pass - Bosphorus 4Kaom-av1: Speed 0 Two-Pass - Bosphorus 4Konednn: IP Shapes 1D - f32 - CPUaom-av1: Speed 9 Realtime - Bosphorus 4Kgnuradio: Five Back to Back FIR Filters1233.729477.61253.54.52127.77.78109.6954.77284.82866.001982.6902984562.1498.511220.09520.69159.615.719.14832.53.23337590813.21915.504527.91383.5149.84.38434.303879.78305.610.4412712.588.6279.6323.4642328.13220.59.4743378.352178167403.83683.2166.25521318122.761014.514.417157333314.797466667354.811.837618.4508.1284.41936.4463123.4231.7817.217.2421.656.3443.63154.16916.57141.195.7684091.0622.6179167.7438.076471.523.0365.4414.3793971.15888862.1501007510705.2234218236295.63635220.71251.928125.67825.72.582.58114789409.46626.34768.62333.49710.9726.38328.428.850.247954630002.986.030.614158.099.461.9669161.3766.215.70685144666716.814636.54161.259158.60838.13323.812603.88.098.70959134.233131.6231934.43162.04334.4012745.941.293.88143415.11162674761.8218.73881.558135.8521.6324782.251728.695245.85780.454115979142.532156.274118626147.957189.97883328.64769.48129.3676.3901950.91720.18423662.218129.436155.817147.04084.18878.537016667.5743654.67621.359631.836187833333662.6539176.99390314.117.70.070.140.140.171.590.059.7117421.68490.43.626476.71266.44.62028.27.83219.5304.85514.91136.052012.9909742962.6598.631219.57820.64161.715.919.14839.83.27304583713.15345.560627.91396.5151.44.41874.316279.87307.020.4422730.789.2778.9823.3472323.53193.29.3942678.292179667401.52482.9167.09621468522.783514.414.517275333314.796833333354.211.913614.5508.2285.21936.6983105.2230.4417.217.3420.456.1553.65153.91416.65841.105.7967190.7222.5089190.4838.116469.423.1395.4624.3734871.32888761.8871002508735.2454231036145.63377220.31257.018126.17925.62.592.58115227407.99526.44776.65332.66311.0126.29028.528.850.108953826672.996.040.612157.599.491.9609189.2366.015.75395159733316.806036.64161.162158.60438.23023.872605.28.078.73084134.559131.6141929.91062.18334.0162743.441.373.88983416.21164334769.8218.31181.693136.1121.5994773.701731.769245.73580.402115854142.37156.188118510147.943189.99843328.34773.76129.2326.3831953.55620.20953662.255129.462155.937146.88984.10878.467020667.8743653.00421.364131.850187790000662.7629174.22390314.117.70.070.140.140.171.590.0513.7323521.37514.73.642489.11284.84.55228.37.92759.5854.81014.84006.102015.3895842963.0278.599222.67020.93159.615.918.90829.43.23256590713.06335.568728.21398.3151.04.37354.346779.12304.140.4382736.889.4179.6823.5492343.53214.69.4108077.691178467404.45283.5167.44621447122.626114.514.417211666714.897233333352.511.878616.9505.1286.14941.9923122.5230.4817.317.2422.856.0273.65154.73016.58241.315.7935391.1622.5359146.5638.256469.523.1175.4384.3602871.46892562.0419286508545.2404213536255.61310221.21256.614126.14925.62.592.59115024409.54826.44786.18332.28011.0126.29728.428.950.075957073332.986.050.614157.999.461.9659168.4266.015.71405150033316.766036.61160.830159.02738.16923.871605.38.078.72047134.386131.9281933.98262.12333.6492749.441.283.88493409.31165004760.4218.64881.534135.9921.6394778.171731.685245.42280.542115780142.292156.447118696147.728190.26873323.74776.38129.4166.3811953.47520.20133666.701129.317155.769147.03284.12878.477015667.4163655.44221.370531.843187753333662.9249175.25390214.117.70.070.140.140.171.590.059.7018822.90498.8OpenBenchmarking.org

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: Zstd Compression 91230.8391.6782.5173.3564.195SE +/- 0.031, N = 3SE +/- 0.028, N = 15SE +/- 0.049, N = 33.7293.6263.642
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: Zstd Compression 9123246810Min: 3.68 / Avg: 3.73 / Max: 3.79Min: 3.53 / Avg: 3.63 / Max: 3.92Min: 3.54 / Avg: 3.64 / Max: 3.7

LuaRadio

LuaRadio is a lightweight software-defined radio (SDR) framework built atop LuaJIT. LuaRadio provides a suite of source, sink, and processing blocks, with a simple API for defining flow graphs, running flow graphs, creating blocks, and creating data types. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Complex Phase123110220330440550SE +/- 13.08, N = 3SE +/- 6.98, N = 3SE +/- 2.38, N = 3477.6476.7489.1
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Complex Phase12390180270360450Min: 455.2 / Avg: 477.57 / Max: 500.5Min: 465.2 / Avg: 476.73 / Max: 489.3Min: 486.1 / Avg: 489.1 / Max: 493.8

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 Speed12330060090012001500SE +/- 6.03, N = 3SE +/- 11.11, N = 3SE +/- 5.17, N = 31253.51266.41284.81. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Compression Speed1232004006008001000Min: 1241.9 / Avg: 1253.47 / Max: 1262.2Min: 1247 / Avg: 1266.37 / Max: 1285.5Min: 1276 / Avg: 1284.83 / Max: 1293.91. (CC) gcc options: -O3 -pthread -lz -llzma

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: MobileNetV2_2241231.03952.0793.11854.1585.1975SE +/- 0.022, N = 3SE +/- 0.028, N = 3SE +/- 0.009, N = 34.5214.6204.552MIN: 4.21 / MAX: 7.71MIN: 4.21 / MAX: 23.77MIN: 4.21 / MAX: 22.391. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: MobileNetV2_224123246810Min: 4.48 / Avg: 4.52 / Max: 4.56Min: 4.56 / Avg: 4.62 / Max: 4.65Min: 4.54 / Avg: 4.55 / Max: 4.571. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOT123714212835SE +/- 0.66, N = 3SE +/- 0.12, N = 3SE +/- 0.12, N = 327.728.228.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOT123612182430Min: 26.4 / Avg: 27.7 / Max: 28.5Min: 28 / Avg: 28.23 / Max: 28.4Min: 28.1 / Avg: 28.33 / Max: 28.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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 3.0Preset: Medium123246810SE +/- 0.0745, N = 3SE +/- 0.0895, N = 3SE +/- 0.0814, N = 37.78107.83217.92751. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.0Preset: Medium1233691215Min: 7.66 / Avg: 7.78 / Max: 7.92Min: 7.74 / Avg: 7.83 / Max: 8.01Min: 7.78 / Avg: 7.93 / Max: 8.071. (CXX) g++ options: -O3 -flto -pthread

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 0.9.0Encoder Speed: 10, Lossless1233691215SE +/- 0.067, N = 3SE +/- 0.031, N = 3SE +/- 0.055, N = 39.6959.5309.5851. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10, Lossless1233691215Min: 9.56 / Avg: 9.7 / Max: 9.78Min: 9.49 / Avg: 9.53 / Max: 9.59Min: 9.48 / Avg: 9.59 / Max: 9.651. (CXX) g++ options: -O3 -fPIC -lm

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian Dragon1231.09242.18483.27724.36965.462SE +/- 0.0034, N = 3SE +/- 0.0336, N = 3SE +/- 0.0066, N = 34.77284.85514.8101MIN: 4.7 / MAX: 4.88MIN: 4.72 / MAX: 4.95MIN: 4.76 / MAX: 4.88
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian Dragon123246810Min: 4.77 / Avg: 4.77 / Max: 4.78Min: 4.79 / Avg: 4.86 / Max: 4.89Min: 4.8 / Avg: 4.81 / Max: 4.82

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian Dragon Obj1231.1052.213.3154.425.525SE +/- 0.0131, N = 3SE +/- 0.0056, N = 3SE +/- 0.0049, N = 34.82864.91134.8400MIN: 4.75 / MAX: 4.93MIN: 4.89 / MAX: 4.97MIN: 4.78 / MAX: 4.93
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian Dragon Obj123246810Min: 4.81 / Avg: 4.83 / Max: 4.85Min: 4.9 / Avg: 4.91 / Max: 4.92Min: 4.83 / Avg: 4.84 / Max: 4.85

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K123246810SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 36.006.056.101. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K123246810Min: 5.95 / Avg: 6 / Max: 6.04Min: 6 / Avg: 6.05 / Max: 6.11Min: 6.09 / Avg: 6.1 / Max: 6.111. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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: 1M123400800120016002000SE +/- 1.33, N = 3SE +/- 0.98, N = 3SE +/- 0.78, N = 31982.62012.92015.31. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s, More Is BetterXmrig 6.12.1Variant: Wownero - Hash Count: 1M123400800120016002000Min: 1980.4 / Avg: 1982.63 / Max: 1985Min: 2011.5 / Avg: 2012.93 / Max: 2014.8Min: 2014.5 / Avg: 2015.33 / Max: 2016.91. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

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 Time1232M4M6M8M10MSE +/- 66915.41, N = 3SE +/- 69756.25, N = 3SE +/- 58988.55, N = 39029845909742989584291. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fprofile-use -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto=jobserver
OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 13Total Time1231.6M3.2M4.8M6.4M8MMin: 8901179 / Avg: 9029845.33 / Max: 9126065Min: 8967911 / Avg: 9097428.67 / Max: 9207095Min: 8864612 / Avg: 8958429.33 / Max: 90672871. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fprofile-use -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto=jobserver

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: inception-v31231428425670SE +/- 0.26, N = 3SE +/- 0.44, N = 3SE +/- 0.43, N = 362.1562.6663.03MIN: 57.76 / MAX: 122.98MIN: 60.57 / MAX: 105.37MIN: 59.95 / MAX: 85.791. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: inception-v31231224364860Min: 61.69 / Avg: 62.15 / Max: 62.6Min: 61.97 / Avg: 62.66 / Max: 63.49Min: 62.24 / Avg: 63.03 / Max: 63.721. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: SqueezeNetV1.0123246810SE +/- 0.045, N = 3SE +/- 0.047, N = 3SE +/- 0.046, N = 38.5118.6318.599MIN: 7.48 / MAX: 26.37MIN: 7.52 / MAX: 51.07MIN: 7.52 / MAX: 26.741. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: SqueezeNetV1.01233691215Min: 8.42 / Avg: 8.51 / Max: 8.57Min: 8.58 / Avg: 8.63 / Max: 8.73Min: 8.52 / Avg: 8.6 / Max: 8.671. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

Timed Wasmer Compilation

This test times how long it takes to compile Wasmer. Wasmer is written in the Rust programming language and is a WebAssembly runtime implementation that supports WASI and EmScripten. This test profile builds Wasmer with the Cranelift and Singlepast compiler features enabled. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Wasmer Compilation 1.0.2Time To Compile12350100150200250SE +/- 1.61, N = 3SE +/- 1.20, N = 3SE +/- 0.89, N = 3220.10219.58222.671. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Wasmer Compilation 1.0.2Time To Compile1234080120160200Min: 217.09 / Avg: 220.1 / Max: 222.59Min: 217.43 / Avg: 219.58 / Max: 221.56Min: 220.92 / Avg: 222.67 / Max: 223.871. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 1080p123510152025SE +/- 0.03, N = 3SE +/- 0.16, N = 3SE +/- 0.14, N = 320.6920.6420.931. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 1080p123510152025Min: 20.63 / Avg: 20.69 / Max: 20.75Min: 20.34 / Avg: 20.64 / Max: 20.9Min: 20.74 / Avg: 20.93 / Max: 21.21. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

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: 8, Long Mode - Compression Speed1234080120160200SE +/- 0.70, N = 3SE +/- 0.88, N = 3SE +/- 0.95, N = 3159.6161.7159.61. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8, Long Mode - Compression Speed123306090120150Min: 158.3 / Avg: 159.6 / Max: 160.7Min: 160.3 / Avg: 161.67 / Max: 163.3Min: 157.7 / Avg: 159.6 / Max: 160.61. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression Speed12348121620SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 315.715.915.91. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression Speed12348121620Min: 15.6 / Avg: 15.73 / Max: 15.9Min: 15.8 / Avg: 15.87 / Max: 15.9Min: 15.8 / Avg: 15.9 / Max: 161. (CC) gcc options: -O3 -pthread -lz -llzma

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K123510152025SE +/- 0.30, N = 3SE +/- 0.08, N = 3SE +/- 0.13, N = 319.1419.1418.901. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K123510152025Min: 18.61 / Avg: 19.14 / Max: 19.64Min: 19.02 / Avg: 19.14 / Max: 19.28Min: 18.67 / Avg: 18.9 / Max: 19.121. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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 Speed1232004006008001000SE +/- 6.39, N = 3SE +/- 2.85, N = 3SE +/- 3.96, N = 3832.5839.8829.41. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3, Long Mode - Compression Speed123150300450600750Min: 820.5 / Avg: 832.53 / Max: 842.3Min: 834.5 / Avg: 839.77 / Max: 844.3Min: 821.8 / Avg: 829.37 / Max: 835.21. (CC) gcc options: -O3 -pthread -lz -llzma

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU1230.73641.47282.20922.94563.682SE +/- 0.02321, N = 3SE +/- 0.00772, N = 3SE +/- 0.01074, N = 33.233373.273043.23256MIN: 2.96MIN: 2.96MIN: 2.961. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU123246810Min: 3.19 / Avg: 3.23 / Max: 3.27Min: 3.26 / Avg: 3.27 / Max: 3.29Min: 3.22 / Avg: 3.23 / Max: 3.251. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 1.4.1Model: Lion12313002600390052006500SE +/- 35.20, N = 3SE +/- 19.35, N = 3SE +/- 51.45, N = 35908583759071. (CXX) g++ options: -O3
OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.4.1Model: Lion12310002000300040005000Min: 5870 / Avg: 5907.67 / Max: 5978Min: 5804 / Avg: 5837 / Max: 5871Min: 5850 / Avg: 5907.33 / Max: 60101. (CXX) g++ options: -O3

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU1233691215SE +/- 0.05, N = 3SE +/- 0.14, N = 3SE +/- 0.07, N = 313.2213.1513.06MIN: 12.54MIN: 12.5MIN: 12.541. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU12348121620Min: 13.16 / Avg: 13.22 / Max: 13.32Min: 12.9 / Avg: 13.15 / Max: 13.39Min: 13 / Avg: 13.06 / Max: 13.21. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian Dragon1231.2532.5063.7595.0126.265SE +/- 0.0240, N = 3SE +/- 0.0128, N = 3SE +/- 0.0093, N = 35.50455.56065.5687MIN: 5.45 / MAX: 5.61MIN: 5.5 / MAX: 5.65MIN: 5.51 / MAX: 5.65
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian Dragon123246810Min: 5.48 / Avg: 5.5 / Max: 5.55Min: 5.54 / Avg: 5.56 / Max: 5.58Min: 5.55 / Avg: 5.57 / Max: 5.58

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-T123714212835SE +/- 0.10, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 227.927.928.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-T123612182430Min: 27.8 / Avg: 27.9 / Max: 28.1Min: 27.8 / Avg: 27.9 / Max: 28Min: 28.1 / Avg: 28.15 / Max: 28.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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: 1M12330060090012001500SE +/- 2.09, N = 3SE +/- 3.37, N = 3SE +/- 4.66, N = 31383.51396.51398.31. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s, More Is BetterXmrig 6.12.1Variant: Monero - Hash Count: 1M1232004006008001000Min: 1379.5 / Avg: 1383.53 / Max: 1386.5Min: 1389.8 / Avg: 1396.47 / Max: 1400.6Min: 1389 / Avg: 1398.3 / Max: 1403.51. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

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: 8 - Compression Speed123306090120150SE +/- 1.13, N = 3SE +/- 2.11, N = 3SE +/- 0.79, N = 3149.8151.4151.01. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Compression Speed123306090120150Min: 147.6 / Avg: 149.83 / Max: 151.3Min: 148.6 / Avg: 151.37 / Max: 155.5Min: 149.5 / Avg: 151 / Max: 152.21. (CC) gcc options: -O3 -pthread -lz -llzma

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian Dragon Obj1230.99421.98842.98263.97684.971SE +/- 0.0034, N = 3SE +/- 0.0434, N = 3SE +/- 0.0062, N = 34.38434.41874.3735MIN: 4.31 / MAX: 4.46MIN: 4.3 / MAX: 4.54MIN: 4.28 / MAX: 4.45
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian Dragon Obj123246810Min: 4.38 / Avg: 4.38 / Max: 4.39Min: 4.37 / Avg: 4.42 / Max: 4.5Min: 4.36 / Avg: 4.37 / Max: 4.38

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Crown1230.9781.9562.9343.9124.89SE +/- 0.0215, N = 3SE +/- 0.0271, N = 3SE +/- 0.0048, N = 34.30384.31624.3467MIN: 4.23 / MAX: 4.42MIN: 4.23 / MAX: 4.42MIN: 4.32 / MAX: 4.42
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Crown123246810Min: 4.28 / Avg: 4.3 / Max: 4.35Min: 4.28 / Avg: 4.32 / Max: 4.37Min: 4.34 / Avg: 4.35 / Max: 4.36

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Summer Nature 4K12320406080100SE +/- 0.61, N = 3SE +/- 0.49, N = 3SE +/- 0.80, N = 379.7879.8779.12MIN: 73.41 / MAX: 96.41MIN: 73.2 / MAX: 95.15MIN: 70.79 / MAX: 96.231. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Summer Nature 4K1231530456075Min: 78.9 / Avg: 79.78 / Max: 80.94Min: 79.15 / Avg: 79.87 / Max: 80.81Min: 78.11 / Avg: 79.12 / Max: 80.71. (CC) gcc options: -pthread -lm

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Chimera 1080p12370140210280350SE +/- 5.12, N = 3SE +/- 3.12, N = 3SE +/- 4.19, N = 3305.61307.02304.14MIN: 221.23 / MAX: 526MIN: 218.37 / MAX: 513.03MIN: 217.43 / MAX: 511.321. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Chimera 1080p12360120180240300Min: 295.82 / Avg: 305.61 / Max: 313.09Min: 303.04 / Avg: 307.02 / Max: 313.17Min: 298.27 / Avg: 304.14 / Max: 312.251. (CC) gcc options: -pthread -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 2021.2Implementation: MPI CPU - Input: water_GMX50_bare1230.09950.1990.29850.3980.4975SE +/- 0.002, N = 3SE +/- 0.002, N = 3SE +/- 0.001, N = 30.4410.4420.4381. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021.2Implementation: MPI CPU - Input: water_GMX50_bare12312345Min: 0.44 / Avg: 0.44 / Max: 0.45Min: 0.44 / Avg: 0.44 / Max: 0.45Min: 0.44 / Avg: 0.44 / Max: 0.441. (CXX) g++ options: -O3 -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 - Decompression Speed1236001200180024003000SE +/- 12.88, N = 3SE +/- 1.45, N = 3SE +/- 3.35, N = 32712.52730.72736.81. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression Speed1235001000150020002500Min: 2691.2 / Avg: 2712.53 / Max: 2735.7Min: 2729.2 / Avg: 2730.7 / Max: 2733.6Min: 2730.8 / Avg: 2736.8 / Max: 2742.41. (CC) gcc options: -O3 -pthread -lz -llzma

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p12320406080100SE +/- 0.68, N = 3SE +/- 0.48, N = 3SE +/- 0.54, N = 388.6289.2789.411. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p12320406080100Min: 87.77 / Avg: 88.62 / Max: 89.97Min: 88.34 / Avg: 89.27 / Max: 89.95Min: 88.41 / Avg: 89.41 / Max: 90.241. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p12320406080100SE +/- 0.04, N = 3SE +/- 0.52, N = 3SE +/- 0.03, N = 379.6378.9879.681. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p1231530456075Min: 79.55 / Avg: 79.63 / Max: 79.7Min: 77.95 / Avg: 78.98 / Max: 79.5Min: 79.65 / Avg: 79.68 / Max: 79.731. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

OpenSCAD

OpenSCAD is a programmer-focused solid 3D CAD modeller. OpenSCAD is free software and allows creating 3D CAD objects in a script-based modelling environment. This test profile will use the system-provided OpenSCAD program otherwise and time how long it takes tn render different SCAD assets to PNG output. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Retro Car123612182430SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 323.4623.3523.551. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Retro Car123510152025Min: 23.33 / Avg: 23.46 / Max: 23.61Min: 23.26 / Avg: 23.35 / Max: 23.43Min: 23.47 / Avg: 23.55 / Max: 23.631. OpenSCAD version 2019.05

GNU Radio

GNU Radio is a free software development toolkit providing signal processing blocks to implement software-defined radios (SDR) and signal processing systems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Signal Source (Cosine)1235001000150020002500SE +/- 15.05, N = 9SE +/- 15.62, N = 5SE +/- 0.80, N = 32328.12323.52343.51. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Signal Source (Cosine)123400800120016002000Min: 2210.9 / Avg: 2328.08 / Max: 2358.8Min: 2263.5 / Avg: 2323.5 / Max: 2353Min: 2342 / Avg: 2343.53 / Max: 2344.71. 3.8.1.0

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: 8 - Decompression Speed1237001400210028003500SE +/- 3.13, N = 3SE +/- 12.27, N = 3SE +/- 7.63, N = 33220.53193.23214.61. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Decompression Speed1236001200180024003000Min: 3215.7 / Avg: 3220.53 / Max: 3226.4Min: 3169.3 / Avg: 3193.2 / Max: 3210Min: 3204.8 / Avg: 3214.57 / Max: 3229.61. (CC) gcc options: -O3 -pthread -lz -llzma

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU1233691215SE +/- 0.08274, N = 3SE +/- 0.01033, N = 3SE +/- 0.02197, N = 39.474339.394269.41080MIN: 8.6MIN: 8.57MIN: 8.571. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU1233691215Min: 9.38 / Avg: 9.47 / Max: 9.64Min: 9.38 / Avg: 9.39 / Max: 9.41Min: 9.37 / Avg: 9.41 / Max: 9.451. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 0.9.0Encoder Speed: 212320406080100SE +/- 0.24, N = 3SE +/- 0.10, N = 3SE +/- 0.10, N = 378.3578.2977.691. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 21231530456075Min: 77.87 / Avg: 78.35 / Max: 78.64Min: 78.1 / Avg: 78.29 / Max: 78.41Min: 77.57 / Avg: 77.69 / Max: 77.891. (CXX) g++ options: -O3 -fPIC -lm

Chia Blockchain VDF

Chia is a blockchain and smart transaction platform based on proofs of space and time rather than proofs of work with other cryptocurrencies. This test profile is benchmarking the CPU performance for Chia VDF performance using the Chia VDF benchmark. The Chia VDF is for the Chia Verifiable Delay Function (Proof of Time). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.1Test: Square Assembly Optimized12340K80K120K160K200KSE +/- 233.33, N = 3SE +/- 1133.33, N = 3SE +/- 1519.14, N = 31781671796671784671. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.1Test: Square Assembly Optimized12330K60K90K120K150KMin: 177800 / Avg: 178166.67 / Max: 178600Min: 177400 / Avg: 179666.67 / Max: 180800Min: 176800 / Avg: 178466.67 / Max: 1815001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread

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 - Decrypt12390180270360450SE +/- 0.27, N = 3SE +/- 2.73, N = 3SE +/- 0.73, N = 3403.84401.52404.451. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Blowfish - Decrypt12370140210280350Min: 403.49 / Avg: 403.84 / Max: 404.37Min: 396.09 / Avg: 401.52 / Max: 404.64Min: 403.4 / Avg: 404.45 / Max: 405.851. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

srsLTE

srsLTE is an open-source LTE software radio suite created by Software Radio Systems (SRS). srsLTE can be used for building your own software defined (SDR) LTE mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsLTE 20.10.1Test: PHY_DL_Test12320406080100SE +/- 0.46, N = 3SE +/- 0.22, N = 3SE +/- 0.15, N = 383.282.983.51. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f
OpenBenchmarking.orgUE Mb/s, More Is BettersrsLTE 20.10.1Test: PHY_DL_Test1231632486480Min: 82.3 / Avg: 83.2 / Max: 83.8Min: 82.5 / Avg: 82.93 / Max: 83.2Min: 83.2 / Avg: 83.47 / Max: 83.71. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f

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 Analysis1234080120160200SE +/- 0.04, N = 3SE +/- 0.37, N = 3SE +/- 0.47, N = 3166.26167.10167.451. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mmpx -mabm -O3 -std=c99 -pedantic -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny Analysis123306090120150Min: 166.19 / Avg: 166.26 / Max: 166.31Min: 166.41 / Avg: 167.1 / Max: 167.69Min: 166.56 / Avg: 167.45 / Max: 168.161. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mmpx -mabm -O3 -std=c99 -pedantic -lm

SecureMark

SecureMark is an objective, standardized benchmarking framework for measuring the efficiency of cryptographic processing solutions developed by EEMBC. SecureMark-TLS is benchmarking Transport Layer Security performance with a focus on IoT/edge computing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLS12350K100K150K200K250KSE +/- 1331.28, N = 3SE +/- 111.50, N = 3SE +/- 260.74, N = 32131812146852144711. (CC) gcc options: -pedantic -O3
OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLS12340K80K120K160K200KMin: 210526.72 / Avg: 213180.54 / Max: 214694.16Min: 214465.53 / Avg: 214684.54 / Max: 214830.39Min: 213953 / Avg: 214471.2 / Max: 214780.811. (CC) gcc options: -pedantic -O3

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU123510152025SE +/- 0.13, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 322.7622.7822.63MIN: 22.36MIN: 22.36MIN: 22.381. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU123510152025Min: 22.6 / Avg: 22.76 / Max: 23.01Min: 22.72 / Avg: 22.78 / Max: 22.84Min: 22.57 / Avg: 22.63 / Max: 22.661. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TT12348121620SE +/- 0.03, N = 3SE +/- 0.05, N = 214.514.414.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TT12348121620Min: 14.4 / Avg: 14.43 / Max: 14.5Min: 14.4 / Avg: 14.45 / Max: 14.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TN12348121620SE +/- 0.03, N = 3SE +/- 0.05, N = 2SE +/- 0.00, N = 214.414.514.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TN12348121620Min: 14.4 / Avg: 14.43 / Max: 14.5Min: 14.4 / Avg: 14.45 / Max: 14.5Min: 14.4 / Avg: 14.4 / Max: 14.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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: 4 - Buffer Length: 256 - Filter Length: 5712340M80M120M160M200MSE +/- 679713.99, N = 3SE +/- 243607.15, N = 3SE +/- 648082.64, N = 31715733331727533331721166671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 4 - Buffer Length: 256 - Filter Length: 5712330M60M90M120M150MMin: 170550000 / Avg: 171573333.33 / Max: 172860000Min: 172490000 / Avg: 172753333.33 / Max: 173240000Min: 170860000 / Avg: 172116666.67 / Max: 1730200001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NT12348121620SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 314.714.714.81. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NT12348121620Min: 14.7 / Avg: 14.73 / Max: 14.8Min: 14.7 / Avg: 14.7 / Max: 14.7Min: 14.8 / Avg: 14.8 / Max: 14.81. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

srsLTE

srsLTE is an open-source LTE software radio suite created by Software Radio Systems (SRS). srsLTE can be used for building your own software defined (SDR) LTE mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSamples / Second, More Is BettersrsLTE 20.10.1Test: OFDM_Test12320M40M60M80M100MSE +/- 272845.09, N = 3SE +/- 536449.23, N = 3SE +/- 480740.17, N = 39746666796833333972333331. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f
OpenBenchmarking.orgSamples / Second, More Is BettersrsLTE 20.10.1Test: OFDM_Test12320M40M60M80M100MMin: 97100000 / Avg: 97466666.67 / Max: 98000000Min: 96200000 / Avg: 96833333.33 / Max: 97900000Min: 96300000 / Avg: 97233333.33 / Max: 979000001. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f

LuaRadio

LuaRadio is a lightweight software-defined radio (SDR) framework built atop LuaJIT. LuaRadio provides a suite of source, sink, and processing blocks, with a simple API for defining flow graphs, running flow graphs, creating blocks, and creating data types. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: FM Deemphasis Filter12380160240320400SE +/- 0.54, N = 3SE +/- 0.95, N = 3SE +/- 3.40, N = 3354.8354.2352.5
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: FM Deemphasis Filter12360120180240300Min: 353.8 / Avg: 354.83 / Max: 355.6Min: 352.6 / Avg: 354.23 / Max: 355.9Min: 345.9 / Avg: 352.47 / Max: 357.3

Basis Universal

Basis Universal is a GPU texture codec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: UASTC Level 01233691215SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 311.8411.9111.881. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: UASTC Level 01233691215Min: 11.77 / Avg: 11.84 / Max: 11.94Min: 11.83 / Avg: 11.91 / Max: 11.96Min: 11.79 / Avg: 11.88 / Max: 11.981. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

LuaRadio

LuaRadio is a lightweight software-defined radio (SDR) framework built atop LuaJIT. LuaRadio provides a suite of source, sink, and processing blocks, with a simple API for defining flow graphs, running flow graphs, creating blocks, and creating data types. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Five Back to Back FIR Filters123130260390520650SE +/- 2.76, N = 3SE +/- 2.46, N = 3SE +/- 4.28, N = 3618.4614.5616.9
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Five Back to Back FIR Filters123110220330440550Min: 615.2 / Avg: 618.4 / Max: 623.9Min: 611.2 / Avg: 614.5 / Max: 619.3Min: 610.5 / Avg: 616.87 / Max: 625

GNU Radio

GNU Radio is a free software development toolkit providing signal processing blocks to implement software-defined radios (SDR) and signal processing systems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FIR Filter123110220330440550SE +/- 2.92, N = 9SE +/- 7.21, N = 5SE +/- 0.26, N = 3508.1508.2505.11. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FIR Filter12390180270360450Min: 497.1 / Avg: 508.12 / Max: 520.9Min: 481 / Avg: 508.24 / Max: 522.7Min: 504.6 / Avg: 505.07 / Max: 505.51. 3.8.1.0

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Summer Nature 1080p12360120180240300SE +/- 2.78, N = 3SE +/- 2.72, N = 3SE +/- 1.38, N = 3284.41285.21286.14MIN: 255.79 / MAX: 317.93MIN: 259.98 / MAX: 316.02MIN: 258.64 / MAX: 317.821. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Summer Nature 1080p12350100150200250Min: 278.95 / Avg: 284.41 / Max: 288.04Min: 279.78 / Avg: 285.21 / Max: 288.17Min: 283.46 / Avg: 286.14 / Max: 288.041. (CC) gcc options: -pthread -lm

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 4 + Zstd Compression 191232004006008001000SE +/- 0.19, N = 3SE +/- 0.48, N = 3SE +/- 3.49, N = 3936.45936.70941.99
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 4 + Zstd Compression 19123170340510680850Min: 936.13 / Avg: 936.45 / Max: 936.78Min: 935.83 / Avg: 936.7 / Max: 937.47Min: 938.44 / Avg: 941.99 / Max: 948.98

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 - Decompression Speed1237001400210028003500SE +/- 7.68, N = 3SE +/- 9.12, N = 3SE +/- 4.11, N = 33123.43105.23122.51. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Decompression Speed1235001000150020002500Min: 3113.3 / Avg: 3123.43 / Max: 3138.5Min: 3087.3 / Avg: 3105.17 / Max: 3117.3Min: 3116 / Avg: 3122.5 / Max: 3130.11. (CC) gcc options: -O3 -pthread -lz -llzma

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Chimera 1080p 10-bit12350100150200250SE +/- 2.07, N = 3SE +/- 2.06, N = 3SE +/- 2.72, N = 3231.78230.44230.48MIN: 171.72 / MAX: 395.29MIN: 175.44 / MAX: 385.15MIN: 172.7 / MAX: 388.11. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Chimera 1080p 10-bit1234080120160200Min: 228.03 / Avg: 231.78 / Max: 235.16Min: 228.19 / Avg: 230.44 / Max: 234.56Min: 225.49 / Avg: 230.48 / Max: 234.841. (CC) gcc options: -pthread -lm

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPY12348121620SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 317.217.217.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPY12348121620Min: 17.1 / Avg: 17.23 / Max: 17.3Min: 17.1 / Avg: 17.17 / Max: 17.2Min: 17.3 / Avg: 17.3 / Max: 17.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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 Speed12348121620SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 317.217.317.21. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression Speed12348121620Min: 17.2 / Avg: 17.23 / Max: 17.3Min: 17.3 / Avg: 17.3 / Max: 17.3Min: 17.2 / Avg: 17.2 / Max: 17.21. (CC) gcc options: -O3 -pthread -lz -llzma

GNU Radio

GNU Radio is a free software development toolkit providing signal processing blocks to implement software-defined radios (SDR) and signal processing systems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Hilbert Transform12390180270360450SE +/- 1.30, N = 9SE +/- 3.03, N = 5SE +/- 3.02, N = 3421.6420.4422.81. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Hilbert Transform12380160240320400Min: 415.9 / Avg: 421.57 / Max: 427.3Min: 413.2 / Avg: 420.42 / Max: 427.5Min: 418.5 / Avg: 422.77 / Max: 428.61. 3.8.1.0

OpenSCAD

OpenSCAD is a programmer-focused solid 3D CAD modeller. OpenSCAD is free software and allows creating 3D CAD objects in a script-based modelling environment. This test profile will use the system-provided OpenSCAD program otherwise and time how long it takes tn render different SCAD assets to PNG output. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Mini-ITX Case1231326395265SE +/- 0.22, N = 3SE +/- 0.15, N = 3SE +/- 0.13, N = 356.3456.1656.031. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Mini-ITX Case1231122334455Min: 56.06 / Avg: 56.34 / Max: 56.77Min: 55.94 / Avg: 56.16 / Max: 56.44Min: 55.85 / Avg: 56.03 / Max: 56.281. OpenSCAD version 2019.05

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p1230.82131.64262.46393.28524.1065SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 33.633.653.651. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p123246810Min: 3.63 / Avg: 3.63 / Max: 3.63Min: 3.65 / Avg: 3.65 / Max: 3.65Min: 3.64 / Avg: 3.65 / Max: 3.651. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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 0.9.0Encoder Speed: 6, Lossless123306090120150SE +/- 0.25, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 3154.17153.91154.731. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, Lossless123306090120150Min: 153.72 / Avg: 154.17 / Max: 154.59Min: 153.81 / Avg: 153.91 / Max: 154Min: 154.55 / Avg: 154.73 / Max: 154.91. (CXX) g++ options: -O3 -fPIC -lm

Helsing

Helsing is an open-source POSIX vampire number generator. This test profile measures the time it takes to generate vampire numbers between varying numbers of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 12 digit12348121620SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 316.5716.6616.581. (CC) gcc options: -O2 -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 12 digit12348121620Min: 16.55 / Avg: 16.57 / Max: 16.59Min: 16.6 / Avg: 16.66 / Max: 16.73Min: 16.45 / Avg: 16.58 / Max: 16.681. (CC) gcc options: -O2 -pthread

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080p123918273645SE +/- 0.01, N = 3SE +/- 0.12, N = 3SE +/- 0.01, N = 341.1941.1041.311. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080p123918273645Min: 41.17 / Avg: 41.19 / Max: 41.21Min: 40.88 / Avg: 41.1 / Max: 41.27Min: 41.3 / Avg: 41.31 / Max: 41.331. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU1231.30432.60863.91295.21726.5215SE +/- 0.00339, N = 3SE +/- 0.01052, N = 3SE +/- 0.01535, N = 35.768405.796715.79353MIN: 5.16MIN: 5.11MIN: 5.111. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU123246810Min: 5.76 / Avg: 5.77 / Max: 5.77Min: 5.78 / Avg: 5.8 / Max: 5.81Min: 5.78 / Avg: 5.79 / Max: 5.821. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080p12320406080100SE +/- 0.09, N = 3SE +/- 0.41, N = 3SE +/- 0.03, N = 391.0690.7291.161. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080p12320406080100Min: 90.88 / Avg: 91.06 / Max: 91.2Min: 89.91 / Avg: 90.72 / Max: 91.27Min: 91.1 / Avg: 91.16 / Max: 91.21. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

OpenSCAD

OpenSCAD is a programmer-focused solid 3D CAD modeller. OpenSCAD is free software and allows creating 3D CAD objects in a script-based modelling environment. This test profile will use the system-provided OpenSCAD program otherwise and time how long it takes tn render different SCAD assets to PNG output. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Leonardo Phone Case Slim123510152025SE +/- 0.10, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 322.6222.5122.541. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Leonardo Phone Case Slim123510152025Min: 22.43 / Avg: 22.62 / Max: 22.76Min: 22.49 / Avg: 22.51 / Max: 22.53Min: 22.46 / Avg: 22.54 / Max: 22.61. OpenSCAD version 2019.05

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU1232K4K6K8K10KSE +/- 4.99, N = 3SE +/- 14.87, N = 3SE +/- 4.56, N = 39167.749190.489146.56MIN: 9107.33MIN: 9115.93MIN: 9085.121. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU12316003200480064008000Min: 9159.47 / Avg: 9167.74 / Max: 9176.71Min: 9160.78 / Avg: 9190.48 / Max: 9206.61Min: 9139.27 / Avg: 9146.56 / Max: 9154.951. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3 + Zstd Compression 19123918273645SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 338.0838.1238.26
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3 + Zstd Compression 19123816243240Min: 38.05 / Avg: 38.08 / Max: 38.09Min: 38.08 / Avg: 38.12 / Max: 38.18Min: 38.13 / Avg: 38.26 / Max: 38.39

GNU Radio

GNU Radio is a free software development toolkit providing signal processing blocks to implement software-defined radios (SDR) and signal processing systems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: IIR Filter123100200300400500SE +/- 1.01, N = 9SE +/- 0.85, N = 5SE +/- 0.51, N = 3471.5469.4469.51. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: IIR Filter12380160240320400Min: 468.6 / Avg: 471.47 / Max: 477.4Min: 467.8 / Avg: 469.4 / Max: 472.1Min: 468.5 / Avg: 469.5 / Max: 470.21. 3.8.1.0

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 1080p123612182430SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 323.0423.1423.121. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 1080p123510152025Min: 22.97 / Avg: 23.04 / Max: 23.11Min: 23.09 / Avg: 23.14 / Max: 23.21Min: 23.07 / Avg: 23.12 / Max: 23.171. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

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 0.9.0Encoder Speed: 101231.2292.4583.6874.9166.145SE +/- 0.012, N = 3SE +/- 0.008, N = 3SE +/- 0.008, N = 35.4415.4625.4381. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10123246810Min: 5.42 / Avg: 5.44 / Max: 5.46Min: 5.45 / Avg: 5.46 / Max: 5.48Min: 5.42 / Avg: 5.44 / Max: 5.451. (CXX) g++ options: -O3 -fPIC -lm

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU1230.98541.97082.95623.94164.927SE +/- 0.01013, N = 3SE +/- 0.00061, N = 3SE +/- 0.00552, N = 34.379394.373484.36028MIN: 3.78MIN: 3.8MIN: 3.81. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU123246810Min: 4.37 / Avg: 4.38 / Max: 4.4Min: 4.37 / Avg: 4.37 / Max: 4.37Min: 4.35 / Avg: 4.36 / Max: 4.371. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p1231632486480SE +/- 0.30, N = 3SE +/- 0.32, N = 3SE +/- 0.32, N = 371.1571.3271.461. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p1231428425670Min: 70.8 / Avg: 71.15 / Max: 71.76Min: 70.97 / Avg: 71.32 / Max: 71.95Min: 71.04 / Avg: 71.46 / Max: 72.091. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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 1.4.1Model: Church Facade1232K4K6K8K10KSE +/- 14.99, N = 3SE +/- 9.64, N = 3SE +/- 22.56, N = 38888888789251. (CXX) g++ options: -O3
OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.4.1Model: Church Facade12315003000450060007500Min: 8860 / Avg: 8888.33 / Max: 8911Min: 8872 / Avg: 8887 / Max: 8905Min: 8887 / Avg: 8924.67 / Max: 89651. (CXX) g++ options: -O3

Xcompact3d Incompact3d

Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per Direction1231428425670SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 362.1561.8962.041. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent -levent_pthreads -lutil -lm -lrt -lz
OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per Direction1231224364860Min: 62.06 / Avg: 62.15 / Max: 62.23Min: 61.83 / Avg: 61.89 / Max: 61.92Min: 61.99 / Avg: 62.04 / Max: 62.111. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent -levent_pthreads -lutil -lm -lrt -lz

PJSIP

PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, Stateless12311K22K33K44K55KSE +/- 85.25, N = 3SE +/- 93.14, N = 3SE +/- 85.08, N = 35107050873508541. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, Stateless1239K18K27K36K45KMin: 50971 / Avg: 51070.33 / Max: 51240Min: 50695 / Avg: 50872.67 / Max: 51010Min: 50713 / Avg: 50854 / Max: 510071. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: mobilenet-v1-1.01231.18012.36023.54034.72045.9005SE +/- 0.023, N = 3SE +/- 0.039, N = 3SE +/- 0.027, N = 35.2235.2455.240MIN: 4.84 / MAX: 8.49MIN: 4.75 / MAX: 24.02MIN: 4.87 / MAX: 9.521. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: mobilenet-v1-1.0123246810Min: 5.18 / Avg: 5.22 / Max: 5.26Min: 5.19 / Avg: 5.24 / Max: 5.32Min: 5.2 / Avg: 5.24 / Max: 5.291. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

BRL-CAD

BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.32.2VGR Performance Metric1239K18K27K36K45K4218242310421351. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -lm -pthread -ldl

Chaos Group V-RAY

This is a test of Chaos Group's V-RAY benchmark. V-RAY is a commercial renderer that can integrate with various creator software products like SketchUp and 3ds Max. The V-RAY benchmark is standalone and supports CPU and NVIDIA CUDA/RTX based rendering. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgvsamples, More Is BetterChaos Group V-RAY 5Mode: CPU1238001600240032004000SE +/- 15.72, N = 3SE +/- 17.10, N = 3SE +/- 10.60, N = 3362936143625
OpenBenchmarking.orgvsamples, More Is BetterChaos Group V-RAY 5Mode: CPU1236001200180024003000Min: 3604 / Avg: 3629 / Max: 3658Min: 3583 / Avg: 3614 / Max: 3642Min: 3612 / Avg: 3625 / Max: 3646

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU1231.26822.53643.80465.07286.341SE +/- 0.02138, N = 3SE +/- 0.01229, N = 3SE +/- 0.00266, N = 35.636355.633775.61310MIN: 5.39MIN: 5.39MIN: 5.391. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU123246810Min: 5.6 / Avg: 5.64 / Max: 5.68Min: 5.62 / Avg: 5.63 / Max: 5.66Min: 5.61 / Avg: 5.61 / Max: 5.621. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

srsLTE

srsLTE is an open-source LTE software radio suite created by Software Radio Systems (SRS). srsLTE can be used for building your own software defined (SDR) LTE mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsLTE 20.10.1Test: PHY_DL_Test12350100150200250SE +/- 0.06, N = 3SE +/- 0.32, N = 3SE +/- 0.32, N = 3220.7220.3221.21. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f
OpenBenchmarking.orgeNb Mb/s, More Is BettersrsLTE 20.10.1Test: PHY_DL_Test1234080120160200Min: 220.6 / Avg: 220.7 / Max: 220.8Min: 219.7 / Avg: 220.33 / Max: 220.7Min: 220.6 / Avg: 221.2 / Max: 221.71. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 15.11Time To Compile12330060090012001500SE +/- 0.13, N = 3SE +/- 1.30, N = 3SE +/- 0.40, N = 31251.931257.021256.61
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 15.11Time To Compile1232004006008001000Min: 1251.72 / Avg: 1251.93 / Max: 1252.16Min: 1255.57 / Avg: 1257.02 / Max: 1259.62Min: 1255.86 / Avg: 1256.61 / Max: 1257.23

OpenSCAD

OpenSCAD is a programmer-focused solid 3D CAD modeller. OpenSCAD is free software and allows creating 3D CAD objects in a script-based modelling environment. This test profile will use the system-provided OpenSCAD program otherwise and time how long it takes tn render different SCAD assets to PNG output. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Projector Mount Swivel123306090120150SE +/- 0.11, N = 3SE +/- 0.31, N = 3SE +/- 0.11, N = 3125.68126.18126.151. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Projector Mount Swivel12320406080100Min: 125.49 / Avg: 125.68 / Max: 125.89Min: 125.79 / Avg: 126.18 / Max: 126.79Min: 125.95 / Avg: 126.15 / Max: 126.311. OpenSCAD version 2019.05

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPY123612182430SE +/- 0.03, N = 3SE +/- 0.15, N = 3SE +/- 0.10, N = 325.725.625.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPY123612182430Min: 25.7 / Avg: 25.73 / Max: 25.8Min: 25.3 / Avg: 25.6 / Max: 25.8Min: 25.4 / Avg: 25.6 / Max: 25.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 0 - Input: Bosphorus 4K1230.58281.16561.74842.33122.914SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.582.592.591. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 0 - Input: Bosphorus 4K123246810Min: 2.57 / Avg: 2.58 / Max: 2.59Min: 2.59 / Avg: 2.59 / Max: 2.59Min: 2.58 / Avg: 2.59 / Max: 2.591. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080p1230.58281.16561.74842.33122.914SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.582.582.591. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080p123246810Min: 2.58 / Avg: 2.58 / Max: 2.59Min: 2.58 / Avg: 2.58 / Max: 2.59Min: 2.59 / Avg: 2.59 / Max: 2.591. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

toyBrot Fractal Generator

ToyBrot is a Mandelbrot fractal generator supporting C++ threads/tasks, OpenMP, Intel Threaded Building Blocks (TBB), and other targets. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: OpenMP12320K40K60K80K100KSE +/- 268.62, N = 3SE +/- 256.16, N = 3SE +/- 189.15, N = 31147891152271150241. (CXX) g++ options: -O3 -lpthread
OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: OpenMP12320K40K60K80K100KMin: 114486 / Avg: 114789.33 / Max: 115325Min: 114791 / Avg: 115227 / Max: 115678Min: 114646 / Avg: 115024 / Max: 1152261. (CXX) g++ options: -O3 -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: Blowfish12390180270360450SE +/- 0.25, N = 3SE +/- 1.19, N = 3SE +/- 0.37, N = 3409.47408.00409.551. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Blowfish12370140210280350Min: 408.99 / Avg: 409.47 / Max: 409.86Min: 405.85 / Avg: 407.99 / Max: 409.95Min: 408.98 / Avg: 409.55 / Max: 410.251. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPY123612182430SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 326.326.426.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPY123612182430Min: 26.2 / Avg: 26.33 / Max: 26.5Min: 26.3 / Avg: 26.37 / Max: 26.4Min: 26.4 / Avg: 26.43 / Max: 26.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU12310002000300040005000SE +/- 3.83, N = 3SE +/- 3.69, N = 3SE +/- 4.86, N = 34768.624776.654786.18MIN: 4732.74MIN: 4731.36MIN: 4747.591. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU1238001600240032004000Min: 4762.13 / Avg: 4768.62 / Max: 4775.39Min: 4770.51 / Avg: 4776.65 / Max: 4783.25Min: 4780.23 / Avg: 4786.18 / Max: 4795.811. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -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: Twofish - Decrypt12370140210280350SE +/- 0.32, N = 3SE +/- 0.50, N = 3SE +/- 0.46, N = 3333.50332.66332.281. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Twofish - Decrypt12360120180240300Min: 332.87 / Avg: 333.5 / Max: 333.85Min: 331.88 / Avg: 332.66 / Max: 333.58Min: 331.75 / Avg: 332.28 / Max: 333.191. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p1233691215SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 310.9711.0111.011. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p1233691215Min: 10.94 / Avg: 10.97 / Max: 10.99Min: 10.99 / Avg: 11.01 / Max: 11.02Min: 10.99 / Avg: 11.01 / Max: 11.031. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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 0.9.0Encoder Speed: 6123612182430SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 326.3826.2926.301. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6123612182430Min: 26.34 / Avg: 26.38 / Max: 26.46Min: 26.23 / Avg: 26.29 / Max: 26.32Min: 26.27 / Avg: 26.3 / Max: 26.321. (CXX) g++ options: -O3 -fPIC -lm

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOT123714212835SE +/- 0.03, N = 3SE +/- 0.00, N = 2SE +/- 0.15, N = 228.428.528.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOT123612182430Min: 28.4 / Avg: 28.43 / Max: 28.5Min: 28.5 / Avg: 28.5 / Max: 28.5Min: 28.2 / Avg: 28.35 / Max: 28.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-N123714212835SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 328.828.828.91. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-N123612182430Min: 28.7 / Avg: 28.77 / Max: 28.8Min: 28.8 / Avg: 28.8 / Max: 28.8Min: 28.8 / Avg: 28.87 / Max: 28.91. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: resnet-v2-501231122334455SE +/- 0.29, N = 3SE +/- 0.45, N = 3SE +/- 0.20, N = 350.2550.1150.08MIN: 48.6 / MAX: 71.15MIN: 48.72 / MAX: 79.21MIN: 45.78 / MAX: 121.751. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: resnet-v2-501231020304050Min: 49.71 / Avg: 50.25 / Max: 50.71Min: 49.65 / Avg: 50.11 / Max: 51.01Min: 49.79 / Avg: 50.08 / Max: 50.471. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -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: 2 - Buffer Length: 256 - Filter Length: 5712320M40M60M80M100MSE +/- 136148.20, N = 3SE +/- 234310.57, N = 3SE +/- 170911.61, N = 39546300095382667957073331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 2 - Buffer Length: 256 - Filter Length: 5712317M34M51M68M85MMin: 95316000 / Avg: 95463000 / Max: 95735000Min: 94917000 / Avg: 95382666.67 / Max: 95661000Min: 95409000 / Avg: 95707333.33 / Max: 960010001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K1230.67281.34562.01842.69123.364SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.982.992.981. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K123246810Min: 2.98 / Avg: 2.98 / Max: 2.99Min: 2.99 / Avg: 2.99 / Max: 2.99Min: 2.98 / Avg: 2.98 / Max: 2.991. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 0 - Input: Bosphorus 1080p123246810SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 36.036.046.051. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 0 - Input: Bosphorus 1080p123246810Min: 5.99 / Avg: 6.03 / Max: 6.08Min: 6.01 / Avg: 6.04 / Max: 6.09Min: 6.01 / Avg: 6.05 / Max: 6.11. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 4K1230.13820.27640.41460.55280.691SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 30.6140.6120.6141. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 4K123246810Min: 0.61 / Avg: 0.61 / Max: 0.61Min: 0.61 / Avg: 0.61 / Max: 0.62Min: 0.61 / Avg: 0.61 / Max: 0.621. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 1080p123306090120150SE +/- 0.26, N = 3SE +/- 0.66, N = 3SE +/- 0.11, N = 3158.09157.59157.991. (CXX) g++ options: -O3 -lpthread -lrt
OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 1080p123306090120150Min: 157.61 / Avg: 158.09 / Max: 158.5Min: 156.31 / Avg: 157.59 / Max: 158.49Min: 157.77 / Avg: 157.99 / Max: 158.141. (CXX) g++ options: -O3 -lpthread -lrt

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p1233691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 39.469.499.461. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p1233691215Min: 9.45 / Avg: 9.46 / Max: 9.48Min: 9.47 / Avg: 9.49 / Max: 9.51Min: 9.41 / Avg: 9.46 / Max: 9.491. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 1080p1230.44240.88481.32721.76962.212SE +/- 0.003, N = 3SE +/- 0.002, N = 3SE +/- 0.004, N = 31.9661.9601.9651. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 1080p123246810Min: 1.96 / Avg: 1.97 / Max: 1.97Min: 1.96 / Avg: 1.96 / Max: 1.96Min: 1.96 / Avg: 1.97 / Max: 1.971. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU1232K4K6K8K10KSE +/- 7.07, N = 3SE +/- 16.27, N = 3SE +/- 8.57, N = 39161.379189.239168.42MIN: 9119.77MIN: 9124.7MIN: 9110.091. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU12316003200480064008000Min: 9147.97 / Avg: 9161.37 / Max: 9171.97Min: 9172.3 / Avg: 9189.23 / Max: 9221.76Min: 9155.89 / Avg: 9168.42 / Max: 9184.811. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

LuaRadio

LuaRadio is a lightweight software-defined radio (SDR) framework built atop LuaJIT. LuaRadio provides a suite of source, sink, and processing blocks, with a simple API for defining flow graphs, running flow graphs, creating blocks, and creating data types. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Hilbert Transform1231530456075SE +/- 0.03, N = 3SE +/- 0.15, N = 3SE +/- 0.06, N = 366.266.066.0
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Hilbert Transform1231326395265Min: 66.2 / Avg: 66.23 / Max: 66.3Min: 65.8 / Avg: 66 / Max: 66.3Min: 65.9 / Avg: 66 / Max: 66.1

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU12348121620SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 315.7115.7515.71MIN: 11.12MIN: 11.16MIN: 11.311. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU12348121620Min: 15.62 / Avg: 15.71 / Max: 15.75Min: 15.73 / Avg: 15.75 / Max: 15.77Min: 15.67 / Avg: 15.71 / Max: 15.751. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -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: 5712311M22M33M44M55MSE +/- 46666.67, N = 3SE +/- 51705.36, N = 3SE +/- 19341.95, N = 35144666751597333515003331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 1 - Buffer Length: 256 - Filter Length: 571239M18M27M36M45MMin: 51360000 / Avg: 51446666.67 / Max: 51520000Min: 51506000 / Avg: 51597333.33 / Max: 51685000Min: 51480000 / Avg: 51500333.33 / Max: 515390001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU12348121620SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 316.8116.8116.77MIN: 15.54MIN: 15.58MIN: 15.551. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU12348121620Min: 16.76 / Avg: 16.81 / Max: 16.86Min: 16.78 / Avg: 16.81 / Max: 16.86Min: 16.73 / Avg: 16.77 / Max: 16.81. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Chimera 1080p 10-bit123816243240SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 336.5436.6436.611. (CXX) g++ options: -O3 -lpthread -lrt
OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Chimera 1080p 10-bit123816243240Min: 36.51 / Avg: 36.54 / Max: 36.57Min: 36.63 / Avg: 36.64 / Max: 36.66Min: 36.56 / Avg: 36.61 / Max: 36.631. (CXX) g++ options: -O3 -lpthread -lrt

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: C75521234080120160200SE +/- 0.92, N = 3SE +/- 0.73, N = 3SE +/- 1.70, N = 3161.26161.16160.831. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE
OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552123306090120150Min: 159.97 / Avg: 161.26 / Max: 163.04Min: 159.81 / Avg: 161.16 / Max: 162.33Min: 158.31 / Avg: 160.83 / Max: 164.081. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

Basis Universal

Basis Universal is a GPU texture codec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: UASTC Level 31234080120160200SE +/- 0.10, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3158.61158.60159.031. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: UASTC Level 3123306090120150Min: 158.5 / Avg: 158.61 / Max: 158.81Min: 158.5 / Avg: 158.6 / Max: 158.68Min: 158.96 / Avg: 159.03 / Max: 159.111. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: ETC1S123918273645SE +/- 0.08, N = 3SE +/- 0.15, N = 3SE +/- 0.10, N = 338.1338.2338.171. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: ETC1S123816243240Min: 37.98 / Avg: 38.13 / Max: 38.21Min: 38 / Avg: 38.23 / Max: 38.5Min: 37.97 / Avg: 38.17 / Max: 38.31. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: Zstd Compression 19123612182430SE +/- 0.02, N = 3SE +/- 0.12, N = 3SE +/- 0.05, N = 323.8123.8723.87
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: Zstd Compression 19123612182430Min: 23.78 / Avg: 23.81 / Max: 23.85Min: 23.63 / Avg: 23.87 / Max: 24.01Min: 23.77 / Avg: 23.87 / Max: 23.94

GNU Radio

GNU Radio is a free software development toolkit providing signal processing blocks to implement software-defined radios (SDR) and signal processing systems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FM Deemphasis Filter123130260390520650SE +/- 2.93, N = 9SE +/- 2.44, N = 5SE +/- 1.11, N = 3603.8605.2605.31. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FM Deemphasis Filter123110220330440550Min: 583.4 / Avg: 603.82 / Max: 611.6Min: 596.5 / Avg: 605.2 / Max: 610.7Min: 603.2 / Avg: 605.27 / Max: 6071. 3.8.1.0

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4K123246810SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 38.098.078.071. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4K1233691215Min: 8.04 / Avg: 8.09 / Max: 8.11Min: 8.03 / Avg: 8.07 / Max: 8.12Min: 8.04 / Avg: 8.07 / Max: 8.121. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU123246810SE +/- 0.01025, N = 3SE +/- 0.00714, N = 3SE +/- 0.01664, N = 38.709598.730848.72047MIN: 7.54MIN: 7.56MIN: 7.521. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU1233691215Min: 8.69 / Avg: 8.71 / Max: 8.72Min: 8.72 / Avg: 8.73 / Max: 8.74Min: 8.7 / Avg: 8.72 / Max: 8.751. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenSCAD

OpenSCAD is a programmer-focused solid 3D CAD modeller. OpenSCAD is free software and allows creating 3D CAD objects in a script-based modelling environment. This test profile will use the system-provided OpenSCAD program otherwise and time how long it takes tn render different SCAD assets to PNG output. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Pistol123306090120150SE +/- 0.37, N = 3SE +/- 0.24, N = 3SE +/- 0.18, N = 3134.23134.56134.391. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Pistol123306090120150Min: 133.69 / Avg: 134.23 / Max: 134.95Min: 134.22 / Avg: 134.56 / Max: 135.02Min: 134.14 / Avg: 134.39 / Max: 134.731. OpenSCAD version 2019.05

Timed Mesa Compilation

This test profile times how long it takes to compile Mesa with Meson/Ninja. For minimizing build dependencies and avoid versioning conflicts, test this is just the core Mesa build without LLVM or the extra Gallium3D/Mesa drivers enabled. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Mesa Compilation 21.0Time To Compile123306090120150SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 3131.62131.61131.93
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Mesa Compilation 21.0Time To Compile12320406080100Min: 131.55 / Avg: 131.62 / Max: 131.71Min: 131.49 / Avg: 131.61 / Max: 131.69Min: 131.69 / Avg: 131.93 / Max: 132.06

Timed LLVM Compilation

This test times how long it takes to build the LLVM compiler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: Ninja123400800120016002000SE +/- 3.42, N = 3SE +/- 0.21, N = 3SE +/- 1.66, N = 31934.431929.911933.98
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: Ninja12330060090012001500Min: 1930.73 / Avg: 1934.43 / Max: 1941.27Min: 1929.68 / Avg: 1929.91 / Max: 1930.33Min: 1931.59 / Avg: 1933.98 / Max: 1937.17

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080p1231428425670SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 362.0462.1862.121. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080p1231224364860Min: 61.95 / Avg: 62.04 / Max: 62.12Min: 62.11 / Avg: 62.18 / Max: 62.21Min: 62.09 / Avg: 62.12 / Max: 62.161. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -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: Twofish12370140210280350SE +/- 0.43, N = 3SE +/- 0.55, N = 3SE +/- 0.51, N = 3334.40334.02333.651. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Twofish12360120180240300Min: 333.65 / Avg: 334.4 / Max: 335.13Min: 332.99 / Avg: 334.02 / Max: 334.86Min: 333.09 / Avg: 333.65 / Max: 334.671. (CXX) g++ options: -fstack-protector -m64 -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, Long Mode - Decompression Speed1236001200180024003000SE +/- 4.15, N = 3SE +/- 4.47, N = 3SE +/- 1.65, N = 32745.92743.42749.41. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Decompression Speed1235001000150020002500Min: 2738.2 / Avg: 2745.93 / Max: 2752.4Min: 2734.5 / Avg: 2743.43 / Max: 2748Min: 2747.2 / Avg: 2749.37 / Max: 2752.61. (CC) gcc options: -O3 -pthread -lz -llzma

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 4K123918273645SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 341.2941.3741.281. (CXX) g++ options: -O3 -lpthread -lrt
OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 4K123918273645Min: 41.23 / Avg: 41.29 / Max: 41.33Min: 41.32 / Avg: 41.37 / Max: 41.4Min: 41.23 / Avg: 41.28 / Max: 41.331. (CXX) g++ options: -O3 -lpthread -lrt

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Crown1230.87521.75042.62563.50084.376SE +/- 0.0284, N = 3SE +/- 0.0226, N = 3SE +/- 0.0393, N = 33.88143.88983.8849MIN: 3.78 / MAX: 3.98MIN: 3.8 / MAX: 3.97MIN: 3.79 / MAX: 3.98
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Crown123246810Min: 3.84 / Avg: 3.88 / Max: 3.93Min: 3.85 / Avg: 3.89 / Max: 3.93Min: 3.81 / Avg: 3.88 / Max: 3.93

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: 8, Long Mode - Decompression Speed1237001400210028003500SE +/- 1.79, N = 3SE +/- 1.54, N = 3SE +/- 6.53, N = 33415.13416.23409.31. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8, Long Mode - Decompression Speed1236001200180024003000Min: 3411.5 / Avg: 3415.07 / Max: 3417.2Min: 3413.2 / Avg: 3416.23 / Max: 3418.2Min: 3396.6 / Avg: 3409.33 / Max: 3418.21. (CC) gcc options: -O3 -pthread -lz -llzma

Chia Blockchain VDF

Chia is a blockchain and smart transaction platform based on proofs of space and time rather than proofs of work with other cryptocurrencies. This test profile is benchmarking the CPU performance for Chia VDF performance using the Chia VDF benchmark. The Chia VDF is for the Chia Verifiable Delay Function (Proof of Time). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.1Test: Square Plain C++12320K40K60K80K100KSE +/- 120.19, N = 3SE +/- 66.67, N = 3SE +/- 115.47, N = 31162671164331165001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.1Test: Square Plain C++12320K40K60K80K100KMin: 116100 / Avg: 116266.67 / Max: 116500Min: 116300 / Avg: 116433.33 / Max: 116500Min: 116300 / Avg: 116500 / Max: 1167001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread

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 Time123100020003000400050004761.84769.84760.41. (CC) gcc options: -O3 -fomit-frame-pointer -lm

Timed Erlang/OTP Compilation

This test times how long it takes to compile Erlang/OTP. Erlang is a programming language and run-time for massively scalable soft real-time systems with high availability requirements. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Erlang/OTP Compilation 23.2Time To Compile12350100150200250SE +/- 0.52, N = 3SE +/- 0.38, N = 3SE +/- 0.06, N = 3218.74218.31218.65
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Erlang/OTP Compilation 23.2Time To Compile1234080120160200Min: 217.77 / Avg: 218.74 / Max: 219.54Min: 217.56 / Avg: 218.31 / Max: 218.78Min: 218.57 / Avg: 218.65 / Max: 218.77

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: KASUMI - Decrypt12320406080100SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.14, N = 381.5681.6981.531. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMI - Decrypt1231632486480Min: 81.39 / Avg: 81.56 / Max: 81.65Min: 81.59 / Avg: 81.69 / Max: 81.83Min: 81.4 / Avg: 81.53 / Max: 81.811. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Chimera 1080p123306090120150SE +/- 0.75, N = 3SE +/- 0.36, N = 3SE +/- 0.54, N = 3135.85136.11135.991. (CXX) g++ options: -O3 -lpthread -lrt
OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Chimera 1080p123306090120150Min: 134.36 / Avg: 135.85 / Max: 136.68Min: 135.4 / Avg: 136.11 / Max: 136.55Min: 134.91 / Avg: 135.99 / Max: 136.561. (CXX) g++ options: -O3 -lpthread -lrt

VOSK Speech Recognition Toolkit

VOSK is an open-source offline speech recognition API/toolkit. VOSK supports speech recognition in 17 languages and has a variety of models available and interfaces for different programming languages. This test profile times the speech-to-text process for a roughly three minute audio recording. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterVOSK Speech Recognition Toolkit 0.3.21123510152025SE +/- 0.19, N = 14SE +/- 0.17, N = 15SE +/- 0.18, N = 1521.6321.6021.64
OpenBenchmarking.orgSeconds, Fewer Is BetterVOSK Speech Recognition Toolkit 0.3.21123510152025Min: 21.37 / Avg: 21.63 / Max: 24.06Min: 21.27 / Avg: 21.6 / Max: 23.98Min: 21.32 / Avg: 21.64 / Max: 24.07

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU12310002000300040005000SE +/- 5.07, N = 3SE +/- 8.68, N = 3SE +/- 6.60, N = 34782.254773.704778.17MIN: 4732.63MIN: 4739.71MIN: 4733.021. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU1238001600240032004000Min: 4777.15 / Avg: 4782.25 / Max: 4792.39Min: 4762.82 / Avg: 4773.7 / Max: 4790.86Min: 4765.47 / Avg: 4778.17 / Max: 4787.641. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Helsing

Helsing is an open-source POSIX vampire number generator. This test profile measures the time it takes to generate vampire numbers between varying numbers of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digit123400800120016002000SE +/- 3.10, N = 3SE +/- 0.38, N = 3SE +/- 4.40, N = 31728.701731.771731.691. (CC) gcc options: -O2 -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digit12330060090012001500Min: 1725.01 / Avg: 1728.69 / Max: 1734.87Min: 1731 / Avg: 1731.77 / Max: 1732.19Min: 1725.56 / Avg: 1731.68 / Max: 1740.231. (CC) gcc options: -O2 -pthread

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To Compile12350100150200250SE +/- 0.63, N = 3SE +/- 0.47, N = 3SE +/- 0.52, N = 3245.86245.74245.42
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To Compile1234080120160200Min: 245.21 / Avg: 245.86 / Max: 247.11Min: 245.19 / Avg: 245.74 / Max: 246.68Min: 244.65 / Avg: 245.42 / Max: 246.41

Basis Universal

Basis Universal is a GPU texture codec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: UASTC Level 212320406080100SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 380.4580.4080.541. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: UASTC Level 21231530456075Min: 80.39 / Avg: 80.45 / Max: 80.57Min: 80.26 / Avg: 80.4 / Max: 80.53Min: 80.43 / Avg: 80.54 / Max: 80.641. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

toyBrot Fractal Generator

ToyBrot is a Mandelbrot fractal generator supporting C++ threads/tasks, OpenMP, Intel Threaded Building Blocks (TBB), and other targets. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ Threads12320K40K60K80K100KSE +/- 52.08, N = 3SE +/- 108.83, N = 3SE +/- 43.62, N = 31159791158541157801. (CXX) g++ options: -O3 -lpthread
OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ Threads12320K40K60K80K100KMin: 115883 / Avg: 115979 / Max: 116062Min: 115728 / Avg: 115854.33 / Max: 116071Min: 115731 / Avg: 115780 / Max: 1158671. (CXX) g++ options: -O3 -lpthread

Timed GDB GNU Debugger Compilation

This test times how long it takes to build the GNU Debugger (GDB) in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 10.2Time To Compile123306090120150SE +/- 0.27, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 3142.53142.37142.29
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 10.2Time To Compile123306090120150Min: 142.03 / Avg: 142.53 / Max: 142.96Min: 142.2 / Avg: 142.37 / Max: 142.47Min: 142.19 / Avg: 142.29 / Max: 142.42

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 0.9.0Encoder Speed: 0123306090120150SE +/- 0.31, N = 3SE +/- 0.49, N = 3SE +/- 0.09, N = 3156.27156.19156.451. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 0123306090120150Min: 155.65 / Avg: 156.27 / Max: 156.64Min: 155.24 / Avg: 156.19 / Max: 156.87Min: 156.34 / Avg: 156.45 / Max: 156.631. (CXX) g++ options: -O3 -fPIC -lm

toyBrot Fractal Generator

ToyBrot is a Mandelbrot fractal generator supporting C++ threads/tasks, OpenMP, Intel Threaded Building Blocks (TBB), and other targets. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ Tasks12330K60K90K120K150KSE +/- 85.55, N = 3SE +/- 130.22, N = 3SE +/- 46.61, N = 31186261185101186961. (CXX) g++ options: -O3 -lpthread
OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ Tasks12320K40K60K80K100KMin: 118467 / Avg: 118626.33 / Max: 118760Min: 118311 / Avg: 118510 / Max: 118755Min: 118603 / Avg: 118696 / Max: 1187481. (CXX) g++ options: -O3 -lpthread

Timed FFmpeg Compilation

This test times how long it takes to build the FFmpeg multimedia library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.4Time To Compile123306090120150SE +/- 0.15, N = 3SE +/- 0.01, N = 3SE +/- 0.12, N = 3147.96147.94147.73
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.4Time To Compile123306090120150Min: 147.7 / Avg: 147.96 / Max: 148.21Min: 147.93 / Avg: 147.94 / Max: 147.96Min: 147.5 / Avg: 147.73 / Max: 147.91

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 3.0Preset: Exhaustive1234080120160200SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.13, N = 3189.98190.00190.271. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.0Preset: Exhaustive123306090120150Min: 189.81 / Avg: 189.98 / Max: 190.09Min: 189.9 / Avg: 190 / Max: 190.09Min: 190.06 / Avg: 190.27 / Max: 190.51. (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: 3, Long Mode - Decompression Speed1237001400210028003500SE +/- 4.13, N = 3SE +/- 5.32, N = 3SE +/- 2.60, N = 33328.63328.33323.71. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3, Long Mode - Decompression Speed1236001200180024003000Min: 3323.9 / Avg: 3328.57 / Max: 3336.8Min: 3320.3 / Avg: 3328.33 / Max: 3338.4Min: 3318.5 / Avg: 3323.7 / Max: 3326.41. (CC) gcc options: -O3 -pthread -lz -llzma

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU12310002000300040005000SE +/- 3.71, N = 3SE +/- 3.51, N = 3SE +/- 1.26, N = 34769.484773.764776.38MIN: 4715.21MIN: 4733.2MIN: 4739.051. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU1238001600240032004000Min: 4762.11 / Avg: 4769.48 / Max: 4773.84Min: 4766.98 / Avg: 4773.76 / Max: 4778.74Min: 4774.45 / Avg: 4776.38 / Max: 4778.751. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -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: CAST-256 - Decrypt123306090120150SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 3129.37129.23129.421. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256 - Decrypt12320406080100Min: 129.27 / Avg: 129.37 / Max: 129.47Min: 129.07 / Avg: 129.23 / Max: 129.47Min: 129.28 / Avg: 129.42 / Max: 129.631. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4K123246810SE +/- 0.027, N = 3SE +/- 0.018, N = 3SE +/- 0.023, N = 36.3906.3836.3811. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4K1233691215Min: 6.34 / Avg: 6.39 / Max: 6.42Min: 6.35 / Avg: 6.38 / Max: 6.4Min: 6.34 / Avg: 6.38 / Max: 6.411. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

Timed LLVM Compilation

This test times how long it takes to build the LLVM compiler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: Unix Makefiles123400800120016002000SE +/- 0.68, N = 3SE +/- 2.14, N = 3SE +/- 1.23, N = 31950.921953.561953.48
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: Unix Makefiles12330060090012001500Min: 1950.08 / Avg: 1950.92 / Max: 1952.26Min: 1950.97 / Avg: 1953.56 / Max: 1957.81Min: 1951.43 / Avg: 1953.48 / Max: 1955.69

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 3.0Preset: Thorough123510152025SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 320.1820.2120.201. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.0Preset: Thorough123510152025Min: 20.17 / Avg: 20.18 / Max: 20.19Min: 20.2 / Avg: 20.21 / Max: 20.22Min: 20.18 / Avg: 20.2 / Max: 20.221. (CXX) g++ options: -O3 -flto -pthread

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 - Decrypt1238001600240032004000SE +/- 3.80, N = 3SE +/- 4.21, N = 3SE +/- 3.36, N = 33662.223662.263666.701. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256 - Decrypt1236001200180024003000Min: 3655.02 / Avg: 3662.22 / Max: 3667.91Min: 3655.43 / Avg: 3662.26 / Max: 3669.95Min: 3660.77 / Avg: 3666.7 / Max: 3672.411. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256123306090120150SE +/- 0.04, N = 3SE +/- 0.15, N = 3SE +/- 0.04, N = 3129.44129.46129.321. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-25612320406080100Min: 129.37 / Avg: 129.44 / Max: 129.52Min: 129.16 / Avg: 129.46 / Max: 129.65Min: 129.26 / Avg: 129.32 / Max: 129.41. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

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: C2670123306090120150SE +/- 0.43, N = 3SE +/- 0.42, N = 3SE +/- 0.15, N = 3155.82155.94155.771. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE
OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670123306090120150Min: 155.1 / Avg: 155.82 / Max: 156.59Min: 155.14 / Avg: 155.94 / Max: 156.57Min: 155.47 / Avg: 155.77 / Max: 155.951. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

Timed HMMer Search

This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.2Pfam Database Search123306090120150SE +/- 0.22, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 3147.04146.89147.031. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.2Pfam Database Search123306090120150Min: 146.7 / Avg: 147.04 / Max: 147.46Min: 146.72 / Avg: 146.89 / Max: 147.03Min: 146.91 / Avg: 147.03 / Max: 147.21. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm -lmpi

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: KASUMI12320406080100SE +/- 0.10, N = 3SE +/- 0.06, N = 3SE +/- 0.14, N = 384.1984.1184.131. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMI1231632486480Min: 83.99 / Avg: 84.19 / Max: 84.29Min: 84.02 / Avg: 84.11 / Max: 84.23Min: 83.85 / Avg: 84.13 / Max: 84.31. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080p12320406080100SE +/- 1.19, N = 3SE +/- 1.05, N = 4SE +/- 1.22, N = 378.5378.4678.471. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080p1231530456075Min: 76.15 / Avg: 78.53 / Max: 79.86Min: 75.3 / Avg: 78.46 / Max: 79.62Min: 76.04 / Avg: 78.47 / Max: 79.71. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

PJSIP

PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, Stateful12315003000450060007500SE +/- 3.21, N = 3SE +/- 2.19, N = 3SE +/- 0.88, N = 37016702070151. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, Stateful12312002400360048006000Min: 7011 / Avg: 7016 / Max: 7022Min: 7016 / Avg: 7020.33 / Max: 7023Min: 7013 / Avg: 7014.67 / Max: 70161. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

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: ChaCha20Poly1305123140280420560700SE +/- 1.52, N = 3SE +/- 2.12, N = 3SE +/- 1.48, N = 3667.57667.87667.421. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305123120240360480600Min: 665.11 / Avg: 667.57 / Max: 670.35Min: 663.65 / Avg: 667.87 / Max: 670.29Min: 664.46 / Avg: 667.42 / Max: 669.011. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-2561238001600240032004000SE +/- 4.99, N = 3SE +/- 5.87, N = 3SE +/- 9.86, N = 33654.683653.003655.441. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-2561236001200180024003000Min: 3644.8 / Avg: 3654.68 / Max: 3660.87Min: 3641.73 / Avg: 3653 / Max: 3661.49Min: 3635.92 / Avg: 3655.44 / Max: 3667.641. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU123510152025SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 321.3621.3621.37MIN: 21MIN: 20.97MIN: 211. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU123510152025Min: 21.33 / Avg: 21.36 / Max: 21.39Min: 21.33 / Avg: 21.36 / Max: 21.42Min: 21.34 / Avg: 21.37 / Max: 21.391. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3123714212835SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 331.8431.8531.84
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3123714212835Min: 31.75 / Avg: 31.84 / Max: 31.99Min: 31.77 / Avg: 31.85 / Max: 31.94Min: 31.82 / Avg: 31.84 / Max: 31.88

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: 8 - Buffer Length: 256 - Filter Length: 5712340M80M120M160M200MSE +/- 72188.03, N = 3SE +/- 76376.26, N = 3SE +/- 75129.52, N = 31878333331877900001877533331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 8 - Buffer Length: 256 - Filter Length: 5712330M60M90M120M150MMin: 187710000 / Avg: 187833333.33 / Max: 187960000Min: 187640000 / Avg: 187790000 / Max: 187890000Min: 187620000 / Avg: 187753333.33 / Max: 1878800001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 - Decrypt123140280420560700SE +/- 0.75, N = 3SE +/- 1.37, N = 3SE +/- 0.22, N = 3662.65662.76662.921. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305 - Decrypt123120240360480600Min: 661.22 / Avg: 662.65 / Max: 663.76Min: 660.04 / Avg: 662.76 / Max: 664.36Min: 662.55 / Avg: 662.92 / Max: 663.311. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU1232K4K6K8K10KSE +/- 10.32, N = 3SE +/- 9.37, N = 3SE +/- 9.54, N = 39176.999174.229175.25MIN: 9097.8MIN: 9112.39MIN: 9123.281. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU12316003200480064008000Min: 9158.16 / Avg: 9176.99 / Max: 9193.74Min: 9155.76 / Avg: 9174.22 / Max: 9186.21Min: 9165.37 / Avg: 9175.25 / Max: 9194.321. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

PJSIP

PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: INVITE1238001600240032004000SE +/- 2.65, N = 3SE +/- 6.12, N = 3SE +/- 5.36, N = 33903390339021. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: INVITE1237001400210028003500Min: 3898 / Avg: 3903 / Max: 3907Min: 3892 / Avg: 3903.33 / Max: 3913Min: 3891 / Avg: 3901.67 / Max: 39081. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NN12348121620SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 314.114.114.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NN12348121620Min: 14 / Avg: 14.07 / Max: 14.1Min: 14.1 / Avg: 14.1 / Max: 14.1Min: 14.1 / Avg: 14.1 / Max: 14.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPY12348121620SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.06, N = 317.717.717.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPY12348121620Min: 17.7 / Avg: 17.7 / Max: 17.7Min: 17.7 / Avg: 17.7 / Max: 17.7Min: 17.6 / Avg: 17.7 / Max: 17.81. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RTLightmap.hdr.4096x40961230.01580.03160.04740.06320.079SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.070.070.07
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RTLightmap.hdr.4096x409612312345Min: 0.07 / Avg: 0.07 / Max: 0.07Min: 0.07 / Avg: 0.07 / Max: 0.07Min: 0.07 / Avg: 0.07 / Max: 0.07

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.ldr_alb_nrm.3840x21601230.03150.0630.09450.1260.1575SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.140.140.14
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.ldr_alb_nrm.3840x216012312345Min: 0.14 / Avg: 0.14 / Max: 0.14Min: 0.14 / Avg: 0.14 / Max: 0.14Min: 0.14 / Avg: 0.14 / Max: 0.14

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.hdr_alb_nrm.3840x21601230.03150.0630.09450.1260.1575SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.140.140.14
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.hdr_alb_nrm.3840x216012312345Min: 0.14 / Avg: 0.14 / Max: 0.14Min: 0.14 / Avg: 0.14 / Max: 0.14Min: 0.14 / Avg: 0.14 / Max: 0.14

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p1230.03830.07660.11490.15320.1915SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.170.170.171. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p12312345Min: 0.17 / Avg: 0.17 / Max: 0.17Min: 0.17 / Avg: 0.17 / Max: 0.17Min: 0.17 / Avg: 0.17 / Max: 0.171. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K1230.35780.71561.07341.43121.789SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.591.591.591. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K123246810Min: 1.59 / Avg: 1.59 / Max: 1.59Min: 1.59 / Avg: 1.59 / Max: 1.6Min: 1.59 / Avg: 1.59 / Max: 1.591. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K1230.01130.02260.03390.04520.0565SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.050.050.051. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K12312345Min: 0.05 / Avg: 0.05 / Max: 0.05Min: 0.05 / Avg: 0.05 / Max: 0.05Min: 0.05 / Avg: 0.05 / Max: 0.051. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU12348121620SE +/- 0.03165, N = 3SE +/- 3.97558, N = 15SE +/- 0.01583, N = 39.7117413.732359.70188MIN: 8.41MIN: 8.41MIN: 8.51. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU12348121620Min: 9.68 / Avg: 9.71 / Max: 9.78Min: 9.72 / Avg: 13.73 / Max: 69.39Min: 9.67 / Avg: 9.7 / Max: 9.721. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K123510152025SE +/- 0.26, N = 6SE +/- 0.18, N = 3SE +/- 0.89, N = 1521.6821.3722.901. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K123510152025Min: 20.82 / Avg: 21.68 / Max: 22.75Min: 21.01 / Avg: 21.37 / Max: 21.55Min: 20.49 / Avg: 22.9 / Max: 33.21. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

GNU Radio

GNU Radio is a free software development toolkit providing signal processing blocks to implement software-defined radios (SDR) and signal processing systems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Five Back to Back FIR Filters123110220330440550SE +/- 11.11, N = 9SE +/- 6.44, N = 5SE +/- 4.57, N = 3490.4514.7498.81. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Five Back to Back FIR Filters12390180270360450Min: 418.9 / Avg: 490.41 / Max: 516Min: 489.1 / Avg: 514.66 / Max: 523.1Min: 490 / Avg: 498.83 / Max: 505.31. 3.8.1.0

177 Results Shown

KTX-Software toktx
LuaRadio
Zstd Compression
Mobile Neural Network
ViennaCL
ASTC Encoder
libavif avifenc
Embree:
  Pathtracer - Asian Dragon
  Pathtracer ISPC - Asian Dragon Obj
AOM AV1
Xmrig
Stockfish
Mobile Neural Network:
  inception-v3
  SqueezeNetV1.0
Timed Wasmer Compilation
VP9 libvpx Encoding
Zstd Compression:
  8, Long Mode - Compression Speed
  19, Long Mode - Compression Speed
AOM AV1
Zstd Compression
oneDNN
Google Draco
oneDNN
Embree
ViennaCL
Xmrig
Zstd Compression
Embree:
  Pathtracer - Asian Dragon Obj
  Pathtracer ISPC - Crown
dav1d:
  Summer Nature 4K
  Chimera 1080p
GROMACS
Zstd Compression
AOM AV1
SVT-VP9
OpenSCAD
GNU Radio
Zstd Compression
oneDNN
libavif avifenc
Chia Blockchain VDF
Botan
srsLTE
Timed MrBayes Analysis
SecureMark
oneDNN
ViennaCL:
  CPU BLAS - dGEMM-TT
  CPU BLAS - dGEMM-TN
Liquid-DSP
ViennaCL
srsLTE
LuaRadio
Basis Universal
LuaRadio
GNU Radio
dav1d
KTX-Software toktx
Zstd Compression
dav1d
ViennaCL
Zstd Compression
GNU Radio
OpenSCAD
AOM AV1
libavif avifenc
Helsing
SVT-HEVC
oneDNN
SVT-HEVC
OpenSCAD
oneDNN
KTX-Software toktx
GNU Radio
SVT-AV1
libavif avifenc
oneDNN
AOM AV1
Google Draco
Xcompact3d Incompact3d
PJSIP
Mobile Neural Network
BRL-CAD
Chaos Group V-RAY
oneDNN
srsLTE
Timed Node.js Compilation
OpenSCAD
ViennaCL
VP9 libvpx Encoding
SVT-HEVC
toyBrot Fractal Generator
Botan
ViennaCL
oneDNN
Botan
AOM AV1
libavif avifenc
ViennaCL:
  CPU BLAS - dDOT
  CPU BLAS - dGEMV-N
Mobile Neural Network
Liquid-DSP
AOM AV1
VP9 libvpx Encoding
SVT-AV1
libgav1
AOM AV1
SVT-AV1
oneDNN
LuaRadio
oneDNN
Liquid-DSP
oneDNN
libgav1
Ngspice
Basis Universal:
  UASTC Level 3
  ETC1S
KTX-Software toktx
GNU Radio
VP9 libvpx Encoding
oneDNN
OpenSCAD
Timed Mesa Compilation
Timed LLVM Compilation
SVT-VP9
Botan
Zstd Compression
libgav1
Embree
Zstd Compression
Chia Blockchain VDF
GNU GMP GMPbench
Timed Erlang/OTP Compilation
Botan
libgav1
VOSK Speech Recognition Toolkit
oneDNN
Helsing
Timed Linux Kernel Compilation
Basis Universal
toyBrot Fractal Generator
Timed GDB GNU Debugger Compilation
libavif avifenc
toyBrot Fractal Generator
Timed FFmpeg Compilation
ASTC Encoder
Zstd Compression
oneDNN
Botan
SVT-AV1
Timed LLVM Compilation
ASTC Encoder
Botan:
  AES-256 - Decrypt
  CAST-256
Ngspice
Timed HMMer Search
Botan
SVT-VP9
PJSIP
Botan:
  ChaCha20Poly1305
  AES-256
oneDNN
KTX-Software toktx
Liquid-DSP
Botan
oneDNN
PJSIP
ViennaCL:
  CPU BLAS - dGEMM-NN
  CPU BLAS - sCOPY
Intel Open Image Denoise:
  RTLightmap.hdr.4096x4096
  RT.ldr_alb_nrm.3840x2160
  RT.hdr_alb_nrm.3840x2160
AOM AV1:
  Speed 0 Two-Pass - Bosphorus 1080p
  Speed 4 Two-Pass - Bosphorus 4K
  Speed 0 Two-Pass - Bosphorus 4K
oneDNN
AOM AV1
GNU Radio