Pop OS 21.04

AMD Ryzen Threadripper 3990X 64-Core testing with a Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS) and AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB on Pop 21.04 vs. 20.04. Pop!_OS benchmarking by Michael Larabel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2107177-IB-POPOS742515
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

Statistics

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

Graph Settings

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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Pop_OS 20.04
July 14 2021
  1 Day, 8 Hours, 47 Minutes
Pop_OS 21.04
July 16 2021
  20 Hours, 25 Minutes
Invert Behavior (Only Show Selected Data)
  1 Day, 2 Hours, 36 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


Pop OS 21.04ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionPop_OS 20.04Pop_OS 21.04AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads)Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS)AMD Starship/Matisse126GBSamsung SSD 970 EVO Plus 500GBAMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (1750/875MHz)AMD Navi 10 HDMI AudioDELL P2415QIntel I211 + Intel Wi-Fi 6 AX200Pop 20.045.11.0-7620-generic (x86_64)GNOME Shell 3.36.7X Server 1.20.94.6 Mesa 21.0.1 (LLVM 11.0.0)1.2.145GCC 9.3.0ext43840x2160Samsung SSD 970 EVO Plus 500GB + 32GB Flash DiskPop 21.04GNOME Shell 3.38.44.6 Mesa 21.0.1 (LLVM 11.0.1)GCC 10.3.0OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- Pop_OS 20.04: --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/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 - Pop_OS 21.04: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-gDeRY6/gcc-10-10.3.0/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-build-config=bootstrap-lto-lean --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- Pop_OS 20.04: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8301039- Pop_OS 21.04: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301039Graphics Details- GLAMORJava Details- Pop_OS 20.04: OpenJDK Runtime Environment (build 11.0.11+9-Ubuntu-0ubuntu2.20.04)- Pop_OS 21.04: OpenJDK Runtime Environment (build 11.0.11+9-Ubuntu-0ubuntu2)Python Details- Pop_OS 20.04: Python 3.8.10- Pop_OS 21.04: Python 3.9.5Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Pop_OS 20.04 vs. Pop_OS 21.04 ComparisonPhoronix Test SuiteBaseline+23.9%+23.9%+47.8%+47.8%+71.7%+71.7%95.4%55.7%39%28%24.5%21.3%18.6%17.2%16.7%14.1%12.9%12.4%10.4%9.9%8.9%8.6%8.3%8.3%8.1%7.4%7.3%7.3%7.2%7.1%7.1%7%6.9%6.8%6.6%6.4%6.1%5.2%5.2%5.1%5%4.9%4.9%4.9%4.6%4.5%4.5%4.4%4.3%4.2%3.9%3.9%3.7%3.6%3.5%3.5%3.5%3.3%3.3%3.3%3.2%3%3%2.9%2.8%2.7%2.6%2.5%2.5%2.4%2.4%2.3%2.3%2.3%2.3%2.3%2.2%2.2%2.1%2.1%2.1%2.1%2%2%2%2%2%2%I.M.D.STime To CompileBosphorus 1080p49.6%nbody41%P.I.O.A.3.V.1.TBosphorus 1080p - Medium36.7%Bosphorus 1080p - Very Fast36.4%Bosphorus 1080p - Slow35.7%Speed 6 Realtime - Bosphorus 4K35.4%Speed 6 Two-Pass - Bosphorus 1080p31.9%Seq Fill29.4%G.A.U.J.F29.3%SVG Files To PNGRand Fill27.5%S.N.126.7%Bosphorus 4K - Very Fast24.8%Time To CompileBosphorus 4K - Medium21.8%Preset 8 - Bosphorus 1080p21.4%resnet-v2-50Bosphorus 4K - Slow20.9%Time To Compile1920 x 1080 - High18.4%Chimera 1080p18.4%C.1.1.b18.3%1920 x 1080 - OpenGL ES17.5%python_startupNon-Exponential10 - Bosphorus 1080p16.2%Update Rand15.6%2560 x 1440 - OpenGL ES15.4%H.2.V.E15.2%Speed 4 Two-Pass - Bosphorus 1080p14.4%7 - Bosphorus 1080p14.2%Wavelet BlurSpeed 8 Realtime - Bosphorus 1080p13.6%Speed 6 Realtime - Bosphorus 1080p13.5%Summer Nature 4K13.3%CPU - alexnetALS Movie Lens12.8%django_templateM.M.B.S.T - u8s8f32 - CPU12.2%V.Q.O - Bosphorus 1080p11.6%Speed 4 Two-Pass - Bosphorus 4K11.3%vklBenchmarkStructuredVolume10.6%FIR FilterP.S.O - Bosphorus 1080p10.2%Speed 9 Realtime - Bosphorus 1080p10.1%Color EnhanceSpeed 6 Two-Pass - Bosphorus 4K9.8%Rand Forest9.6%OPTIONS, Stateful9.3%D.B.s - f32 - CPU9.1%VMAF Optimized - Bosphorus 1080p9%Complex PhaseApache Spark ALS8.8%Antialiasmobilenet-v1-1.08.5%PistolTime To Compilepickle_pure_python3840 x 2160 - OpenGL ES7.5%Tile Glass6, LosslessTime To CompileResizingReflectTime To CompileQ.1.C.E.5SpaceshipHairBosphorus 4K - Ultra Fast6.7%Defaultblosclz6.6%Water CausticC26706.3%Total Time6.1%CPU - DenseNetPreset 4 - Bosphorus 1080p6%Preset 8 - Bosphorus 4K5.9%Speed 8 Realtime - Bosphorus 4K5.4%CPU5.2%IP Shapes 3D - f32 - CPUCartoonR.C.a.P - CPUC.B.S.A - f32 - CPUSpeed 0 - Bosphorus 1080pmobilenetV32048 x 2048 - Total TimefloatRotate 90 Degreesregex_compilevklBenchmarkVdbVolume4.4%R.R.W.RSpeed 5 - Bosphorus 1080p104.2%Q.1.H.C4.2%Danish Mood - CPUC75524%CPU - regnety_400m3.9%SharpenETC1 - HighestSwirl3.9%Bosphorus 4K3.9%P.P.B3.7%CPU - resnet503.7%ETC2 RGB - HighestF.B.t.B.F.F3.6%DefaultETC1M.M.B.S.T - f32 - CPUScala DottyQ.1.L.CP.R.W.S.S.MD.B.s - f32 - CPUCrop2to3Rand ReadBosphorus 1080p - Ultra Fast2.9%F.B.t.B.F.Fsqueezenetv1.1EnhancedIP Shapes 3D - u8s8f32 - CPU2.7%Preset 4 - Bosphorus 4K2.7%gopathlib62.5%CPU - mnasnetDLSC - CPU2.4%VoiceMark_10010, Lossless2.4%2560 x 1440OPTIONS, StatelessWavelet Contour - 3840 x 2160Wavelet Contour - 3840 x 21603840 x 2160T.T.F.S.SSavina Reactors.IO2.3%Speed 5 - Bosphorus 4K2.2%Rand Fill SyncMaterial Tester1920 x 1080CPU - SqueezeNet v1.1INVITEEP.CNASA Streamlines - Path Tracer2.1%Hilbert TransformETC2SMP Parallel5.P.T.2.P.S.2.Q2%L.P.C.SApache Spark Bayes2%Wavelet Volume - 3840 x 2160WAV To MP32%Wavelet Volume - 3840 x 2160WAV To Ogg2%52%crypto_pyaes2%RenaissanceTimed Wasmer Compilationx265PyPerformanceG'MICKvazaarKvazaarKvazaarAOM AV1AOM AV1Facebook RocksDBRenaissancelibrsvgFacebook RocksDBdav1dKvazaarTimed GDB GNU Debugger CompilationKvazaarSVT-AV1Mobile Neural NetworkKvazaarTimed FFmpeg CompilationUnvanquisheddav1ddav1dGravityMarkPyPerformanceTungsten RendererSVT-HEVCFacebook RocksDBGravityMarkx264AOM AV1SVT-HEVCGEGLAOM AV1AOM AV1dav1dNCNNRenaissancePyPerformanceoneDNNSVT-VP9AOM AV1OpenVKLGNU RadioSVT-VP9AOM AV1GEGLAOM AV1RenaissancePJSIPoneDNNSVT-VP9LuaRadioRenaissanceGEGLMobile Neural NetworkOpenSCADTimed Linux Kernel CompilationPyPerformanceGravityMarkGEGLlibavif avifencTimed Mesa CompilationGraphicsMagickGEGLTimed Godot Game Engine CompilationWebP2 Image EncodeNatronTungsten RendererKvazaarWebP Image EncodeC-BloscTungsten RendererNgspiceTachyonTNNSVT-AV1SVT-AV1AOM AV1DeepSpeechoneDNNGEGLLuxCoreRenderoneDNNVP9 libvpx EncodingMobile Neural NetworkAOBenchPyPerformanceGEGLPyPerformanceOpenVKLFacebook RocksDBVP9 libvpx Encodinglibavif avifencWebP Image EncodeLuxCoreRenderNgspiceNCNNGraphicsMagickBetsy GPU CompressorGraphicsMagickx265LibRawNCNNBetsy GPU CompressorGNU RadioWebP2 Image EncodeEtcpakoneDNNRenaissanceWebP2 Image EncodeTTSIOD 3D RendereroneDNNGEGLPyPerformanceFacebook RocksDBKvazaarLuaRadioMobile Neural NetworkGraphicsMagickoneDNNSVT-AV1PyPerformancePyPerformancelibavif avifencNCNNLuxCoreRenderGoogle SynthMarklibavif avifencET: LegacyPJSIPParaViewParaViewET: LegacyYafaRayRenaissanceVP9 libvpx EncodingFacebook RocksDBAppleseedET: LegacyTNNPJSIPNAS Parallel BenchmarksOSPrayGNU RadioEtcpakRadiance BenchmarksrsRANOpenSCADRenaissanceParaViewLAME MP3 EncodingParaViewOgg Audio Encodingrav1ePyPerformancePop_OS 20.04Pop_OS 21.04

Pop OS 21.04renaissance: In-Memory Database Shootoutbuild-wasmer: Time To Compilex265: Bosphorus 1080ppyperformance: nbodygmic: Plotting Isosurface Of A 3D Volume, 1000 Timeskvazaar: Bosphorus 1080p - Mediumkvazaar: Bosphorus 1080p - Very Fastkvazaar: Bosphorus 1080p - Slowaom-av1: Speed 6 Realtime - Bosphorus 4Kaom-av1: Speed 6 Two-Pass - Bosphorus 1080procksdb: Seq Fillrenaissance: Genetic Algorithm Using Jenetics + Futuresrsvg: SVG Files To PNGrocksdb: Rand Filldav1d: Summer Nature 1080pkvazaar: Bosphorus 4K - Very Fastbuild-gdb: Time To Compilekvazaar: Bosphorus 4K - Mediumsvt-av1: Preset 8 - Bosphorus 1080pmnn: resnet-v2-50kvazaar: Bosphorus 4K - Slowbuild-ffmpeg: Time To Compiledav1d: Chimera 1080pdav1d: Chimera 1080p 10-bitgravitymark: 1920 x 1080 - OpenGL ESpyperformance: python_startuptungsten: Non-Exponentialsvt-hevc: 10 - Bosphorus 1080procksdb: Update Randgravitymark: 2560 x 1440 - OpenGL ESx264: H.264 Video Encodingaom-av1: Speed 4 Two-Pass - Bosphorus 1080psvt-hevc: 7 - Bosphorus 1080pgegl: Wavelet Bluraom-av1: Speed 8 Realtime - Bosphorus 1080paom-av1: Speed 6 Realtime - Bosphorus 1080pdav1d: Summer Nature 4Krenaissance: ALS Movie Lenspyperformance: django_templateonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUsvt-vp9: Visual Quality Optimized - Bosphorus 1080paom-av1: Speed 4 Two-Pass - Bosphorus 4Kopenvkl: vklBenchmarkStructuredVolumegnuradio: FIR Filtersvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080paom-av1: Speed 9 Realtime - Bosphorus 1080pgegl: Color Enhanceaom-av1: Speed 6 Two-Pass - Bosphorus 4Krenaissance: Rand Forestpjsip: OPTIONS, Statefulonednn: Deconvolution Batch shapes_1d - f32 - CPUluaradio: Complex Phaserenaissance: Apache Spark ALSgegl: Antialiasmnn: mobilenet-v1-1.0openscad: Pistolbuild-linux-kernel: Time To Compilepyperformance: pickle_pure_pythongravitymark: 3840 x 2160 - OpenGL ESgegl: Tile Glassavifenc: 6, Losslessbuild-mesa: Time To Compilegegl: Reflectbuild-godot: Time To Compilewebp2: Quality 100, Compression Effort 5natron: Spaceshiptungsten: Hairkvazaar: Bosphorus 4K - Ultra Fastwebp: Defaultblosc: blosclztungsten: Water Causticngspice: C2670tnn: CPU - DenseNetsvt-av1: Preset 4 - Bosphorus 1080psvt-av1: Preset 8 - Bosphorus 4Kaom-av1: Speed 8 Realtime - Bosphorus 4Kdeepspeech: CPUonednn: IP Shapes 3D - f32 - CPUgegl: Cartoononednn: Convolution Batch Shapes Auto - f32 - CPUvpxenc: Speed 0 - Bosphorus 1080pmnn: mobilenetV3aobench: 2048 x 2048 - Total Timepyperformance: floatgegl: Rotate 90 Degreespyperformance: regex_compileopenvkl: vklBenchmarkVdbVolumerocksdb: Read Rand Write Randvpxenc: Speed 5 - Bosphorus 1080pavifenc: 10webp: Quality 100, Highest Compressionluxcorerender: Danish Mood - CPUngspice: C7552ncnn: CPU - regnety_400mgraphics-magick: Sharpengraphics-magick: Swirlx265: Bosphorus 4Klibraw: Post-Processing Benchmarkncnn: CPU - resnet50gnuradio: Five Back to Back FIR Filterswebp2: Defaultetcpak: ETC1onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUrenaissance: Scala Dottywebp2: Quality 100, Lossless Compressionttsiod-renderer: Phong Rendering With Soft-Shadow Mappingonednn: Deconvolution Batch shapes_3d - f32 - CPUgegl: Croppyperformance: 2to3rocksdb: Rand Readkvazaar: Bosphorus 1080p - Ultra Fastluaradio: Five Back to Back FIR Filtersmnn: squeezenetv1.1graphics-magick: Enhancedonednn: IP Shapes 3D - u8s8f32 - CPUsvt-av1: Preset 4 - Bosphorus 4Kpyperformance: gopyperformance: pathlibavifenc: 6luxcorerender: DLSC - CPUsynthmark: VoiceMark_100avifenc: 10, Losslessetlegacy: 2560 x 1440pjsip: OPTIONS, Statelessparaview: Wavelet Contour - 3840 x 2160paraview: Wavelet Contour - 3840 x 2160etlegacy: 3840 x 2160yafaray: Total Time For Sample Scenerenaissance: Savina Reactors.IOvpxenc: Speed 5 - Bosphorus 4Krocksdb: Rand Fill Syncappleseed: Material Testeretlegacy: 1920 x 1080tnn: CPU - SqueezeNet v1.1pjsip: INVITEnpb: EP.Cospray: NASA Streamlines - Path Tracergnuradio: Hilbert Transformetcpak: ETC2radiance: SMP Parallelsrsran: 5G PHY_DL_NR Test 270 PRB SISO 256-QAMopenscad: Leonardo Phone Case Slimrenaissance: Apache Spark Bayesencode-mp3: WAV To MP3encode-ogg: WAV To Oggrav1e: 5pyperformance: crypto_pyaesgraphics-magick: Noise-Gaussianaom-av1: Speed 0 Two-Pass - Bosphorus 1080pvpxenc: Speed 0 - Bosphorus 4Kappleseed: Disney Materialc-ray: Total Time - 4K, 16 Rays Per Pixelsrsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMmnn: inception-v3onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUrenaissance: Akka Unbalanced Cobwebbed Treewebp: Quality 100, Lossless, Highest Compressionpovray: Trace Timemnn: MobileNetV2_224npb: CG.Cnpb: SP.Bgegl: Scalebasis: UASTC Level 2blender: BMW27 - CPU-Onlywebp: Quality 100ncnn: CPU - resnet18basis: UASTC Level 0webp2: Quality 75, Compression Effort 7luajit: Sparse Matrix Multiplysrsran: OFDM_Testtesseract-ocr: Time To OCR 7 Imagesgraphics-magick: HWB Color Spaceluxcorerender: LuxCore Benchmark - CPUncnn: CPU - shufflenet-v2onednn: IP Shapes 1D - u8s8f32 - CPUtoktx: UASTC 3onednn: Recurrent Neural Network Training - u8s8f32 - CPUtnn: CPU - MobileNet v2rav1e: 6paraview: Wavelet Contour - 1920 x 1080paraview: Wavelet Contour - 1920 x 1080onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUopenscad: Mini-ITX Caseencode-ape: WAV To APEwebp: Quality 100, Losslessgravitymark: 2560 x 1440 - OpenGLonednn: Recurrent Neural Network Training - f32 - CPUncnn: CPU - yolov4-tinyhugin: Panorama Photo Assistant + Stitching Timerocksdb: Read While Writingetcpak: DXT1rnnoise: onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUbrl-cad: VGR Performance Metricgraphics-magick: Rotategnuradio: FM Deemphasis Filterneatbench: CPUtesseract: 3840 x 2160onednn: Convolution Batch Shapes Auto - u8s8f32 - CPUastcenc: Exhaustivepyperformance: chaosgnuradio: IIR Filterluajit: Fast Fourier Transformdcraw: RAW To PPM Image Conversiononednn: Recurrent Neural Network Inference - u8s8f32 - CPUespeak: Text-To-Speech Synthesisrav1e: 1onednn: Recurrent Neural Network Inference - f32 - CPUsrsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMsrsran: 5G PHY_DL_NR Test 270 PRB SISO 256-QAMavifenc: 0encode-opus: WAV To Opus Encodeneatbench: Allradiance: Serialopenvkl: vklBenchmarkUnstructuredVolumesrsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMncnn: CPU - squeezenet_ssdoidn: RT.ldr_alb_nrm.3840x2160oidn: RT.hdr_alb_nrm.3840x2160liquid-dsp: 64 - 256 - 57blender: Pabellon Barcelona - CPU-Onlytnn: CPU - SqueezeNet v2pybench: Total For Average Test Timesaom-av1: Speed 9 Realtime - Bosphorus 4Knpb: FT.Cencode-flac: WAV To FLACsvt-hevc: 1 - Bosphorus 1080pluaradio: Hilbert Transformnpb: IS.Dgnuradio: Signal Source (Cosine)v-ray: CPUgravitymark: 3840 x 2160 - Vulkanblender: Classroom - CPU-Onlyospray: San Miguel - Path Tracerembree: Pathtracer - Asian Dragonliquid-dsp: 32 - 256 - 57srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMnpb: LU.Ctungsten: Volumetric Causticmnn: SqueezeNetV1.0renaissance: Finagle HTTP Requestsluajit: Monte Carloncnn: CPU - googlenetembree: Pathtracer ISPC - Asian Dragonluajit: Jacobi Successive Over-Relaxationtoktx: Zstd Compression 9luxcorerender: Orange Juice - CPUpyperformance: raytracetesseract: 2560 x 1440openvkl: vklBenchmarkpyperformance: json_loadsavifenc: 2gravitymark: 2560 x 1440 - Vulkanrenaissance: Apache Spark PageRankvosk: astcenc: Mediumbasis: UASTC Level 3paraview: Many Spheres - 3840 x 2160paraview: Many Spheres - 3840 x 2160luajit: Dense LU Matrix Factorizationliquid-dsp: 128 - 256 - 57tjbench: Decompression Throughputtoktx: Zstd Compression 19astcenc: Thoroughsrsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMembree: Pathtracer - Crownnpb: BT.Cluaradio: FM Deemphasis Filterluajit: Compositenpb: SP.Cembree: Pathtracer ISPC - Asian Dragon Objnpb: MG.Cnpb: EP.Dappleseed: Emilyblender: Fishy Cat - CPU-Onlyliquid-dsp: 16 - 256 - 57srsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMbasis: ETC1Stoktx: UASTC 3 + Zstd Compression 19srsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMparaview: Many Spheres - 1920 x 1080gravitymark: 3840 x 2160 - OpenGLparaview: Many Spheres - 1920 x 1080indigobench: CPU - Supercarindigobench: CPU - Bedroomonednn: IP Shapes 1D - f32 - CPUncnn: CPU - mobilenetrav1e: 10srsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUgravitymark: 1920 x 1080 - OpenGLgravitymark: 1920 x 1080 - Vulkanwebp2: Quality 95, Compression Effort 7srsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMetcpak: ETC1 + Ditheringtesseract: 1920 x 1080ncnn: CPU - vgg16gmic: 3D Elevated Function In Rand Colors, 100 Timesembree: Pathtracer ISPC - Crownsrsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMdraco: Church Facadephpbench: PHP Benchmark Suiteembree: Pathtracer - Asian Dragon Objdraco: Lionrawtherapee: Total Benchmark Timeblender: Barbershop - CPU-Onlyrays1bench: Large Scenetoktx: UASTC 4 + Zstd Compression 19encode-wavpack: WAV To WavPackocrmypdf: Processing 60 Page PDF Documentgimp: unsharp-maskgimp: auto-levelsgimp: rotategimp: resizencnn: CPU - blazefaceocrmypdf: Processing 60 Page PDF Documentinkscape: SVG Files To PNGgimp: unsharp-maskgimp: auto-levelsgimp: rotategimp: resizedarktable: Server Room - CPU-onlydarktable: Server Rack - CPU-onlydarktable: Masskrug - CPU-onlydarktable: Boat - CPU-onlymontage: Mosaic of M17, K band, 1.5 deg x 1.5 degoidn: RTLightmap.hdr.4096x4096aom-av1: Speed 0 Two-Pass - Bosphorus 4Kospray: Magnetic Reconnection - Path Tracerospray: Magnetic Reconnection - SciVisospray: XFrog Forest - Path Tracerospray: NASA Streamlines - SciVisospray: XFrog Forest - SciVisospray: San Miguel - SciVislibgav1: Chimera 1080p 10-bitlibgav1: Summer Nature 1080plibgav1: Summer Nature 4Klibgav1: Chimera 1080pncnn: CPU - alexnetncnn: CPU - efficientnet-b0ncnn: CPU - mnasnetncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU-v2-v2 - mobilenet-v2tachyon: Total Timesvt-vp9: VMAF Optimized - Bosphorus 1080pgraphics-magick: Resizingluxcorerender: Rainbow Colors and Prism - CPUparaview: Wavelet Volume - 3840 x 2160paraview: Wavelet Volume - 3840 x 2160paraview: Wavelet Volume - 1920 x 1080paraview: Wavelet Volume - 1920 x 1080unvanquished: 1920 x 1080 - Highbetsy: ETC2 RGB - Highestbetsy: ETC1 - HighestPop_OS 20.04Pop_OS 21.0410849.457.45372.6110519.05072.15180.0569.6718.0023.575734091802.827.9905641161213.4739.3050.12917.00162.06932.79616.7120.9801094.86787.9096.68.032.35866540.4752163683.1219.197.63329.8163.100106.7126.81442.286928.347.90.670480289.864.94118998667670.1362.21120.2856.5819.55763.174896.77481591.21551.439.4364.138117.22426.03744260.131.46529.22519.48030.08753.4016.6202.96.0136365.211.49424538.218.2906128.6842955.32211.02554.32240.4974.083536.3291593.1430.94786812.982.99932.67511438.65916424920938335570627.143.1677.1756.42102.89145.94664221527.3841.4242.16446.52.725291.4710.440750834.8392.655955.7112.135027.588305247139141257.49747.65.41110250.9989824.90023616.29.4848.79801.1135.706469.441065226.032355.521466.055.28810039.613.0022928149.869341470.9247.42840054825.7920.41530.3172.875177.168113.319.931903.07.32319.1311.3511027260.546.0955.7730410.294103.930.7073756.2721356.833.6419.3135.3688282.4546938.835.47510.30432.262.28322.186.306115.3981253.8812426666724.84914217.1315.041.451804.1903749.46284.8801.790326.083398.1640.90319449.81411.76816.31874.03740.1640.5545.06389962621264.31318.769746.631790257631920.037.1357.19916.5372515.0411105617.1263.1741.080745.00328.2330.467746.097377.971.248.3847.39037.5569.8361892634350.831.241.351.35289480000096.7568.27595748.3128368.467.84639.69103.91132.813704.85477155.883.715.6244.1501182006666756.665164.123.540307.4343561.9507.3530.7743.04912053.552.49413.01448489.231345623.425.44375.43718.018.6224.015814.7975397.34453.843820.813106400000211.51373121.6686.1333223.754.694467929.32418.51579.7523049.8136.911726683.504967.59130.7184145.97998603333380.223.94611.224155.35710.24553.656.9622.58210.4601.2880131.093.850353.40.96351385.286.6214.155266.3277.953487.547356.7485.07649.8697144.9891466233038.5995560847.881121.24301.6362.46312.8436.7115.10026.45617.31614.01011.3367.2802.2500.0882.8875.49275.1950.670.22333.3345.457.1690.9113.5871.4343.94144.7050.99112.6916.5519.3215.0414.5715.8417.0700361.1476322.953496.394218.525672.666354.54294.37.5756.7165551.936.89348.5414813.70252.79131.9851.3513.2917.874432972330.221.868442301958.0831.4840.26513.96133.45827.04513.8217.696924.58666.2282.26.852.02073465.0545108472.0190.266.67288.7255.31693.9323.62390.307812.942.60.75259259.694.44107633594740.0328.69109.2251.4798.70836.468527.39013643.61687.536.3144.488108.23824.04440955.929.30427.22718.16328.08549.8786.1873.15.6297261.131.40223027.917.1950136.7762786.56310.40251.27838.4077.936866.0164088.5430.90246013.622.85931.15110936.99615723861640350204428.313.3017.4776.69107.04647.74690213226.3639.9443.71430.92.631301.6500.425879806.7379.980987.4172.066497.352296254600246250.18769.05.26310531.026174.77223015.89.7208.58820.5315.844480.542021231.232409.667476.754.05810266.012.7223431146.662234480.8242.34340884924.9420541.1176.390173.666111.119.545920.87.46719.5071.3251047400.536.2056.77505210.477102.130.1763691.7320996.134.2039.4675.4558416.1247693.915.56210.46432.762.24921.856.401113.6891272.4212246666724.49814417.2315.251.472054.2473699.43281.2011.767330.313442.2210.89166749.18911.62116.52173.13694.7340.0644.52188933821278.93218.556738.179799167638930.137.5353.47406.4704514.8979106622.8265.5841.453738.47028.4780.463739.796381.171.847.9837.33137.8574.3551878141353.531.011.361.36291423333397.3868.70796348.0128545.477.89239.46104.51139.093724.95447856.184.165.5944.3854182976666756.965506.863.521827.4733579.7504.8330.6242.83952043.702.50613.07446487.093245823.525.54675.73732.718.5504.031014.8535416.69754.033833.513096233333212.20648721.5986.1527223.054.863368134.54417.31584.0123109.2436.820826748.894979.50131.0090245.87996436667379.423.89711.247155.05721.23653.757.0622.54310.4781.2901631.143.844353.90.96473385.386.5213.909266.6277.655488.068856.8085.14449.9089145.0890866273938.5768560547.905121.19301.5262.44712.84512.72218.52914.16011.2178.1686.710.670.22333.3345.457.1690.9113.5871.4314.6619.4214.6814.7615.8118.1192331.4081824.133565.118222.825717.318357.33248.57.3076.463OpenBenchmarking.org

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: In-Memory Database ShootoutPop_OS 20.04Pop_OS 21.042K4K6K8K10KSE +/- 89.60, N = 12SE +/- 35.71, N = 310849.45551.9MIN: 10225.86 / MAX: 12906.34MIN: 5391.37 / MAX: 6291.05

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 CompilePop_OS 20.04Pop_OS 21.041326395265SE +/- 0.13, N = 3SE +/- 0.08, N = 357.4536.891. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pPop_OS 20.04Pop_OS 21.041632486480SE +/- 0.23, N = 3SE +/- 0.08, N = 372.6148.541. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: nbodyPop_OS 20.04Pop_OS 21.04306090120150SE +/- 1.15, N = 3105148

G'MIC

G'MIC is an open-source framework for image processing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterG'MICTest: Plotting Isosurface Of A 3D Volume, 1000 TimesPop_OS 20.04Pop_OS 21.04510152025SE +/- 0.04, N = 3SE +/- 0.02, N = 319.0513.701. Version 2.4.5, Copyright (c) 2008-2019, David Tschumperle.

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: MediumPop_OS 20.04Pop_OS 21.041632486480SE +/- 0.12, N = 3SE +/- 0.13, N = 372.1552.791. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Very FastPop_OS 20.04Pop_OS 21.044080120160200SE +/- 0.39, N = 3SE +/- 0.27, N = 3180.05131.981. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: SlowPop_OS 20.04Pop_OS 21.041632486480SE +/- 0.31, N = 3SE +/- 0.21, N = 369.6751.351. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KPop_OS 20.04Pop_OS 21.0448121620SE +/- 0.01, N = 3SE +/- 0.04, N = 318.0013.291. (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 1080pPop_OS 20.04Pop_OS 21.04612182430SE +/- 0.03, N = 3SE +/- 0.07, N = 323.5717.871. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.22.1Test: Sequential FillPop_OS 20.04Pop_OS 21.04120K240K360K480K600KSE +/- 786.81, N = 3SE +/- 1145.38, N = 35734094432971. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Genetic Algorithm Using Jenetics + FuturesPop_OS 20.04Pop_OS 21.045001000150020002500SE +/- 10.87, N = 3SE +/- 31.23, N = 31802.82330.2MIN: 1704.06 / MAX: 1903.23MIN: 2144.91 / MAX: 2468.1

librsvg

RSVG/librsvg is an SVG vector graphics library. This test profile times how long it takes to complete various operations by rsvg-convert. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterlibrsvgOperation: SVG Files To PNGPop_OS 20.04Pop_OS 21.04714212835SE +/- 0.12, N = 3SE +/- 0.03, N = 327.9921.871. Pop_OS 20.04: rsvg-convert version 2.48.92. Pop_OS 21.04: rsvg-convert version 2.50.3

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.22.1Test: Random FillPop_OS 20.04Pop_OS 21.04120K240K360K480K600KSE +/- 4881.43, N = 3SE +/- 611.32, N = 35641164423011. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -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 0.9.0Video Input: Summer Nature 1080pPop_OS 20.04Pop_OS 21.0430060090012001500SE +/- 7.62, N = 3SE +/- 1.93, N = 31213.47958.08MIN: 547.29 / MAX: 1355.01MIN: 444.78 / MAX: 1062.981. (CC) gcc options: -pthread -lm

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Very FastPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.31, N = 3SE +/- 0.07, N = 339.3031.481. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Timed GDB GNU Debugger Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 10.2Time To CompilePop_OS 20.04Pop_OS 21.041122334455SE +/- 0.05, N = 3SE +/- 0.01, N = 350.1340.27

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: MediumPop_OS 20.04Pop_OS 21.0448121620SE +/- 0.05, N = 3SE +/- 0.02, N = 317.0013.961. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 1080pPop_OS 20.04Pop_OS 21.044080120160200SE +/- 0.44, N = 3SE +/- 0.90, N = 3162.07133.461. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

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 1.2Model: resnet-v2-50Pop_OS 20.04Pop_OS 21.04816243240SE +/- 0.10, N = 3SE +/- 0.02, N = 332.8027.05MIN: 29.14 / MAX: 35.63MIN: 26.03 / MAX: 30.951. (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

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: SlowPop_OS 20.04Pop_OS 21.0448121620SE +/- 0.04, N = 3SE +/- 0.01, N = 316.7113.821. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Timed FFmpeg Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.4Time To CompilePop_OS 20.04Pop_OS 21.04510152025SE +/- 0.12, N = 3SE +/- 0.03, N = 320.9817.70

dav1d

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

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Chimera 1080pPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 1.64, N = 3SE +/- 1.18, N = 31094.86924.58MIN: 653.74 / MAX: 1402.09MIN: 539.12 / MAX: 1186.831. (CC) gcc options: -pthread -lm

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.0Video Input: Chimera 1080p 10-bitPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 2.61, N = 3SE +/- 1.58, N = 3787.90666.22MIN: 550.77 / MAX: 1026.42MIN: 470.44 / MAX: 868.581. (CC) gcc options: -pthread -lm

GravityMark

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.2Resolution: 1920 x 1080 - Renderer: OpenGL ESPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.36, N = 3SE +/- 0.35, N = 396.682.2

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: python_startupPop_OS 20.04Pop_OS 21.04246810SE +/- 0.01, N = 3SE +/- 0.00, N = 38.036.85

Tungsten Renderer

Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Non-ExponentialPop_OS 20.04Pop_OS 21.040.53071.06141.59212.12282.6535SE +/- 0.01320, N = 3SE +/- 0.01588, N = 32.358662.020731. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl

SVT-HEVC

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pPop_OS 20.04Pop_OS 21.04120240360480600SE +/- 5.01, N = 3SE +/- 3.50, N = 3540.47465.051. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.22.1Test: Update RandomPop_OS 20.04Pop_OS 21.04110K220K330K440K550KSE +/- 1240.41, N = 3SE +/- 696.39, N = 35216364510841. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

GravityMark

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.2Resolution: 2560 x 1440 - Renderer: OpenGL ESPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.25, N = 3SE +/- 0.10, N = 383.172.0

x264

This is a simple test of the x264 encoder run on the CPU (OpenCL support disabled) with a sample video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2019-12-17H.264 Video EncodingPop_OS 20.04Pop_OS 21.0450100150200250SE +/- 2.70, N = 15SE +/- 2.61, N = 12219.19190.261. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize

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 1080pPop_OS 20.04Pop_OS 21.04246810SE +/- 0.03, N = 3SE +/- 0.09, N = 37.636.671. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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 1080pPop_OS 20.04Pop_OS 21.0470140210280350SE +/- 3.84, N = 4SE +/- 1.99, N = 3329.81288.721. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Wavelet BlurPop_OS 20.04Pop_OS 21.041428425670SE +/- 0.02, N = 3SE +/- 0.11, N = 363.1055.32

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 1080pPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.36, N = 3SE +/- 0.32, N = 3106.7193.931. (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 1080pPop_OS 20.04Pop_OS 21.04612182430SE +/- 0.13, N = 3SE +/- 0.14, N = 326.8123.621. (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 0.9.0Video Input: Summer Nature 4KPop_OS 20.04Pop_OS 21.04100200300400500SE +/- 0.76, N = 3SE +/- 0.88, N = 3442.28390.30MIN: 209.45 / MAX: 470.99MIN: 197.57 / MAX: 415.761. (CC) gcc options: -pthread -lm

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: ALS Movie LensPop_OS 20.04Pop_OS 21.042K4K6K8K10KSE +/- 58.09, N = 3SE +/- 53.91, N = 36928.37812.9MIN: 6802.77 / MAX: 7617.87MIN: 7538.23 / MAX: 8536.4

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: django_templatePop_OS 20.04Pop_OS 21.041122334455SE +/- 0.15, N = 3SE +/- 0.09, N = 347.942.6

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUPop_OS 20.04Pop_OS 21.040.16930.33860.50790.67720.8465SE +/- 0.002087, N = 3SE +/- 0.001697, N = 30.6704800.752590MIN: 0.64MIN: 0.71. (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 1080pPop_OS 20.04Pop_OS 21.0460120180240300SE +/- 3.92, N = 3SE +/- 2.64, N = 3289.86259.691. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -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 4 Two-Pass - Input: Bosphorus 4KPop_OS 20.04Pop_OS 21.041.11152.2233.33454.4465.5575SE +/- 0.03, N = 3SE +/- 0.05, N = 34.944.441. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkStructuredVolumePop_OS 20.04Pop_OS 21.0430M60M90M120M150MSE +/- 999512.98, N = 8SE +/- 999801.47, N = 3118998667107633594MIN: 961600 / MAX: 1332000648MIN: 977646 / MAX: 969548832

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 FilterPop_OS 20.04Pop_OS 21.04160320480640800SE +/- 1.33, N = 4SE +/- 0.79, N = 3670.1740.01. Pop_OS 20.04: 3.8.1.02. Pop_OS 21.04: 3.8.2.0

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 1080pPop_OS 20.04Pop_OS 21.0480160240320400SE +/- 3.78, N = 3SE +/- 0.27, N = 3362.21328.691. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -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 1080pPop_OS 20.04Pop_OS 21.04306090120150SE +/- 0.20, N = 3SE +/- 0.63, N = 3120.28109.221. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Color EnhancePop_OS 20.04Pop_OS 21.041326395265SE +/- 0.06, N = 3SE +/- 0.05, N = 356.5851.48

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 4KPop_OS 20.04Pop_OS 21.043691215SE +/- 0.06, N = 3SE +/- 0.06, N = 39.558.701. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Random ForestPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 3.92, N = 3SE +/- 10.70, N = 3763.1836.4MIN: 683.84 / MAX: 933.51MIN: 733.68 / MAX: 1005.37

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, StatefulPop_OS 20.04Pop_OS 21.0416003200480064008000SE +/- 85.98, N = 3SE +/- 13.58, N = 374896852-lopus1. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

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: CPUPop_OS 20.04Pop_OS 21.04246810SE +/- 0.03891, N = 3SE +/- 0.03891, N = 36.774817.39013MIN: 5.83MIN: 6.511. (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: Complex PhasePop_OS 20.04Pop_OS 21.04140280420560700SE +/- 10.19, N = 3SE +/- 0.67, N = 3591.2643.6

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Apache Spark ALSPop_OS 20.04Pop_OS 21.04400800120016002000SE +/- 21.52, N = 3SE +/- 10.82, N = 151551.41687.5MIN: 1442.6 / MAX: 1700.48MIN: 1455.08 / MAX: 2039.34

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: AntialiasPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.04, N = 3SE +/- 0.01, N = 339.4436.31

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 1.2Model: mobilenet-v1-1.0Pop_OS 20.04Pop_OS 21.041.00982.01963.02944.03925.049SE +/- 0.022, N = 3SE +/- 0.087, N = 34.1384.488MIN: 3.96 / MAX: 4.55MIN: 4.02 / MAX: 4.891. (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

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: PistolPop_OS 20.04Pop_OS 21.04306090120150SE +/- 0.04, N = 3SE +/- 0.31, N = 3117.22108.241. Pop_OS 20.04: OpenSCAD version 2019.052. Pop_OS 21.04: OpenSCAD version 2021.01

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 CompilePop_OS 20.04Pop_OS 21.04612182430SE +/- 0.20, N = 10SE +/- 0.24, N = 626.0424.04

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pickle_pure_pythonPop_OS 20.04Pop_OS 21.04100200300400500442409

GravityMark

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.2Resolution: 3840 x 2160 - Renderer: OpenGL ESPop_OS 20.04Pop_OS 21.041326395265SE +/- 0.17, N = 3SE +/- 0.71, N = 360.155.9

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Tile GlassPop_OS 20.04Pop_OS 21.04714212835SE +/- 0.06, N = 3SE +/- 0.00, N = 331.4729.30

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, LosslessPop_OS 20.04Pop_OS 21.04714212835SE +/- 0.07, N = 3SE +/- 0.12, N = 329.2327.231. (CXX) g++ options: -O3 -fPIC -lm

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 CompilePop_OS 20.04Pop_OS 21.04510152025SE +/- 0.05, N = 3SE +/- 0.06, N = 319.4818.16

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: ReflectPop_OS 20.04Pop_OS 21.04714212835SE +/- 0.05, N = 3SE +/- 0.04, N = 330.0928.09

Timed Godot Game Engine Compilation

This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3Time To CompilePop_OS 20.04Pop_OS 21.041224364860SE +/- 0.41, N = 15SE +/- 0.02, N = 353.4049.88

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Compression Effort 5Pop_OS 20.04Pop_OS 21.04246810SE +/- 0.003, N = 3SE +/- 0.009, N = 36.6206.187-lOpenGL -lGLX -lGLU -lglut -lXmu -lXi-lgif1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux

Natron

Natron is an open-source, cross-platform compositing software for visual effects (VFX) and motion graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterNatron 2.4Input: SpaceshipPop_OS 20.04Pop_OS 21.040.69751.3952.09252.793.4875SE +/- 0.03, N = 3SE +/- 0.02, N = 152.93.1

Tungsten Renderer

Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: HairPop_OS 20.04Pop_OS 21.04246810SE +/- 0.03885, N = 3SE +/- 0.03388, N = 36.013635.629721. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Ultra FastPop_OS 20.04Pop_OS 21.041530456075SE +/- 0.56, N = 8SE +/- 0.06, N = 365.2161.131. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

WebP Image Encode

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

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: DefaultPop_OS 20.04Pop_OS 21.040.33620.67241.00861.34481.681SE +/- 0.005, N = 3SE +/- 0.005, N = 31.4941.4021. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

C-Blosc

A simple, compressed, fast and persistent data store library for C. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterC-Blosc 2.0Compressor: blosclzPop_OS 20.04Pop_OS 21.045K10K15K20K25KSE +/- 268.39, N = 3SE +/- 294.39, N = 324538.223027.91. (CC) gcc options: -std=gnu99 -O3 -pthread -lrt -lm

Tungsten Renderer

Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Water CausticPop_OS 20.04Pop_OS 21.04510152025SE +/- 0.00, N = 3SE +/- 0.02, N = 318.2917.201. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl

Ngspice

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

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670Pop_OS 20.04Pop_OS 21.04306090120150SE +/- 0.92, N = 3SE +/- 1.30, N = 12128.68136.781. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: DenseNetPop_OS 20.04Pop_OS 21.046001200180024003000SE +/- 3.63, N = 3SE +/- 14.98, N = 32955.322786.56MIN: 2904.89 / MAX: 3027.73MIN: 2539.42 / MAX: 3065.521. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 1080pPop_OS 20.04Pop_OS 21.043691215SE +/- 0.02, N = 3SE +/- 0.06, N = 311.0310.401. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4KPop_OS 20.04Pop_OS 21.041224364860SE +/- 0.45, N = 8SE +/- 0.21, N = 354.3251.281. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

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 4KPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.07, N = 3SE +/- 0.14, N = 340.4938.401. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

DeepSpeech

Mozilla DeepSpeech is a speech-to-text engine powered by TensorFlow for machine learning and derived from Baidu's Deep Speech research paper. 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 BetterDeepSpeech 0.6Acceleration: CPUPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.26, N = 3SE +/- 0.50, N = 374.0877.94

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: IP Shapes 3D - Data Type: f32 - Engine: CPUPop_OS 20.04Pop_OS 21.04246810SE +/- 0.02719, N = 3SE +/- 0.00247, N = 36.329156.01640MIN: 6.07MIN: 5.871. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: CartoonPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.30, N = 3SE +/- 0.06, N = 393.1488.54

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: CPUPop_OS 20.04Pop_OS 21.040.21330.42660.63990.85321.0665SE +/- 0.010129, N = 15SE +/- 0.008147, N = 150.9478680.902460MIN: 0.83MIN: 0.811. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

VP9 libvpx Encoding

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

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 0 - Input: Bosphorus 1080pPop_OS 20.04Pop_OS 21.0448121620SE +/- 0.02, N = 3SE +/- 0.03, N = 312.9813.621. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

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 1.2Model: mobilenetV3Pop_OS 20.04Pop_OS 21.040.67481.34962.02442.69923.374SE +/- 0.011, N = 3SE +/- 0.030, N = 32.9992.859MIN: 2.94 / MAX: 3.07MIN: 2.76 / MAX: 2.941. (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

AOBench

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

OpenBenchmarking.orgSeconds, Fewer Is BetterAOBenchSize: 2048 x 2048 - Total TimePop_OS 20.04Pop_OS 21.04816243240SE +/- 0.03, N = 3SE +/- 0.02, N = 332.6831.151. (CC) gcc options: -lm -O3

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: floatPop_OS 20.04Pop_OS 21.04306090120150SE +/- 1.15, N = 3114109

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Rotate 90 DegreesPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.05, N = 3SE +/- 0.01, N = 338.6637.00

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: regex_compilePop_OS 20.04Pop_OS 21.044080120160200SE +/- 0.33, N = 3SE +/- 0.33, N = 3164157

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkVdbVolumePop_OS 20.04Pop_OS 21.045M10M15M20M25MSE +/- 126223.63, N = 3SE +/- 144705.32, N = 32492093823861640MIN: 832564 / MAX: 140531328MIN: 890158 / MAX: 139227048

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.22.1Test: Read Random Write RandomPop_OS 20.04Pop_OS 21.04800K1600K2400K3200K4000KSE +/- 5042.25, N = 3SE +/- 34734.40, N = 3335570635020441. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

VP9 libvpx Encoding

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

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 1080pPop_OS 20.04Pop_OS 21.04714212835SE +/- 0.30, N = 4SE +/- 0.10, N = 327.1428.311. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

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: 10Pop_OS 20.04Pop_OS 21.040.74271.48542.22812.97083.7135SE +/- 0.007, N = 3SE +/- 0.024, N = 33.1673.3011. (CXX) g++ options: -O3 -fPIC -lm

WebP Image Encode

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

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest CompressionPop_OS 20.04Pop_OS 21.04246810SE +/- 0.029, N = 3SE +/- 0.044, N = 37.1757.4771. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Danish Mood - Acceleration: CPUPop_OS 20.04Pop_OS 21.04246810SE +/- 0.03, N = 3SE +/- 0.09, N = 36.426.69MIN: 2.57 / MAX: 7.51MIN: 2.8 / MAX: 7.77

Ngspice

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

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552Pop_OS 20.04Pop_OS 21.0420406080100SE +/- 1.11, N = 3SE +/- 1.27, N = 15102.89107.051. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: regnety_400mPop_OS 20.04Pop_OS 21.041122334455SE +/- 0.60, N = 15SE +/- 0.24, N = 645.9447.74MIN: 41.77 / MAX: 141.29MIN: 45.17 / MAX: 178.011. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

GraphicsMagick

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

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SharpenPop_OS 20.04Pop_OS 21.04150300450600750SE +/- 2.08, N = 3SE +/- 1.86, N = 36646901. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SwirlPop_OS 20.04Pop_OS 21.045001000150020002500SE +/- 6.81, N = 3SE +/- 5.24, N = 3221521321. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KPop_OS 20.04Pop_OS 21.04612182430SE +/- 0.14, N = 3SE +/- 0.13, N = 327.3826.361. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

LibRaw

LibRaw is a RAW image decoder for digital camera photos. This test profile runs LibRaw's post-processing benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing BenchmarkPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.10, N = 3SE +/- 0.09, N = 341.4239.941. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: resnet50Pop_OS 20.04Pop_OS 21.041020304050SE +/- 0.27, N = 15SE +/- 0.31, N = 642.1643.71MIN: 38.53 / MAX: 142.69MIN: 40.1 / MAX: 56.621. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

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 FiltersPop_OS 20.04Pop_OS 21.04100200300400500SE +/- 5.06, N = 4SE +/- 2.64, N = 3446.5430.91. Pop_OS 20.04: 3.8.1.02. Pop_OS 21.04: 3.8.2.0

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: DefaultPop_OS 20.04Pop_OS 21.040.61311.22621.83932.45243.0655SE +/- 0.022, N = 3SE +/- 0.023, N = 152.7252.631-lOpenGL -lGLX -lGLU -lglut -lXmu -lXi-lgif1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC1Pop_OS 20.04Pop_OS 21.0470140210280350SE +/- 1.26, N = 3SE +/- 0.26, N = 3291.47301.651. (CXX) g++ options: -O3 -march=native -std=c++11 -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: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUPop_OS 20.04Pop_OS 21.040.09920.19840.29760.39680.496SE +/- 0.003002, N = 3SE +/- 0.004692, N = 50.4407500.425879MIN: 0.4MIN: 0.41. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Scala DottyPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 12.42, N = 15SE +/- 12.15, N = 15834.8806.7MIN: 669.74 / MAX: 1135.69MIN: 668.92 / MAX: 1155.3

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Lossless CompressionPop_OS 20.04Pop_OS 21.0490180270360450SE +/- 0.30, N = 3SE +/- 0.86, N = 3392.66379.98-lOpenGL -lGLX -lGLU -lglut -lXmu -lXi-lgif1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux

TTSIOD 3D Renderer

A portable GPL 3D software renderer that supports OpenMP and Intel Threading Building Blocks with many different rendering modes. This version does not use OpenGL but is entirely CPU/software based. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterTTSIOD 3D Renderer 2.3bPhong Rendering With Soft-Shadow MappingPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 9.02, N = 6SE +/- 10.89, N = 15955.71987.421. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++

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: CPUPop_OS 20.04Pop_OS 21.040.48040.96081.44121.92162.402SE +/- 0.00603, N = 3SE +/- 0.00490, N = 32.135022.06649MIN: 2.04MIN: 2.021. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: CropPop_OS 20.04Pop_OS 21.04246810SE +/- 0.028, N = 3SE +/- 0.017, N = 37.5887.352

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: 2to3Pop_OS 20.04Pop_OS 21.0470140210280350SE +/- 0.33, N = 3305296

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.22.1Test: Random ReadPop_OS 20.04Pop_OS 21.0450M100M150M200M250MSE +/- 2251596.95, N = 3SE +/- 3547928.94, N = 32471391412546002461. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Ultra FastPop_OS 20.04Pop_OS 21.0460120180240300SE +/- 1.49, N = 3SE +/- 0.47, N = 3257.49250.181. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 FiltersPop_OS 20.04Pop_OS 21.04170340510680850SE +/- 8.38, N = 3SE +/- 1.01, N = 3747.6769.0

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 1.2Model: squeezenetv1.1Pop_OS 20.04Pop_OS 21.041.21752.4353.65254.876.0875SE +/- 0.039, N = 3SE +/- 0.066, N = 35.4115.263MIN: 5.21 / MAX: 5.98MIN: 4.86 / MAX: 6.311. (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

GraphicsMagick

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

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: EnhancedPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 1.20, N = 3SE +/- 1.67, N = 3102510531. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -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: u8s8f32 - Engine: CPUPop_OS 20.04Pop_OS 21.040.23090.46180.69270.92361.1545SE +/- 0.003236, N = 3SE +/- 0.001940, N = 30.9989821.026170MIN: 0.94MIN: 0.961. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 4KPop_OS 20.04Pop_OS 21.041.10252.2053.30754.415.5125SE +/- 0.017, N = 3SE +/- 0.004, N = 34.9004.7721. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: goPop_OS 20.04Pop_OS 21.0450100150200250SE +/- 0.33, N = 3236230

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pathlibPop_OS 20.04Pop_OS 21.0448121620SE +/- 0.03, N = 3SE +/- 0.03, N = 316.215.8

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: 6Pop_OS 20.04Pop_OS 21.043691215SE +/- 0.020, N = 3SE +/- 0.026, N = 39.4849.7201. (CXX) g++ options: -O3 -fPIC -lm

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: DLSC - Acceleration: CPUPop_OS 20.04Pop_OS 21.04246810SE +/- 0.05, N = 3SE +/- 0.04, N = 38.798.58MIN: 8.51 / MAX: 9.61MIN: 8.31 / MAX: 9.57

Google SynthMark

SynthMark is a cross platform tool for benchmarking CPU performance under a variety of real-time audio workloads. It uses a polyphonic synthesizer model to provide standardized tests for latency, jitter and computational throughput. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100Pop_OS 20.04Pop_OS 21.042004006008001000SE +/- 1.06, N = 3SE +/- 0.35, N = 3801.11820.531. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

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, LosslessPop_OS 20.04Pop_OS 21.041.31492.62983.94475.25966.5745SE +/- 0.018, N = 3SE +/- 0.036, N = 35.7065.8441. (CXX) g++ options: -O3 -fPIC -lm

ET: Legacy

ETLegacy is an open-source engine evolution of Wolfenstein: Enemy Territory, a World War II era first person shooter that was released for free by Splash Damage using the id Tech 3 engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.77Resolution: 2560 x 1440Pop_OS 20.04Pop_OS 21.04100200300400500SE +/- 0.38, N = 3SE +/- 0.55, N = 3469.4480.5

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, StatelessPop_OS 20.04Pop_OS 21.049K18K27K36K45KSE +/- 27.00, N = 3SE +/- 189.23, N = 34106542021-lopus1. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 3840 x 2160Pop_OS 20.04Pop_OS 21.0450100150200250SE +/- 0.12, N = 3SE +/- 0.08, N = 3226.03231.23

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 3840 x 2160Pop_OS 20.04Pop_OS 21.045001000150020002500SE +/- 1.28, N = 3SE +/- 0.86, N = 32355.522409.67

ET: Legacy

ETLegacy is an open-source engine evolution of Wolfenstein: Enemy Territory, a World War II era first person shooter that was released for free by Splash Damage using the id Tech 3 engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.77Resolution: 3840 x 2160Pop_OS 20.04Pop_OS 21.04100200300400500SE +/- 1.01, N = 3SE +/- 1.44, N = 3466.0476.7

YafaRay

YafaRay is an open-source physically based montecarlo ray-tracing engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterYafaRay 3.4.1Total Time For Sample ScenePop_OS 20.04Pop_OS 21.041224364860SE +/- 0.55, N = 6SE +/- 0.29, N = 355.2954.061. (CXX) g++ options: -std=c++11 -O3 -ffast-math -rdynamic -ldl -lImath -lIlmImf -lIex -lHalf -lz -lIlmThread -lxml2 -lfreetype -lpthread

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Savina Reactors.IOPop_OS 20.04Pop_OS 21.042K4K6K8K10KSE +/- 148.02, N = 12SE +/- 70.36, N = 310039.610266.0MIN: 9225.64 / MAX: 21675.88MIN: 10181.21 / MAX: 15583.21

VP9 libvpx Encoding

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

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4KPop_OS 20.04Pop_OS 21.043691215SE +/- 0.11, N = 3SE +/- 0.01, N = 313.0012.721. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.22.1Test: Random Fill SyncPop_OS 20.04Pop_OS 21.045K10K15K20K25KSE +/- 254.81, N = 4SE +/- 44.11, N = 322928234311. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Material TesterPop_OS 20.04Pop_OS 21.04306090120150149.87146.66

ET: Legacy

ETLegacy is an open-source engine evolution of Wolfenstein: Enemy Territory, a World War II era first person shooter that was released for free by Splash Damage using the id Tech 3 engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.77Resolution: 1920 x 1080Pop_OS 20.04Pop_OS 21.04100200300400500SE +/- 3.98, N = 3SE +/- 3.80, N = 3470.9480.8

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.1Pop_OS 20.04Pop_OS 21.0450100150200250SE +/- 1.32, N = 3SE +/- 0.23, N = 3247.43242.34MIN: 245.1 / MAX: 291.42MIN: 241.29 / MAX: 244.831. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

PJSIP

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

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: INVITEPop_OS 20.04Pop_OS 21.049001800270036004500SE +/- 7.57, N = 3SE +/- 3.38, N = 340054088-lopus1. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CPop_OS 20.04Pop_OS 21.0411002200330044005500SE +/- 5.97, N = 3SE +/- 52.99, N = 34825.794924.94-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerPop_OS 20.04Pop_OS 21.04510152025SE +/- 0.00, N = 320.4120.00MIN: 19.61 / MAX: 20.83MIN: 19.61 / MAX: 20.83

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 TransformPop_OS 20.04Pop_OS 21.04120240360480600SE +/- 1.54, N = 4SE +/- 2.31, N = 3530.3541.11. Pop_OS 20.04: 3.8.1.02. Pop_OS 21.04: 3.8.2.0

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC2Pop_OS 20.04Pop_OS 21.044080120160200SE +/- 0.05, N = 3SE +/- 0.24, N = 3172.88176.391. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

Radiance Benchmark

This is a benchmark of NREL Radiance, a synthetic imaging system that is open-source and developed by the Lawrence Berkeley National Laboratory in California. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRadiance Benchmark 5.0Test: SMP ParallelPop_OS 20.04Pop_OS 21.044080120160200177.17173.67

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.04Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAMPop_OS 20.04Pop_OS 21.04306090120150SE +/- 0.06, N = 3SE +/- 0.09, N = 3113.3111.11. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

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 SlimPop_OS 20.04Pop_OS 21.04510152025SE +/- 0.09, N = 3SE +/- 0.11, N = 319.9319.551. Pop_OS 20.04: OpenSCAD version 2019.052. Pop_OS 21.04: OpenSCAD version 2021.01

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Apache Spark BayesPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 1.62, N = 3SE +/- 9.59, N = 4903.0920.8MIN: 652.77 / MAX: 906.2MIN: 659.31 / MAX: 1117.6

LAME MP3 Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3Pop_OS 20.04Pop_OS 21.04246810SE +/- 0.024, N = 3SE +/- 0.016, N = 37.3237.4671. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lm

Ogg Audio Encoding

This test times how long it takes to encode a sample WAV file to Ogg format using the reference Xiph.org tools/libraries. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOgg Audio Encoding 1.3.4WAV To OggPop_OS 20.04Pop_OS 21.04510152025SE +/- 0.04, N = 3SE +/- 0.02, N = 319.1319.511. (CC) gcc options: -O2 -ffast-math -fsigned-char

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 5Pop_OS 20.04Pop_OS 21.040.3040.6080.9121.2161.52SE +/- 0.001, N = 3SE +/- 0.002, N = 31.3511.325

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: crypto_pyaesPop_OS 20.04Pop_OS 21.0420406080100102104

GraphicsMagick

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

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Noise-GaussianPop_OS 20.04Pop_OS 21.04160320480640800SE +/- 1.53, N = 37267401. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -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 0 Two-Pass - Input: Bosphorus 1080pPop_OS 20.04Pop_OS 21.040.12150.2430.36450.4860.6075SE +/- 0.00, N = 3SE +/- 0.00, N = 30.540.531. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

VP9 libvpx Encoding

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

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 0 - Input: Bosphorus 4KPop_OS 20.04Pop_OS 21.04246810SE +/- 0.02, N = 3SE +/- 0.03, N = 36.096.201. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Disney MaterialPop_OS 20.04Pop_OS 21.04132639526555.7756.78

C-Ray

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

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per PixelPop_OS 20.04Pop_OS 21.043691215SE +/- 0.02, N = 3SE +/- 0.04, N = 310.2910.481. (CC) gcc options: -lm -lpthread -O3

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.04Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.17, N = 3SE +/- 0.30, N = 3103.9102.11. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

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 1.2Model: inception-v3Pop_OS 20.04Pop_OS 21.04714212835SE +/- 0.12, N = 3SE +/- 0.41, N = 330.7130.18MIN: 29.89 / MAX: 32.31MIN: 29.1 / MAX: 31.181. (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: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUPop_OS 20.04Pop_OS 21.048001600240032004000SE +/- 1.36, N = 3SE +/- 1.23, N = 33756.273691.73MIN: 3722.83MIN: 3676.011. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Akka Unbalanced Cobwebbed TreePop_OS 20.04Pop_OS 21.045K10K15K20K25KSE +/- 106.43, N = 3SE +/- 122.55, N = 321356.820996.1MIN: 16918.56 / MAX: 21569.17MIN: 16904.54 / MAX: 21229.63

WebP Image Encode

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

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest CompressionPop_OS 20.04Pop_OS 21.04816243240SE +/- 0.03, N = 3SE +/- 0.07, N = 333.6434.201. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

POV-Ray

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

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace TimePop_OS 20.04Pop_OS 21.043691215SE +/- 0.009, N = 3SE +/- 0.014, N = 39.3139.467-lImath -lHalf -lIex -lIexMath -lpthread-lIlmImf-2_5 -lImath-2_5 -lHalf-2_5 -lIex-2_5 -lIexMath-2_5 -lIlmThread-2_51. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSDL -lXpm -lSM -lICE -lX11 -lIlmImf -lIlmThread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

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 1.2Model: MobileNetV2_224Pop_OS 20.04Pop_OS 21.041.22742.45483.68224.90966.137SE +/- 0.061, N = 3SE +/- 0.027, N = 35.3685.455MIN: 5.13 / MAX: 5.68MIN: 5.21 / MAX: 6.211. (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

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CPop_OS 20.04Pop_OS 21.042K4K6K8K10KSE +/- 7.63, N = 3SE +/- 32.24, N = 38282.458416.12-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.BPop_OS 20.04Pop_OS 21.0410K20K30K40K50KSE +/- 486.42, N = 3SE +/- 92.18, N = 346938.8347693.91-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: ScalePop_OS 20.04Pop_OS 21.041.25152.5033.75455.0066.2575SE +/- 0.007, N = 3SE +/- 0.020, N = 35.4755.562

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 2Pop_OS 20.04Pop_OS 21.043691215SE +/- 0.01, N = 3SE +/- 0.04, N = 310.3010.461. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -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-OnlyPop_OS 20.04Pop_OS 21.04816243240SE +/- 0.25, N = 3SE +/- 0.20, N = 332.2632.76

WebP Image Encode

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

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100Pop_OS 20.04Pop_OS 21.040.51371.02741.54112.05482.5685SE +/- 0.002, N = 3SE +/- 0.009, N = 32.2832.2491. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: resnet18Pop_OS 20.04Pop_OS 21.04510152025SE +/- 0.21, N = 15SE +/- 0.11, N = 622.1821.85MIN: 19.89 / MAX: 118.25MIN: 20.95 / MAX: 23.431. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

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 0Pop_OS 20.04Pop_OS 21.04246810SE +/- 0.006, N = 3SE +/- 0.016, N = 36.3066.4011. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 75, Compression Effort 7Pop_OS 20.04Pop_OS 21.04306090120150SE +/- 0.21, N = 3SE +/- 0.23, N = 3115.40113.69-lOpenGL -lGLX -lGLU -lglut -lXmu -lXi-lgif1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux

LuaJIT

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

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Sparse Matrix MultiplyPop_OS 20.04Pop_OS 21.0430060090012001500SE +/- 1.65, N = 3SE +/- 1.94, N = 31253.881272.421. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSamples / Second, More Is BettersrsRAN 21.04Test: OFDM_TestPop_OS 20.04Pop_OS 21.0430M60M90M120M150MSE +/- 233333.33, N = 3SE +/- 392994.20, N = 31242666671224666671. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

Tesseract OCR

Tesseract-OCR is the open-source optical character recognition (OCR) engine for the conversion of text within images to raw text output. This test profile relies upon a system-supplied Tesseract installation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTesseract OCR 4.1.1Time To OCR 7 ImagesPop_OS 20.04Pop_OS 21.04612182430SE +/- 0.06, N = 3SE +/- 0.06, N = 324.8524.50

GraphicsMagick

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

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: HWB Color SpacePop_OS 20.04Pop_OS 21.0430060090012001500SE +/- 2.33, N = 3SE +/- 7.17, N = 3142114411. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: LuxCore Benchmark - Acceleration: CPUPop_OS 20.04Pop_OS 21.04246810SE +/- 0.07, N = 3SE +/- 0.09, N = 37.137.23MIN: 2.73 / MAX: 8.37MIN: 2.86 / MAX: 8.57

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: shufflenet-v2Pop_OS 20.04Pop_OS 21.0448121620SE +/- 0.21, N = 15SE +/- 0.25, N = 615.0415.25MIN: 13.31 / MAX: 108.55MIN: 13.97 / MAX: 107.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -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: CPUPop_OS 20.04Pop_OS 21.040.33120.66240.99361.32481.656SE +/- 0.01216, N = 3SE +/- 0.01507, N = 31.451801.47205MIN: 1.25MIN: 1.291. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

KTX-Software toktx

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

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3Pop_OS 20.04Pop_OS 21.040.95561.91122.86683.82244.778SE +/- 0.003, N = 3SE +/- 0.010, N = 34.1904.247

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: CPUPop_OS 20.04Pop_OS 21.048001600240032004000SE +/- 11.90, N = 3SE +/- 9.59, N = 33749.463699.43MIN: 3698.5MIN: 3671.591. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v2Pop_OS 20.04Pop_OS 21.0460120180240300SE +/- 0.52, N = 3SE +/- 3.19, N = 4284.88281.20MIN: 274.2 / MAX: 321.88MIN: 262.99 / MAX: 332.871. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 6Pop_OS 20.04Pop_OS 21.040.40280.80561.20841.61122.014SE +/- 0.002, N = 3SE +/- 0.002, N = 31.7901.767

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 1920 x 1080Pop_OS 20.04Pop_OS 21.0470140210280350SE +/- 1.63, N = 3SE +/- 0.23, N = 3326.08330.31

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 1920 x 1080Pop_OS 20.04Pop_OS 21.047001400210028003500SE +/- 16.99, N = 3SE +/- 2.40, N = 33398.163442.22

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: CPUPop_OS 20.04Pop_OS 21.040.20320.40640.60960.81281.016SE +/- 0.000498, N = 3SE +/- 0.001655, N = 30.9031940.891667MIN: 0.84MIN: 0.831. (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 CasePop_OS 20.04Pop_OS 21.041122334455SE +/- 0.07, N = 3SE +/- 0.18, N = 349.8149.191. Pop_OS 20.04: OpenSCAD version 2019.052. Pop_OS 21.04: OpenSCAD version 2021.01

Monkey Audio Encoding

This test times how long it takes to encode a sample WAV file to Monkey's Audio APE format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMonkey Audio Encoding 3.99.6WAV To APEPop_OS 20.04Pop_OS 21.043691215SE +/- 0.02, N = 5SE +/- 0.04, N = 511.7711.621. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt

WebP Image Encode

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

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessPop_OS 20.04Pop_OS 21.0448121620SE +/- 0.03, N = 3SE +/- 0.07, N = 316.3216.521. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

GravityMark

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.2Resolution: 2560 x 1440 - Renderer: OpenGLPop_OS 20.04Pop_OS 21.041632486480SE +/- 0.47, N = 3SE +/- 0.12, N = 374.073.1

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 Training - Data Type: f32 - Engine: CPUPop_OS 20.04Pop_OS 21.048001600240032004000SE +/- 10.65, N = 3SE +/- 11.88, N = 33740.163694.73MIN: 3691.91MIN: 3660.931. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: yolov4-tinyPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.14, N = 15SE +/- 0.06, N = 640.5540.06MIN: 38.32 / MAX: 94MIN: 38.75 / MAX: 47.41. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Hugin

Hugin is an open-source, cross-platform panorama photo stitcher software package. This test profile times how long it takes to run the assistant and panorama photo stitching on a set of images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHuginPanorama Photo Assistant + Stitching TimePop_OS 20.04Pop_OS 21.041020304050SE +/- 0.25, N = 3SE +/- 0.32, N = 345.0644.52

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.22.1Test: Read While WritingPop_OS 20.04Pop_OS 21.042M4M6M8M10MSE +/- 79157.64, N = 8SE +/- 98426.00, N = 3899626288933821. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: DXT1Pop_OS 20.04Pop_OS 21.0430060090012001500SE +/- 3.86, N = 3SE +/- 0.72, N = 31264.311278.931. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

RNNoise

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

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 2020-06-28Pop_OS 20.04Pop_OS 21.04510152025SE +/- 0.02, N = 3SE +/- 0.05, N = 318.7718.561. (CC) gcc options: -O2 -pedantic -fvisibility=hidden

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: CPUPop_OS 20.04Pop_OS 21.04160320480640800SE +/- 1.19, N = 3SE +/- 1.28, N = 3746.63738.18MIN: 729.06MIN: 726.581. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

BRL-CAD

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

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.32.2VGR Performance MetricPop_OS 20.04Pop_OS 21.04200K400K600K800K1000K7902577991671. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -pthread -ldl -lm

GraphicsMagick

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

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: RotatePop_OS 20.04Pop_OS 21.04140280420560700SE +/- 1.33, N = 36316381. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GNU Radio

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

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FM Deemphasis FilterPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 2.41, N = 4SE +/- 2.18, N = 3920.0930.11. Pop_OS 20.04: 3.8.1.02. Pop_OS 21.04: 3.8.2.0

NeatBench

NeatBench is a benchmark of the cross-platform Neat Video software on the CPU and optional GPU (OpenCL / CUDA) support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterNeatBench 5Acceleration: CPUPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.12, N = 3SE +/- 0.15, N = 337.137.5

Tesseract

Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 3840 x 2160Pop_OS 20.04Pop_OS 21.0480160240320400SE +/- 1.18, N = 3SE +/- 1.49, N = 3357.20353.47

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: CPUPop_OS 20.04Pop_OS 21.04246810SE +/- 0.01383, N = 3SE +/- 0.01801, N = 36.537256.47045MIN: 6.41MIN: 6.371. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.0Preset: ExhaustivePop_OS 20.04Pop_OS 21.0448121620SE +/- 0.04, N = 3SE +/- 0.04, N = 315.0414.901. (CXX) g++ options: -O3 -flto -pthread

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: chaosPop_OS 20.04Pop_OS 21.0420406080100105106

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 FilterPop_OS 20.04Pop_OS 21.04130260390520650SE +/- 1.53, N = 4SE +/- 2.08, N = 3617.1622.81. Pop_OS 20.04: 3.8.1.02. Pop_OS 21.04: 3.8.2.0

LuaJIT

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

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Fast Fourier TransformPop_OS 20.04Pop_OS 21.0460120180240300SE +/- 1.05, N = 3SE +/- 0.38, N = 3263.17265.581. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

dcraw

This test times how long it takes to convert several high-resolution RAW NEF image files to PPM image format using dcraw. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterdcrawRAW To PPM Image ConversionPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.05, N = 3SE +/- 0.08, N = 341.0841.451. (CC) gcc options: -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: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUPop_OS 20.04Pop_OS 21.04160320480640800SE +/- 0.56, N = 3SE +/- 1.34, N = 3745.00738.47MIN: 727.63MIN: 726.911. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

eSpeak-NG Speech Engine

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

OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech SynthesisPop_OS 20.04Pop_OS 21.04714212835SE +/- 0.09, N = 4SE +/- 0.04, N = 428.2328.481. (CC) gcc options: -O2 -std=c99

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 1Pop_OS 20.04Pop_OS 21.040.10510.21020.31530.42040.5255SE +/- 0.001, N = 3SE +/- 0.000, N = 30.4670.463

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUPop_OS 20.04Pop_OS 21.04160320480640800SE +/- 2.72, N = 3SE +/- 1.51, N = 3746.10739.80MIN: 725.78MIN: 727.711. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.04Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAMPop_OS 20.04Pop_OS 21.0480160240320400SE +/- 0.95, N = 3SE +/- 0.07, N = 3377.9381.11. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.04Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAMPop_OS 20.04Pop_OS 21.041632486480SE +/- 0.06, N = 3SE +/- 0.06, N = 371.271.81. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

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: 0Pop_OS 20.04Pop_OS 21.041122334455SE +/- 0.11, N = 3SE +/- 0.11, N = 348.3847.981. (CXX) g++ options: -O3 -fPIC -lm

Opus Codec Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus EncodePop_OS 20.04Pop_OS 21.04246810SE +/- 0.008, N = 5SE +/- 0.024, N = 57.3907.3311. (CXX) g++ options: -fvisibility=hidden -logg -lm

NeatBench

NeatBench is a benchmark of the cross-platform Neat Video software on the CPU and optional GPU (OpenCL / CUDA) support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterNeatBench 5Acceleration: AllPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.15, N = 3SE +/- 0.09, N = 337.537.8

Radiance Benchmark

This is a benchmark of NREL Radiance, a synthetic imaging system that is open-source and developed by the Lawrence Berkeley National Laboratory in California. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRadiance Benchmark 5.0Test: SerialPop_OS 20.04Pop_OS 21.04120240360480600569.84574.36

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkUnstructuredVolumePop_OS 20.04Pop_OS 21.04400K800K1200K1600K2000KSE +/- 3997.25, N = 3SE +/- 1722.35, N = 318926341878141MIN: 22800 / MAX: 6289959MIN: 22798 / MAX: 6163438

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.04Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAMPop_OS 20.04Pop_OS 21.0480160240320400SE +/- 1.68, N = 3SE +/- 0.38, N = 3350.8353.51. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: squeezenet_ssdPop_OS 20.04Pop_OS 21.04714212835SE +/- 0.09, N = 15SE +/- 0.14, N = 631.2431.01MIN: 29.29 / MAX: 138.86MIN: 30.09 / MAX: 36.51. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Intel Open Image Denoise

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

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.ldr_alb_nrm.3840x2160Pop_OS 20.04Pop_OS 21.040.3060.6120.9181.2241.53SE +/- 0.00, N = 3SE +/- 0.00, N = 31.351.36

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.hdr_alb_nrm.3840x2160Pop_OS 20.04Pop_OS 21.040.3060.6120.9181.2241.53SE +/- 0.00, N = 3SE +/- 0.00, N = 31.351.36

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: 64 - Buffer Length: 256 - Filter Length: 57Pop_OS 20.04Pop_OS 21.04600M1200M1800M2400M3000MSE +/- 12563571.68, N = 3SE +/- 21310899.66, N = 3289480000029142333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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: Pabellon Barcelona - Compute: CPU-OnlyPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.35, N = 3SE +/- 0.28, N = 396.7597.38

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v2Pop_OS 20.04Pop_OS 21.041530456075SE +/- 0.15, N = 3SE +/- 0.09, N = 368.2868.71MIN: 67.33 / MAX: 75.66MIN: 68.11 / MAX: 69.851. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

PyBench

This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 6.06, N = 3SE +/- 2.65, N = 3957963

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 4KPop_OS 20.04Pop_OS 21.041122334455SE +/- 0.13, N = 3SE +/- 0.05, N = 348.3148.011. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CPop_OS 20.04Pop_OS 21.046K12K18K24K30KSE +/- 86.75, N = 3SE +/- 3.25, N = 328368.4628545.47-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

FLAC Audio Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.2WAV To FLACPop_OS 20.04Pop_OS 21.04246810SE +/- 0.014, N = 5SE +/- 0.009, N = 57.8467.8921. (CXX) g++ options: -O2 -fvisibility=hidden -logg -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: 1 - Input: Bosphorus 1080pPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.12, N = 3SE +/- 0.20, N = 339.6939.461. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

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 TransformPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.06, N = 3SE +/- 0.07, N = 3103.9104.5

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 0.70, N = 3SE +/- 4.10, N = 31132.811139.09-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

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)Pop_OS 20.04Pop_OS 21.048001600240032004000SE +/- 23.63, N = 4SE +/- 72.46, N = 33704.83724.91. Pop_OS 20.04: 3.8.1.02. Pop_OS 21.04: 3.8.2.0

Chaos Group V-RAY

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

OpenBenchmarking.orgvsamples, More Is BetterChaos Group V-RAY 5Mode: CPUPop_OS 20.04Pop_OS 21.0412K24K36K48K60KSE +/- 130.56, N = 3SE +/- 483.62, N = 75477154478

GravityMark

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.2Resolution: 3840 x 2160 - Renderer: VulkanPop_OS 20.04Pop_OS 21.041326395265SE +/- 0.00, N = 3SE +/- 0.03, N = 355.856.1

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-OnlyPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.06, N = 3SE +/- 0.34, N = 383.7184.16

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerPop_OS 20.04Pop_OS 21.041.26452.5293.79355.0586.3225SE +/- 0.02, N = 3SE +/- 0.00, N = 35.625.59MIN: 5.56 / MAX: 5.75MIN: 5.52 / MAX: 5.71

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 DragonPop_OS 20.04Pop_OS 21.041020304050SE +/- 0.03, N = 3SE +/- 0.05, N = 344.1544.39MIN: 43.29 / MAX: 46.06MIN: 43.47 / MAX: 46.24

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 32 - Buffer Length: 256 - Filter Length: 57Pop_OS 20.04Pop_OS 21.04400M800M1200M1600M2000MSE +/- 3829853.49, N = 3SE +/- 5572651.47, N = 3182006666718297666671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.04Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMPop_OS 20.04Pop_OS 21.041326395265SE +/- 0.03, N = 3SE +/- 0.12, N = 356.656.91. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CPop_OS 20.04Pop_OS 21.0414K28K42K56K70KSE +/- 80.83, N = 3SE +/- 74.35, N = 365164.1265506.86-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

Tungsten Renderer

Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Volumetric CausticPop_OS 20.04Pop_OS 21.040.79661.59322.38983.18643.983SE +/- 0.00398, N = 3SE +/- 0.00790, N = 33.540303.521821. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -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 1.2Model: SqueezeNetV1.0Pop_OS 20.04Pop_OS 21.04246810SE +/- 0.030, N = 3SE +/- 0.078, N = 37.4347.473MIN: 7.28 / MAX: 7.7MIN: 7.16 / MAX: 8.11. (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

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Finagle HTTP RequestsPop_OS 20.04Pop_OS 21.048001600240032004000SE +/- 24.83, N = 3SE +/- 12.74, N = 33561.93579.7MIN: 3287.21 / MAX: 3611.28MIN: 3308.88 / MAX: 3636.68

LuaJIT

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

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Monte CarloPop_OS 20.04Pop_OS 21.04110220330440550SE +/- 1.90, N = 3SE +/- 0.51, N = 3507.35504.831. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: googlenetPop_OS 20.04Pop_OS 21.04714212835SE +/- 0.26, N = 15SE +/- 0.67, N = 630.7730.62MIN: 26.71 / MAX: 106.87MIN: 27.79 / MAX: 88.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

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 DragonPop_OS 20.04Pop_OS 21.041020304050SE +/- 0.05, N = 3SE +/- 0.07, N = 343.0542.84MIN: 42.21 / MAX: 44.8MIN: 42.02 / MAX: 44.65

LuaJIT

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

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Jacobi Successive Over-RelaxationPop_OS 20.04Pop_OS 21.04400800120016002000SE +/- 6.58, N = 3SE +/- 3.69, N = 32053.552043.701. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

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 9Pop_OS 20.04Pop_OS 21.040.56391.12781.69172.25562.8195SE +/- 0.001, N = 3SE +/- 0.010, N = 32.4942.506

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Orange Juice - Acceleration: CPUPop_OS 20.04Pop_OS 21.043691215SE +/- 0.03, N = 3SE +/- 0.04, N = 313.0113.07MIN: 11.1 / MAX: 15.13MIN: 11.13 / MAX: 15.14

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: raytracePop_OS 20.04Pop_OS 21.04100200300400500448446

Tesseract

Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 2560 x 1440Pop_OS 20.04Pop_OS 21.04110220330440550SE +/- 2.78, N = 3SE +/- 5.41, N = 3489.23487.09

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkPop_OS 20.04Pop_OS 21.04100200300400500SE +/- 1.45, N = 3SE +/- 3.18, N = 3456458MIN: 1 / MAX: 1288MIN: 1 / MAX: 1303

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: json_loadsPop_OS 20.04Pop_OS 21.04612182430SE +/- 0.03, N = 3SE +/- 0.03, N = 323.423.5

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: 2Pop_OS 20.04Pop_OS 21.04612182430SE +/- 0.02, N = 3SE +/- 0.07, N = 325.4425.551. (CXX) g++ options: -O3 -fPIC -lm

GravityMark

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.2Resolution: 2560 x 1440 - Renderer: VulkanPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.25, N = 3SE +/- 0.12, N = 375.475.7

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Apache Spark PageRankPop_OS 20.04Pop_OS 21.048001600240032004000SE +/- 30.53, N = 3SE +/- 51.35, N = 33718.03732.7MIN: 3375.82 / MAX: 3824.71MIN: 3305.1 / MAX: 3820.58

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 0.3.21Pop_OS 20.04Pop_OS 21.04510152025SE +/- 0.04, N = 3SE +/- 0.08, N = 318.6218.55

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.0Preset: MediumPop_OS 20.04Pop_OS 21.040.9071.8142.7213.6284.535SE +/- 0.0059, N = 3SE +/- 0.0113, N = 34.01584.03101. (CXX) g++ options: -O3 -flto -pthread

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 3Pop_OS 20.04Pop_OS 21.0448121620SE +/- 0.01, N = 3SE +/- 0.01, N = 314.8014.851. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 3840 x 2160Pop_OS 20.04Pop_OS 21.0412002400360048006000SE +/- 1.60, N = 3SE +/- 0.12, N = 35397.345416.70

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 3840 x 2160Pop_OS 20.04Pop_OS 21.041224364860SE +/- 0.02, N = 3SE +/- 0.00, N = 353.8454.03

LuaJIT

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

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Dense LU Matrix FactorizationPop_OS 20.04Pop_OS 21.048001600240032004000SE +/- 9.52, N = 3SE +/- 2.96, N = 33820.813833.511. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

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: 128 - Buffer Length: 256 - Filter Length: 57Pop_OS 20.04Pop_OS 21.04700M1400M2100M2800M3500MSE +/- 12611238.37, N = 3SE +/- 3485366.23, N = 3310640000030962333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

libjpeg-turbo tjbench

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

OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.1.0Test: Decompression ThroughputPop_OS 20.04Pop_OS 21.0450100150200250SE +/- 0.26, N = 3SE +/- 0.29, N = 3211.51212.211. (CC) gcc options: -O3 -rdynamic

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 19Pop_OS 20.04Pop_OS 21.04510152025SE +/- 0.05, N = 3SE +/- 0.05, N = 321.6721.60

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.0Preset: ThoroughPop_OS 20.04Pop_OS 21.04246810SE +/- 0.0107, N = 3SE +/- 0.0070, N = 36.13336.15271. (CXX) g++ options: -O3 -flto -pthread

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.04Test: 4G PHY_DL_Test 100 PRB SISO 64-QAMPop_OS 20.04Pop_OS 21.0450100150200250SE +/- 0.57, N = 3SE +/- 0.37, N = 3223.7223.01. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

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: CrownPop_OS 20.04Pop_OS 21.041224364860SE +/- 0.13, N = 3SE +/- 0.16, N = 354.6954.86MIN: 53.54 / MAX: 57.58MIN: 53.74 / MAX: 56.96

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CPop_OS 20.04Pop_OS 21.0415K30K45K60K75KSE +/- 22.10, N = 3SE +/- 9.87, N = 367929.3268134.54-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

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 FilterPop_OS 20.04Pop_OS 21.0490180270360450SE +/- 2.42, N = 3SE +/- 8.12, N = 3418.5417.3

LuaJIT

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

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: CompositePop_OS 20.04Pop_OS 21.0430060090012001500SE +/- 4.11, N = 3SE +/- 0.26, N = 31579.751584.011. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CPop_OS 20.04Pop_OS 21.045K10K15K20K25KSE +/- 2.62, N = 3SE +/- 9.77, N = 323049.8123109.24-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

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 ObjPop_OS 20.04Pop_OS 21.04816243240SE +/- 0.02, N = 3SE +/- 0.01, N = 336.9136.82MIN: 36.33 / MAX: 38.42MIN: 36.26 / MAX: 38.32

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CPop_OS 20.04Pop_OS 21.046K12K18K24K30KSE +/- 23.28, N = 3SE +/- 18.19, N = 326683.5026748.89-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.DPop_OS 20.04Pop_OS 21.0411002200330044005500SE +/- 9.94, N = 3SE +/- 14.84, N = 34967.594979.50-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Pop_OS 20.04: Open MPI 4.0.33. Pop_OS 21.04: Open MPI 4.1.0

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: EmilyPop_OS 20.04Pop_OS 21.04306090120150130.72131.01

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-OnlyPop_OS 20.04Pop_OS 21.041020304050SE +/- 0.17, N = 3SE +/- 0.09, N = 345.9745.87

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: 16 - Buffer Length: 256 - Filter Length: 57Pop_OS 20.04Pop_OS 21.04200M400M600M800M1000MSE +/- 813271.86, N = 3SE +/- 1337813.31, N = 39986033339964366671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.04Test: 4G PHY_DL_Test 100 PRB SISO 256-QAMPop_OS 20.04Pop_OS 21.0480160240320400SE +/- 0.52, N = 3SE +/- 0.72, N = 3380.2379.41. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

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: ETC1SPop_OS 20.04Pop_OS 21.04612182430SE +/- 0.04, N = 3SE +/- 0.17, N = 323.9523.901. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

KTX-Software toktx

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

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3 + Zstd Compression 19Pop_OS 20.04Pop_OS 21.043691215SE +/- 0.01, N = 3SE +/- 0.04, N = 311.2211.25

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.04Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAMPop_OS 20.04Pop_OS 21.04306090120150SE +/- 0.12, N = 3SE +/- 0.12, N = 3155.3155.01. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 1920 x 1080Pop_OS 20.04Pop_OS 21.0412002400360048006000SE +/- 1.44, N = 3SE +/- 0.39, N = 35710.255721.24

GravityMark

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.2Resolution: 3840 x 2160 - Renderer: OpenGLPop_OS 20.04Pop_OS 21.041224364860SE +/- 0.37, N = 3SE +/- 0.44, N = 353.653.7

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 1920 x 1080Pop_OS 20.04Pop_OS 21.041326395265SE +/- 0.02, N = 3SE +/- 0.00, N = 356.9657.06

IndigoBench

This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: SupercarPop_OS 20.04Pop_OS 21.04510152025SE +/- 0.03, N = 3SE +/- 0.07, N = 322.5822.54

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: BedroomPop_OS 20.04Pop_OS 21.043691215SE +/- 0.03, N = 3SE +/- 0.01, N = 310.4610.48

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: CPUPop_OS 20.04Pop_OS 21.040.29030.58060.87091.16121.4515SE +/- 0.00919, N = 12SE +/- 0.01037, N = 91.288011.29016MIN: 1.21MIN: 1.21. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: mobilenetPop_OS 20.04Pop_OS 21.04714212835SE +/- 0.22, N = 15SE +/- 0.31, N = 631.0931.14MIN: 28.17 / MAX: 109.78MIN: 29.1 / MAX: 36.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 10Pop_OS 20.04Pop_OS 21.040.86631.73262.59893.46524.3315SE +/- 0.005, N = 3SE +/- 0.023, N = 33.8503.844

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.04Test: 4G PHY_DL_Test 100 PRB SISO 64-QAMPop_OS 20.04Pop_OS 21.0480160240320400SE +/- 0.79, N = 3SE +/- 0.52, N = 3353.4353.91. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

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: CPUPop_OS 20.04Pop_OS 21.040.21710.43420.65130.86841.0855SE +/- 0.003567, N = 3SE +/- 0.003063, N = 30.9635130.964733MIN: 0.92MIN: 0.931. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

GravityMark

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.2Resolution: 1920 x 1080 - Renderer: OpenGLPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.49, N = 3SE +/- 0.57, N = 385.285.3

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.2Resolution: 1920 x 1080 - Renderer: VulkanPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.37, N = 3SE +/- 0.21, N = 386.686.5

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 95, Compression Effort 7Pop_OS 20.04Pop_OS 21.0450100150200250SE +/- 0.39, N = 3SE +/- 1.21, N = 3214.16213.91-lOpenGL -lGLX -lGLU -lglut -lXmu -lXi-lgif1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.04Test: 4G PHY_DL_Test 100 PRB SISO 256-QAMPop_OS 20.04Pop_OS 21.0460120180240300SE +/- 0.74, N = 3SE +/- 0.29, N = 3266.3266.61. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC1 + DitheringPop_OS 20.04Pop_OS 21.0460120180240300SE +/- 0.21, N = 3SE +/- 0.10, N = 3277.95277.661. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

Tesseract

Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 1920 x 1080Pop_OS 20.04Pop_OS 21.04110220330440550SE +/- 5.29, N = 4SE +/- 5.63, N = 4487.55488.07

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: vgg16Pop_OS 20.04Pop_OS 21.041326395265SE +/- 0.30, N = 15SE +/- 0.33, N = 656.7456.80MIN: 50.91 / MAX: 146.45MIN: 52.7 / MAX: 301.821. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

G'MIC

G'MIC is an open-source framework for image processing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterG'MICTest: 3D Elevated Function In Random Colors, 100 TimesPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.07, N = 3SE +/- 0.03, N = 385.0885.141. Version 2.4.5, Copyright (c) 2008-2019, David Tschumperle.

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: CrownPop_OS 20.04Pop_OS 21.041122334455SE +/- 0.08, N = 3SE +/- 0.05, N = 349.8749.91MIN: 48.99 / MAX: 51.99MIN: 49.03 / MAX: 51.98

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.04Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAMPop_OS 20.04Pop_OS 21.04306090120150SE +/- 0.47, N = 3SE +/- 0.20, N = 3144.9145.01. (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 -lbladeRF -lm -lfftw3f -lmbedcrypto

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 FacadePop_OS 20.04Pop_OS 21.042K4K6K8K10KSE +/- 28.67, N = 3SE +/- 11.72, N = 3891489081. (CXX) g++ options: -O3

PHPBench

PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuitePop_OS 20.04Pop_OS 21.04140K280K420K560K700KSE +/- 4579.44, N = 3SE +/- 2903.75, N = 3662330662739

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 ObjPop_OS 20.04Pop_OS 21.04918273645SE +/- 0.03, N = 3SE +/- 0.04, N = 338.6038.58MIN: 37.96 / MAX: 40.49MIN: 37.89 / MAX: 40.22

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: LionPop_OS 20.04Pop_OS 21.0412002400360048006000SE +/- 12.53, N = 3SE +/- 13.05, N = 3560856051. (CXX) g++ options: -O3

RawTherapee

RawTherapee is a cross-platform, open-source multi-threaded RAW image processing program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRawTherapeeTotal Benchmark TimePop_OS 20.04Pop_OS 21.041122334455SE +/- 0.09, N = 3SE +/- 0.21, N = 347.8847.911. RawTherapee, version 5.8, command line.

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: Barbershop - Compute: CPU-OnlyPop_OS 20.04Pop_OS 21.04306090120150SE +/- 0.12, N = 3SE +/- 0.04, N = 3121.24121.19

rays1bench

This is a test of rays1bench, a simple path-tracer / ray-tracing that supports SSE and AVX instructions, multi-threading, and other features. This test profile is measuring the performance of the "large scene" in rays1bench. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmrays/s, More Is Betterrays1bench 2020-01-09Large ScenePop_OS 20.04Pop_OS 21.0470140210280350SE +/- 0.33, N = 3SE +/- 0.34, N = 3301.63301.52

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 19Pop_OS 20.04Pop_OS 21.041428425670SE +/- 0.16, N = 3SE +/- 0.14, N = 362.4662.45

WavPack Audio Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPackPop_OS 20.04Pop_OS 21.043691215SE +/- 0.01, N = 5SE +/- 0.00, N = 512.8412.851. (CXX) g++ options: -rdynamic

OCRMyPDF

OCRMyPDF is an optical character recognition (OCR) text layer to scanned PDF files, producing new PDFs with the text now selectable/searchable/copy-paste capable. OCRMyPDF leverages the Tesseract OCR engine and is written in Python. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOCRMyPDF 10.3.1+dfsgProcessing 60 Page PDF DocumentPop_OS 21.043691215SE +/- 0.06, N = 312.72

GIMP

GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.22Test: unsharp-maskPop_OS 21.04510152025SE +/- 0.02, N = 318.53

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.22Test: auto-levelsPop_OS 21.0448121620SE +/- 0.09, N = 314.16

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.22Test: rotatePop_OS 21.043691215SE +/- 0.09, N = 311.22

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.22Test: resizePop_OS 21.04246810SE +/- 0.058, N = 38.168

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: blazefacePop_OS 20.04Pop_OS 21.04246810SE +/- 0.08, N = 15SE +/- 0.04, N = 66.716.71MIN: 5.91 / MAX: 85.91MIN: 6.41 / MAX: 8.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OCRMyPDF

OCRMyPDF is an optical character recognition (OCR) text layer to scanned PDF files, producing new PDFs with the text now selectable/searchable/copy-paste capable. OCRMyPDF leverages the Tesseract OCR engine and is written in Python. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOCRMyPDF 9.6.0+dfsgProcessing 60 Page PDF DocumentPop_OS 20.0448121620SE +/- 0.10, N = 315.10

Inkscape

Inkscape is an open-source vector graphics editor. This test profile times how long it takes to complete various operations by Inkscape. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterInkscapeOperation: SVG Files To PNGPop_OS 20.04612182430SE +/- 0.04, N = 326.461. Inkscape 0.92.5 (2060ec1f9f, 2020-04-08)

GIMP

GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.18Test: unsharp-maskPop_OS 20.0448121620SE +/- 0.03, N = 317.32

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.18Test: auto-levelsPop_OS 20.0448121620SE +/- 0.05, N = 314.01

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.18Test: rotatePop_OS 20.043691215SE +/- 0.03, N = 311.34

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.18Test: resizePop_OS 20.04246810SE +/- 0.022, N = 37.280

Darktable

Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.0.1Test: Server Room - Acceleration: CPU-onlyPop_OS 20.040.50631.01261.51892.02522.5315SE +/- 0.004, N = 32.250

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.0.1Test: Server Rack - Acceleration: CPU-onlyPop_OS 20.040.01980.03960.05940.07920.099SE +/- 0.001, N = 50.088

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.0.1Test: Masskrug - Acceleration: CPU-onlyPop_OS 20.040.64961.29921.94882.59843.248SE +/- 0.006, N = 32.887

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.0.1Test: Boat - Acceleration: CPU-onlyPop_OS 20.041.23572.47143.70714.94286.1785SE +/- 0.023, N = 35.492

Montage Astronomical Image Mosaic Engine

Montage is an open-source astronomical image mosaic engine. This BSD-licensed astronomy software is developed by the California Institute of Technology, Pasadena. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMontage Astronomical Image Mosaic Engine 6.0Mosaic of M17, K band, 1.5 deg x 1.5 degPop_OS 20.0420406080100SE +/- 0.40, N = 375.201. (CC) gcc options: -std=gnu99 -lcfitsio -lm -O2

Intel Open Image Denoise

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

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RTLightmap.hdr.4096x4096Pop_OS 20.04Pop_OS 21.040.15080.30160.45240.60320.754SE +/- 0.00, N = 3SE +/- 0.00, N = 30.670.67

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 4KPop_OS 20.04Pop_OS 21.040.04950.0990.14850.1980.2475SE +/- 0.00, N = 3SE +/- 0.00, N = 30.220.221. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: Path TracerPop_OS 20.04Pop_OS 21.0470140210280350SE +/- 0.00, N = 3SE +/- 0.00, N = 3333.33333.33MIN: 250 / MAX: 500MIN: 250 / MAX: 500

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: SciVisPop_OS 20.04Pop_OS 21.041020304050SE +/- 0.00, N = 3SE +/- 0.00, N = 345.4545.45MIN: 43.48 / MAX: 47.62MIN: 43.48 / MAX: 47.62

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerPop_OS 20.04Pop_OS 21.04246810SE +/- 0.02, N = 3SE +/- 0.02, N = 37.167.16MIN: 7.09 / MAX: 7.35MIN: 7.04 / MAX: 7.35

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: SciVisPop_OS 20.04Pop_OS 21.0420406080100SE +/- 0.00, N = 3SE +/- 0.00, N = 390.9190.91MAX: 100MIN: 83.33 / MAX: 100

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVisPop_OS 20.04Pop_OS 21.043691215SE +/- 0.06, N = 3SE +/- 0.06, N = 313.5813.58MIN: 13.16 / MAX: 13.89MIN: 13.33 / MAX: 13.89

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisPop_OS 20.04Pop_OS 21.041632486480SE +/- 0.00, N = 3SE +/- 0.00, N = 371.4371.43MIN: 66.67 / MAX: 76.92MIN: 66.67 / MAX: 76.92

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-bitPop_OS 20.041020304050SE +/- 0.04, N = 343.941. (CXX) g++ options: -O3 -lpthread -lrt

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 1080pPop_OS 20.04306090120150SE +/- 0.28, N = 3144.701. (CXX) g++ options: -O3 -lpthread -lrt

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 4KPop_OS 20.041224364860SE +/- 0.06, N = 350.991. (CXX) g++ options: -O3 -lpthread -lrt

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Chimera 1080pPop_OS 20.04306090120150SE +/- 0.30, N = 3112.691. (CXX) g++ options: -O3 -lpthread -lrt

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: alexnetPop_OS 20.04Pop_OS 21.0448121620SE +/- 0.42, N = 15SE +/- 0.04, N = 616.5514.66MIN: 13.79 / MAX: 108.4MIN: 14.17 / MAX: 19.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: efficientnet-b0Pop_OS 20.04Pop_OS 21.04510152025SE +/- 0.31, N = 15SE +/- 0.40, N = 619.3219.42MIN: 17.29 / MAX: 106.05MIN: 18.1 / MAX: 24.41. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: mnasnetPop_OS 20.04Pop_OS 21.0448121620SE +/- 0.57, N = 15SE +/- 0.30, N = 615.0414.68MIN: 13.05 / MAX: 107.7MIN: 13.72 / MAX: 18.571. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU-v3-v3 - Model: mobilenet-v3Pop_OS 20.04Pop_OS 21.0448121620SE +/- 0.41, N = 15SE +/- 0.27, N = 614.5714.76MIN: 12.97 / MAX: 113.13MIN: 13.69 / MAX: 157.81. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU-v2-v2 - Model: mobilenet-v2Pop_OS 20.04Pop_OS 21.0448121620SE +/- 0.59, N = 15SE +/- 0.29, N = 615.8415.81MIN: 13.99 / MAX: 96.38MIN: 14.73 / MAX: 20.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Tachyon

This is a test of the threaded Tachyon, a parallel ray-tracing system, measuring the time to ray-trace a sample scene. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.99b6Total TimePop_OS 20.04Pop_OS 21.0448121620SE +/- 0.52, N = 12SE +/- 0.59, N = 1217.0718.121. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080pPop_OS 20.04Pop_OS 21.0480160240320400SE +/- 14.31, N = 12SE +/- 3.86, N = 3361.14331.401. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

GraphicsMagick

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

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: ResizingPop_OS 20.04Pop_OS 21.042004006008001000SE +/- 11.06, N = 15SE +/- 13.08, N = 157638181. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Rainbow Colors and Prism - Acceleration: CPUPop_OS 20.04Pop_OS 21.04612182430SE +/- 0.95, N = 12SE +/- 0.95, N = 1522.9524.13MIN: 19.43 / MAX: 28.22MIN: 17.89 / MAX: 28.43

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiVoxels / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 3840 x 2160Pop_OS 20.04Pop_OS 21.048001600240032004000SE +/- 44.13, N = 3SE +/- 61.51, N = 153496.393565.12

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 3840 x 2160Pop_OS 20.04Pop_OS 21.0450100150200250SE +/- 2.76, N = 3SE +/- 3.84, N = 15218.52222.82

OpenBenchmarking.orgMiVoxels / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 1920 x 1080Pop_OS 20.04Pop_OS 21.0412002400360048006000SE +/- 84.30, N = 12SE +/- 229.45, N = 155672.675717.32

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 1920 x 1080Pop_OS 20.04Pop_OS 21.0480160240320400SE +/- 5.27, N = 12SE +/- 14.34, N = 15354.54357.33

Unvanquished

Unvanquished is a modern fork of the Tremulous first person shooter. Unvanquished is powered by the Daemon engine, a combination of the ioquake3 engine with the graphically-beautiful XreaL engine. Unvanquished supports a modern OpenGL 3 renderer and other advanced graphics features for this open-source game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 1920 x 1080 - Effects Quality: HighPop_OS 20.04Pop_OS 21.0460120180240300SE +/- 2.60, N = 2SE +/- 11.05, N = 12294.3248.5

Betsy GPU Compressor

Betsy is an open-source GPU compressor of various GPU compression techniques. Betsy is written in GLSL for Vulkan/OpenGL (compute shader) support for GPU-based texture compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC2 RGB - Quality: HighestPop_OS 20.04Pop_OS 21.04246810SE +/- 0.190, N = 15SE +/- 0.191, N = 157.5757.3071. (CXX) g++ options: -O3 -O2 -lpthread -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC1 - Quality: HighestPop_OS 20.04Pop_OS 21.04246810SE +/- 0.194, N = 15SE +/- 0.195, N = 156.7166.4631. (CXX) g++ options: -O3 -O2 -lpthread -ldl

354 Results Shown

Renaissance
Timed Wasmer Compilation
x265
PyPerformance
G'MIC
Kvazaar:
  Bosphorus 1080p - Medium
  Bosphorus 1080p - Very Fast
  Bosphorus 1080p - Slow
AOM AV1:
  Speed 6 Realtime - Bosphorus 4K
  Speed 6 Two-Pass - Bosphorus 1080p
Facebook RocksDB
Renaissance
librsvg
Facebook RocksDB
dav1d
Kvazaar
Timed GDB GNU Debugger Compilation
Kvazaar
SVT-AV1
Mobile Neural Network
Kvazaar
Timed FFmpeg Compilation
dav1d:
  Chimera 1080p
  Chimera 1080p 10-bit
GravityMark
PyPerformance
Tungsten Renderer
SVT-HEVC
Facebook RocksDB
GravityMark
x264
AOM AV1
SVT-HEVC
GEGL
AOM AV1:
  Speed 8 Realtime - Bosphorus 1080p
  Speed 6 Realtime - Bosphorus 1080p
dav1d
Renaissance
PyPerformance
oneDNN
SVT-VP9
AOM AV1
OpenVKL
GNU Radio
SVT-VP9
AOM AV1
GEGL
AOM AV1
Renaissance
PJSIP
oneDNN
LuaRadio
Renaissance
GEGL
Mobile Neural Network
OpenSCAD
Timed Linux Kernel Compilation
PyPerformance
GravityMark
GEGL
libavif avifenc
Timed Mesa Compilation
GEGL
Timed Godot Game Engine Compilation
WebP2 Image Encode
Natron
Tungsten Renderer
Kvazaar
WebP Image Encode
C-Blosc
Tungsten Renderer
Ngspice
TNN
SVT-AV1:
  Preset 4 - Bosphorus 1080p
  Preset 8 - Bosphorus 4K
AOM AV1
DeepSpeech
oneDNN
GEGL
oneDNN
VP9 libvpx Encoding
Mobile Neural Network
AOBench
PyPerformance
GEGL
PyPerformance
OpenVKL
Facebook RocksDB
VP9 libvpx Encoding
libavif avifenc
WebP Image Encode
LuxCoreRender
Ngspice
NCNN
GraphicsMagick:
  Sharpen
  Swirl
x265
LibRaw
NCNN
GNU Radio
WebP2 Image Encode
Etcpak
oneDNN
Renaissance
WebP2 Image Encode
TTSIOD 3D Renderer
oneDNN
GEGL
PyPerformance
Facebook RocksDB
Kvazaar
LuaRadio
Mobile Neural Network
GraphicsMagick
oneDNN
SVT-AV1
PyPerformance:
  go
  pathlib
libavif avifenc
LuxCoreRender
Google SynthMark
libavif avifenc
ET: Legacy
PJSIP
ParaView:
  Wavelet Contour - 3840 x 2160:
    Frames / Sec
    MiPolys / Sec
ET: Legacy
YafaRay
Renaissance
VP9 libvpx Encoding
Facebook RocksDB
Appleseed
ET: Legacy
TNN
PJSIP
NAS Parallel Benchmarks
OSPray
GNU Radio
Etcpak
Radiance Benchmark
srsRAN
OpenSCAD
Renaissance
LAME MP3 Encoding
Ogg Audio Encoding
rav1e
PyPerformance
GraphicsMagick
AOM AV1
VP9 libvpx Encoding
Appleseed
C-Ray
srsRAN
Mobile Neural Network
oneDNN
Renaissance
WebP Image Encode
POV-Ray
Mobile Neural Network
NAS Parallel Benchmarks:
  CG.C
  SP.B
GEGL
Basis Universal
Blender
WebP Image Encode
NCNN
Basis Universal
WebP2 Image Encode
LuaJIT
srsRAN
Tesseract OCR
GraphicsMagick
LuxCoreRender
NCNN
oneDNN
KTX-Software toktx
oneDNN
TNN
rav1e
ParaView:
  Wavelet Contour - 1920 x 1080:
    Frames / Sec
    MiPolys / Sec
oneDNN
OpenSCAD
Monkey Audio Encoding
WebP Image Encode
GravityMark
oneDNN
NCNN
Hugin
Facebook RocksDB
Etcpak
RNNoise
oneDNN
BRL-CAD
GraphicsMagick
GNU Radio
NeatBench
Tesseract
oneDNN
ASTC Encoder
PyPerformance
GNU Radio
LuaJIT
dcraw
oneDNN
eSpeak-NG Speech Engine
rav1e
oneDNN
srsRAN:
  4G PHY_DL_Test 100 PRB MIMO 256-QAM
  5G PHY_DL_NR Test 270 PRB SISO 256-QAM
libavif avifenc
Opus Codec Encoding
NeatBench
Radiance Benchmark
OpenVKL
srsRAN
NCNN
Intel Open Image Denoise:
  RT.ldr_alb_nrm.3840x2160
  RT.hdr_alb_nrm.3840x2160
Liquid-DSP
Blender
TNN
PyBench
AOM AV1
NAS Parallel Benchmarks
FLAC Audio Encoding
SVT-HEVC
LuaRadio
NAS Parallel Benchmarks
GNU Radio
Chaos Group V-RAY
GravityMark
Blender
OSPray
Embree
Liquid-DSP
srsRAN
NAS Parallel Benchmarks
Tungsten Renderer
Mobile Neural Network
Renaissance
LuaJIT
NCNN
Embree
LuaJIT
KTX-Software toktx
LuxCoreRender
PyPerformance
Tesseract
OpenVKL
PyPerformance
libavif avifenc
GravityMark
Renaissance
VOSK Speech Recognition Toolkit
ASTC Encoder
Basis Universal
ParaView:
  Many Spheres - 3840 x 2160:
    MiPolys / Sec
    Frames / Sec
LuaJIT
Liquid-DSP
libjpeg-turbo tjbench
KTX-Software toktx
ASTC Encoder
srsRAN
Embree
NAS Parallel Benchmarks
LuaRadio
LuaJIT
NAS Parallel Benchmarks
Embree
NAS Parallel Benchmarks:
  MG.C
  EP.D
Appleseed
Blender
Liquid-DSP
srsRAN
Basis Universal
KTX-Software toktx
srsRAN
ParaView
GravityMark
ParaView
IndigoBench:
  CPU - Supercar
  CPU - Bedroom
oneDNN
NCNN
rav1e
srsRAN
oneDNN
GravityMark:
  1920 x 1080 - OpenGL
  1920 x 1080 - Vulkan
WebP2 Image Encode
srsRAN
Etcpak
Tesseract
NCNN
G'MIC
Embree
srsRAN
Google Draco
PHPBench
Embree
Google Draco
RawTherapee
Blender
rays1bench
KTX-Software toktx
WavPack Audio Encoding
OCRMyPDF
GIMP:
  unsharp-mask
  auto-levels
  rotate
  resize
NCNN
OCRMyPDF
Inkscape
GIMP:
  unsharp-mask
  auto-levels
  rotate
  resize
Darktable:
  Server Room - CPU-only
  Server Rack - CPU-only
  Masskrug - CPU-only
  Boat - CPU-only
Montage Astronomical Image Mosaic Engine
Intel Open Image Denoise
AOM AV1
OSPray:
  Magnetic Reconnection - Path Tracer
  Magnetic Reconnection - SciVis
  XFrog Forest - Path Tracer
  NASA Streamlines - SciVis
  XFrog Forest - SciVis
  San Miguel - SciVis
libgav1:
  Chimera 1080p 10-bit
  Summer Nature 1080p
  Summer Nature 4K
  Chimera 1080p
NCNN:
  CPU - alexnet
  CPU - efficientnet-b0
  CPU - mnasnet
  CPU-v3-v3 - mobilenet-v3
  CPU-v2-v2 - mobilenet-v2
Tachyon
SVT-VP9
GraphicsMagick
LuxCoreRender
ParaView:
  Wavelet Volume - 3840 x 2160:
    MiVoxels / Sec
    Frames / Sec
  Wavelet Volume - 1920 x 1080:
    MiVoxels / Sec
    Frames / Sec
Unvanquished
Betsy GPU Compressor:
  ETC2 RGB - Highest
  ETC1 - Highest