6800K May

Intel Core i7-6800K testing with a MSI X99A WORKSTATION (MS-7A54) v1.0 (1.10 BIOS) and Zotac NVIDIA NV137 2GB 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 2105106-IB-6800KMAY414
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 4 Tests
Bioinformatics 3 Tests
C++ Boost Tests 3 Tests
Timed Code Compilation 6 Tests
C/C++ Compiler Tests 11 Tests
CPU Massive 14 Tests
Creator Workloads 18 Tests
Cryptocurrency Benchmarks, CPU Mining Tests 2 Tests
Cryptography 3 Tests
Encoding 6 Tests
Game Development 5 Tests
HPC - High Performance Computing 8 Tests
Imaging 4 Tests
Machine Learning 2 Tests
Molecular Dynamics 3 Tests
MPI Benchmarks 4 Tests
Multi-Core 19 Tests
NVIDIA GPU Compute 3 Tests
Intel oneAPI 2 Tests
OpenMPI Tests 6 Tests
Programmer / Developer System Benchmarks 7 Tests
Python Tests 6 Tests
Scientific Computing 6 Tests
Software Defined Radio 4 Tests
Server CPU Tests 13 Tests
Single-Threaded 2 Tests
Telephony 2 Tests
Texture Compression 4 Tests
Video Encoding 6 Tests
Common Workstation Benchmarks 2 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
May 08 2021
  6 Hours, 17 Minutes
2
May 08 2021
  23 Hours, 14 Minutes
3
May 09 2021
  19 Hours, 46 Minutes
Invert Hiding All Results Option
  16 Hours, 26 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):

1

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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 Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0xb000038
Python Notes: Python 3.8.6
Security Notes: 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: Not affected + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable

2

3

Processor: Intel Core i7-6800K @ 3.80GHz (6 Cores / 12 Threads), Motherboard: MSI X99A WORKSTATION (MS-7A54) v1.0 (1.10 BIOS), Chipset: Intel Xeon E7 v4/Xeon, Memory: 16GB, Disk: 120GB TOSHIBA TR150, Graphics: Zotac NVIDIA NV137 2GB, Audio: Realtek ALC1150, Monitor: G237HL, Network: Intel I218-LM + Intel I210

OS: Ubuntu 20.10, Kernel: 5.8.0-41-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: nouveau, OpenGL: 4.3 Mesa 20.2.1, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1080

6800K MayProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen Resolution123Intel Core i7-6800K @ 3.80GHz (6 Cores / 12 Threads)MSI X99A WORKSTATION (MS-7A54) v1.0 (1.10 BIOS)Intel Xeon E7 v4/Xeon16GB120GB TOSHIBA TR150Zotac NVIDIA NV137 2GBRealtek ALC1150G237HLIntel I218-LM + Intel I210Ubuntu 20.105.8.0-41-generic (x86_64)GNOME Shell 3.38.1X Server 1.20.9nouveau4.3 Mesa 20.2.1GCC 10.2.0ext41920x1080OpenBenchmarking.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_cpufreq ondemand - CPU Microcode: 0xb000038Python 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: Not affected + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable

123Result OverviewPhoronix Test Suite100%103%105%108%110%VOSK Speech Recognition ToolkitTimed MrBayes AnalysisSecureMarkStockfishLiquid-DSPGNU GMP GMPbenchTimed Linux Kernel CompilationMobile Neural NetworkCP2K Molecular DynamicsTimed Mesa CompilationJPEG XL DecodingSVT-HEVCEmbreeSVT-VP9QMCPACKStreamlibjpeg-turbo tjbenchTimed Node.js CompilationChia Blockchain VDFXcompact3d Incompact3dLuaRadiolibavif avifencsrsLTETimed Wasmer CompilationGNU RadioBlenderXmrigViennaCLBasis UniversalJPEG XLoneDNNTimed HMMer SearchHelsingSysbenchPJSIPASTC EncoderAOM AV1GROMACSOpenSCADTimed Erlang/OTP Compilationdav1dlibgav1Zstd CompressionKTX-Software toktxGoogle DracoTimed LLVM Compilation

6800K Mayvosk: liquid-dsp: 8 - 256 - 57onednn: IP Shapes 3D - u8s8f32 - CPUgnuradio: Five Back to Back FIR Filtersincompact3d: input.i3d 129 Cells Per Directionincompact3d: input.i3d 193 Cells Per Directionsvt-vp9: VMAF Optimized - Bosphorus 1080pmrbayes: Primate Phylogeny Analysisviennacl: CPU BLAS - sDOTgnuradio: IIR Filtersvt-hevc: 7 - Bosphorus 1080pjpegxl: JPEG - 8jpegxl: JPEG - 5mnn: resnet-v2-50onednn: Deconvolution Batch shapes_3d - f32 - CPUembree: Pathtracer - Asian Dragon Objstockfish: Total Timeviennacl: CPU BLAS - dDOTembree: Pathtracer ISPC - Asian Dragon Objgnuradio: FIR Filterjpegxl-decode: Allluaradio: FM Deemphasis Filtercompress-zstd: 19, Long Mode - Compression Speedgnuradio: Signal Source (Cosine)viennacl: CPU BLAS - sAXPYcompress-zstd: 8 - Compression Speedgmpbench: Total Timecompress-zstd: 3 - Compression Speedbuild-linux-kernel: Time To Compilechia-vdf: Square Assembly Optimizedavifenc: 10mnn: inception-v3jpegxl: PNG - 8srslte: OFDM_Testviennacl: CPU BLAS - dGEMV-Tviennacl: CPU BLAS - dGEMV-Nliquid-dsp: 2 - 256 - 57draco: Church Facadejpegxl-decode: 1srslte: PHY_DL_Testembree: Pathtracer - Crownstream: Addtoktx: UASTC 3avifenc: 10, Losslessstream: Scaleliquid-dsp: 4 - 256 - 57build-mesa: Time To Compilebasis: UASTC Level 0onednn: IP Shapes 3D - f32 - CPUcompress-zstd: 8 - Decompression Speedcompress-zstd: 19 - Compression Speedqmcpack: simple-H2Oopenscad: Retro Carluaradio: Five Back to Back FIR Filterstoktx: UASTC 3 + Zstd Compression 19viennacl: CPU BLAS - sCOPYjpegxl: PNG - 5jpegxl: JPEG - 7cp2k: Fayalite-FISTtjbench: Decompression Throughputaom-av1: Speed 8 Realtime - Bosphorus 1080pviennacl: CPU BLAS - dAXPYliquid-dsp: 1 - 256 - 57stream: Copyxmrig: Monero - 1Monednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUcompress-zstd: 3, Long Mode - Decompression Speedonednn: Recurrent Neural Network Training - u8s8f32 - CPUgnuradio: FM Deemphasis Filtercompress-zstd: 19 - Decompression Speedstream: Triadsrslte: PHY_DL_Testluaradio: Complex Phasecompress-zstd: 3, Long Mode - Compression Speedembree: Pathtracer - Asian Dragoncompress-zstd: 19, Long Mode - Decompression Speedhelsing: 12 digitavifenc: 0astcenc: Mediumonednn: IP Shapes 1D - u8s8f32 - CPUopenscad: Mini-ITX Caseliquid-dsp: 12 - 256 - 57pjsip: OPTIONS, Statelesssysbench: RAM / Memoryonednn: IP Shapes 1D - f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUsvt-vp9: Visual Quality Optimized - Bosphorus 1080pembree: Pathtracer ISPC - Crownbasis: UASTC Level 3avifenc: 6, Losslessbasis: ETC1Sopenscad: Leonardo Phone Case Slimtoktx: Zstd Compression 9aom-av1: Speed 6 Realtime - Bosphorus 1080pdav1d: Chimera 1080psvt-hevc: 10 - Bosphorus 1080ponednn: Recurrent Neural Network Inference - u8s8f32 - CPUluaradio: Hilbert Transformonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUcompress-zstd: 8, Long Mode - Compression Speedbuild-wasmer: Time To Compileaom-av1: Speed 8 Realtime - Bosphorus 4Kblender: BMW27 - CPU-Onlyviennacl: CPU BLAS - dCOPYavifenc: 2aom-av1: Speed 9 Realtime - Bosphorus 1080paom-av1: Speed 6 Realtime - Bosphorus 4Kblender: Fishy Cat - CPU-Onlylibgav1: Chimera 1080ponednn: Recurrent Neural Network Training - f32 - CPUblender: Classroom - CPU-Onlyaom-av1: Speed 6 Two-Pass - Bosphorus 1080ponednn: Convolution Batch Shapes Auto - u8s8f32 - CPUopenscad: Projector Mount Swivelhmmer: Pfam Database Searchavifenc: 6openscad: Pistolaom-av1: Speed 6 Two-Pass - Bosphorus 4Konednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUlibgav1: Chimera 1080p 10-bitembree: Pathtracer ISPC - Asian Dragononednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUdav1d: Chimera 1080p 10-bitbasis: UASTC Level 2aom-av1: Speed 4 Two-Pass - Bosphorus 1080pbuild-llvm: Unix Makefilesonednn: Deconvolution Batch shapes_1d - f32 - CPUgnuradio: Hilbert Transformbuild-erlang: Time To Compilecompress-zstd: 3 - Decompression Speedxmrig: Wownero - 1Mpjsip: OPTIONS, Statefuldav1d: Summer Nature 4Kgromacs: water_GMX50_barejpegxl: PNG - 7aom-av1: Speed 9 Realtime - Bosphorus 4Khelsing: 14 digittoktx: Zstd Compression 19build-llvm: Ninjapjsip: INVITEcompress-zstd: 8, Long Mode - Decompression Speedonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUdav1d: Summer Nature 1080pbuild-nodejs: Time To Compilelibgav1: Summer Nature 1080ponednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUchia-vdf: Square Plain C++libgav1: Summer Nature 4Kastcenc: Exhaustivedraco: Lionastcenc: Thoroughsysbench: CPUsvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080ptoktx: UASTC 4 + Zstd Compression 19svt-hevc: 1 - Bosphorus 1080paom-av1: Speed 0 Two-Pass - Bosphorus 1080paom-av1: Speed 4 Two-Pass - Bosphorus 4Kaom-av1: Speed 0 Two-Pass - Bosphorus 4Kmnn: mobilenet-v1-1.0mnn: MobileNetV2_224mnn: SqueezeNetV1.0viennacl: CPU BLAS - dGEMM-TTviennacl: CPU BLAS - dGEMM-TNviennacl: CPU BLAS - dGEMM-NTviennacl: CPU BLAS - dGEMM-NNsecuremark: SecureMark-TLS12325.6972558300003.36519504.841.1415367156.71374589.33173.5441843655.6819.54359.0416.07557.1061397374123.18.0582532.9129.91365.621.32194.129.2317.24608.31739.9176.8991505005.13374.2990.597950000024.324.698947000904829.1778.56.418629342.323.0518.54527218.3188270000100.10310.8999.846082996.426.755.88324.251651.629.17919.939.8842.451347.639171.67991977.2733.65202100043325.42565.916.95589.701893161.711018.35932810.629294197481.3673.27.54882783.810.27122.9578.04469.0994158.3622846700004889316736.8912.79427328.2475.257.2156113.168111.76635.08923.4383.41213.38369.44121.247353.482.97382.83336.2165.54424.26285.3522.663.79792.817.4401.68112.2311040.2856.9510.8917.1722129.726151.6921.756139.3453.5210.762235.58.86697.1132587.2759.0313.81394.79219.4447335.3169.1462988.73445.76406104.30.5665.7734.84918.41621.5321368.45937793215.610.3818332.31888.501135.3711062.711900039.91198.6593633525.938410354.1995.25655.5673.840.211.870.077.0426.37112.36420.919.52020.220598323.4082772566673.57796490.943.4190303149.26389692.51174.59217.4449.857.4320.1041.9257.33215.66696.90951360592323.57.8579519.7128.81366.420.82176.428.6317.846081773.5173.5521533005.22773.5240.598046666724.424.898514333904529.6377.36.337729554.423.2208.60327311.619073500098.81910.9239.904943031.726.755.27924.442654.728.90419.940.2642.301335.09171.39763276.8733.95241200043524.32580.716.89389.781943135.911106.7595.32802.329324.2195.5482.7669.87.54632766.110.304123.7088.07479.0387558.3222865366674913216779.8212.73427369.7375.167.1725113.846112.32535.22823.3573.41213.42371.37121.877368.7782.87367.46337.5164.76624.15286.6422.563.95492.427.38400.78111.7911061.1858.1310.8717.1720129.572151.32121.716139.8003.5310.784235.408.89197.1111487.1359.0763.811397.41319.4434334.6168.7842982.23438.56419104.380.5655.7634.83918.95221.5261370.70237753216.110.3787332.59889.312135.3311053.111890039.94198.6162633325.934510352.6095.30655.4393.840.211.870.077.0586.20312.12718.921.319.620.720608523.3272691626673.56832478.042.5190188150.01359592.74180.16317.7448.957.3719.7641.7457.52615.62277.04861373768823.77.9602525.1126.70357.421.12226.429.1323.64519.21755.8174.2771532335.17373.0040.608083333324.024.497365167918929.40786.433929759.523.3748.66327580.019043166798.89811.0409.967253031.626.455.58224.181658.628.87720.139.8942.051341.451170.10045476.5633.75247766743699.52587.917.03789.768103159.311089.2597.72788.729521.1196.8479.1668.37.60122785.310.234123.7978.09959.0516858.7132860833334920716673.4512.71547372.4174.807.1957113.366112.43535.02323.4913.39413.35369.97121.557333.1882.57347.50337.8165.45924.21285.7122.664.06892.467.37400.10112.1811082.8860.1210.8517.2335130.000151.19521.787139.5033.5310.753835.428.88537.0934087.0359.1893.801398.36719.3952335.4169.1862983.73443.96414104.170.5665.7634.89919.96721.4961370.14837813211.510.3930332.73889.461135.2411060.211893339.94198.5244633125.950510358.7995.25655.4623.840.211.870.077.8676.03911.51919.621.519.119.6199607OpenBenchmarking.org

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 0.3.21123612182430SE +/- 0.21, N = 15SE +/- 0.20, N = 1525.7023.4123.33
OpenBenchmarking.orgSeconds, Fewer Is BetterVOSK Speech Recognition Toolkit 0.3.21123612182430Min: 22.89 / Avg: 23.41 / Max: 25.78Min: 22.99 / Avg: 23.33 / Max: 25.84

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: 5712360M120M180M240M300MSE +/- 2271668.50, N = 3SE +/- 2636553.03, N = 152558300002772566672691626671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 8 - Buffer Length: 256 - Filter Length: 5712350M100M150M200M250MMin: 273040000 / Avg: 277256666.67 / Max: 280830000Min: 248880000 / Avg: 269162666.67 / Max: 2814500001. (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: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU1230.8051.612.4153.224.025SE +/- 0.02094, N = 3SE +/- 0.01805, N = 33.365193.577963.56832MIN: 3.1MIN: 3.12MIN: 3.051. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU123246810Min: 3.54 / Avg: 3.58 / Max: 3.6Min: 3.55 / Avg: 3.57 / Max: 3.61. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 +/- 6.36, N = 9SE +/- 4.01, N = 3504.8490.9478.01. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Five Back to Back FIR Filters12390180270360450Min: 460 / Avg: 490.89 / Max: 526.1Min: 471.1 / Avg: 478 / Max: 4851. 3.8.1.0

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 Direction1231020304050SE +/- 0.29, N = 15SE +/- 0.35, N = 941.1443.4242.521. (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 Direction123918273645Min: 41.79 / Avg: 43.42 / Max: 45.11Min: 41.74 / Avg: 42.52 / Max: 44.691. (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 193 Cells Per Direction123306090120150SE +/- 0.63, N = 3SE +/- 0.64, N = 3156.71149.26150.011. (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 193 Cells Per Direction123306090120150Min: 148.29 / Avg: 149.26 / Max: 150.44Min: 148.91 / Avg: 150.01 / Max: 151.131. (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

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 +/- 0.97, N = 5SE +/- 0.97, N = 589.3392.5192.741. (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 1080p12320406080100Min: 88.66 / Avg: 92.51 / Max: 93.74Min: 88.89 / Avg: 92.74 / Max: 94.211. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

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 +/- 1.69, N = 3SE +/- 1.97, N = 5173.54174.59180.161. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny Analysis123306090120150Min: 171.78 / Avg: 174.59 / Max: 177.63Min: 174.49 / Avg: 180.16 / Max: 186.81. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -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 - sDOT12348121620SE +/- 0.55, N = 3SE +/- 0.17, N = 1218.017.417.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOT123510152025Min: 16.4 / Avg: 17.4 / Max: 18.3Min: 16.4 / Avg: 17.65 / Max: 18.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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 +/- 1.40, N = 9SE +/- 2.30, N = 3436.0449.8448.91. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: IIR Filter12380160240320400Min: 441.7 / Avg: 449.82 / Max: 456.1Min: 444.5 / Avg: 448.93 / Max: 452.21. 3.8.1.0

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 1080p1231326395265SE +/- 0.00, N = 3SE +/- 0.03, N = 355.6857.4357.371. (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 1080p1231122334455Min: 57.42 / Avg: 57.43 / Max: 57.43Min: 57.32 / Avg: 57.37 / Max: 57.431. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

JPEG XL

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: JPEG - Encode Speed: 8123510152025SE +/- 0.10, N = 3SE +/- 0.20, N = 319.5020.1019.761. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: JPEG - Encode Speed: 8123510152025Min: 19.9 / Avg: 20.1 / Max: 20.21Min: 19.36 / Avg: 19.76 / Max: 20.021. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: JPEG - Encode Speed: 51231020304050SE +/- 0.25, N = 3SE +/- 0.14, N = 343.0041.9241.741. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: JPEG - Encode Speed: 5123918273645Min: 41.49 / Avg: 41.92 / Max: 42.37Min: 41.46 / Avg: 41.74 / Max: 41.931. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl

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-v2-501231326395265SE +/- 0.31, N = 9SE +/- 0.82, N = 359.0457.3357.53MIN: 47.53 / MAX: 248.6MIN: 42.93 / MAX: 183.34MIN: 43.37 / MAX: 182.61. (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-v2-501231224364860Min: 56.4 / Avg: 57.33 / Max: 59.42Min: 56.19 / Avg: 57.53 / Max: 59.021. (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

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 +/- 0.09, N = 316.0815.6715.62MIN: 11.92MIN: 11.93MIN: 11.891. (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: 15.6 / Avg: 15.67 / Max: 15.71Min: 15.44 / Avg: 15.62 / Max: 15.731. (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 - Model: Asian Dragon Obj123246810SE +/- 0.0326, N = 3SE +/- 0.0278, N = 37.10606.90957.0486MIN: 7.08 / MAX: 7.17MIN: 6.79 / MAX: 7.12MIN: 6.93 / MAX: 7.17
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian Dragon Obj1233691215Min: 6.86 / Avg: 6.91 / Max: 6.97Min: 7 / Avg: 7.05 / Max: 7.1

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 Time1233M6M9M12M15MSE +/- 72340.52, N = 3SE +/- 62832.49, N = 31397374113605923137376881. (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 Time1232M4M6M8M10MMin: 13499128 / Avg: 13605923 / Max: 13743852Min: 13629500 / Avg: 13737688 / Max: 138471471. (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

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 - dDOT123612182430SE +/- 0.50, N = 3SE +/- 0.18, N = 1223.123.523.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOT123612182430Min: 22.5 / Avg: 23.5 / Max: 24.1Min: 22.9 / Avg: 23.72 / Max: 24.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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 Dragon Obj123246810SE +/- 0.0092, N = 3SE +/- 0.0144, N = 38.05827.85797.9602MIN: 8.02 / MAX: 8.14MIN: 7.79 / MAX: 8MIN: 7.9 / MAX: 8.09
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian Dragon Obj1233691215Min: 7.85 / Avg: 7.86 / Max: 7.88Min: 7.94 / Avg: 7.96 / Max: 7.99

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 Filter123120240360480600SE +/- 5.14, N = 9SE +/- 2.59, N = 3532.9519.7525.11. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FIR Filter12390180270360450Min: 479 / Avg: 519.67 / Max: 528.7Min: 521.2 / Avg: 525.1 / Max: 5301. 3.8.1.0

JPEG XL Decoding

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding 0.3.3CPU Threads: All123306090120150SE +/- 0.06, N = 3SE +/- 0.08, N = 3129.91128.81126.70
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding 0.3.3CPU Threads: All12320406080100Min: 128.68 / Avg: 128.81 / Max: 128.88Min: 126.57 / Avg: 126.7 / Max: 126.85

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 +/- 0.27, N = 3SE +/- 11.52, N = 3365.6366.4357.4
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: FM Deemphasis Filter12370140210280350Min: 366 / Avg: 366.37 / Max: 366.9Min: 334.4 / Avg: 357.43 / Max: 369.5

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 1.4.9Compression Level: 19, Long Mode - Compression Speed123510152025SE +/- 0.06, N = 3SE +/- 0.12, N = 321.320.821.11. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Compression Speed123510152025Min: 20.7 / Avg: 20.8 / Max: 20.9Min: 20.9 / Avg: 21.07 / Max: 21.31. (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: Signal Source (Cosine)1235001000150020002500SE +/- 20.50, N = 9SE +/- 13.74, N = 32194.12176.42226.41. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Signal Source (Cosine)123400800120016002000Min: 2034.8 / Avg: 2176.4 / Max: 2259.6Min: 2202.5 / Avg: 2226.37 / Max: 2250.11. 3.8.1.0

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 - sAXPY123714212835SE +/- 0.20, N = 3SE +/- 0.11, N = 1229.228.629.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPY123612182430Min: 28.2 / Avg: 28.57 / Max: 28.9Min: 28.6 / Avg: 29.13 / Max: 29.91. (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 1.4.9Compression Level: 8 - Compression Speed12370140210280350SE +/- 1.37, N = 3SE +/- 0.90, N = 3317.2317.8323.61. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8 - Compression Speed12360120180240300Min: 315.1 / Avg: 317.8 / Max: 319.6Min: 322 / Avg: 323.6 / Max: 325.11. (CC) gcc options: -O3 -pthread -lz -llzma

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 Time123100020003000400050004608.34608.04519.21. (CC) gcc options: -O3 -fomit-frame-pointer -lm

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 1.4.9Compression Level: 3 - Compression Speed123400800120016002000SE +/- 9.22, N = 3SE +/- 9.91, N = 31739.91773.51755.81. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3 - Compression Speed12330060090012001500Min: 1758.1 / Avg: 1773.53 / Max: 1790Min: 1745.1 / Avg: 1755.8 / Max: 1775.61. (CC) gcc options: -O3 -pthread -lz -llzma

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 Compile1234080120160200SE +/- 1.19, N = 3SE +/- 1.62, N = 3176.90173.55174.28
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To Compile123306090120150Min: 172.31 / Avg: 173.55 / Max: 175.93Min: 172.49 / Avg: 174.28 / Max: 177.5

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 Optimized12330K60K90K120K150KSE +/- 1001.67, N = 3SE +/- 1328.32, N = 31505001533001532331. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.1Test: Square Assembly Optimized12330K60K90K120K150KMin: 151400 / Avg: 153300 / Max: 154800Min: 150900 / Avg: 153233.33 / Max: 1555001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -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: 101231.17612.35223.52834.70445.8805SE +/- 0.051, N = 3SE +/- 0.006, N = 35.1335.2275.1731. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10123246810Min: 5.16 / Avg: 5.23 / Max: 5.33Min: 5.17 / Avg: 5.17 / Max: 5.191. (CXX) g++ options: -O3 -fPIC -lm

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-v31231632486480SE +/- 0.45, N = 9SE +/- 0.90, N = 374.3073.5273.00MIN: 63.07 / MAX: 247.5MIN: 57.58 / MAX: 234.51MIN: 56.13 / MAX: 217.151. (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-v31231428425670Min: 71.92 / Avg: 73.52 / Max: 75.82Min: 71.3 / Avg: 73 / Max: 74.341. (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

JPEG XL

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: PNG - Encode Speed: 81230.1350.270.4050.540.675SE +/- 0.00, N = 3SE +/- 0.00, N = 30.590.590.601. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: PNG - Encode Speed: 8123246810Min: 0.58 / Avg: 0.59 / Max: 0.59Min: 0.59 / Avg: 0.6 / Max: 0.61. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -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.orgSamples / Second, More Is BettersrsLTE 20.10.1Test: OFDM_Test12320M40M60M80M100MSE +/- 633333.33, N = 3SE +/- 384418.75, N = 37950000080466667808333331. (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_Test12314M28M42M56M70MMin: 79200000 / Avg: 80466666.67 / Max: 81100000Min: 80100000 / Avg: 80833333.33 / Max: 814000001. (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

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-T123612182430SE +/- 0.26, N = 3SE +/- 0.15, N = 1024.324.424.01. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-T123612182430Min: 23.9 / Avg: 24.37 / Max: 24.8Min: 23.2 / Avg: 24 / Max: 24.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-N123612182430SE +/- 0.29, N = 3SE +/- 0.10, N = 1224.624.824.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-N123612182430Min: 24.3 / Avg: 24.83 / Max: 25.3Min: 23.7 / Avg: 24.38 / Max: 24.91. (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: 2 - Buffer Length: 256 - Filter Length: 5712320M40M60M80M100MSE +/- 471745.81, N = 3SE +/- 1085656.09, N = 129894700098514333973651671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 2 - Buffer Length: 256 - Filter Length: 5712320M40M60M80M100MMin: 97591000 / Avg: 98514333.33 / Max: 99144000Min: 86392000 / Avg: 97365166.67 / Max: 991900001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 +/- 7.51, N = 3SE +/- 112.19, N = 159048904591891. (CXX) g++ options: -O3
OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.4.1Model: Church Facade12316003200480064008000Min: 9032 / Avg: 9044.67 / Max: 9058Min: 9031 / Avg: 9188.73 / Max: 107221. (CXX) g++ options: -O3

JPEG XL Decoding

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding 0.3.3CPU Threads: 1123714212835SE +/- 0.01, N = 3SE +/- 0.06, N = 329.1729.6329.40
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding 0.3.3CPU Threads: 1123714212835Min: 29.61 / Avg: 29.63 / Max: 29.65Min: 29.28 / Avg: 29.4 / Max: 29.48

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 +/- 0.24, N = 378.577.378.01. (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_Test1231530456075Min: 76.8 / Avg: 77.27 / Max: 77.61. (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

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: Crown123246810SE +/- 0.0011, N = 3SE +/- 0.0158, N = 36.41866.33776.4339MIN: 6.38 / MAX: 6.5MIN: 6.28 / MAX: 6.47MIN: 6.37 / MAX: 6.54
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Crown1233691215Min: 6.34 / Avg: 6.34 / Max: 6.34Min: 6.41 / Avg: 6.43 / Max: 6.46

Stream

This is a benchmark of Stream, the popular system memory (RAM) benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: Add1236K12K18K24K30KSE +/- 15.91, N = 5SE +/- 12.57, N = 529342.329554.429759.51. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: Add1235K10K15K20K25KMin: 29524.6 / Avg: 29554.42 / Max: 29615Min: 29712.6 / Avg: 29759.5 / Max: 29782.61. (CC) gcc options: -O3 -march=native -fopenmp

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 3123612182430SE +/- 0.12, N = 3SE +/- 0.28, N = 323.0523.2223.37
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3123510152025Min: 23.07 / Avg: 23.22 / Max: 23.45Min: 23.08 / Avg: 23.37 / Max: 23.94

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, Lossless123246810SE +/- 0.030, N = 3SE +/- 0.010, N = 38.5458.6038.6631. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10, Lossless1233691215Min: 8.54 / Avg: 8.6 / Max: 8.64Min: 8.64 / Avg: 8.66 / Max: 8.681. (CXX) g++ options: -O3 -fPIC -lm

Stream

This is a benchmark of Stream, the popular system memory (RAM) benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: Scale1236K12K18K24K30KSE +/- 41.98, N = 5SE +/- 6.54, N = 527218.327311.627580.01. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: Scale1235K10K15K20K25KMin: 27200.2 / Avg: 27311.62 / Max: 27456.5Min: 27564.9 / Avg: 27579.96 / Max: 27599.61. (CC) gcc options: -O3 -march=native -fopenmp

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 +/- 1897283.58, N = 6SE +/- 1876216.79, N = 61882700001907350001904316671. (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: 183240000 / Avg: 190735000 / Max: 197720000Min: 185140000 / Avg: 190431666.67 / Max: 1966800001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 Compile12320406080100SE +/- 0.33, N = 3SE +/- 0.09, N = 3100.1098.8298.90
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Mesa Compilation 21.0Time To Compile12320406080100Min: 98.3 / Avg: 98.82 / Max: 99.43Min: 98.79 / Avg: 98.9 / Max: 99.08

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 +/- 0.00, N = 3SE +/- 0.11, N = 1410.9010.9211.041. (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: 10.92 / Avg: 10.92 / Max: 10.93Min: 10.89 / Avg: 11.04 / Max: 12.511. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -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 3D - Data Type: f32 - Engine: CPU1233691215SE +/- 0.04476, N = 3SE +/- 0.04181, N = 39.846089.904949.96725MIN: 8.59MIN: 8.69MIN: 8.681. (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: CPU1233691215Min: 9.82 / Avg: 9.9 / Max: 9.97Min: 9.92 / Avg: 9.97 / Max: 10.051. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 1.4.9Compression Level: 8 - Decompression Speed1237001400210028003500SE +/- 1.79, N = 3SE +/- 1.06, N = 32996.43031.73031.61. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8 - Decompression Speed1235001000150020002500Min: 3028.5 / Avg: 3031.67 / Max: 3034.7Min: 3029.5 / Avg: 3031.57 / Max: 30331. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression Speed123612182430SE +/- 0.29, N = 3SE +/- 0.27, N = 326.726.726.41. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression Speed123612182430Min: 26.2 / Avg: 26.73 / Max: 27.2Min: 25.9 / Avg: 26.43 / Max: 26.71. (CC) gcc options: -O3 -pthread -lz -llzma

QMCPACK

QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. QMCPACK is an open-source production level many-body ab initio Quantum Monte Carlo code for computing the electronic structure of atoms, molecules, and solids. QMCPACK is supported by the U.S. Department of Energy. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.11Input: simple-H2O1231326395265SE +/- 0.21, N = 3SE +/- 0.69, N = 355.8855.2855.581. (CXX) g++ options: -fopenmp -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -fomit-frame-pointer -ffast-math -pthread -lm
OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.11Input: simple-H2O1231122334455Min: 55.03 / Avg: 55.28 / Max: 55.69Min: 54.23 / Avg: 55.58 / Max: 56.441. (CXX) g++ options: -fopenmp -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -fomit-frame-pointer -ffast-math -pthread -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 +/- 0.20, N = 14SE +/- 0.01, N = 324.2524.4424.181. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Retro Car123612182430Min: 24.15 / Avg: 24.44 / Max: 27.06Min: 24.17 / Avg: 24.18 / Max: 24.191. OpenSCAD version 2019.05

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 Filters123140280420560700SE +/- 4.23, N = 3SE +/- 0.20, N = 3651.6654.7658.6
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Five Back to Back FIR Filters123120240360480600Min: 649.8 / Avg: 654.67 / Max: 663.1Min: 658.2 / Avg: 658.57 / Max: 658.9

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 19123714212835SE +/- 0.02, N = 3SE +/- 0.02, N = 329.1828.9028.88
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3 + Zstd Compression 19123612182430Min: 28.88 / Avg: 28.9 / Max: 28.94Min: 28.84 / Avg: 28.88 / Max: 28.9

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 - sCOPY123510152025SE +/- 0.10, N = 3SE +/- 0.14, N = 1219.919.920.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPY123510152025Min: 19.8 / Avg: 19.9 / Max: 20.1Min: 19.5 / Avg: 20.14 / Max: 21.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

JPEG XL

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: PNG - Encode Speed: 5123918273645SE +/- 0.20, N = 3SE +/- 0.07, N = 339.8840.2639.891. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: PNG - Encode Speed: 5123816243240Min: 39.86 / Avg: 40.26 / Max: 40.51Min: 39.77 / Avg: 39.89 / Max: 40.021. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: JPEG - Encode Speed: 71231020304050SE +/- 0.44, N = 3SE +/- 0.16, N = 342.4542.3042.051. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: JPEG - Encode Speed: 7123918273645Min: 41.74 / Avg: 42.3 / Max: 43.17Min: 41.84 / Avg: 42.05 / Max: 42.371. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl

CP2K Molecular Dynamics

CP2K is an open-source molecular dynamics software package focused on quantum chemistry and solid-state physics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 8.1Input: Fayalite-FIST123300600900120015001347.641335.091341.45

libjpeg-turbo tjbench

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

OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.1.0Test: Decompression Throughput1234080120160200SE +/- 0.61, N = 3SE +/- 0.71, N = 3171.68171.40170.101. (CC) gcc options: -O3 -rdynamic
OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.1.0Test: Decompression Throughput123306090120150Min: 170.18 / Avg: 171.4 / Max: 172.05Min: 168.88 / Avg: 170.1 / Max: 171.331. (CC) gcc options: -O3 -rdynamic

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 1080p12320406080100SE +/- 0.01, N = 3SE +/- 0.30, N = 377.2776.8776.561. (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 1080p1231530456075Min: 76.85 / Avg: 76.87 / Max: 76.89Min: 76.01 / Avg: 76.56 / Max: 77.041. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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 - dAXPY123816243240SE +/- 0.35, N = 3SE +/- 0.12, N = 1233.633.933.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPY123714212835Min: 33.2 / Avg: 33.9 / Max: 34.3Min: 33.1 / Avg: 33.73 / Max: 34.41. (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: 1 - Buffer Length: 256 - Filter Length: 5712311M22M33M44M55MSE +/- 44508.43, N = 3SE +/- 13678.86, N = 35202100052412000524776671. (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: 52323000 / Avg: 52412000 / Max: 52458000Min: 52463000 / Avg: 52477666.67 / Max: 525050001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Stream

This is a benchmark of Stream, the popular system memory (RAM) benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: Copy1239K18K27K36K45KSE +/- 41.44, N = 5SE +/- 128.05, N = 543325.443524.343699.51. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: Copy1238K16K24K32K40KMin: 43458.1 / Avg: 43524.28 / Max: 43677.6Min: 43477.2 / Avg: 43699.5 / Max: 44128.21. (CC) gcc options: -O3 -march=native -fopenmp

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: 1M1236001200180024003000SE +/- 3.21, N = 3SE +/- 4.53, N = 32565.92580.72587.91. (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: 1M1235001000150020002500Min: 2574.3 / Avg: 2580.67 / Max: 2584.6Min: 2582.1 / Avg: 2587.87 / Max: 2596.81. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

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: CPU12348121620SE +/- 0.13, N = 3SE +/- 0.07, N = 316.9616.8917.04MIN: 14.02MIN: 13.96MIN: 13.851. (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: CPU12348121620Min: 16.69 / Avg: 16.89 / Max: 17.12Min: 16.91 / Avg: 17.04 / Max: 17.141. (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: CPU1233691215SE +/- 0.03538, N = 3SE +/- 0.03293, N = 39.701899.781949.76810MIN: 4.61MIN: 4.62MIN: 4.621. (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: 9.74 / Avg: 9.78 / Max: 9.85Min: 9.7 / Avg: 9.77 / Max: 9.811. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 1.4.9Compression Level: 3, Long Mode - Decompression Speed1237001400210028003500SE +/- 19.54, N = 3SE +/- 1.14, N = 33161.73135.93159.31. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3, Long Mode - Decompression Speed1235001000150020002500Min: 3096.8 / Avg: 3135.87 / Max: 3156Min: 3157 / Avg: 3159.27 / Max: 3160.61. (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: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU1232K4K6K8K10KSE +/- 27.42, N = 3SE +/- 20.98, N = 311018.311106.711089.2MIN: 10752.6MIN: 10805MIN: 10779.61. (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: CPU1232K4K6K8K10KMin: 11066.7 / Avg: 11106.73 / Max: 11159.2Min: 11056.5 / Avg: 11089.17 / Max: 11128.31. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 +/- 4.77, N = 9SE +/- 8.03, N = 3593.0595.3597.71. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FM Deemphasis Filter123110220330440550Min: 560.4 / Avg: 595.28 / Max: 607.1Min: 581.9 / Avg: 597.73 / Max: 6081. 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 1.4.9Compression Level: 19 - Decompression Speed1236001200180024003000SE +/- 3.02, N = 3SE +/- 10.31, N = 32810.62802.32788.71. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Decompression Speed1235001000150020002500Min: 2796.4 / Avg: 2802.3 / Max: 2806.4Min: 2768.3 / Avg: 2788.7 / Max: 2801.51. (CC) gcc options: -O3 -pthread -lz -llzma

Stream

This is a benchmark of Stream, the popular system memory (RAM) benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: Triad1236K12K18K24K30KSE +/- 4.27, N = 5SE +/- 12.68, N = 529294.029324.229521.11. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: Triad1235K10K15K20K25KMin: 29315.2 / Avg: 29324.18 / Max: 29340.2Min: 29488.4 / Avg: 29521.12 / Max: 29552.61. (CC) gcc options: -O3 -march=native -fopenmp

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_Test1234080120160200SE +/- 1.19, N = 3SE +/- 0.27, N = 3197.0195.5196.81. (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: 193.1 / Avg: 195.47 / Max: 196.9Min: 196.3 / Avg: 196.83 / Max: 197.21. (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

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 Phase123100200300400500SE +/- 0.93, N = 3SE +/- 2.75, N = 3481.3482.7479.1
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Complex Phase12390180270360450Min: 481.4 / Avg: 482.7 / Max: 484.5Min: 475.2 / Avg: 479.1 / Max: 484.4

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 1.4.9Compression Level: 3, Long Mode - Compression Speed123150300450600750SE +/- 3.37, N = 3SE +/- 5.03, N = 3673.2669.8668.31. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3, Long Mode - Compression Speed123120240360480600Min: 663.1 / Avg: 669.77 / Max: 674Min: 658.3 / Avg: 668.3 / Max: 674.21. (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 Dragon123246810SE +/- 0.0542, N = 3SE +/- 0.0353, N = 37.54887.54637.6012MIN: 7.51 / MAX: 7.62MIN: 7.46 / MAX: 7.75MIN: 7.51 / MAX: 7.76
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian Dragon1233691215Min: 7.49 / Avg: 7.55 / Max: 7.65Min: 7.54 / Avg: 7.6 / Max: 7.67

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 1.4.9Compression Level: 19, Long Mode - Decompression Speed1236001200180024003000SE +/- 17.80, N = 3SE +/- 0.75, N = 32783.82766.12785.31. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Decompression Speed1235001000150020002500Min: 2731 / Avg: 2766.1 / Max: 2788.8Min: 2784 / Avg: 2785.27 / Max: 2786.61. (CC) gcc options: -O3 -pthread -lz -llzma

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 digit1233691215SE +/- 0.05, N = 3SE +/- 0.01, N = 310.2710.3010.231. (CC) gcc options: -O2 -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 12 digit1233691215Min: 10.24 / Avg: 10.3 / Max: 10.4Min: 10.22 / Avg: 10.23 / Max: 10.261. (CC) gcc options: -O2 -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: 0123306090120150SE +/- 0.51, N = 3SE +/- 0.22, N = 3122.96123.71123.801. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 012320406080100Min: 122.77 / Avg: 123.71 / Max: 124.53Min: 123.48 / Avg: 123.8 / Max: 124.221. (CXX) g++ options: -O3 -fPIC -lm

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: Medium123246810SE +/- 0.0129, N = 3SE +/- 0.0322, N = 38.04468.07478.09951. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: Medium1233691215Min: 8.05 / Avg: 8.07 / Max: 8.09Min: 8.04 / Avg: 8.1 / Max: 8.151. (CXX) g++ options: -O3 -flto -pthread

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: CPU1233691215SE +/- 0.02692, N = 3SE +/- 0.03152, N = 39.099419.038759.05168MIN: 3.91MIN: 3.85MIN: 3.921. (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: CPU1233691215Min: 8.99 / Avg: 9.04 / Max: 9.07Min: 9 / Avg: 9.05 / Max: 9.111. (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: Mini-ITX Case1231326395265SE +/- 0.09, N = 3SE +/- 0.50, N = 358.3658.3258.711. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Mini-ITX Case1231224364860Min: 58.2 / Avg: 58.32 / Max: 58.5Min: 58.09 / Avg: 58.71 / Max: 59.711. OpenSCAD version 2019.05

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: 12 - Buffer Length: 256 - Filter Length: 5712360M120M180M240M300MSE +/- 112150.69, N = 3SE +/- 163333.33, N = 32846700002865366672860833331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 12 - Buffer Length: 256 - Filter Length: 5712350M100M150M200M250MMin: 286370000 / Avg: 286536666.67 / Max: 286750000Min: 285780000 / Avg: 286083333.33 / Max: 2863400001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 +/- 96.85, N = 3SE +/- 35.35, N = 34889349132492071. (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: 48951 / Avg: 49132.33 / Max: 49282Min: 49149 / Avg: 49207 / Max: 492711. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

Sysbench

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

OpenBenchmarking.orgMiB/sec, More Is BetterSysbench 1.0.20Test: RAM / Memory1234K8K12K16K20KSE +/- 47.60, N = 3SE +/- 92.41, N = 316736.8916779.8216673.451. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
OpenBenchmarking.orgMiB/sec, More Is BetterSysbench 1.0.20Test: RAM / Memory1233K6K9K12K15KMin: 16693.94 / Avg: 16779.82 / Max: 16858.34Min: 16543.99 / Avg: 16673.45 / Max: 16852.411. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -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: f32 - Engine: CPU1233691215SE +/- 0.09, N = 3SE +/- 0.03, N = 312.7912.7312.72MIN: 8.26MIN: 6.1MIN: 7.471. (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: 12.58 / Avg: 12.73 / Max: 12.91Min: 12.67 / Avg: 12.72 / Max: 12.761. (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: CPU12316003200480064008000SE +/- 9.15, N = 3SE +/- 4.84, N = 37328.247369.737372.41MIN: 7044.31MIN: 7085.09MIN: 7094.31. (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: CPU12313002600390052006500Min: 7351.48 / Avg: 7369.73 / Max: 7380.05Min: 7363.5 / Avg: 7372.41 / Max: 7380.121. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 1080p12320406080100SE +/- 0.05, N = 3SE +/- 0.23, N = 375.2575.1674.801. (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 1080p1231428425670Min: 75.07 / Avg: 75.16 / Max: 75.23Min: 74.37 / Avg: 74.8 / Max: 75.181. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -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 ISPC - Model: Crown123246810SE +/- 0.0186, N = 3SE +/- 0.0160, N = 37.21567.17257.1957MIN: 7.17 / MAX: 7.32MIN: 7.09 / MAX: 7.31MIN: 7.12 / MAX: 7.32
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Crown1233691215Min: 7.15 / Avg: 7.17 / Max: 7.21Min: 7.18 / Avg: 7.2 / Max: 7.23

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 3123306090120150SE +/- 0.60, N = 3SE +/- 0.12, N = 3113.17113.85113.371. (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 312320406080100Min: 113.23 / Avg: 113.85 / Max: 115.05Min: 113.19 / Avg: 113.37 / Max: 113.581. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -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 +/- 0.28, N = 3SE +/- 0.51, N = 3111.77112.33112.441. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, Lossless12320406080100Min: 112.02 / Avg: 112.32 / Max: 112.88Min: 111.45 / Avg: 112.43 / Max: 113.171. (CXX) g++ options: -O3 -fPIC -lm

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: ETC1S123816243240SE +/- 0.19, N = 3SE +/- 0.09, N = 335.0935.2335.021. (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: 34.88 / Avg: 35.23 / Max: 35.55Min: 34.88 / Avg: 35.02 / Max: 35.191. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

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 Slim123612182430SE +/- 0.02, N = 3SE +/- 0.03, N = 323.4423.3623.491. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Leonardo Phone Case Slim123510152025Min: 23.34 / Avg: 23.36 / Max: 23.39Min: 23.46 / Avg: 23.49 / Max: 23.541. OpenSCAD version 2019.05

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 91230.76771.53542.30313.07083.8385SE +/- 0.006, N = 3SE +/- 0.005, N = 33.4123.4123.394
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: Zstd Compression 9123246810Min: 3.41 / Avg: 3.41 / Max: 3.42Min: 3.39 / Avg: 3.39 / Max: 3.4

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 +/- 0.02, N = 313.3813.4213.351. (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 1080p12348121620Min: 13.38 / Avg: 13.42 / Max: 13.45Min: 13.31 / Avg: 13.35 / Max: 13.391. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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 0.8.2Video Input: Chimera 1080p12380160240320400SE +/- 0.08, N = 3SE +/- 0.78, N = 3369.44371.37369.97MIN: 273.31 / MAX: 554.87MIN: 273.86 / MAX: 598.69MIN: 272.55 / MAX: 578.081. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Chimera 1080p12370140210280350Min: 371.26 / Avg: 371.37 / Max: 371.53Min: 368.43 / Avg: 369.97 / Max: 370.881. (CC) gcc options: -pthread -lm

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 1080p123306090120150SE +/- 0.24, N = 3SE +/- 0.29, N = 3121.24121.87121.551. (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: 121.41 / Avg: 121.87 / Max: 122.22Min: 121.21 / Avg: 121.55 / Max: 122.121. (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: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU12316003200480064008000SE +/- 11.77, N = 3SE +/- 10.21, N = 37353.407368.777333.18MIN: 7100.35MIN: 7106.81MIN: 7032.311. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU12313002600390052006500Min: 7349.17 / Avg: 7368.77 / Max: 7389.87Min: 7319.2 / Avg: 7333.18 / Max: 7353.051. (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 Transform12320406080100SE +/- 0.12, N = 3SE +/- 0.38, N = 382.982.882.5
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Hilbert Transform1231632486480Min: 82.6 / Avg: 82.8 / Max: 83Min: 81.8 / Avg: 82.53 / Max: 83.1

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: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU12316003200480064008000SE +/- 8.78, N = 3SE +/- 6.08, N = 37382.837367.467347.50MIN: 7128.16MIN: 7086.46MIN: 7067.911. (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: CPU12313002600390052006500Min: 7358.15 / Avg: 7367.46 / Max: 7385.01Min: 7335.38 / Avg: 7347.5 / Max: 7354.491. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 1.4.9Compression Level: 8, Long Mode - Compression Speed12370140210280350SE +/- 1.06, N = 3SE +/- 0.55, N = 3336.2337.5337.81. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Compression Speed12360120180240300Min: 335.9 / Avg: 337.5 / Max: 339.5Min: 337.1 / Avg: 337.83 / Max: 338.91. (CC) gcc options: -O3 -pthread -lz -llzma

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 Compile1234080120160200SE +/- 0.63, N = 3SE +/- 1.08, N = 3165.54164.77165.461. (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 Compile123306090120150Min: 163.93 / Avg: 164.77 / Max: 165.99Min: 164.34 / Avg: 165.46 / Max: 167.611. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil

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 4K123612182430SE +/- 0.05, N = 3SE +/- 0.03, N = 324.2624.1524.211. (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 4K123612182430Min: 24.05 / Avg: 24.15 / Max: 24.21Min: 24.18 / Avg: 24.21 / Max: 24.271. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CPU-Only12360120180240300SE +/- 0.16, N = 3SE +/- 0.07, N = 3285.35286.64285.71
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CPU-Only12350100150200250Min: 286.37 / Avg: 286.64 / Max: 286.92Min: 285.58 / Avg: 285.71 / Max: 285.78

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 - dCOPY123510152025SE +/- 0.15, N = 3SE +/- 0.07, N = 1222.622.522.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPY123510152025Min: 22.3 / Avg: 22.53 / Max: 22.8Min: 22.3 / Avg: 22.6 / Max: 23.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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: 21231428425670SE +/- 0.31, N = 3SE +/- 0.04, N = 363.8063.9564.071. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 21231326395265Min: 63.43 / Avg: 63.95 / Max: 64.52Min: 64.01 / Avg: 64.07 / Max: 64.151. (CXX) g++ options: -O3 -fPIC -lm

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 +/- 0.14, N = 3SE +/- 0.09, N = 392.8192.4292.461. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p12320406080100Min: 92.17 / Avg: 92.42 / Max: 92.67Min: 92.37 / Avg: 92.46 / Max: 92.631. (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 4K123246810SE +/- 0.01, N = 3SE +/- 0.01, N = 37.407.387.371. (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 4K1233691215Min: 7.36 / Avg: 7.38 / Max: 7.39Min: 7.36 / Avg: 7.37 / Max: 7.391. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: CPU-Only12390180270360450SE +/- 0.27, N = 3SE +/- 0.22, N = 3401.68400.78400.10
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: CPU-Only12370140210280350Min: 400.4 / Avg: 400.78 / Max: 401.31Min: 399.77 / Avg: 400.1 / Max: 400.52

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 +/- 0.31, N = 3SE +/- 0.35, N = 3112.23111.79112.181. (CXX) g++ options: -O3 -lpthread -lrt
OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Chimera 1080p12320406080100Min: 111.42 / Avg: 111.79 / Max: 112.41Min: 111.48 / Avg: 112.18 / Max: 112.631. (CXX) g++ options: -O3 -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: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU1232K4K6K8K10KSE +/- 13.40, N = 3SE +/- 10.08, N = 311040.211061.111082.8MIN: 10813.1MIN: 10800.2MIN: 10778.51. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU1232K4K6K8K10KMin: 11036 / Avg: 11061.1 / Max: 11081.8Min: 11069.6 / Avg: 11082.8 / Max: 11102.61. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-Only1232004006008001000SE +/- 0.62, N = 3SE +/- 1.77, N = 3856.95858.13860.12
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-Only123150300450600750Min: 857.27 / Avg: 858.13 / Max: 859.32Min: 857.52 / Avg: 860.12 / Max: 863.5

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 +/- 0.00, N = 3SE +/- 0.00, N = 310.8910.8710.851. (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: 10.87 / Avg: 10.87 / Max: 10.88Min: 10.85 / Avg: 10.85 / Max: 10.861. (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: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU12348121620SE +/- 0.01, N = 3SE +/- 0.04, N = 317.1717.1717.23MIN: 12.97MIN: 13.2MIN: 12.691. (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: CPU12348121620Min: 17.15 / Avg: 17.17 / Max: 17.18Min: 17.16 / Avg: 17.23 / Max: 17.291. (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: Projector Mount Swivel123306090120150SE +/- 0.12, N = 3SE +/- 0.54, N = 3129.73129.57130.001. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Projector Mount Swivel12320406080100Min: 129.38 / Avg: 129.57 / Max: 129.78Min: 129.03 / Avg: 130 / Max: 130.881. OpenSCAD version 2019.05

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 +/- 0.11, N = 3SE +/- 0.10, N = 3151.69151.32151.201. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.2Pfam Database Search123306090120150Min: 151.16 / Avg: 151.32 / Max: 151.53Min: 151.02 / Avg: 151.2 / Max: 151.361. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm -lmpi

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: 6123510152025SE +/- 0.11, N = 3SE +/- 0.05, N = 321.7621.7221.791. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6123510152025Min: 21.57 / Avg: 21.72 / Max: 21.94Min: 21.72 / Avg: 21.79 / Max: 21.881. (CXX) g++ options: -O3 -fPIC -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: Pistol123306090120150SE +/- 0.29, N = 3SE +/- 0.20, N = 3139.35139.80139.501. OpenSCAD version 2019.05
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Pistol123306090120150Min: 139.25 / Avg: 139.8 / Max: 140.22Min: 139.14 / Avg: 139.5 / Max: 139.811. 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 6 Two-Pass - Input: Bosphorus 4K1230.79431.58862.38293.17723.9715SE +/- 0.00, N = 3SE +/- 0.00, N = 33.523.533.531. (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: 3.53 / Avg: 3.53 / Max: 3.53Min: 3.52 / Avg: 3.53 / Max: 3.531. (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: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU1233691215SE +/- 0.04, N = 3SE +/- 0.06, N = 310.7610.7810.75MIN: 8.2MIN: 8.22MIN: 8.211. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU1233691215Min: 10.72 / Avg: 10.78 / Max: 10.87Min: 10.64 / Avg: 10.75 / Max: 10.861. (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 +/- 0.01, N = 3SE +/- 0.02, N = 335.5035.4035.421. (CXX) g++ options: -O3 -lpthread -lrt
OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Chimera 1080p 10-bit123816243240Min: 35.38 / Avg: 35.4 / Max: 35.42Min: 35.39 / Avg: 35.42 / Max: 35.461. (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 ISPC - Model: Asian Dragon123246810SE +/- 0.0200, N = 3SE +/- 0.0342, N = 38.86698.89198.8853MIN: 8.83 / MAX: 8.95MIN: 8.83 / MAX: 9.03MIN: 8.78 / MAX: 9.04
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian Dragon1233691215Min: 8.87 / Avg: 8.89 / Max: 8.93Min: 8.82 / Avg: 8.89 / Max: 8.94

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: CPU123246810SE +/- 0.01221, N = 3SE +/- 0.02595, N = 37.113257.111147.09340MIN: 3.36MIN: 3.38MIN: 3.381. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU1233691215Min: 7.09 / Avg: 7.11 / Max: 7.13Min: 7.06 / Avg: 7.09 / Max: 7.151. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 0.8.2Video Input: Chimera 1080p 10-bit12320406080100SE +/- 0.05, N = 3SE +/- 0.01, N = 387.2787.1387.03MIN: 57.49 / MAX: 206.69MIN: 57.31 / MAX: 206.69MIN: 57.18 / MAX: 209.791. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Chimera 1080p 10-bit12320406080100Min: 87.06 / Avg: 87.13 / Max: 87.23Min: 87.01 / Avg: 87.03 / Max: 87.051. (CC) gcc options: -pthread -lm

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 21231326395265SE +/- 0.02, N = 3SE +/- 0.02, N = 359.0359.0859.191. (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 21231224364860Min: 59.06 / Avg: 59.08 / Max: 59.11Min: 59.16 / Avg: 59.19 / Max: 59.211. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

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 1080p1230.85731.71462.57193.42924.2865SE +/- 0.00, N = 3SE +/- 0.01, N = 33.803.813.801. (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: 3.8 / Avg: 3.81 / Max: 3.81Min: 3.79 / Avg: 3.8 / Max: 3.811. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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 Makefiles12330060090012001500SE +/- 5.10, N = 3SE +/- 1.12, N = 31394.791397.411398.37
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: Unix Makefiles1232004006008001000Min: 1391.23 / Avg: 1397.41 / Max: 1407.53Min: 1396.15 / Avg: 1398.37 / Max: 1399.74

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: CPU123510152025SE +/- 0.03, N = 3SE +/- 0.10, N = 319.4419.4419.40MIN: 9.35MIN: 8.16MIN: 8.121. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU123510152025Min: 19.4 / Avg: 19.44 / Max: 19.51Min: 19.19 / Avg: 19.4 / Max: 19.511. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 Transform12370140210280350SE +/- 1.51, N = 9SE +/- 1.35, N = 3335.3334.6335.41. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Hilbert Transform12360120180240300Min: 325.1 / Avg: 334.58 / Max: 339.2Min: 333.5 / Avg: 335.37 / Max: 3381. 3.8.1.0

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 Compile1234080120160200SE +/- 0.21, N = 3SE +/- 0.21, N = 3169.15168.78169.19
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Erlang/OTP Compilation 23.2Time To Compile123306090120150Min: 168.41 / Avg: 168.78 / Max: 169.15Min: 168.78 / Avg: 169.19 / Max: 169.47

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 1.4.9Compression Level: 3 - Decompression Speed1236001200180024003000SE +/- 0.78, N = 3SE +/- 2.22, N = 32988.72982.22983.71. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3 - Decompression Speed1235001000150020002500Min: 2981.1 / Avg: 2982.2 / Max: 2983.7Min: 2980.4 / Avg: 2983.67 / Max: 2987.91. (CC) gcc options: -O3 -pthread -lz -llzma

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: 1M1237001400210028003500SE +/- 13.36, N = 3SE +/- 3.71, N = 33445.73438.53443.91. (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: 1M1236001200180024003000Min: 3424.3 / Avg: 3438.5 / Max: 3465.2Min: 3436.5 / Avg: 3443.87 / Max: 3448.31. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

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, Stateful12314002800420056007000SE +/- 4.04, N = 3SE +/- 0.88, N = 36406641964141. (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, Stateful12311002200330044005500Min: 6411 / Avg: 6419 / Max: 6424Min: 6413 / Avg: 6414.33 / Max: 64161. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

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 0.8.2Video Input: Summer Nature 4K12320406080100SE +/- 0.16, N = 3SE +/- 0.15, N = 3104.30104.38104.17MIN: 98.53 / MAX: 116.86MIN: 98.48 / MAX: 117.64MIN: 98.31 / MAX: 117.281. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 4K12320406080100Min: 104.09 / Avg: 104.38 / Max: 104.64Min: 103.91 / Avg: 104.17 / Max: 104.431. (CC) gcc options: -pthread -lm

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing on the CPU with the water_GMX50 data. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021Input: water_GMX50_bare1230.12740.25480.38220.50960.637SE +/- 0.006, N = 3SE +/- 0.007, N = 40.5660.5650.5661. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021Input: water_GMX50_bare123246810Min: 0.55 / Avg: 0.57 / Max: 0.57Min: 0.55 / Avg: 0.57 / Max: 0.581. (CXX) g++ options: -O3 -pthread

JPEG XL

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: PNG - Encode Speed: 71231.29832.59663.89495.19326.4915SE +/- 0.01, N = 3SE +/- 0.01, N = 35.775.765.761. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.3Input: PNG - Encode Speed: 7123246810Min: 5.74 / Avg: 5.76 / Max: 5.78Min: 5.75 / Avg: 5.76 / Max: 5.771. (CXX) g++ options: -funwind-tables -O3 -O2 -pthread -fPIE -pie -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 4K123816243240SE +/- 0.06, N = 3SE +/- 0.03, N = 334.8434.8334.891. (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 4K123714212835Min: 34.73 / Avg: 34.83 / Max: 34.94Min: 34.83 / Avg: 34.89 / Max: 34.931. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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 digit1232004006008001000SE +/- 0.54, N = 3SE +/- 0.13, N = 3918.42918.95919.971. (CC) gcc options: -O2 -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digit123160320480640800Min: 917.9 / Avg: 918.95 / Max: 919.73Min: 919.73 / Avg: 919.97 / Max: 920.171. (CC) gcc options: -O2 -pthread

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 19123510152025SE +/- 0.03, N = 3SE +/- 0.02, N = 321.5321.5321.50
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: Zstd Compression 19123510152025Min: 21.49 / Avg: 21.53 / Max: 21.59Min: 21.47 / Avg: 21.5 / Max: 21.52

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: Ninja12330060090012001500SE +/- 1.68, N = 3SE +/- 0.07, N = 31368.461370.701370.15
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: Ninja1232004006008001000Min: 1368.46 / Avg: 1370.7 / Max: 1374Min: 1370.02 / Avg: 1370.15 / Max: 1370.22

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 +/- 1.20, N = 3SE +/- 3.28, N = 33779377537811. (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: 3773 / Avg: 3774.67 / Max: 3777Min: 3776 / Avg: 3780.67 / Max: 37871. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

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 1.4.9Compression Level: 8, Long Mode - Decompression Speed1237001400210028003500SE +/- 2.82, N = 3SE +/- 4.71, N = 33215.63216.13211.51. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Decompression Speed1236001200180024003000Min: 3212.4 / Avg: 3216.07 / Max: 3221.6Min: 3202.6 / Avg: 3211.5 / Max: 3218.61. (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_1d - Data Type: u8s8f32 - Engine: CPU1233691215SE +/- 0.01, N = 3SE +/- 0.01, N = 310.3810.3810.39MIN: 5.16MIN: 5.14MIN: 5.151. (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: CPU1233691215Min: 10.37 / Avg: 10.38 / Max: 10.39Min: 10.38 / Avg: 10.39 / Max: 10.411. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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 0.8.2Video Input: Summer Nature 1080p12370140210280350SE +/- 0.62, N = 3SE +/- 1.02, N = 3332.31332.59332.73MIN: 294.32 / MAX: 361.09MIN: 290.31 / MAX: 364.3MIN: 281.58 / MAX: 363.481. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 1080p12360120180240300Min: 331.4 / Avg: 332.59 / Max: 333.51Min: 330.78 / Avg: 332.73 / Max: 334.221. (CC) gcc options: -pthread -lm

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 Compile1232004006008001000SE +/- 0.14, N = 3SE +/- 0.33, N = 3888.50889.31889.46
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 15.11Time To Compile123160320480640800Min: 889.05 / Avg: 889.31 / Max: 889.51Min: 888.81 / Avg: 889.46 / Max: 889.85

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 +/- 0.05, N = 3SE +/- 0.54, N = 3135.37135.33135.241. (CXX) g++ options: -O3 -lpthread -lrt
OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 1080p123306090120150Min: 135.23 / Avg: 135.33 / Max: 135.4Min: 134.18 / Avg: 135.24 / Max: 135.931. (CXX) g++ options: -O3 -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: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU1232K4K6K8K10KSE +/- 16.13, N = 3SE +/- 10.66, N = 311062.711053.111060.2MIN: 10779.5MIN: 10763.5MIN: 10798.11. (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: CPU1232K4K6K8K10KMin: 11026 / Avg: 11053.07 / Max: 11081.8Min: 11039.5 / Avg: 11060.2 / Max: 110751. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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++12330K60K90K120K150KSE +/- 57.74, N = 3SE +/- 33.33, N = 31190001189001189331. (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: 118800 / Avg: 118900 / Max: 119000Min: 118900 / Avg: 118933.33 / Max: 1190001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread

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 +/- 0.05, N = 3SE +/- 0.03, N = 339.9139.9439.941. (CXX) g++ options: -O3 -lpthread -lrt
OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 4K123816243240Min: 39.86 / Avg: 39.94 / Max: 40.03Min: 39.9 / Avg: 39.94 / Max: 401. (CXX) g++ options: -O3 -lpthread -lrt

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 2.4Preset: Exhaustive1234080120160200SE +/- 0.06, N = 3SE +/- 0.11, N = 3198.66198.62198.521. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: Exhaustive1234080120160200Min: 198.51 / Avg: 198.62 / Max: 198.7Min: 198.41 / Avg: 198.52 / Max: 198.751. (CXX) g++ options: -O3 -flto -pthread

Google Draco

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

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.4.1Model: Lion12314002800420056007000SE +/- 15.84, N = 3SE +/- 8.84, N = 36335633363311. (CXX) g++ options: -O3
OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.4.1Model: Lion12311002200330044005500Min: 6317 / Avg: 6333.33 / Max: 6365Min: 6314 / Avg: 6331.33 / Max: 63431. (CXX) g++ options: -O3

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 2.4Preset: Thorough123612182430SE +/- 0.03, N = 3SE +/- 0.01, N = 325.9425.9325.951. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: Thorough123612182430Min: 25.88 / Avg: 25.93 / Max: 25.99Min: 25.94 / Avg: 25.95 / Max: 25.961. (CXX) g++ options: -O3 -flto -pthread

Sysbench

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

OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPU1232K4K6K8K10KSE +/- 2.89, N = 3SE +/- 2.02, N = 310354.1910352.6010358.791. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPU1232K4K6K8K10KMin: 10348.5 / Avg: 10352.6 / Max: 10358.17Min: 10355.1 / Avg: 10358.79 / Max: 10362.071. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm

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 +/- 0.15, N = 3SE +/- 0.22, N = 395.2595.3095.251. (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 1080p12320406080100Min: 95.1 / Avg: 95.3 / Max: 95.59Min: 94.9 / Avg: 95.25 / Max: 95.661. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -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 19123140280420560700SE +/- 0.06, N = 3SE +/- 0.07, N = 3655.57655.44655.46
OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 4 + Zstd Compression 19123120240360480600Min: 655.37 / Avg: 655.44 / Max: 655.57Min: 655.37 / Avg: 655.46 / Max: 655.6

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 1080p1230.8641.7282.5923.4564.32SE +/- 0.00, N = 3SE +/- 0.00, N = 33.843.843.841. (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: 3.83 / Avg: 3.84 / Max: 3.84Min: 3.84 / Avg: 3.84 / Max: 3.851. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -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 0 Two-Pass - Input: Bosphorus 1080p1230.04730.09460.14190.18920.2365SE +/- 0.00, N = 3SE +/- 0.00, N = 30.210.210.211. (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: 0.21 / Avg: 0.21 / Max: 0.21Min: 0.21 / Avg: 0.21 / Max: 0.211. (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 4K1230.42080.84161.26241.68322.104SE +/- 0.00, N = 3SE +/- 0.00, N = 31.871.871.871. (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: 1.87 / Avg: 1.87 / Max: 1.87Min: 1.86 / Avg: 1.87 / Max: 1.871. (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 4K1230.01580.03160.04740.06320.079SE +/- 0.00, N = 3SE +/- 0.00, N = 30.070.070.071. (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: 0.07 / Avg: 0.07 / Max: 0.07Min: 0.07 / Avg: 0.07 / Max: 0.071. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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-v1-1.0123246810SE +/- 0.054, N = 9SE +/- 0.799, N = 37.0427.0587.867MIN: 6.48 / MAX: 25.93MIN: 4.84 / MAX: 31.92MIN: 4.68 / MAX: 31.891. (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-v1-1.01233691215Min: 6.77 / Avg: 7.06 / Max: 7.32Min: 6.85 / Avg: 7.87 / Max: 9.441. (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_224123246810SE +/- 0.130, N = 9SE +/- 0.012, N = 36.3716.2036.039MIN: 3.99 / MAX: 30.98MIN: 3.99 / MAX: 30.96MIN: 5.58 / MAX: 23.611. (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_2241233691215Min: 6.01 / Avg: 6.2 / Max: 7.23Min: 6.03 / Avg: 6.04 / Max: 6.061. (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: SqueezeNetV1.01233691215SE +/- 0.35, N = 9SE +/- 0.14, N = 312.3612.1311.52MIN: 7.66 / MAX: 47.8MIN: 7.02 / MAX: 50.71MIN: 8.33 / MAX: 53.251. (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: SqueezeNetV1.012348121620Min: 11.16 / Avg: 12.13 / Max: 14.22Min: 11.27 / Avg: 11.52 / Max: 11.761. (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.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TT123510152025SE +/- 0.49, N = 3SE +/- 0.38, N = 1220.918.919.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TT123510152025Min: 18.1 / Avg: 18.9 / Max: 19.8Min: 17.6 / Avg: 19.63 / Max: 221. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TN123510152025SE +/- 1.18, N = 3SE +/- 0.42, N = 1119.521.321.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TN123510152025Min: 19.1 / Avg: 21.33 / Max: 23.1Min: 19.1 / Avg: 21.45 / Max: 23.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NT123510152025SE +/- 0.46, N = 3SE +/- 0.39, N = 1120.019.619.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NT123510152025Min: 18.8 / Avg: 19.6 / Max: 20.4Min: 16.6 / Avg: 19.12 / Max: 21.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NN123510152025SE +/- 0.70, N = 3SE +/- 0.40, N = 1020.220.719.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NN123510152025Min: 19.4 / Avg: 20.7 / Max: 21.8Min: 18.1 / Avg: 19.63 / Max: 21.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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-TLS12340K80K120K160K200KSE +/- 181.65, N = 3SE +/- 5046.21, N = 92059832060851996071. (CC) gcc options: -pedantic -O3
OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLS12340K80K120K160K200KMin: 205721.33 / Avg: 206084.63 / Max: 206268.44Min: 161242.58 / Avg: 199607.08 / Max: 206402.361. (CC) gcc options: -pedantic -O3

167 Results Shown

VOSK Speech Recognition Toolkit
Liquid-DSP
oneDNN
GNU Radio
Xcompact3d Incompact3d:
  input.i3d 129 Cells Per Direction
  input.i3d 193 Cells Per Direction
SVT-VP9
Timed MrBayes Analysis
ViennaCL
GNU Radio
SVT-HEVC
JPEG XL:
  JPEG - 8
  JPEG - 5
Mobile Neural Network
oneDNN
Embree
Stockfish
ViennaCL
Embree
GNU Radio
JPEG XL Decoding
LuaRadio
Zstd Compression
GNU Radio
ViennaCL
Zstd Compression
GNU GMP GMPbench
Zstd Compression
Timed Linux Kernel Compilation
Chia Blockchain VDF
libavif avifenc
Mobile Neural Network
JPEG XL
srsLTE
ViennaCL:
  CPU BLAS - dGEMV-T
  CPU BLAS - dGEMV-N
Liquid-DSP
Google Draco
JPEG XL Decoding
srsLTE
Embree
Stream
KTX-Software toktx
libavif avifenc
Stream
Liquid-DSP
Timed Mesa Compilation
Basis Universal
oneDNN
Zstd Compression:
  8 - Decompression Speed
  19 - Compression Speed
QMCPACK
OpenSCAD
LuaRadio
KTX-Software toktx
ViennaCL
JPEG XL:
  PNG - 5
  JPEG - 7
CP2K Molecular Dynamics
libjpeg-turbo tjbench
AOM AV1
ViennaCL
Liquid-DSP
Stream
Xmrig
oneDNN:
  Convolution Batch Shapes Auto - f32 - CPU
  Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU
Zstd Compression
oneDNN
GNU Radio
Zstd Compression
Stream
srsLTE
LuaRadio
Zstd Compression
Embree
Zstd Compression
Helsing
libavif avifenc
ASTC Encoder
oneDNN
OpenSCAD
Liquid-DSP
PJSIP
Sysbench
oneDNN:
  IP Shapes 1D - f32 - CPU
  Recurrent Neural Network Inference - f32 - CPU
SVT-VP9
Embree
Basis Universal
libavif avifenc
Basis Universal
OpenSCAD
KTX-Software toktx
AOM AV1
dav1d
SVT-HEVC
oneDNN
LuaRadio
oneDNN
Zstd Compression
Timed Wasmer Compilation
AOM AV1
Blender
ViennaCL
libavif avifenc
AOM AV1:
  Speed 9 Realtime - Bosphorus 1080p
  Speed 6 Realtime - Bosphorus 4K
Blender
libgav1
oneDNN
Blender
AOM AV1
oneDNN
OpenSCAD
Timed HMMer Search
libavif avifenc
OpenSCAD
AOM AV1
oneDNN
libgav1
Embree
oneDNN
dav1d
Basis Universal
AOM AV1
Timed LLVM Compilation
oneDNN
GNU Radio
Timed Erlang/OTP Compilation
Zstd Compression
Xmrig
PJSIP
dav1d
GROMACS
JPEG XL
AOM AV1
Helsing
KTX-Software toktx
Timed LLVM Compilation
PJSIP
Zstd Compression
oneDNN
dav1d
Timed Node.js Compilation
libgav1
oneDNN
Chia Blockchain VDF
libgav1
ASTC Encoder
Google Draco
ASTC Encoder
Sysbench
SVT-VP9
KTX-Software toktx
SVT-HEVC
AOM AV1:
  Speed 0 Two-Pass - Bosphorus 1080p
  Speed 4 Two-Pass - Bosphorus 4K
  Speed 0 Two-Pass - Bosphorus 4K
Mobile Neural Network:
  mobilenet-v1-1.0
  MobileNetV2_224
  SqueezeNetV1.0
ViennaCL:
  CPU BLAS - dGEMM-TT
  CPU BLAS - dGEMM-TN
  CPU BLAS - dGEMM-NT
  CPU BLAS - dGEMM-NN
SecureMark

1

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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 Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0xb000038
Python Notes: Python 3.8.6
Security Notes: 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: Not affected + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable

Testing initiated at 8 May 2021 14:44 by user phoronix.

2

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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 Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0xb000038
Python Notes: Python 3.8.6
Security Notes: 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: Not affected + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable

Testing initiated at 8 May 2021 21:01 by user phoronix.

3

Processor: Intel Core i7-6800K @ 3.80GHz (6 Cores / 12 Threads), Motherboard: MSI X99A WORKSTATION (MS-7A54) v1.0 (1.10 BIOS), Chipset: Intel Xeon E7 v4/Xeon, Memory: 16GB, Disk: 120GB TOSHIBA TR150, Graphics: Zotac NVIDIA NV137 2GB, Audio: Realtek ALC1150, Monitor: G237HL, Network: Intel I218-LM + Intel I210

OS: Ubuntu 20.10, Kernel: 5.8.0-41-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: nouveau, OpenGL: 4.3 Mesa 20.2.1, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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 Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0xb000038
Python Notes: Python 3.8.6
Security Notes: 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: Not affected + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable

Testing initiated at 9 May 2021 14:59 by user phoronix.