AMD Threadripper 3960X vs. Intel Core i9 10980XE

Intel Core i9-10980XE benchmarks against AMD Ryzen Threadripper 3960X. Benchmarks by Michael Larabel on Ubuntu 19.10 Linux.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1911270-HU-THREADRIP09
Jump To Table - Results

View

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

Limit displaying results to tests within:

Audio Encoding 3 Tests
AV1 4 Tests
Bioinformatics 3 Tests
C++ Boost Tests 3 Tests
Web Browsers 1 Tests
Chess Test Suite 7 Tests
Timed Code Compilation 6 Tests
C/C++ Compiler Tests 32 Tests
Compression Tests 4 Tests
CPU Massive 66 Tests
Creator Workloads 34 Tests
Database Test Suite 7 Tests
Encoding 11 Tests
Game Development 2 Tests
HPC - High Performance Computing 18 Tests
Imaging 4 Tests
Java 5 Tests
Common Kernel Benchmarks 10 Tests
LAPACK (Linear Algebra Pack) Tests 2 Tests
Linear Algebra 2 Tests
Machine Learning 2 Tests
Memory Test Suite 5 Tests
Molecular Dynamics 5 Tests
MPI Benchmarks 8 Tests
Multi-Core 51 Tests
NVIDIA GPU Compute 8 Tests
Intel oneAPI 4 Tests
OpenCL 3 Tests
OpenMPI Tests 8 Tests
Productivity 3 Tests
Programmer / Developer System Benchmarks 12 Tests
Python Tests 6 Tests
Raytracing 4 Tests
Renderers 13 Tests
Rust Tests 2 Tests
Scientific Computing 12 Tests
Server 11 Tests
Server CPU Tests 41 Tests
Single-Threaded 22 Tests
Video Encoding 8 Tests
Common Workstation Benchmarks 8 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Threadripper 3960X
November 23 2019
  1 Day, 8 Hours, 55 Minutes
Core i9 10980XE
November 25 2019
  1 Day, 8 Hours, 20 Minutes
Invert Hiding All Results Option
  1 Day, 8 Hours, 37 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):


AMD Threadripper 3960X vs. Intel Core i9 10980XEOpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen Threadripper 3960X 24-Core @ 3.80GHz (24 Cores / 48 Threads)Intel Core i9-10980XE @ 4.60GHz (18 Cores / 36 Threads)MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS)Gigabyte X299X DESIGNARE 10G (F1 BIOS)AMD Starship/MatisseIntel Sky Lake-E DMI3 Registers4 x 8192 MB DDR4-3200MT/s240GB Force MP510AMD Navi 10 8GB (2100/875MHz)AMD Navi 10 HDMI AudioRealtek ALC1220ASUS PB278Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Device 27232 x Intel 10G X550T + Intel Device 2723Ubuntu 19.105.3.0-23-generic (x86_64)GNOME Shell 3.34.1X Server 1.20.5modesetting 1.20.54.5 Mesa 19.2.1 (LLVM 9.0.0)GCC 9.2.1 20191008ext42560x1440ProcessorsMotherboardsChipsetsMemoryDiskGraphicsAudiosMonitorNetworksOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionAMD Threadripper 3960X Vs. Intel Core I9 10980XE BenchmarksSystem Logs- --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,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 - NONE / errors=remount-ro,relatime,rw- Threadripper 3960X: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025- Core i9 10980XE: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- OpenJDK Runtime Environment (build 11.0.5-ea+10-post-Ubuntu-0ubuntu1)- Python 2.7.17rc1 + Python 3.7.5rc1- Threadripper 3960X: 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 + tsx_async_abort: Not affected - Core i9 10980XE: itlb_multihit: KVM: Mitigation of Split huge pages + 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 Enhanced IBRS IBPB: conditional RSB filling + tsx_async_abort: Mitigation of TSX disabled

Threadripper 3960X vs. Core i9 10980XE ComparisonPhoronix Test SuiteBaseline+39.5%+39.5%+79%+79%+118.5%+118.5%103%87.6%74.7%63.1%61.6%58.3%49.2%45.6%44.9%43.8%42.5%39.8%37.3%35.7%34.2%33%32.7%31%29.3%28%27.8%27.3%27.2%26.8%26.7%26.6%25.8%24.9%24.6%23.1%22.5%21.8%19.6%18.7%17.8%17.6%15.1%14.6%13.6%13.4%13.4%13.3%13.2%13%12.4%12.3%12%12%11.3%10.3%9.9%8.1%8.1%7.7%7.7%7.2%6.7%6.7%6.6%6.4%6.2%6.2%5.8%5.6%5.3%5.2%5%4.9%4.6%4.4%4.3%4%3.9%3.6%3.6%3%2.8%2.8%2.7%O.M.G157.9%v.V.1.V.E148.9%8 - Thread135.6%IP Batch 1D - f32135.6%Trace Time130.3%Fayalite-FIST Data127.7%R.R.B127.5%S.N.1127.4%Memory Copy - 4096 MiB122.9%8 - Process121.3%Wavelet Contour - 2560 x 1440119.9%Wavelet Contour - 2560 x 1440119.9%Memory Copy - 1024 MiB116.6%SP.B112.8%Chimera 1080p112.3%Tradebeans109.7%Memory Copy - 512 MiB109.3%Time To Solve104.5%P.N.T.T.2.0.0C.B.c - f3296.1%Swirl94.7%V.P.M93.2%Elapsed Time93%4 - Process90.3%Non-Exponential90.1%16 - Thread88.2%Reads4 - Thread85.2%16 - Process82.7%C.1.1.b80.7%H280.4%Crypto80%Total Time - 4.1.R.P.P78.5%Classroom - CPU-Only77.9%Chimera 1080p77.8%Volumetric Caustic76.1%S.N.174.8%200OpenMP CUTCP74.1%Pabellon Barcelona - CPU-Only73.3%Summer Nature 4K70.5%Fishy Cat - CPU-Only70.2%Float + SSE - 2D FFT Size 51268.8%Summer Nature 4K68.8%CoreMark Size 666 - I.P.S68.8%P.P.A67.9%Enc Mode 8 - 1080p66.6%G.C.S.F65.6%Water Benchmark64.2%leblancbigsedovbigFloat + SSE - 2D FFT Size 102460.6%Rand60%BMW27 - CPU-Only59.9%Sharpen59.9%C.1.1.b59.3%tConvolve OpenMP - Gridding58.8%R.R.LG.I.R.1.S58.2%Barbershop - CPU-Only57.3%Pathtracer - Crown57.1%G.Q.D.S57%NASA Streamlines - Path Tracer56.7%HWB Color Space56.5%H.2.1.V.E56.5%Time To Compile56.1%C.S.T55.9%Enhanced55%Enc Mode 4 - 1080p54.7%Noise-Gaussian54.4%Total Time54.3%XFrog Forest - SciVis54.2%G-Ffte53.8%G-Ffte53.8%MG.C53.7%Float + SSE - 2D FFT Size 25652.7%Maze Solver - Google Chrome52.5%Time To Compile52%Triad - Integer51.8%Resizing51.6%Time To Compile51.5%Time To Compile51%OpenMP LavaMD50.9%Hair50.8%Maze Solver - Firefox50%CPU50%Add - Floating Point49.3%Float + SSE - 2D FFT Size 204849.3%Standard MemsetRand Read49%Bedroom48.7%H.2.V.E48.4%XFrog Forest - Path Tracer48.2%Socket Activity47.9%Total Time47.7%json47.6%Add - Integer46.5%Boat - CPU-only45.9%Seq FillSavina Reactors.IO45.3%CPU45.3%D.B.d - f32M.C.F.B.S - 4096 MiB44.3%S.C.c.j44.3%R.N.N.T - f32NASA Streamlines - SciVis43.8%Writes43.6%Float + SSE - 2D FFT Size 12843.5%Tradesoap43.4%Pathtracer - Asian Dragon43.4%Vector Math42.9%Buffer Test - Normal Load - Read Only42.7%Pathtracer - Asian Dragon Obj42.6%EP-STREAM TriadEmily41.7%1.H.M.2.D41.5%Buffer Test - Normal Load - Read Write41.2%Create ThreadsAverage - Floating Point39.6%tConvolve MT - Degridding39%tConvolve MPI - Degridding38.8%Disney Material38.3%I.M.D.S38.3%San Miguel - SciVis37.9%M.C.F.B.S - 512 MiB37.7%F.E.TS.C.m.j36.9%M.C.F.B.S - 1024 MiB36%Pathtracer ISPC - Crown35.8%sqrtComposite35.4%35.2%Time To Compile34.2%Launch ProgramsATPase Simulation - 327,506 Atoms34.1%rotate33.6%G-Rand Access33.4%O.BWavelet Volume - 2560 x 144032.9%Wavelet Volume - 2560 x 144032.9%CPU Stress32.9%ReadScale - Integer32.4%Stock - 2D FFT Size 204831.7%M.C.F.B.S - 128 MiB31.1%Masskrug - CPU-only31%tConvolve OpenMP - DegriddingScale - Floating Point30.7%Supercar30.6%build30.2%resize30.1%Triad - Floating Point30.1%ForkingV.Q.O - Bosphorus 1080p29.2%WAV To MP328.5%StyleBench - Firefox28.5%OpenMP CFD Solver28.3%28.3%BT.C28.2%S.W.P.SCopyCopy - Integer27.8%W.O.E.O.D27.8%27.4%VMAF Optimized - Bosphorus 1080p27.4%ffsAverage - Integer27.2%Create ProcessesffsllStock - 1D FFT Size 3226.7%P.S.O - Bosphorus 1080p26.7%C.S.TD.B.d - f32Stock - 1D FFT Size 6426.6%Stock - 2D FFT Size 6426.5%Ctypepthread_oncePSPDFKit WASM - Google Chrome23.9%Float + SSE - 2D FFT Size 409623.8%1.T.A.V.E23.7%123.5%Copy - Floating Point23.2%Stock - 2D FFT Size 3223.1%Context SwitchingPSPDFKit WASM - Firefox23%Stock - 2D FFT Size 409622.5%Float + SSE - 2D FFT Size 6422.5%modfG.I.I.S22.3%Time To Compile22.3%WAV To FLAC22.2%R.C.a.Punsharp-mask21.8%Speedometer - Firefox21.5%Memory Copy - 128 MiB21.5%tConvolve MT - Gridding20.6%LU.C20.5%Pathtracer ISPC - Asian Dragon Obj20.1%Pathtracer ISPC - Asian Dragon19.6%Cifar10Stock - 1D FFT Size 25619.6%T.T.C.S.P.M19.1%A.S.P18.9%tConvolve MPI - Gridding18.9%Speedometer - Google Chrome18.8%Mixed 1:3Float + SSE - 2D FFT Size 3218.4%R.M.W18.3%P.B.SM.R - SciVisServer Room - CPU-only17.5%S.F.P.R17.3%32 - Process17.2%Stock - 1D FFT Size 12816.8%C.B.c - f3216.7%Read While Writing16.6%3000 Fall16.4%SMP Parallel16.2%San Miguel - Path Tracer15.7%Function ObjectsP.P.S14.9%Stock - 2D FFT Size 51214.9%Rand Numbersauto-levels14.2%Float + SSE - 1D FFT Size 1024Stock - 2D FFT Size 102413.4%136 Ragdolls13.4%SETC.B.c - f32M.R - Path Tracer1 - ProcessStock - 2D FFT Size 25613.1%QuasigroupARES-6 - Firefox12.7%Jetstream - Firefox12.6%Float + SSE - 1D FFT Size 32R.T.P.I.CT.T.C.C.G.C12.3%Water Caustic12.2%Solitaire12.1%OpenMP Stencil12.1%EP.DRotateC.u.1.0.3.s.i.i.C.L.111.4%MemorialD.B.d - f32P.R.W.S.S.M10.2%Small10.1%Standard Memcpy10%Serial10%Apache Spark ALSMemory Allocations9.8%log29.7%Server Rack - CPU-only9.2%C.B.c - f329.1%Float + SSE - 1D FFT Size 1289%Redis8.7%1000 Stack8.6%Triad8.4%Float + SSE - 1D FFT Size 20481 - ThreadAdd7.9%Float + SSE - 1D FFT Size 512Octane - Firefox7.7%EP.CPrim Trimesh7.5%Jython7.4%Graceful7.3%EP-DGEMM2 - Thread7%ARES-6 - Google Chrome7%M.P.P.B6.9%2.D.T.P6.9%Scale6.9%T.F.A.T.TCreate FilesIP Batch All - f32S.W.P.S6.6%AtolConvex Trimesh6.4%sincos6.2%CompositeWrite8S.A5.8%Raytests5.8%Material Testertanh5.6%G-Ptrans5.6%L.S.T.A.T.t.gA.C.PJetstream - Google Chrome5.2%SemaphoresEnc Mode 0 - 1080p4.9%cosMatrix Math4.9%Apache Spark Bayes4.8%BLASFT.CDLSCT.T.S.S4.2%sinhttp4%StyleBench - Google Chrome4%1000 Convex4%A.U.C.TDhrystone 2Float + SSE - 1D FFT Size 642 - Process3.1%Float + SSE - 1D FFT Size 4096Float + SSE - 1D FFT Size 256Octane - Google Chrome2.8%GETStock - 2D FFT Size 128garbage2.3%ParboilVP9 libvpx EncodingHackbenchMKL-DNN DNNLPOV-RayCP2K Molecular DynamicsHPC Challengedav1dMBWHackbenchParaViewParaViewMBWNAS Parallel Benchmarksdav1dDaCapo BenchmarkMBWm-queensRust Prime BenchmarkMKL-DNN DNNLGraphicsMagickBRL-CADN-QueensHackbenchTungsten RendererHackbenchApache CassandraHackbenchHackbenchdav1dDaCapo BenchmarkStress-NGC-RayBlenderlibgav1Tungsten Rendererlibgav1Apache SiegeParboilBlenderdav1dBlenderFFTWlibgav1CoremarkTimed MrBayes AnalysisSVT-AV1Stress-NGGROMACSPennantPennantFFTWLeelaChessZeroBlenderGraphicsMagicklibgav1ASKAPHPC ChallengeSmallptBlenderEmbreeStress-NGOSPrayGraphicsMagickx265Timed GCC Compilation7-Zip CompressionGraphicsMagickSVT-AV1GraphicsMagickStockfishOSPrayHPC ChallengeHPC ChallengeNAS Parallel BenchmarksFFTWSeleniumTimed Linux Kernel CompilationRAMspeed SMPGraphicsMagickTimed LLVM CompilationTimed ImageMagick CompilationRodiniaTungsten RendererSeleniumChaos Group V-RAYRAMspeed SMPFFTWTinymembenchFacebook RocksDBIndigoBenchx264OSPrayStress-NGTachyonGo BenchmarksRAMspeed SMPDarktableFacebook RocksDBRenaissanceNeatBenchMKL-DNN DNNLMBWSPECjbb 2015MKL-DNN DNNLOSPrayApache CassandraFFTWDaCapo BenchmarkEmbreeStress-NGPostgreSQL pgbenchEmbreeHPC ChallengeAppleseedasmFishPostgreSQL pgbenchOSBenchRAMspeed SMPASKAPASKAPAppleseedRenaissanceOSPrayMBWBork File EncrypterSPECjbb 2015MBWEmbreeglibc benchJava SciMarkGNU Octave BenchmarkBuild2OSBenchNAMDGIMPHPC ChallengeOptcarrotParaViewParaViewStress-NGCacheBenchRAMspeed SMPFFTWMBWDarktableASKAPRAMspeed SMPIndigoBenchGo BenchmarksGIMPRAMspeed SMPStress-NGSVT-VP9LAME MP3 EncodingSeleniumRodiniaXsbenchNAS Parallel BenchmarksNGINX BenchmarkStreamRAMspeed SMPOpus Codec EncodingOpenCV BenchmarkSVT-VP9glibc benchRAMspeed SMPOSBenchglibc benchFFTWSVT-VP9ctx_clockMKL-DNN DNNLFFTWFFTWCppPerformanceBenchmarksglibc benchQMCPACKSeleniumFFTWrav1eSQLiteRAMspeed SMPFFTWStress-NGSeleniumFFTWFFTWglibc benchSunflow Rendering SystemTimed PHP CompilationFLAC Audio EncodingLuxCoreRenderGIMPSeleniumMBWASKAPNAS Parallel BenchmarksEmbreeEmbreeTensorflowFFTWRust MandelbrotRenaissanceASKAPSeleniumApache CassandraFFTWCacheBenchPHPBenchOSPrayDarktableACES DGEMMHackbenchFFTWMKL-DNN DNNLFacebook RocksDBBullet Physics EngineRadiance BenchmarkOSPrayCppPerformanceBenchmarksHimeno BenchmarkFFTWCppPerformanceBenchmarksGIMPFFTWFFTWBullet Physics EngineRedisMKL-DNN DNNLOSPrayHackbenchFFTWMinionSeleniumSeleniumFFTWdcrawGitTungsten RendererMinionParboilNAS Parallel BenchmarksGraphicsMagickZstd CompressionIntel Open Image DenoiseMKL-DNN DNNLTTSIOD 3D RendererminiFETinymembenchRadiance BenchmarkRenaissanceOSBenchglibc benchDarktableMKL-DNN DNNLFFTWMemtier_benchmarkBullet Physics EngineStreamFFTWHackbenchStreamFFTWSeleniumNAS Parallel BenchmarksBullet Physics EngineDaCapo BenchmarkMinionHPC ChallengeHackbenchSeleniumHPC ChallengeLibreOfficeStreamPyBenchOSBenchMKL-DNN DNNLApache BenchmarkCppPerformanceBenchmarksBullet Physics Engineglibc benchSciMarkCacheBenchSQLiteCppPerformanceBenchmarksBullet Physics EngineAppleseedglibc benchHPC ChallengeGzip CompressionTSCPSeleniumStress-NGSVT-AV1glibc benchStress-NGRenaissanceLeelaChessZeroNAS Parallel BenchmarksLuxCoreRendereSpeak Speech Engineglibc benchGo BenchmarksSeleniumBullet Physics EngineRenaissanceBYTE Unix BenchmarkFFTWHackbenchFFTWFFTWSeleniumRedisFFTWGo BenchmarksThreadripper 3960XCore i9 10980XE

AMD Threadripper 3960X vs. Intel Core i9 10980XEpennant: sedovbigpennant: leblancbigspec-jbb2015: SPECjbb2015-Composite critical-jOPSspec-jbb2015: SPECjbb2015-Composite max-jOPShpcc: G-HPLcassandra: Mixed 1:3cpp-perf-bench: Rand Numbersbuild-gcc: Time To Compilemkl-dnn: Convolution Batch conv_all - f32libgav1: Chimera 1080p 10-bitqmcpack: svt-av1: Enc Mode 0 - 1080pcassandra: Readspgbench: Buffer Test - Normal Load - Read Writemkl-dnn: Deconvolution Batch deconv_all - f32blender: Barbershop - CPU-Onlybrl-cad: VGR Performance Metricfftw: Float + SSE - 2D FFT Size 4096memtier-benchmark: Rediscpp-perf-bench: Math Libraryblender: Pabellon Barcelona - CPU-Onlylibgav1: Chimera 1080prenaissance: Savina Reactors.IOcp2k: Fayalite-FIST Datacompress-lzma: 256MB File Compressiontinymembench: Standard Memsettinymembench: Standard Memcpylibgav1: Summer Nature 4Kdav1d: Chimera 1080p 10-bitblender: Classroom - CPU-Onlyminion: Solitaireradiance: Serialselenium: Jetstream - Firefoxfftw: Stock - 2D FFT Size 4096selenium: Jetstream - Google Chromesqlite: 8luxcorerender: Rainbow Colors and Prismhackbench: 32 - Processselenium: ARES-6 - Firefoxasmfish: 1024 Hash Memory, 26 Depthappleseed: Emilymkl-dnn: Convolution Batch conv_googlenet_v3 - f32cachebench: Read / Modify / Writecachebench: Writecachebench: Readospray: XFrog Forest - SciVisbyte: Dhrystone 2ospray: XFrog Forest - Path Tracerblender: Fishy Cat - CPU-Onlyvpxenc: vpxenc VP9 1080p Video Encodepovray: Trace Timeselenium: StyleBench - Google Chromeospray: San Miguel - Path Tracerminion: Quasigroupgromacs: Water Benchmarkospray: San Miguel - SciViscassandra: Writesmrbayes: Primate Phylogeny Analysisrocksdb: Seq Fillstress-ng: Glibc C String Functionsstress-ng: Context Switchingrenaissance: Akka Unbalanced Cobwebbed Treeselenium: Speedometer - Firefoxappleseed: Material Testerparboil: OpenMP MRI Griddingapache-siege: 200hackbench: 16 - Threadpgbench: Buffer Test - Normal Load - Read Onlycpp-perf-bench: Stepanov Vectorblender: BMW27 - CPU-Onlyopencv-bench: askap: tConvolve MT - Degriddingaskap: tConvolve MT - Griddingbuild-linux-kernel: Time To Compileappleseed: Disney Materialselenium: StyleBench - Firefoxlibgav1: Summer Nature 1080pnpb: EP.Dselenium: Speedometer - Google Chromefftw: Float + SSE - 2D FFT Size 2048hackbench: 8 - Processv-ray: CPUospray: NASA Streamlines - Path Tracerbuild2: Time To Compilenpb: BT.Crenaissance: Apache Spark PageRankindigobench: Bedroomradiance: SMP Parallelluxcorerender: DLSCindigobench: Supercarrocksdb: Read While Writinggraphics-magick: Enhancedgraphics-magick: Resizinggraphics-magick: Noise-Gaussianrocksdb: Rand Readgraphics-magick: Sharpengraphics-magick: Swirlgraphics-magick: Rotategraphics-magick: HWB Color Spacebuild-llvm: Time To Compilesqlite-speedtest: Timed Time - Size 1,000stockfish: Total Timenpb: SP.Bcpp-perf-bench: Atollibreoffice: 20 Documents To PDFnamd: ATPase Simulation - 327,506 Atomsaskap: tConvolve MPI - Degriddingaskap: tConvolve MPI - Griddingrenaissance: Apache Spark Bayesespeak: Text-To-Speech Synthesismbw: Memory Copy - 4096 MiBnginx: Static Web Page Servingrav1e: 1080p To AV1 Video Encodelczero: BLASselenium: ARES-6 - Google Chromemkl-dnn: IP Batch All - f32selenium: Octane - Firefoxfftw: Stock - 2D FFT Size 2048mbw: Memory Copy, Fixed Block Size - 4096 MiBnpb: LU.Cbuild-php: Time To Compileminion: Gracefulselenium: Octane - Google Chromejava-scimark2: Compositecompress-7zip: Compress Speed Testhimeno: Poisson Pressure Solverrenaissance: In-Memory Database Shootoutmkl-dnn: Convolution Batch conv_3d - f32stream: Copyembree: Pathtracer - Asian Dragon Objminife: Smalldcraw: RAW To PPM Image Conversionmkl-dnn: Recurrent Neural Network Training - f32embree: Pathtracer ISPC - Asian Dragon Objrust-mandel: Time To Complete Serial/Parallel Mandelbrotoctave-benchmark: m-queens: Time To Solvecompress-gzip: Linux Source Tree Archiving To .tar.gzmbw: Memory Copy - 1024 MiBhackbench: 16 - Processredis: SETredis: GETgo-benchmark: httphackbench: 4 - Threadapache: Static Web Page Servingcoremark: CoreMark Size 666 - Iterations Per Secondglibc-bench: expdav1d: Chimera 1080paskap: tConvolve OpenMP - Degriddingaskap: tConvolve OpenMP - Griddingstress-ng: CPU Stressstress-ng: Semaphoresstress-ng: Cryptostress-ng: Forkingstress-ng: Vector Mathstress-ng: Matrix Mathstress-ng: Socket Activitystress-ng: Glibc Qsort Data Sortingcpp-perf-bench: Ctypec-ray: Total Time - 4K, 16 Rays Per Pixelmt-dgemm: Sustained Floating-Point Ratecpp-perf-bench: Stepanov Abstractionhackbench: 4 - Processttsiod-renderer: Phong Rendering With Soft-Shadow Mappingembree: Pathtracer - Crownrenaissance: Apache Spark ALStungsten: Non-Exponentialembree: Pathtracer ISPC - Crownscimark2: Compositemkl-dnn: IP Batch 1D - f32tesseract-ocr: Time To OCR 7 Imagesembree: Pathtracer - Asian Dragonnpb: FT.Ctensorflow: Cifar10gimp: unsharp-maskgo-benchmark: buildtungsten: Water Causticcrafty: Elapsed Timeembree: Pathtracer ISPC - Asian Dragonselenium: PSPDFKit WASM - Firefoxpybench: Total For Average Test Timesmkl-dnn: Deconvolution Batch deconv_1d - f32ospray: Magnetic Reconnection - SciVisglibc-bench: cosglibc-bench: sincosglibc-bench: sincompress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9fftw: Float + SSE - 2D FFT Size 1024selenium: PSPDFKit WASM - Google Chromehackbench: 8 - Threadselenium: Maze Solver - Firefoxramspeed: Triad - Floating Pointgimp: auto-levelsbuild-imagemagick: Time To Compileramspeed: Average - Integerfftw: Float + SSE - 1D FFT Size 256ramspeed: Copy - Floating Pointsqlite: 1ramspeed: Scale - Floating Pointramspeed: Add - Floating Pointramspeed: Average - Floating Pointramspeed: Add - Integerramspeed: Copy - Integerramspeed: Scale - Integerramspeed: Triad - Integerparaview: Many Spheres - 2560 x 1440paraview: Many Spheres - 2560 x 1440dav1d: Summer Nature 4Kdacapobench: Tradebeansospray: NASA Streamlines - SciVismkl-dnn: Convolution Batch conv_alexnet - f32xsbench: selenium: Maze Solver - Google Chromefftw: Stock - 1D FFT Size 64encode-flac: WAV To FLACgimp: rotatephpbench: PHP Benchmark Suitesvt-av1: Enc Mode 4 - 1080pneatbench: CPUglibc-bench: ffscpp-perf-bench: Function Objectsgo-benchmark: jsonrodinia: OpenMP LavaMDtungsten: Hairdacapobench: Tradesoapcompress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19dacapobench: H2encode-opus: WAV, Opus Encode, Opus Decodex265: H.265 1080p Video Encodingrust-prime: Prime Number Test To 200,000,000mbw: Memory Copy, Fixed Block Size - 1024 MiBlczero: Randnpb: EP.Cfftw: Stock - 2D FFT Size 1024rodinia: OpenMP CFD Solverglibc-bench: atanhglibc-bench: sqrtglibc-bench: modfglibc-bench: log2glibc-bench: tanhglibc-bench: pthread_onceglibc-bench: asinhglibc-bench: sinhglibc-bench: ffsllgimp: resizesvt-vp9: VMAF Optimized - Bosphorus 1080pnpb: MG.Cparboil: OpenMP Stencilgo-benchmark: garbagedarktable: Boat - CPU-onlydav1d: Summer Nature 1080pfftw: Float + SSE - 2D FFT Size 64fftw: Float + SSE - 1D FFT Size 64encode-mp3: WAV To MP3fftw: Stock - 2D FFT Size 32fftw: Float + SSE - 2D FFT Size 32n-queens: Elapsed Timesunflow: Global Illumination + Image Synthesishackbench: 1 - Processdacapobench: Jythonfftw: Float + SSE - 1D FFT Size 4096ospray: Magnetic Reconnection - Path Tracerbork: File Encryption Timefftw: Float + SSE - 2D FFT Size 512svt-av1: Enc Mode 8 - 1080px264: H.264 Video Encodingparaview: Wavelet Volume - 2560 x 1440paraview: Wavelet Volume - 2560 x 1440tungsten: Volumetric Causticmbw: Memory Copy, Fixed Block Size - 512 MiBfftw: Float + SSE - 1D FFT Size 2048paraview: Wavelet Contour - 2560 x 1440paraview: Wavelet Contour - 2560 x 1440osbench: Create Filesoidn: Memorialgit: Time To Complete Common Git Commandsosbench: Memory Allocationsosbench: Create Processesosbench: Launch Programsosbench: Create Threadsfftw: Stock - 1D FFT Size 4096fftw: Float + SSE - 1D FFT Size 1024fftw: Stock - 2D FFT Size 64darktable: Masskrug - CPU-onlysmallpt: Global Illumination Renderer; 128 Samplesfftw: Float + SSE - 1D FFT Size 512fftw: Float + SSE - 2D FFT Size 256mbw: Memory Copy - 512 MiBfftw: Stock - 2D FFT Size 512bullet: Raytestshackbench: 2 - Threadfftw: Stock - 2D FFT Size 256fftw: Stock - 1D FFT Size 2048fftw: Stock - 2D FFT Size 128fftw: Float + SSE - 2D FFT Size 128hackbench: 2 - Processfftw: Stock - 1D FFT Size 128darktable: Server Room - CPU-onlyfftw: Stock - 1D FFT Size 256fftw: Float + SSE - 1D FFT Size 32fftw: Stock - 1D FFT Size 32mkl-dnn: Deconvolution Batch deconv_3d - f32fftw: Stock - 1D FFT Size 1024optcarrot: Optimized Benchmarkhackbench: 1 - Threadfftw: Stock - 1D FFT Size 512svt-vp9: Visual Quality Optimized - Bosphorus 1080pfftw: Float + SSE - 1D FFT Size 128svt-vp9: PSNR/SSIM Optimized - Bosphorus 1080pdarktable: Server Rack - CPU-onlytachyon: Total Timetscp: AI Chess Performancembw: Memory Copy - 128 MiBparboil: OpenMP CUTCPmbw: Memory Copy, Fixed Block Size - 128 MiBctx-clock: Context Switch Timebullet: Convex Trimeshbullet: Prim Trimeshbullet: 136 Ragdollsbullet: 1000 Convexbullet: 1000 Stackbullet: 3000 Fallhpcc: Max Ping Pong Bandwidthhpcc: Rand Ring Bandwidthhpcc: Rand Ring Latencyhpcc: G-Rand Accesshpcc: EP-STREAM Triadhpcc: G-Ptranshpcc: EP-DGEMMhpcc: G-Fftehpcc: G-Fftestream: Addstream: Triadstream: ScaleThreadripper 3960XCore i9 10980XE6502.5646305.170264155265960.460731711105.427598.9121070.5420.461875.80.0643007813768.0507791772.78237.55393502223142912967.74270.690191.4849.6714138.775419.693226.23517756.910538.022.9987.19150.8960.510068558.724217.576784.1217.5061.4102.2944.56150.6775540266160.6052257.283061655.49410027032059.6992957463063.547.0845423513.13.7284.00201.3616.41036.62.88107.1804942.48638.4623061971.13310303682449871.8315705353.8612584.85697.1136.80498649.06779655723.6322.955662564.29652876.64757.9964.1183307.001911.4128.53887.46424979.074.292214.271333410411.1463940410.7554.08352747.6217245.1193.926172.0792.798.1044877714569226356514368827037913517101462141.55758.4307543452823242.0857.7616.7880.722353311.011865.033355.90727.81119697.47536127.851.36932.734919.9716.7995367397864.311493.07051711.8839.64441.781302474333033.261509154748.2936066759.0326.8473533465.128.78587790.6739.498226.81927.739432.69813.30623.28133.75419435.38222.7371834121.402649622.7086196.03234193.151069837.2286394.84041587.813915.534617.479167.113239494.257072.9380851.0667241.37151388.0518734.34380.7333.08821.0598.54388728.3045.468780.04429.76736643.0793.5642228.3694714.922.8176924.13632.152017832.3917.0420.5111929101375919.3355904172932.151313249682.596112542.020612.137041.416820.35740897150811.6664.437623.8517.12514.53936961.633826035940.8915.73837097.9943043.8440240.9843028.4437204.2538106.3843894.265900.23258.85291.09464352.63142.22750295284.09362.27.83412.3166584088.04335.31.7592714.677190107810.1039.79737336410.22736946.27265.7114.99911380.3221069082208.118097.99.03710.01722.162352.219845.8371710.51971.743958.420527.826341.753078.609382.2225120.087.2434407025006.279677.0647733189007.11711113461234.4690.6852.762396358875333.338.0184304978.792193.762811.522175.724.1564611306.184611122578.928247.4711.53892619.594.76570.59605941.03660648.14624820.8433478818.8599379585.43.4023.675512404261619400.4328684.62.1143483.7368674.19048.18902.3469703.5298569.22.2428883.815042112372.863909356.2106.882.8589114.0326.4127460392.870.1301.6899134141210893.6781.26465710863.4211520.9551880.7987752.1816233.813674.1109413.41694122107.9133.364000.456830.142921.822285.5402231.6369010.7307710.7307736666.436725.133514.64023.5403865.417183113847159.38110203964.613934.7311168.4312.841505.10.061564269752.3847671400.03373.59203716180172679835.95273.810331.8527.9320542.519955.802222.94926498.89578.013.6248.25268.4667.861407614.512193.305536.2206.7158.0332.7952.22357.0853386196227.6508966.841752123.62252014334036.9397560954064.634.5947073027.62.51142.9780.9137.79235.22.4994.8832431.51427.88160569119.40315004551479518.8619329847.8912112.80979.9129.57724126.54025897360.6143.198464274.74445178.06592.7481.7082378.471584.3343.379120.96102361.542.512479.621122284424.670262746.8672.58941131.7320511.3072.640199.9352.916.2074184688367149336696435581237694795934214.45857.9684888835710924.4354.2687.2540.968972385.571568.433516.99028.9838836.11146249.341.10734.226421.3715.7565341075969.37965.85242920.1648.48644.821723461422240.80967964133.5484679344.83313.425442773.020.18427074.2535.174157.69223.094038.93717.98447.60032.0478972.17241.5462079297.962722488.67896211.17132075.14633910.0843094.91715276.895128.682908.066897.803402838.843929.12104506.2547063.29144369.1012670.41242.4526.31137.5877.28081829.94810.407707.55818.94566041.9656.7765220.8831759.076.6373923.85622.422718608.7514.2524.9852511189633121.7014919454326.883516289071.7921229.4140.066112.888039.819820.12625471186927.4906.628912.6519.55921.94729049.733934529165.4219.43228379.328825.9728822.829367.5529115.628779.328909.785851.91658.37170.75973636.60125.46339192806.17395.59.57016.4507755155.19824.31.3824412.746280687615.24814.7793482511.39366658.01341.997.3888370.68366812.82377.797139.011.59810.16951.593311.812556.4058511.10571.396568.547107.742191.3821711.204300.0716348.608.1211547189329.160297.8038956195779.1439025.3389398.6270.8382.439425760652377.785.8392550147.291130.572115.006132.197.319398210.910660811173.005112.5610.81632521.815.35077.53531132.25326535.87961214.9122888725.2680937578.54.4575.81455196279049270.7497561.62.2368703.9997671.48921.49138.5327223.6387338.52.6357430.5169128866.72.596959279.2142.132.6449149.0252.7225195310.090.1422.496314115298969.5842.2014428285.5551201.0158920.8590132.4737123.9644534.4665053.97716220672.6951.478510.288610.107102.596325.2482333.918736.977526.9775233985.033869.131362.1OpenBenchmarking.org

Pennant

Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigCore i9 10980XEThreadripper 3960X14002800420056007000SE +/- 0.50, N = 3SE +/- 2.00, N = 34023.546502.561. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbigCore i9 10980XEThreadripper 3960X14002800420056007000SE +/- 0.31, N = 3SE +/- 12.26, N = 33865.426305.171. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi

SPECjbb 2015

This is a benchmark of SPECjbb 2015. For this test profile to work, you must have a valid license/copy of the SPECjbb 2015 ISO (SPECjbb2015-1.02.iso) in your Phoronix Test Suite download cache. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgjOPS, More Is BetterSPECjbb 2015SPECjbb2015-Composite critical-jOPSThreadripper 3960XCore i9 10980XE6K12K18K24K30K2641518311

OpenBenchmarking.orgjOPS, More Is BetterSPECjbb 2015SPECjbb2015-Composite max-jOPSThreadripper 3960XCore i9 10980XE11K22K33K44K55K5265938471

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-HPLThreadripper 3960XCore i9 10980XE1428425670SE +/- 0.16, N = 3SE +/- 0.51, N = 360.4659.381. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

Apache Cassandra

This is a benchmark of the Apache Cassandra NoSQL database management system making use of cassandra-stress. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 3.11.4Test: Mixed 1:3Core i9 10980XEThreadripper 3960X4080120160200SE +/- 10.30, N = 9SE +/- 5.83, N = 7203171

CppPerformanceBenchmarks

CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCppPerformanceBenchmarks 9Test: Random NumbersCore i9 10980XEThreadripper 3960X2004006008001000SE +/- 0.10, N = 3SE +/- 0.65, N = 3964.611105.431. (CXX) g++ options: -std=c++11 -O3

Timed GCC Compilation

This test times how long it takes to build the GNU Compiler Collection (GCC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GCC Compilation 8.2Time To CompileThreadripper 3960XCore i9 10980XE2004006008001000SE +/- 1.15, N = 3SE +/- 1.54, N = 3598.91934.73

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_all - Data Type: f32Threadripper 3960XCore i9 10980XE30060090012001500SE +/- 2.64, N = 3SE +/- 0.32, N = 31070.541168.43MIN: 997.44MIN: 1161.471. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

libgav1

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

OpenBenchmarking.orgFPS, More Is Betterlibgav1 2019-10-05Video Input: Chimera 1080p 10-bitThreadripper 3960XCore i9 10980XE510152025SE +/- 0.11, N = 3SE +/- 0.01, N = 320.4612.841. (CXX) g++ options: -O3 -lpthread

QMCPACK

QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.8Core i9 10980XEThreadripper 3960X4008001200160020001505.11875.81. (CXX) g++ options: -fopenmp -fomit-frame-pointer -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -ffast-math -lm

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.7Encoder Mode: Enc Mode 0 - Input: 1080pThreadripper 3960XCore i9 10980XE0.01440.02880.04320.05760.072SE +/- 0.000, N = 3SE +/- 0.000, N = 60.0640.0611. (CXX) g++ options: -fPIE -fPIC -pie

Apache Cassandra

This is a benchmark of the Apache Cassandra NoSQL database management system making use of cassandra-stress. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 3.11.4Test: ReadsCore i9 10980XEThreadripper 3960X12K24K36K48K60KSE +/- 10790.29, N = 9SE +/- 12892.81, N = 95642630078

PostgreSQL pgbench

This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read WriteThreadripper 3960XCore i9 10980XE3K6K9K12K15KSE +/- 1254.90, N = 12SE +/- 109.44, N = 1513768.059752.381. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_all - Data Type: f32Core i9 10980XEThreadripper 3960X400800120016002000SE +/- 0.30, N = 3SE +/- 6.22, N = 31400.031772.78MIN: 1395.48MIN: 16771. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.81Blend File: Barbershop - Compute: CPU-OnlyThreadripper 3960XCore i9 10980XE80160240320400SE +/- 0.15, N = 3SE +/- 0.27, N = 3237.55373.59

BRL-CAD

BRL-CAD 7.28.0 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 7.28.0VGR Performance MetricThreadripper 3960XCore i9 10980XE80K160K240K320K400K3935022037161. (CXX) g++ options: -std=c++98 -pipe -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -rdynamic -lSM -lICE -lXi -lGLU -lGL -lX11 -lXext -lXrender -lpthread -ldl -luuid -lm

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096Threadripper 3960XCore i9 10980XE5K10K15K20K25KSE +/- 60.75, N = 3SE +/- 95.47, N = 322314180171. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Memtier_benchmark

Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool. This current test profile currently just stresses the Redis protocol and basic options exposed wotj a 1:1 Set/Get ratio, 30 pipeline, 100 clients per thread, and thread count equal to the number of CPU cores/threads present. Patches to extend the test are welcome as always. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemtier_benchmark 1.2.17Protocol: RedisThreadripper 3960XCore i9 10980XE600K1200K1800K2400K3000KSE +/- 26919.58, N = 3SE +/- 42755.58, N = 152912967.742679835.951. (CXX) g++ options: -O2 -levent -lpthread -lz -lpcre

CppPerformanceBenchmarks

CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCppPerformanceBenchmarks 9Test: Math LibraryThreadripper 3960XCore i9 10980XE60120180240300SE +/- 2.14, N = 3SE +/- 0.43, N = 3270.69273.811. (CXX) g++ options: -std=c++11 -O3

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.81Blend File: Pabellon Barcelona - Compute: CPU-OnlyThreadripper 3960XCore i9 10980XE70140210280350SE +/- 0.19, N = 3SE +/- 0.26, N = 3191.48331.85

libgav1

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

OpenBenchmarking.orgFPS, More Is Betterlibgav1 2019-10-05Video Input: Chimera 1080pThreadripper 3960XCore i9 10980XE1122334455SE +/- 0.18, N = 3SE +/- 0.01, N = 349.6727.931. (CXX) g++ options: -O3 -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 0.9.0Test: Savina Reactors.IOThreadripper 3960XCore i9 10980XE4K8K12K16K20KSE +/- 209.59, N = 40SE +/- 347.62, N = 4014138.7820542.52

CP2K Molecular Dynamics

CP2K is an open-source molecular dynamics software package focused on quantum chemistry and solid-state physics. This test profile currently makes use of the OpenMP implementation and using the Fayalite-FIST molecular dynamics run and measures the total time to complete. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 6.1Fayalite-FIST DataThreadripper 3960XCore i9 10980XE2004006008001000419.69955.80

LZMA Compression

This test measures the time needed to compress a file using LZMA compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterLZMA Compression256MB File CompressionCore i9 10980XEThreadripper 3960X50100150200250SE +/- 1.12, N = 3SE +/- 0.29, N = 3222.95226.241. (CXX) g++ options: -O2

Tinymembench

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemsetCore i9 10980XEThreadripper 3960X6K12K18K24K30KSE +/- 390.55, N = 3SE +/- 18.62, N = 326498.817756.91. (CC) gcc options: -O2 -lm

OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemcpyThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 165.90, N = 3SE +/- 147.18, N = 310538.09578.01. (CC) gcc options: -O2 -lm

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 2019-10-05Video Input: Summer Nature 4KThreadripper 3960XCore i9 10980XE612182430SE +/- 0.05, N = 3SE +/- 0.01, N = 322.9913.621. (CXX) g++ options: -O3 -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 0.5.0Video Input: Chimera 1080p 10-bitThreadripper 3960XCore i9 10980XE20406080100SE +/- 0.14, N = 3SE +/- 0.10, N = 387.1948.25MIN: 54.26 / MAX: 180.02MIN: 32.68 / MAX: 103.581. (CC) gcc options: -pthread

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.81Blend File: Classroom - Compute: CPU-OnlyThreadripper 3960XCore i9 10980XE60120180240300SE +/- 0.20, N = 3SE +/- 0.25, N = 3150.89268.46

Minion

Minion is an open-source constraint solver that is designed to be very scalable. This test profile uses Minion's integrated benchmarking problems to solve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMinion 1.8Benchmark: SolitaireThreadripper 3960XCore i9 10980XE1530456075SE +/- 0.35, N = 3SE +/- 0.91, N = 1560.5167.861. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic

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: SerialThreadripper 3960XCore i9 10980XE130260390520650558.72614.51

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: Jetstream - Browser: FirefoxThreadripper 3960XCore i9 10980XE50100150200250SE +/- 0.36, N = 3SE +/- 0.63, N = 3217.57193.301. firefox 70.0.1

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 4096Threadripper 3960XCore i9 10980XE15003000450060007500SE +/- 8.75, N = 3SE +/- 21.00, N = 36784.15536.21. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: Jetstream - Browser: Google ChromeThreadripper 3960XCore i9 10980XE50100150200250SE +/- 0.77, N = 3SE +/- 0.52, N = 3217.50206.711. chrome 78.0.3904.108

SQLite

This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.30.1Threads / Copies: 8Core i9 10980XEThreadripper 3960X1428425670SE +/- 4.82, N = 15SE +/- 0.32, N = 358.0361.411. (CC) gcc options: -O2 -lz -lm -ldl -lpthread

LuxCoreRender

LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on the CPU as opposed to the OpenCL version. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.2Scene: Rainbow Colors and PrismCore i9 10980XEThreadripper 3960X0.62781.25561.88342.51123.139SE +/- 0.03, N = 3SE +/- 0.15, N = 122.792.29MIN: 2.71 / MAX: 2.88MIN: 1.36 / MAX: 2.94

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 32 - Type: ProcessThreadripper 3960XCore i9 10980XE1224364860SE +/- 0.08, N = 3SE +/- 2.07, N = 1544.5652.221. (CC) gcc options: -lpthread

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: ARES-6 - Browser: FirefoxThreadripper 3960XCore i9 10980XE1326395265SE +/- 0.41, N = 3SE +/- 0.60, N = 350.6757.081. firefox 70.0.1

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthThreadripper 3960XCore i9 10980XE16M32M48M64M80MSE +/- 575831.34, N = 3SE +/- 332072.50, N = 37554026653386196

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: EmilyThreadripper 3960XCore i9 10980XE50100150200250160.61227.65

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32Threadripper 3960XCore i9 10980XE1530456075SE +/- 0.28, N = 3SE +/- 0.05, N = 357.2866.84MIN: 51.08MIN: 66.111. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

CacheBench

This is a performance test of CacheBench, which is part of LLCbench. CacheBench is designed to test the memory and cache bandwidth performance Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterCacheBenchTest: Read / Modify / WriteThreadripper 3960XCore i9 10980XE13K26K39K52K65KSE +/- 95.04, N = 3SE +/- 18.18, N = 361655.4952123.62MIN: 55797.19 / MAX: 66291.68MIN: 44592.74 / MAX: 54711.161. (CC) gcc options: -lrt

OpenBenchmarking.orgMB/s, More Is BetterCacheBenchTest: WriteCore i9 10980XEThreadripper 3960X7K14K21K28K35KSE +/- 1.58, N = 3SE +/- 55.58, N = 334036.9432059.70MIN: 30183.01 / MAX: 36380.95MIN: 28046.66 / MAX: 33459.41. (CC) gcc options: -lrt

OpenBenchmarking.orgMB/s, More Is BetterCacheBenchTest: ReadCore i9 10980XEThreadripper 3960X9001800270036004500SE +/- 0.05, N = 3SE +/- 3.47, N = 34064.633063.54MIN: 4055.03 / MAX: 4068.9MIN: 3047.77 / MAX: 3069.921. (CC) gcc options: -lrt

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: XFrog Forest - Renderer: SciVisThreadripper 3960XCore i9 10980XE246810SE +/- 0.01, N = 4SE +/- 0.00, N = 127.084.59MIN: 6.94 / MAX: 7.19MIN: 4.35 / MAX: 4.63

BYTE Unix Benchmark

This is a test of BYTE. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgLPS, More Is BetterBYTE Unix Benchmark 3.6Computational Test: Dhrystone 2Core i9 10980XEThreadripper 3960X10M20M30M40M50MSE +/- 229834.08, N = 3SE +/- 405676.33, N = 347073027.645423513.1

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: XFrog Forest - Renderer: Path TracerThreadripper 3960XCore i9 10980XE0.8371.6742.5113.3484.185SE +/- 0.01, N = 3SE +/- 0.00, N = 63.722.51MIN: 3.66 / MAX: 3.76MIN: 2.44 / MAX: 2.53

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.81Blend File: Fishy Cat - Compute: CPU-OnlyThreadripper 3960XCore i9 10980XE306090120150SE +/- 0.11, N = 3SE +/- 0.12, N = 384.00142.97

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9/WebM format using a sample 1080p video. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.8.1vpxenc VP9 1080p Video EncodeThreadripper 3960XCore i9 10980XE4080120160200SE +/- 0.93, N = 3SE +/- 0.63, N = 3201.3680.911. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11

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 TimeThreadripper 3960XCore i9 10980XE918273645SE +/- 0.06, N = 3SE +/- 1.83, N = 1516.4137.791. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSDL -lXpm -lSM -lICE -lX11 -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lpthread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRuns / Minute, More Is BetterSeleniumBenchmark: StyleBench - Browser: Google ChromeThreadripper 3960XCore i9 10980XE816243240SE +/- 0.09, N = 3SE +/- 0.12, N = 336.635.21. chrome 78.0.3904.108

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 TracerThreadripper 3960XCore i9 10980XE0.6481.2961.9442.5923.24SE +/- 0.00, N = 3SE +/- 0.00, N = 32.882.49MIN: 2.87 / MAX: 2.92MIN: 2.42 / MAX: 2.51

Minion

Minion is an open-source constraint solver that is designed to be very scalable. This test profile uses Minion's integrated benchmarking problems to solve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMinion 1.8Benchmark: QuasigroupCore i9 10980XEThreadripper 3960X20406080100SE +/- 0.02, N = 3SE +/- 0.62, N = 394.88107.181. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic

GROMACS

The Gromacs molecular dynamics package testing on the CPU with the water_GMX50 data. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2019.4Water BenchmarkThreadripper 3960XCore i9 10980XE0.55941.11881.67822.23762.797SE +/- 0.004, N = 3SE +/- 0.001, N = 32.4861.514-mavx2-mavx512f1. (CXX) g++ options: -mfma -std=c++11 -O3 -funroll-all-loops -pthread -lrt -lpthread -lm

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: SciVisThreadripper 3960XCore i9 10980XE918273645SE +/- 0.00, N = 12SE +/- 0.10, N = 838.4627.88MIN: 35.71 / MAX: 40MIN: 26.32 / MAX: 28.57

Apache Cassandra

This is a benchmark of the Apache Cassandra NoSQL database management system making use of cassandra-stress. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 3.11.4Test: WritesThreadripper 3960XCore i9 10980XE50K100K150K200K250KSE +/- 3283.40, N = 4SE +/- 1499.53, N = 3230619160569

Timed MrBayes Analysis

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisThreadripper 3960XCore i9 10980XE306090120150SE +/- 0.28, N = 3SE +/- 0.43, N = 371.13119.40-msse4a -msha-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm

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 6.3.6Test: Sequential FillCore i9 10980XEThreadripper 3960X300K600K900K1200K1500KSE +/- 20977.69, N = 15SE +/- 13038.81, N = 4150045510303681. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Glibc C String FunctionsThreadripper 3960XCore i9 10980XE500K1000K1500K2000K2500KSE +/- 28050.29, N = 3SE +/- 28903.19, N = 152449871.831479518.861. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Context SwitchingCore i9 10980XEThreadripper 3960X4M8M12M16M20MSE +/- 307514.51, N = 3SE +/- 214670.64, N = 1519329847.8915705353.861. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

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 0.9.0Test: Akka Unbalanced Cobwebbed TreeCore i9 10980XEThreadripper 3960X3K6K9K12K15KSE +/- 114.18, N = 32SE +/- 127.16, N = 812112.8112584.86

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRuns Per Minute, More Is BetterSeleniumBenchmark: Speedometer - Browser: FirefoxThreadripper 3960XCore i9 10980XE20406080100SE +/- 0.27, N = 3SE +/- 0.66, N = 397.179.91. firefox 70.0.1

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 TesterCore i9 10980XEThreadripper 3960X306090120150129.58136.80

Parboil

The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP MRI GriddingThreadripper 3960XCore i9 10980XE306090120150SE +/- 0.16, N = 3SE +/- 1.07, N = 349.07126.541. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp

Apache Siege

This is a test of the Apache web server performance being facilitated by the Siege web serverb enchmark program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTransactions Per Second, More Is BetterApache Siege 2.4.29Concurrent Users: 200Core i9 10980XEThreadripper 3960X20K40K60K80K100KSE +/- 512.43, N = 3SE +/- 2566.16, N = 1297360.6155723.631. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ThreadThreadripper 3960XCore i9 10980XE1020304050SE +/- 0.01, N = 3SE +/- 0.41, N = 1022.9643.201. (CC) gcc options: -lpthread

PostgreSQL pgbench

This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read OnlyThreadripper 3960XCore i9 10980XE140K280K420K560K700KSE +/- 1102.78, N = 3SE +/- 432.10, N = 3662564.30464274.741. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

CppPerformanceBenchmarks

CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCppPerformanceBenchmarks 9Test: Stepanov VectorThreadripper 3960XCore i9 10980XE20406080100SE +/- 0.17, N = 3SE +/- 0.02, N = 376.6578.071. (CXX) g++ options: -std=c++11 -O3

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.81Blend File: BMW27 - Compute: CPU-OnlyThreadripper 3960XCore i9 10980XE20406080100SE +/- 0.06, N = 3SE +/- 0.14, N = 357.9992.74

OpenCV Benchmark

Stress benchmark tests to measure time consumed by the OpenCV libraries installed Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenCV Benchmark 3.3.0Threadripper 3960XCore i9 10980XE20406080100SE +/- 0.13, N = 3SE +/- 0.43, N = 364.1281.711. (CXX) g++ options: -std=c++11 -rdynamic

ASKAP

This is a CUDA benchmark of ATNF's ASKAP Benchmark with currently using the tConvolveCuda sub-test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 2018-11-10Test: tConvolve MT - DegriddingThreadripper 3960XCore i9 10980XE7001400210028003500SE +/- 8.26, N = 3SE +/- 0.68, N = 33307.002378.471. (CXX) g++ options: -lpthread

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 2018-11-10Test: tConvolve MT - GriddingThreadripper 3960XCore i9 10980XE400800120016002000SE +/- 9.48, N = 3SE +/- 0.52, N = 31911.411584.331. (CXX) g++ options: -lpthread

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.4Time To CompileThreadripper 3960XCore i9 10980XE1020304050SE +/- 0.39, N = 4SE +/- 0.47, N = 728.5443.38

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 MaterialThreadripper 3960XCore i9 10980XE30609012015087.46120.96

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRuns / Minute, More Is BetterSeleniumBenchmark: StyleBench - Browser: FirefoxThreadripper 3960XCore i9 10980XE20406080100SE +/- 0.32, N = 3SE +/- 0.32, N = 379.061.51. firefox 70.0.1

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 2019-10-05Video Input: Summer Nature 1080pThreadripper 3960XCore i9 10980XE1632486480SE +/- 0.15, N = 3SE +/- 0.03, N = 374.2942.511. (CXX) g++ options: -O3 -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: EP.DCore i9 10980XEThreadripper 3960X5001000150020002500SE +/- 33.55, N = 4SE +/- 7.75, N = 32479.622214.271. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 3.1.3

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRuns Per Minute, More Is BetterSeleniumBenchmark: Speedometer - Browser: Google ChromeThreadripper 3960XCore i9 10980XE306090120150SE +/- 0.33, N = 3SE +/- 0.33, N = 31331121. chrome 78.0.3904.108

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 2048Threadripper 3960XCore i9 10980XE7K14K21K28K35KSE +/- 375.01, N = 3SE +/- 68.19, N = 334104228441. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ProcessThreadripper 3960XCore i9 10980XE612182430SE +/- 0.07, N = 3SE +/- 0.27, N = 1511.1524.671. (CC) gcc options: -lpthread

Chaos Group V-RAY

This is a test of Chaos Group's V-RAY benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgKsamples, More Is BetterChaos Group V-RAY 4.10.03Mode: CPUThreadripper 3960XCore i9 10980XE8K16K24K32K40KSE +/- 88.94, N = 3SE +/- 28.26, N = 33940426274

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 TracerThreadripper 3960XCore i9 10980XE3691215SE +/- 0.00, N = 12SE +/- 0.01, N = 410.756.86MIN: 10.53 / MAX: 10.99MIN: 6.21 / MAX: 7.04

Build2

This test profile measures the time to bootstrap/install the build2 C++ build toolchain from source. Build2 is a cross-platform build toolchain for C/C++ code and features Cargo-like features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.12Time To CompileThreadripper 3960XCore i9 10980XE1632486480SE +/- 0.10, N = 3SE +/- 0.20, N = 354.0872.59

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.CThreadripper 3960XCore i9 10980XE11K22K33K44K55KSE +/- 720.74, N = 3SE +/- 21.23, N = 352747.6241131.731. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 3.1.3

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 0.9.0Test: Apache Spark PageRankThreadripper 3960XCore i9 10980XE4K8K12K16K20KSE +/- 108.55, N = 8SE +/- 134.35, N = 817245.1220511.31

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 4.0.64Scene: BedroomThreadripper 3960XCore i9 10980XE0.88341.76682.65023.53364.417SE +/- 0.010, N = 3SE +/- 0.002, N = 33.9262.640

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 ParallelThreadripper 3960XCore i9 10980XE4080120160200172.08199.94

LuxCoreRender

LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on the CPU as opposed to the OpenCL version. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.2Scene: DLSCCore i9 10980XEThreadripper 3960X0.65481.30961.96442.61923.274SE +/- 0.03, N = 3SE +/- 0.01, N = 32.912.79MIN: 2.79 / MAX: 3.05MIN: 2.72 / MAX: 2.85

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 4.0.64Scene: SupercarThreadripper 3960XCore i9 10980XE246810SE +/- 0.003, N = 3SE +/- 0.008, N = 38.1046.207

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 6.3.6Test: Read While WritingThreadripper 3960XCore i9 10980XE1000K2000K3000K4000K5000KSE +/- 40449.58, N = 3SE +/- 69881.71, N = 3487771441846881. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

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: EnhancedThreadripper 3960XCore i9 10980XE120240360480600SE +/- 0.67, N = 35693671. (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: ResizingThreadripper 3960XCore i9 10980XE5001000150020002500SE +/- 12.39, N = 3SE +/- 4.84, N = 3226314931. (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: Noise-GaussianThreadripper 3960XCore i9 10980XE120240360480600SE +/- 0.33, N = 35653661. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -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 6.3.6Test: Random ReadThreadripper 3960XCore i9 10980XE30M60M90M120M150MSE +/- 1282448.37, N = 3SE +/- 337442.71, N = 3143688270964355811. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

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: SharpenThreadripper 3960XCore i9 10980XE80160240320400SE +/- 2.19, N = 33792371. (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: SwirlThreadripper 3960XCore i9 10980XE30060090012001500SE +/- 5.49, N = 313516941. (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: RotateCore i9 10980XEThreadripper 3960X2004006008001000SE +/- 13.69, N = 3SE +/- 3.84, N = 37957101. (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: HWB Color SpaceThreadripper 3960XCore i9 10980XE30060090012001500SE +/- 1.86, N = 3SE +/- 1.20, N = 314629341. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 6.0.1Time To CompileThreadripper 3960XCore i9 10980XE50100150200250141.56214.46

SQLite Speedtest

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

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Core i9 10980XEThreadripper 3960X1326395265SE +/- 0.04, N = 3SE +/- 0.15, N = 357.9758.431. (CC) gcc options: -O2 -ldl -lz -lpthread

Stockfish

This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 9Total TimeThreadripper 3960XCore i9 10980XE16M32M48M64M80MSE +/- 740558.08, N = 3SE +/- 616852.34, N = 375434528488883571. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -msse -msse3 -mpopcnt -flto

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.BThreadripper 3960XCore i9 10980XE5K10K15K20K25KSE +/- 269.46, N = 15SE +/- 19.65, N = 323242.0810924.431. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 3.1.3

CppPerformanceBenchmarks

CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCppPerformanceBenchmarks 9Test: AtolCore i9 10980XEThreadripper 3960X1326395265SE +/- 0.00, N = 3SE +/- 0.19, N = 354.2757.761. (CXX) g++ options: -std=c++11 -O3

LibreOffice

Various benchmarking operations with the LibreOffice open-source office suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterLibreOfficeTest: 20 Documents To PDFThreadripper 3960XCore i9 10980XE246810SE +/- 0.054, N = 24SE +/- 0.075, N = 236.7887.2541. LibreOffice 6.3.3.2 30(Build:2)

NAMD

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.13b1ATPase Simulation - 327,506 AtomsThreadripper 3960XCore i9 10980XE0.2180.4360.6540.8721.09SE +/- 0.00128, N = 3SE +/- 0.00236, N = 30.722350.96897

ASKAP

This is a CUDA benchmark of ATNF's ASKAP Benchmark with currently using the tConvolveCuda sub-test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 2018-11-10Test: tConvolve MPI - DegriddingThreadripper 3960XCore i9 10980XE7001400210028003500SE +/- 9.46, N = 3SE +/- 0.69, N = 33311.012385.571. (CXX) g++ options: -lpthread

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 2018-11-10Test: tConvolve MPI - GriddingThreadripper 3960XCore i9 10980XE400800120016002000SE +/- 4.29, N = 3SE +/- 0.17, N = 31865.031568.431. (CXX) g++ options: -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 0.9.0Test: Apache Spark BayesThreadripper 3960XCore i9 10980XE8001600240032004000SE +/- 30.14, N = 40SE +/- 33.13, N = 83355.913516.99

eSpeak 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak Speech Engine 1.48.04Text-To-Speech SynthesisThreadripper 3960XCore i9 10980XE714212835SE +/- 0.48, N = 3SE +/- 0.30, N = 827.8128.981. (CXX) g++ options: -lstdc++ -lespeak -O2 -fpic -fvisibility=hidden -pedantic -fno-exceptions

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 4096 MiBThreadripper 3960XCore i9 10980XE4K8K12K16K20KSE +/- 33.30, N = 3SE +/- 109.62, N = 519697.488836.111. (CC) gcc options: -O3 -march=native

NGINX Benchmark

This is a test of ab, which is the Apache Benchmark program running against nginx. This test profile measures how many requests per second a given system can sustain when carrying out 2,000,000 requests with 500 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterNGINX Benchmark 1.9.9Static Web Page ServingCore i9 10980XEThreadripper 3960X10K20K30K40K50KSE +/- 190.20, N = 3SE +/- 62.41, N = 346249.3436127.851. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native

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 0.11080p To AV1 Video EncodeThreadripper 3960XCore i9 10980XE0.3080.6160.9241.2321.54SE +/- 0.002, N = 3SE +/- 0.003, N = 31.3691.107

LeelaChessZero

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

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.22.0Backend: BLASCore i9 10980XEThreadripper 3960X816243240SE +/- 0.79, N = 15SE +/- 0.94, N = 1234.2332.731. (CXX) g++ options: -lpthread

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: ARES-6 - Browser: Google ChromeThreadripper 3960XCore i9 10980XE510152025SE +/- 0.05, N = 3SE +/- 0.10, N = 319.9721.371. chrome 78.0.3904.108

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch All - Data Type: f32Core i9 10980XEThreadripper 3960X48121620SE +/- 0.03, N = 3SE +/- 0.12, N = 315.7616.80MIN: 15.35MIN: 15.381. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGeometric Mean, More Is BetterSeleniumBenchmark: Octane - Browser: FirefoxThreadripper 3960XCore i9 10980XE8K16K24K32K40KSE +/- 203.61, N = 3SE +/- 244.23, N = 336739341071. firefox 70.0.1

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 2048Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 3.11, N = 3SE +/- 88.60, N = 37864.35969.31. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiBThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 4.86, N = 3SE +/- 103.12, N = 311493.077965.851. (CC) gcc options: -O3 -march=native

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.CThreadripper 3960XCore i9 10980XE11K22K33K44K55KSE +/- 232.22, N = 3SE +/- 9.92, N = 351711.8842920.161. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 3.1.3

Timed PHP Compilation

This test times how long it takes to build PHP 5 with the Zend engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 7.1.9Time To CompileThreadripper 3960XCore i9 10980XE1122334455SE +/- 0.04, N = 3SE +/- 0.13, N = 339.6448.491. (CC) gcc options: -O2 -pedantic -ldl -lz -lm

Minion

Minion is an open-source constraint solver that is designed to be very scalable. This test profile uses Minion's integrated benchmarking problems to solve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMinion 1.8Benchmark: GracefulThreadripper 3960XCore i9 10980XE1020304050SE +/- 0.26, N = 3SE +/- 0.29, N = 341.7844.821. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGeometric Mean, More Is BetterSeleniumBenchmark: Octane - Browser: Google ChromeThreadripper 3960XCore i9 10980XE10K20K30K40K50KSE +/- 57.21, N = 3SE +/- 212.58, N = 347433461421. chrome 78.0.3904.108

Java SciMark

This test runs the Java version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This benchmark is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterJava SciMark 2.0Computational Test: CompositeThreadripper 3960XCore i9 10980XE7001400210028003500SE +/- 3.56, N = 4SE +/- 4.50, N = 43033.262240.80

7-Zip Compression

This is a test of 7-Zip using p7zip with its integrated benchmark feature or upstream 7-Zip for the Windows x64 build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 16.02Compress Speed TestThreadripper 3960XCore i9 10980XE30K60K90K120K150KSE +/- 539.47, N = 3SE +/- 568.50, N = 3150915967961. (CXX) g++ options: -pipe -lpthread

Himeno Benchmark

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

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverThreadripper 3960XCore i9 10980XE10002000300040005000SE +/- 50.65, N = 3SE +/- 16.83, N = 34748.294133.551. (CC) gcc options: -O3 -mavx2

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 0.9.0Test: In-Memory Database ShootoutThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 52.85, N = 8SE +/- 83.32, N = 86759.039344.83

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_3d - Data Type: f32Threadripper 3960XCore i9 10980XE3691215SE +/- 0.08494, N = 3SE +/- 0.01201, N = 36.8473513.42540MIN: 5.9MIN: 13.241. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Stream

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: CopyCore i9 10980XEThreadripper 3960X9K18K27K36K45KSE +/- 6.45, N = 5SE +/- 15.88, N = 542773.033465.11. (CC) gcc options: -O3 -march=native -fopenmp

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer - Model: Asian Dragon ObjThreadripper 3960XCore i9 10980XE714212835SE +/- 0.01, N = 3SE +/- 0.01, N = 328.7920.18MIN: 28.45 / MAX: 29.34MIN: 20.1 / MAX: 20.39

miniFE

MiniFE Finite Element is an application for unstructured implicit finite element codes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 16.76, N = 3SE +/- 2.02, N = 37790.677074.251. (CXX) g++ options: -O3 -fopenmp -pthread -lmpi_cxx -lmpi

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 ConversionCore i9 10980XEThreadripper 3960X918273645SE +/- 0.02, N = 3SE +/- 0.06, N = 335.1739.501. (CC) gcc options: -lm

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Recurrent Neural Network Training - Data Type: f32Core i9 10980XEThreadripper 3960X50100150200250SE +/- 0.06, N = 3SE +/- 1.80, N = 3157.69226.82MIN: 156.3MIN: 196.671. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Asian Dragon ObjThreadripper 3960XCore i9 10980XE714212835SE +/- 0.02, N = 3SE +/- 0.01, N = 327.7423.09MIN: 27.42 / MAX: 28.3MIN: 22.98 / MAX: 23.33

Rust Mandelbrot

This test profile is of the combined time for the serial and parallel Mandelbrot sets written in Rustlang via willi-kappler/mandel-rust. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRust MandelbrotTime To Complete Serial/Parallel MandelbrotThreadripper 3960XCore i9 10980XE918273645SE +/- 0.11, N = 3SE +/- 0.01, N = 332.7038.941. (CC) gcc options: -m64 -pie -nodefaultlibs -lutil -ldl -lrt -lpthread -lgcc_s -lc -lm

GNU Octave Benchmark

This test profile measures how long it takes to complete several reference GNU Octave files via octave-benchmark. GNU Octave is used for numerical computations and is an open-source alternative to MATLAB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGNU Octave Benchmark 4.4.1Threadripper 3960XCore i9 10980XE48121620SE +/- 0.11, N = 12SE +/- 0.07, N = 313.3117.98

m-queens

A solver for the N-queens problem with multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterm-queens 1.2Time To SolveThreadripper 3960XCore i9 10980XE1122334455SE +/- 0.02, N = 3SE +/- 0.01, N = 323.2847.601. (CXX) g++ options: -fopenmp -O2 -march=native

Gzip Compression

This test measures the time needed to archive/compress two copies of the Linux 4.13 kernel source tree using Gzip compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGzip CompressionLinux Source Tree Archiving To .tar.gzCore i9 10980XEThreadripper 3960X816243240SE +/- 0.07, N = 3SE +/- 0.01, N = 332.0533.75

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 1024 MiBThreadripper 3960XCore i9 10980XE4K8K12K16K20KSE +/- 19.92, N = 3SE +/- 72.16, N = 1519435.388972.171. (CC) gcc options: -O3 -march=native

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ProcessThreadripper 3960XCore i9 10980XE918273645SE +/- 0.07, N = 3SE +/- 0.35, N = 322.7441.551. (CC) gcc options: -lpthread

Redis

Redis is an open-source data structure server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 5.0.5Test: SETCore i9 10980XEThreadripper 3960X400K800K1200K1600K2000KSE +/- 17605.46, N = 3SE +/- 31778.19, N = 152079297.961834121.401. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 5.0.5Test: GETCore i9 10980XEThreadripper 3960X600K1200K1800K2400K3000KSE +/- 15007.86, N = 3SE +/- 58191.94, N = 152722488.672649622.701. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

Go Benchmarks

Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: httpThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 59.54, N = 3SE +/- 99.49, N = 1586198962

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ThreadThreadripper 3960XCore i9 10980XE3691215SE +/- 0.034, N = 3SE +/- 0.134, N = 156.03211.1711. (CC) gcc options: -lpthread

Apache Benchmark

This is a test of ab, which is the Apache benchmark program. This test profile measures how many requests per second a given system can sustain when carrying out 1,000,000 requests with 100 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.29Static Web Page ServingThreadripper 3960XCore i9 10980XE7K14K21K28K35KSE +/- 49.18, N = 3SE +/- 128.22, N = 334193.1532075.141. (CC) gcc options: -shared -fPIC -O2 -pthread

Coremark

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

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondThreadripper 3960XCore i9 10980XE200K400K600K800K1000KSE +/- 601.18, N = 3SE +/- 768.42, N = 31069837.23633910.081. (CC) gcc options: -O2 -lrt" -lrt

glibc bench

The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: expThreadripper 3960XCore i9 10980XE1.10642.21283.31924.42565.532SE +/- 0.05687, N = 3SE +/- 0.03939, N = 34.840414.91715

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 0.5.0Video Input: Chimera 1080pThreadripper 3960XCore i9 10980XE130260390520650SE +/- 1.35, N = 3SE +/- 0.19, N = 3587.81276.89MIN: 448.01 / MAX: 721.98MIN: 216.47 / MAX: 339.921. (CC) gcc options: -pthread

ASKAP

This is a CUDA benchmark of ATNF's ASKAP Benchmark with currently using the tConvolveCuda sub-test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 2018-11-10Test: tConvolve OpenMP - DegriddingCore i9 10980XEThreadripper 3960X11002200330044005500SE +/- 8.37, N = 12SE +/- 0.00, N = 35128.683915.531. (CXX) g++ options: -lpthread

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 2018-11-10Test: tConvolve OpenMP - GriddingThreadripper 3960XCore i9 10980XE10002000300040005000SE +/- 26.85, N = 3SE +/- 74.03, N = 124617.472908.061. (CXX) g++ options: -lpthread

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: CPU StressThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 16.14, N = 3SE +/- 4.48, N = 39167.116897.801. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: SemaphoresCore i9 10980XEThreadripper 3960X700K1400K2100K2800K3500KSE +/- 9846.32, N = 3SE +/- 8385.90, N = 33402838.843239494.251. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: CryptoThreadripper 3960XCore i9 10980XE15003000450060007500SE +/- 27.32, N = 3SE +/- 1.60, N = 37072.933929.121. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: ForkingCore i9 10980XEThreadripper 3960X20K40K60K80K100KSE +/- 269.92, N = 3SE +/- 775.10, N = 3104506.2580851.061. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Vector MathThreadripper 3960XCore i9 10980XE14K28K42K56K70KSE +/- 2.23, N = 3SE +/- 7.64, N = 367241.3747063.291. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Matrix MathThreadripper 3960XCore i9 10980XE30K60K90K120K150KSE +/- 505.47, N = 3SE +/- 470.59, N = 3151388.05144369.101. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Socket ActivityThreadripper 3960XCore i9 10980XE4K8K12K16K20KSE +/- 48.18, N = 3SE +/- 55.64, N = 318734.3412670.411. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Glibc Qsort Data SortingThreadripper 3960XCore i9 10980XE80160240320400SE +/- 1.42, N = 3SE +/- 0.29, N = 3380.73242.451. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc

CppPerformanceBenchmarks

CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCppPerformanceBenchmarks 9Test: CtypeCore i9 10980XEThreadripper 3960X816243240SE +/- 0.02, N = 3SE +/- 0.33, N = 326.3133.091. (CXX) g++ options: -std=c++11 -O3

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 PixelThreadripper 3960XCore i9 10980XE918273645SE +/- 0.03, N = 3SE +/- 0.01, N = 321.0637.591. (CC) gcc options: -lm -lpthread -O3

ACES DGEMM

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

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateThreadripper 3960XCore i9 10980XE246810SE +/- 0.028924, N = 3SE +/- 0.063804, N = 38.5438877.2808181. (CC) gcc options: -O3 -march=native -fopenmp

CppPerformanceBenchmarks

CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCppPerformanceBenchmarks 9Test: Stepanov AbstractionThreadripper 3960XCore i9 10980XE714212835SE +/- 0.15, N = 3SE +/- 0.01, N = 328.3029.951. (CXX) g++ options: -std=c++11 -O3

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ProcessThreadripper 3960XCore i9 10980XE3691215SE +/- 0.008, N = 3SE +/- 0.121, N = 155.46810.4071. (CC) gcc options: -lpthread

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 MappingThreadripper 3960XCore i9 10980XE2004006008001000SE +/- 9.82, N = 4SE +/- 1.98, N = 3780.04707.561. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer - Model: CrownThreadripper 3960XCore i9 10980XE714212835SE +/- 0.07, N = 3SE +/- 0.03, N = 329.7718.95MIN: 29.2 / MAX: 30.55MIN: 18.77 / MAX: 19.18

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 0.9.0Test: Apache Spark ALSCore i9 10980XEThreadripper 3960X14002800420056007000SE +/- 39.16, N = 8SE +/- 43.26, N = 86041.976643.08

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-ExponentialThreadripper 3960XCore i9 10980XE246810SE +/- 0.10003, N = 15SE +/- 0.14330, N = 153.564226.77652-march=znver1 -msse4a -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a1. (CXX) g++ options: -std=c++0x -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512pf -mno-avx512er -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: CrownThreadripper 3960XCore i9 10980XE714212835SE +/- 0.02, N = 3SE +/- 0.02, N = 328.3720.88MIN: 27.81 / MAX: 29.1MIN: 20.68 / MAX: 21.18

SciMark

This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: CompositeCore i9 10980XEThreadripper 3960X160320480640800SE +/- 0.42, N = 3SE +/- 1.06, N = 3759.07714.921. (CC) gcc options: -lm

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch 1D - Data Type: f32Threadripper 3960XCore i9 10980XE246810SE +/- 0.03365, N = 6SE +/- 0.09051, N = 42.817696.63739MIN: 2.36MIN: 5.831. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

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 4.1.0Time To OCR 7 ImagesCore i9 10980XEThreadripper 3960X612182430SE +/- 0.02, N = 3SE +/- 0.05, N = 323.8624.14

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer - Model: Asian DragonThreadripper 3960XCore i9 10980XE714212835SE +/- 0.01, N = 3SE +/- 0.02, N = 332.1522.42MIN: 31.73 / MAX: 32.8MIN: 22.33 / MAX: 22.61

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.CCore i9 10980XEThreadripper 3960X4K8K12K16K20KSE +/- 27.55, N = 3SE +/- 33.84, N = 318608.7517832.391. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 3.1.3

Tensorflow

This is a benchmark of the Tensorflow deep learning framework using the CIFAR10 data set. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTensorflow 2017-02-03Build: Cifar10Core i9 10980XEThreadripper 3960X48121620SE +/- 0.02, N = 3SE +/- 0.04, N = 314.2517.04

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 2.10.8Test: unsharp-maskThreadripper 3960XCore i9 10980XE612182430SE +/- 0.04, N = 3SE +/- 0.27, N = 320.5124.99

Go Benchmarks

Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: buildThreadripper 3960XCore i9 10980XE5000M10000M15000M20000M25000MSE +/- 83434537.14, N = 3SE +/- 310952042.40, N = 31929101375925111896331

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 CausticThreadripper 3960XCore i9 10980XE510152025SE +/- 0.03, N = 3SE +/- 0.05, N = 319.3421.70-march=znver1 -msse4a -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a1. (CXX) g++ options: -std=c++0x -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512pf -mno-avx512er -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl

Crafty

This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed TimeCore i9 10980XEThreadripper 3960X2M4M6M8M10MSE +/- 5589.11, N = 3SE +/- 53093.30, N = 3919454390417291. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Asian DragonThreadripper 3960XCore i9 10980XE714212835SE +/- 0.02, N = 3SE +/- 0.03, N = 332.1526.88MIN: 31.68 / MAX: 32.72MIN: 26.72 / MAX: 27.17

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, Fewer Is BetterSeleniumBenchmark: PSPDFKit WASM - Browser: FirefoxThreadripper 3960XCore i9 10980XE30060090012001500SE +/- 5.46, N = 3SE +/- 27.69, N = 3132416281. firefox 70.0.1

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 TimesCore i9 10980XEThreadripper 3960X2004006008001000SE +/- 10.69, N = 3907968

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_1d - Data Type: f32Core i9 10980XEThreadripper 3960X0.58411.16821.75232.33642.9205SE +/- 0.00157, N = 3SE +/- 0.03171, N = 31.792122.59611MIN: 1.76MIN: 2.271. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

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: SciVisCore i9 10980XEThreadripper 3960X714212835SE +/- 0.00, N = 1229.4125.00MIN: 27.03 / MAX: 30.3MIN: 24.39 / MAX: 25.64

glibc bench

The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: cosCore i9 10980XEThreadripper 3960X1020304050SE +/- 0.04, N = 3SE +/- 0.04, N = 340.0742.02

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: sincosThreadripper 3960XCore i9 10980XE3691215SE +/- 0.01, N = 3SE +/- 0.01, N = 312.1412.89

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: sinCore i9 10980XEThreadripper 3960X918273645SE +/- 0.12, N = 3SE +/- 0.34, N = 339.8241.42

XZ Compression

This test measures the time needed to compress a sample file (an Ubuntu file-system image) using XZ compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXZ Compression 5.2.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9Core i9 10980XEThreadripper 3960X510152025SE +/- 0.13, N = 3SE +/- 0.04, N = 320.1320.361. (CC) gcc options: -pthread -fvisibility=hidden -O2

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 1024Threadripper 3960XCore i9 10980XE9K18K27K36K45KSE +/- 199.23, N = 3SE +/- 167.23, N = 340897254711. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, Fewer Is BetterSeleniumBenchmark: PSPDFKit WASM - Browser: Google ChromeThreadripper 3960XCore i9 10980XE400800120016002000SE +/- 3.28, N = 3SE +/- 14.84, N = 3150818691. chrome 78.0.3904.108

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ThreadThreadripper 3960XCore i9 10980XE612182430SE +/- 0.04, N = 3SE +/- 0.45, N = 311.6727.491. (CC) gcc options: -lpthread

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSeleniumBenchmark: Maze Solver - Browser: FirefoxThreadripper 3960XCore i9 10980XE246810SE +/- 0.07, N = 3SE +/- 0.00, N = 34.46.61. firefox 70.0.1

RAMspeed SMP

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Triad - Benchmark: Floating PointThreadripper 3960XCore i9 10980XE8K16K24K32K40K37623.8528912.651. (CC) gcc options: -O3 -march=native

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 2.10.8Test: auto-levelsThreadripper 3960XCore i9 10980XE510152025SE +/- 0.05, N = 3SE +/- 0.09, N = 317.1319.56

Timed ImageMagick Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed ImageMagick Compilation 6.9.0Time To CompileThreadripper 3960XCore i9 10980XE510152025SE +/- 0.12, N = 3SE +/- 0.07, N = 314.5421.95

RAMspeed SMP

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Average - Benchmark: IntegerThreadripper 3960XCore i9 10980XE8K16K24K32K40K36961.6329049.731. (CC) gcc options: -O3 -march=native

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 256Core i9 10980XEThreadripper 3960X8K16K24K32K40KSE +/- 931.89, N = 12SE +/- 412.34, N = 1539345382601. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

RAMspeed SMP

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Copy - Benchmark: Floating PointThreadripper 3960XCore i9 10980XE8K16K24K32K40K35940.8929165.421. (CC) gcc options: -O3 -march=native

SQLite

This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.30.1Threads / Copies: 1Threadripper 3960XCore i9 10980XE510152025SE +/- 0.10, N = 3SE +/- 0.05, N = 315.7419.431. (CC) gcc options: -O2 -lz -lm -ldl -lpthread

RAMspeed SMP

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Scale - Benchmark: Floating PointThreadripper 3960XCore i9 10980XE8K16K24K32K40K37097.9928379.301. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Add - Benchmark: Floating PointThreadripper 3960XCore i9 10980XE9K18K27K36K45K43043.8428825.971. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Average - Benchmark: Floating PointThreadripper 3960XCore i9 10980XE9K18K27K36K45K40240.9828822.801. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Add - Benchmark: IntegerThreadripper 3960XCore i9 10980XE9K18K27K36K45K43028.4429367.551. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Copy - Benchmark: IntegerThreadripper 3960XCore i9 10980XE8K16K24K32K40K37204.2529115.601. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Scale - Benchmark: IntegerThreadripper 3960XCore i9 10980XE8K16K24K32K40K38106.3828779.301. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Triad - Benchmark: IntegerThreadripper 3960XCore i9 10980XE9K18K27K36K45K43894.2628909.781. (CC) gcc options: -O3 -march=native

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.4.1Test: Many Spheres - Resolution: 2560 x 1440Threadripper 3960XCore i9 10980XE13002600390052006500SE +/- 4.91, N = 3SE +/- 3.10, N = 35900.235851.92

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.4.1Test: Many Spheres - Resolution: 2560 x 1440Threadripper 3960XCore i9 10980XE1326395265SE +/- 0.05, N = 3SE +/- 0.03, N = 358.8558.37

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 0.5.0Video Input: Summer Nature 4KThreadripper 3960XCore i9 10980XE60120180240300SE +/- 0.85, N = 3SE +/- 1.19, N = 3291.09170.75MIN: 179.18 / MAX: 309.08MIN: 103.66 / MAX: 183.341. (CC) gcc options: -pthread

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 13.42, N = 4SE +/- 59.76, N = 446439736

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: SciVisThreadripper 3960XCore i9 10980XE1224364860SE +/- 0.00, N = 12SE +/- 0.44, N = 352.6336.60MIN: 43.48 / MAX: 55.56MIN: 31.25 / MAX: 37.04

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_alexnet - Data Type: f32Core i9 10980XEThreadripper 3960X306090120150SE +/- 0.11, N = 3SE +/- 1.31, N = 3125.46142.23MIN: 124.99MIN: 123.191. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Xsbench

XSBench is a mini-app representing a key computational kernel of the Monte Carlo neutronics application OpenMC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgLookups/s, More Is BetterXsbench 2017-07-06Threadripper 3960XCore i9 10980XE1.1M2.2M3.3M4.4M5.5MSE +/- 3935.07, N = 3SE +/- 2612.15, N = 3502952839192801. (CC) gcc options: -std=gnu99 -fopenmp -O3 -lm

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSeleniumBenchmark: Maze Solver - Browser: Google ChromeThreadripper 3960XCore i9 10980XE246810SE +/- 0.03, N = 3SE +/- 0.03, N = 34.06.11. chrome 78.0.3904.108

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 64Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 231.46, N = 15SE +/- 254.66, N = 159362.27395.51. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

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 FLACThreadripper 3960XCore i9 10980XE3691215SE +/- 0.013, N = 5SE +/- 0.016, N = 57.8349.5701. (CXX) g++ options: -O2 -fvisibility=hidden -logg -lm

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 2.10.8Test: rotateThreadripper 3960XCore i9 10980XE48121620SE +/- 0.02, N = 3SE +/- 0.17, N = 312.3216.45

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. The number of iterations used is 1,000,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteCore i9 10980XEThreadripper 3960X170K340K510K680K850KSE +/- 613.52, N = 3SE +/- 5664.15, N = 3775515658408

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.7Encoder Mode: Enc Mode 4 - Input: 1080pThreadripper 3960XCore i9 10980XE246810SE +/- 0.027, N = 3SE +/- 0.021, N = 38.0435.1981. (CXX) g++ options: -fPIE -fPIC -pie

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: CPUThreadripper 3960XCore i9 10980XE816243240SE +/- 0.18, N = 3SE +/- 0.31, N = 535.324.3

glibc bench

The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: ffsCore i9 10980XEThreadripper 3960X0.39580.79161.18741.58321.979SE +/- 0.00052, N = 5SE +/- 0.00277, N = 31.382441.75927

CppPerformanceBenchmarks

CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCppPerformanceBenchmarks 9Test: Function ObjectsCore i9 10980XEThreadripper 3960X48121620SE +/- 0.01, N = 3SE +/- 0.12, N = 312.7514.681. (CXX) g++ options: -std=c++11 -O3

Go Benchmarks

Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: jsonThreadripper 3960XCore i9 10980XE600K1200K1800K2400K3000KSE +/- 13151.09, N = 3SE +/- 8421.06, N = 319010782806876

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP LavaMDThreadripper 3960XCore i9 10980XE48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 310.1015.251. (CXX) g++ options: -O2 -lOpenCL

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: HairThreadripper 3960XCore i9 10980XE48121620SE +/- 0.02317, N = 3SE +/- 0.00669, N = 39.7973714.77930-march=znver1 -msse4a -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a1. (CXX) g++ options: -std=c++0x -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512pf -mno-avx512er -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradesoapThreadripper 3960XCore i9 10980XE10002000300040005000SE +/- 29.66, N = 4SE +/- 31.54, N = 433644825

Zstd Compression

This test measures the time needed to compress a sample file (an Ubuntu file-system image) using Zstd compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterZstd Compression 1.3.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19Threadripper 3960XCore i9 10980XE3691215SE +/- 0.15, N = 4SE +/- 0.05, N = 310.2311.391. (CC) gcc options: -O3 -pthread -lz -llzma

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2Threadripper 3960XCore i9 10980XE14002800420056007000SE +/- 28.55, N = 4SE +/- 41.51, N = 436946665

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 and then to decode the generated Opus file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.0.1WAV, Opus Encode, Opus DecodeThreadripper 3960XCore i9 10980XE246810SE +/- 0.013, N = 5SE +/- 0.009, N = 56.2728.0131. (CC) gcc options: -O3 -ffast-math -fPIE -pie -logg -lm -lopus

x265

This is a simple test of the x265 encoder run on the CPU with a sample 1080p video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.1.2H.265 1080p Video EncodingThreadripper 3960XCore i9 10980XE1530456075SE +/- 0.22, N = 3SE +/- 0.19, N = 365.7141.991. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Rust Prime Benchmark

Based on petehunt/rust-benchmark, this is a prime number benchmark that is multi-threaded and written in Rustlang. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRust Prime BenchmarkPrime Number Test To 200,000,000Core i9 10980XEThreadripper 3960X48121620SE +/- 0.007, N = 3SE +/- 0.023, N = 37.38814.9991. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiBThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 13.80, N = 3SE +/- 124.76, N = 311380.328370.681. (CC) gcc options: -O3 -march=native

LeelaChessZero

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

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.22.0Backend: RandomThreadripper 3960XCore i9 10980XE20K40K60K80K100KSE +/- 513.36, N = 3SE +/- 427.06, N = 3106908.066812.81. (CXX) g++ options: -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: EP.CCore i9 10980XEThreadripper 3960X5001000150020002500SE +/- 55.63, N = 12SE +/- 5.95, N = 32377.792208.111. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 3.1.3

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 1024Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 4.47, N = 3SE +/- 71.24, N = 38097.97139.01. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP CFD SolverThreadripper 3960XCore i9 10980XE3691215SE +/- 0.025, N = 3SE +/- 0.061, N = 39.03711.5981. (CXX) g++ options: -O2 -lOpenCL

glibc bench

The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: atanhThreadripper 3960XCore i9 10980XE3691215SE +/- 0.00, N = 3SE +/- 0.01, N = 310.0210.17

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: sqrtCore i9 10980XEThreadripper 3960X0.48650.9731.45951.9462.4325SE +/- 0.00098, N = 3SE +/- 0.01495, N = 31.593312.16235

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: modfCore i9 10980XEThreadripper 3960X0.49950.9991.49851.9982.4975SE +/- 0.00271, N = 3SE +/- 0.00220, N = 31.812552.21984

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: log2Threadripper 3960XCore i9 10980XE246810SE +/- 0.00530, N = 3SE +/- 0.00446, N = 35.837176.40585

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: tanhThreadripper 3960XCore i9 10980XE3691215SE +/- 0.11, N = 3SE +/- 0.02, N = 310.5211.11

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: pthread_onceCore i9 10980XEThreadripper 3960X0.39240.78481.17721.56961.962SE +/- 0.00039, N = 3SE +/- 0.00112, N = 31.396561.74395

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: asinhThreadripper 3960XCore i9 10980XE246810SE +/- 0.00200, N = 3SE +/- 0.00708, N = 38.420528.54710

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: sinhCore i9 10980XEThreadripper 3960X246810SE +/- 0.00713, N = 3SE +/- 0.07381, N = 37.742197.82634

OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: ffsllCore i9 10980XEThreadripper 3960X0.39440.78881.18321.57761.972SE +/- 0.00114, N = 3SE +/- 0.00365, N = 31.382171.75307

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 2.10.8Test: resizeThreadripper 3960XCore i9 10980XE3691215SE +/- 0.135, N = 3SE +/- 0.141, N = 38.60911.204

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 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: VMAF Optimized - Input: Bosphorus 1080pThreadripper 3960XCore i9 10980XE80160240320400SE +/- 4.06, N = 13SE +/- 5.35, N = 12382.22300.071. (CC) gcc options: -fPIE -fPIC -fvisibility=hidden -O3 -pie -rdynamic -lpthread -lrt -lm

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.CThreadripper 3960XCore i9 10980XE5K10K15K20K25KSE +/- 41.78, N = 3SE +/- 46.20, N = 325120.0816348.601. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 3.1.3

Parboil

The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP StencilThreadripper 3960XCore i9 10980XE246810SE +/- 0.018289, N = 3SE +/- 0.027934, N = 37.2434408.1211541. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp

Go Benchmarks

Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: garbageThreadripper 3960XCore i9 10980XE150K300K450K600K750KSE +/- 2346.17, N = 3SE +/- 2158.09, N = 3702500718932

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 2.6.0Test: Boat - Acceleration: CPU-onlyThreadripper 3960XCore i9 10980XE3691215SE +/- 0.046, N = 3SE +/- 0.004, N = 36.2799.160

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 0.5.0Video Input: Summer Nature 1080pThreadripper 3960XCore i9 10980XE150300450600750SE +/- 1.15, N = 3SE +/- 1.40, N = 3677.06297.80MIN: 426.02 / MAX: 738.11MIN: 186.02 / MAX: 327.581. (CC) gcc options: -pthread

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 64Threadripper 3960XCore i9 10980XE10K20K30K40K50KSE +/- 259.03, N = 3SE +/- 683.04, N = 1547733389561. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 64Core i9 10980XEThreadripper 3960X4K8K12K16K20KSE +/- 282.40, N = 3SE +/- 715.56, N = 1519577189001. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

LAME MP3 Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3Threadripper 3960XCore i9 10980XE3691215SE +/- 0.003, N = 3SE +/- 0.003, N = 37.1179.1431. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lm

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 32Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 17.13, N = 3SE +/- 281.35, N = 1511113.09025.31. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 32Threadripper 3960XCore i9 10980XE10K20K30K40K50KSE +/- 88.57, N = 3SE +/- 1023.30, N = 1546123389391. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

N-Queens

This is a test of the OpenMP version of a test that solves the N-queens problem. The board problem size is 18. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterN-Queens 1.0Elapsed TimeThreadripper 3960XCore i9 10980XE246810SE +/- 0.012, N = 3SE +/- 0.001, N = 44.4698.6271. (CC) gcc options: -static -fopenmp -O3 -march=native

Sunflow Rendering System

This test runs benchmarks of the Sunflow Rendering System. The Sunflow Rendering System is an open-source render engine for photo-realistic image synthesis with a ray-tracing core. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSunflow Rendering System 0.07.2Global Illumination + Image SynthesisThreadripper 3960XCore i9 10980XE0.18860.37720.56580.75440.943SE +/- 0.010, N = 3SE +/- 0.014, N = 30.6850.838

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ProcessCore i9 10980XEThreadripper 3960X0.62151.2431.86452.4863.1075SE +/- 0.006, N = 3SE +/- 0.027, N = 142.4392.7621. (CC) gcc options: -lpthread

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonThreadripper 3960XCore i9 10980XE9001800270036004500SE +/- 10.28, N = 4SE +/- 15.42, N = 439634257

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 4096Core i9 10980XEThreadripper 3960X13K26K39K52K65KSE +/- 332.60, N = 3SE +/- 410.08, N = 360652588751. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

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 TracerCore i9 10980XEThreadripper 3960X80160240320400SE +/- 19.70, N = 15SE +/- 0.00, N = 12377.78333.33MIN: 333.33 / MAX: 500MIN: 166.67 / MAX: 500

Bork File Encrypter

Bork is a small, cross-platform file encryption utility. It is written in Java and designed to be included along with the files it encrypts for long-term storage. This test measures the amount of time it takes to encrypt a sample file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBork File Encrypter 1.4File Encryption TimeCore i9 10980XEThreadripper 3960X246810SE +/- 0.036, N = 3SE +/- 0.021, N = 35.8398.018

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 512Threadripper 3960XCore i9 10980XE9K18K27K36K45KSE +/- 137.80, N = 3SE +/- 239.71, N = 343049255011. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.7Encoder Mode: Enc Mode 8 - Input: 1080pThreadripper 3960XCore i9 10980XE20406080100SE +/- 0.12, N = 3SE +/- 0.09, N = 378.7947.291. (CXX) g++ options: -fPIE -fPIC -pie

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 2018-09-25H.264 Video EncodingThreadripper 3960XCore i9 10980XE4080120160200SE +/- 1.86, N = 3SE +/- 1.50, N = 6193.76130.571. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiVoxels / Sec, More Is BetterParaView 5.4.1Test: Wavelet Volume - Resolution: 2560 x 1440Threadripper 3960XCore i9 10980XE6001200180024003000SE +/- 15.86, N = 3SE +/- 21.64, N = 152811.522115.01

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.4.1Test: Wavelet Volume - Resolution: 2560 x 1440Threadripper 3960XCore i9 10980XE4080120160200SE +/- 0.99, N = 3SE +/- 1.35, N = 15175.72132.19

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 CausticThreadripper 3960XCore i9 10980XE246810SE +/- 0.00428, N = 3SE +/- 0.05152, N = 34.156467.31939-march=znver1 -msse4a -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a1. (CXX) g++ options: -std=c++0x -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512pf -mno-avx512er -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 512 MiBThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 6.27, N = 3SE +/- 25.02, N = 311306.188210.911. (CC) gcc options: -O3 -march=native

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 2048Core i9 10980XEThreadripper 3960X14K28K42K56K70KSE +/- 364.62, N = 3SE +/- 623.77, N = 366081611121. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.4.1Test: Wavelet Contour - Resolution: 2560 x 1440Threadripper 3960XCore i9 10980XE6001200180024003000SE +/- 1.76, N = 3SE +/- 14.90, N = 32578.931173.01

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.4.1Test: Wavelet Contour - Resolution: 2560 x 1440Threadripper 3960XCore i9 10980XE50100150200250SE +/- 0.17, N = 3SE +/- 1.43, N = 3247.47112.56

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create FilesCore i9 10980XEThreadripper 3960X3691215SE +/- 0.04, N = 3SE +/- 0.01, N = 310.8211.541. (CC) gcc options: -lm

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 1.0.0Scene: MemorialCore i9 10980XEThreadripper 3960X510152025SE +/- 0.08, N = 3SE +/- 0.05, N = 321.8119.59

Git

This test measures the time needed to carry out some sample Git operations on an example, static repository that happens to be a copy of the GNOME GTK tool-kit repository. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGitTime To Complete Common Git CommandsThreadripper 3960XCore i9 10980XE1.20382.40763.61144.81526.019SE +/- 0.035, N = 3SE +/- 0.024, N = 34.7655.3501. git version 2.20.1

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory AllocationsThreadripper 3960XCore i9 10980XE20406080100SE +/- 0.15, N = 3SE +/- 0.31, N = 370.6077.541. (CC) gcc options: -lm

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ProcessesCore i9 10980XEThreadripper 3960X918273645SE +/- 0.40, N = 3SE +/- 0.55, N = 332.2541.041. (CC) gcc options: -lm

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Launch ProgramsCore i9 10980XEThreadripper 3960X1122334455SE +/- 0.01, N = 3SE +/- 0.47, N = 335.8848.151. (CC) gcc options: -lm

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ThreadsCore i9 10980XEThreadripper 3960X510152025SE +/- 0.06, N = 3SE +/- 0.25, N = 314.9120.841. (CC) gcc options: -lm

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 4096Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 31.06, N = 3SE +/- 59.01, N = 38818.88725.21. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 1024Core i9 10980XEThreadripper 3960X15K30K45K60K75KSE +/- 904.98, N = 4SE +/- 370.51, N = 368093599371. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 64Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 21.91, N = 3SE +/- 86.93, N = 69585.47578.51. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

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 2.6.0Test: Masskrug - Acceleration: CPU-onlyThreadripper 3960XCore i9 10980XE1.00282.00563.00844.01125.014SE +/- 0.027, N = 3SE +/- 0.026, N = 33.4024.457

Smallpt

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

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 128 SamplesThreadripper 3960XCore i9 10980XE1.30822.61643.92465.23286.541SE +/- 0.005, N = 3SE +/- 0.009, N = 33.6755.8141. (CXX) g++ options: -fopenmp -O3

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 512Core i9 10980XEThreadripper 3960X12K24K36K48K60KSE +/- 684.50, N = 4SE +/- 123.47, N = 355196512401. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 256Threadripper 3960XCore i9 10980XE9K18K27K36K45KSE +/- 310.57, N = 3SE +/- 130.18, N = 342616279041. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 512 MiBThreadripper 3960XCore i9 10980XE4K8K12K16K20KSE +/- 62.41, N = 3SE +/- 1.43, N = 319400.439270.751. (CC) gcc options: -O3 -march=native

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 512Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 17.50, N = 3SE +/- 66.31, N = 38684.67561.61. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Bullet Physics Engine

This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: RaytestsThreadripper 3960XCore i9 10980XE0.50331.00661.50992.01322.5165SE +/- 0.004788, N = 3SE +/- 0.005216, N = 32.1143482.2368701. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ThreadThreadripper 3960XCore i9 10980XE0.89981.79962.69943.59924.499SE +/- 0.057, N = 3SE +/- 0.013, N = 33.7363.9991. (CC) gcc options: -lpthread

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 256Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 8.89, N = 3SE +/- 43.39, N = 38674.17671.41. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 2048Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 21.61, N = 3SE +/- 20.08, N = 39048.18921.41. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 128Core i9 10980XEThreadripper 3960X2K4K6K8K10KSE +/- 7.40, N = 3SE +/- 14.00, N = 39138.58902.31. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 128Threadripper 3960XCore i9 10980XE10K20K30K40K50KSE +/- 237.97, N = 3SE +/- 32.10, N = 346970327221. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ProcessThreadripper 3960XCore i9 10980XE0.81861.63722.45583.27444.093SE +/- 0.014, N = 3SE +/- 0.015, N = 33.5293.6381. (CC) gcc options: -lpthread

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 128Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 77.36, N = 3SE +/- 18.83, N = 38569.27338.51. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

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 2.6.0Test: Server Room - Acceleration: CPU-onlyThreadripper 3960XCore i9 10980XE0.59291.18581.77872.37162.9645SE +/- 0.016, N = 3SE +/- 0.006, N = 32.2422.635

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 256Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 12.22, N = 3SE +/- 38.68, N = 38883.87430.51. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 32Core i9 10980XEThreadripper 3960X4K8K12K16K20KSE +/- 159.08, N = 3SE +/- 208.40, N = 416912150421. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 32Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 5.51, N = 3SE +/- 0.00, N = 311237.08866.71. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) 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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_3d - Data Type: f32Core i9 10980XEThreadripper 3960X0.64441.28881.93322.57763.222SE +/- 0.00924, N = 3SE +/- 0.01436, N = 32.596952.86390MIN: 2.55MIN: 2.451. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 1024Threadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 5.17, N = 3SE +/- 79.13, N = 39356.29279.21. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Optcarrot

Optcarrot is an NES emulator benchmark for the Ruby language. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOptcarrotOptimized BenchmarkCore i9 10980XEThreadripper 3960X306090120150SE +/- 1.08, N = 4SE +/- 0.64, N = 4142.13106.881. ruby 2.5.5p157 (2019-03-15 revision 67260) [x86_64-linux-gnu]

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ThreadCore i9 10980XEThreadripper 3960X0.64311.28621.92932.57243.2155SE +/- 0.009, N = 3SE +/- 0.030, N = 32.6442.8581. (CC) gcc options: -lpthread

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 512Core i9 10980XEThreadripper 3960X2K4K6K8K10KSE +/- 95.94, N = 3SE +/- 39.54, N = 39149.09114.01. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: Visual Quality Optimized - Input: Bosphorus 1080pThreadripper 3960XCore i9 10980XE70140210280350SE +/- 2.33, N = 3SE +/- 1.64, N = 3326.41252.721. (CC) gcc options: -fPIE -fPIC -fvisibility=hidden -O3 -pie -rdynamic -lpthread -lrt -lm

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 128Threadripper 3960XCore i9 10980XE6K12K18K24K30KSE +/- 98.24, N = 3SE +/- 370.47, N = 327460251951. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pThreadripper 3960XCore i9 10980XE90180270360450SE +/- 2.37, N = 3SE +/- 1.22, N = 3392.87310.091. (CC) gcc options: -fPIE -fPIC -fvisibility=hidden -O3 -pie -rdynamic -lpthread -lrt -lm

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 2.6.0Test: Server Rack - Acceleration: CPU-onlyThreadripper 3960XCore i9 10980XE0.0320.0640.0960.1280.16SE +/- 0.002, N = 15SE +/- 0.002, N = 30.1300.142

Tachyon

This is a test of the threaded Tachyon, a parallel ray-tracing system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.98.9Total TimeThreadripper 3960XCore i9 10980XE0.56171.12341.68512.24682.8085SE +/- 0.0039, N = 3SE +/- 0.0032, N = 31.68992.49631. (CC) gcc options: -m32 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread

TSCP

This is a performance test of TSCP, Tom Kerrigan's Simple Chess Program, which has a built-in performance benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess PerformanceCore i9 10980XEThreadripper 3960X300K600K900K1200K1500KSE +/- 722.40, N = 5SE +/- 2224.51, N = 5141152913414121. (CC) gcc options: -O3 -march=native

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 128 MiBThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 18.64, N = 3SE +/- 109.28, N = 510893.688969.581. (CC) gcc options: -O3 -march=native

Parboil

The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP CUTCPThreadripper 3960XCore i9 10980XE0.49530.99061.48591.98122.4765SE +/- 0.000632, N = 3SE +/- 0.007167, N = 31.2646572.2014421. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 128 MiBThreadripper 3960XCore i9 10980XE2K4K6K8K10KSE +/- 15.80, N = 3SE +/- 103.21, N = 310863.428285.561. (CC) gcc options: -O3 -march=native

ctx_clock

Ctx_clock is a simple test program to measure the context switch time in clock cycles. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgClocks, Fewer Is Betterctx_clockContext Switch TimeCore i9 10980XEThreadripper 3960X306090120150SE +/- 0.67, N = 3120152

Bullet Physics Engine

This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: Convex TrimeshThreadripper 3960XCore i9 10980XE0.22860.45720.68580.91441.143SE +/- 0.001348, N = 3SE +/- 0.004634, N = 30.9551881.0158921. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: Prim TrimeshThreadripper 3960XCore i9 10980XE0.19330.38660.57990.77320.9665SE +/- 0.000720, N = 3SE +/- 0.002200, N = 30.7987750.8590131. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 136 RagdollsThreadripper 3960XCore i9 10980XE0.55661.11321.66982.22642.783SE +/- 0.002207, N = 3SE +/- 0.001436, N = 32.1816232.4737121. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 1000 ConvexThreadripper 3960XCore i9 10980XE0.8921.7842.6763.5684.46SE +/- 0.013041, N = 3SE +/- 0.035249, N = 33.8136703.9644531. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 1000 StackThreadripper 3960XCore i9 10980XE1.0052.013.0154.025.025SE +/- 0.006196, N = 3SE +/- 0.006379, N = 34.1109414.4665051. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 3000 FallThreadripper 3960XCore i9 10980XE0.89491.78982.68473.57964.4745SE +/- 0.005941, N = 3SE +/- 0.007341, N = 33.4169413.9771621. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Max Ping Pong BandwidthThreadripper 3960XCore i9 10980XE5K10K15K20K25KSE +/- 366.52, N = 3SE +/- 124.91, N = 322107.9120672.701. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Random Ring BandwidthThreadripper 3960XCore i9 10980XE0.75691.51382.27073.02763.7845SE +/- 0.00987, N = 3SE +/- 0.00664, N = 33.364001.478511. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

OpenBenchmarking.orgusecs, Fewer Is BetterHPC Challenge 1.5.0Test / Class: Random Ring LatencyCore i9 10980XEThreadripper 3960X0.10280.20560.30840.41120.514SE +/- 0.00081, N = 3SE +/- 0.00045, N = 30.288610.456831. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

OpenBenchmarking.orgGUP/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-Random AccessThreadripper 3960XCore i9 10980XE0.03220.06440.09660.12880.161SE +/- 0.00070, N = 3SE +/- 0.00089, N = 30.142920.107101. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: EP-STREAM TriadCore i9 10980XEThreadripper 3960X0.58421.16841.75262.33682.921SE +/- 0.00602, N = 3SE +/- 0.00309, N = 32.596321.822281. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-PtransThreadripper 3960XCore i9 10980XE1.24652.4933.73954.9866.2325SE +/- 0.01609, N = 3SE +/- 0.00274, N = 35.540225.248231. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: EP-DGEMMCore i9 10980XEThreadripper 3960X816243240SE +/- 0.13, N = 3SE +/- 0.45, N = 333.9231.641. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

OpenBenchmarking.orgGFLOP/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-FfteThreadripper 3960XCore i9 10980XE3691215SE +/- 0.18914, N = 3SE +/- 0.02340, N = 310.730776.977521. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-FfteThreadripper 3960XCore i9 10980XE3691215SE +/- 0.18914, N = 3SE +/- 0.02340, N = 310.730776.977521. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 3.1.3

Stream

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: AddThreadripper 3960XCore i9 10980XE8K16K24K32K40KSE +/- 12.82, N = 5SE +/- 2.52, N = 536666.433985.01. (CC) gcc options: -O3 -march=native -fopenmp

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: TriadThreadripper 3960XCore i9 10980XE8K16K24K32K40KSE +/- 19.57, N = 5SE +/- 2.25, N = 536725.133869.11. (CC) gcc options: -O3 -march=native -fopenmp

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: ScaleThreadripper 3960XCore i9 10980XE7K14K21K28K35KSE +/- 16.43, N = 5SE +/- 3.34, N = 533514.631362.11. (CC) gcc options: -O3 -march=native -fopenmp

334 Results Shown

Pennant:
  sedovbig
  leblancbig
SPECjbb 2015:
  SPECjbb2015-Composite critical-jOPS
  SPECjbb2015-Composite max-jOPS
HPC Challenge
Apache Cassandra
CppPerformanceBenchmarks
Timed GCC Compilation
MKL-DNN DNNL
libgav1
QMCPACK
SVT-AV1
Apache Cassandra
PostgreSQL pgbench
MKL-DNN DNNL
Blender
BRL-CAD
FFTW
Memtier_benchmark
CppPerformanceBenchmarks
Blender
libgav1
Renaissance
CP2K Molecular Dynamics
LZMA Compression
Tinymembench:
  Standard Memset
  Standard Memcpy
libgav1
dav1d
Blender
Minion
Radiance Benchmark
Selenium
FFTW
Selenium
SQLite
LuxCoreRender
Hackbench
Selenium
asmFish
Appleseed
MKL-DNN DNNL
CacheBench:
  Read / Modify / Write
  Write
  Read
OSPray
BYTE Unix Benchmark
OSPray
Blender
VP9 libvpx Encoding
POV-Ray
Selenium
OSPray
Minion
GROMACS
OSPray
Apache Cassandra
Timed MrBayes Analysis
Facebook RocksDB
Stress-NG:
  Glibc C String Functions
  Context Switching
Renaissance
Selenium
Appleseed
Parboil
Apache Siege
Hackbench
PostgreSQL pgbench
CppPerformanceBenchmarks
Blender
OpenCV Benchmark
ASKAP:
  tConvolve MT - Degridding
  tConvolve MT - Gridding
Timed Linux Kernel Compilation
Appleseed
Selenium
libgav1
NAS Parallel Benchmarks
Selenium
FFTW
Hackbench
Chaos Group V-RAY
OSPray
Build2
NAS Parallel Benchmarks
Renaissance
IndigoBench
Radiance Benchmark
LuxCoreRender
IndigoBench
Facebook RocksDB
GraphicsMagick:
  Enhanced
  Resizing
  Noise-Gaussian
Facebook RocksDB
GraphicsMagick:
  Sharpen
  Swirl
  Rotate
  HWB Color Space
Timed LLVM Compilation
SQLite Speedtest
Stockfish
NAS Parallel Benchmarks
CppPerformanceBenchmarks
LibreOffice
NAMD
ASKAP:
  tConvolve MPI - Degridding
  tConvolve MPI - Gridding
Renaissance
eSpeak Speech Engine
MBW
NGINX Benchmark
rav1e
LeelaChessZero
Selenium
MKL-DNN DNNL
Selenium
FFTW
MBW
NAS Parallel Benchmarks
Timed PHP Compilation
Minion
Selenium
Java SciMark
7-Zip Compression
Himeno Benchmark
Renaissance
MKL-DNN DNNL
Stream
Embree
miniFE
dcraw
MKL-DNN DNNL
Embree
Rust Mandelbrot
GNU Octave Benchmark
m-queens
Gzip Compression
MBW
Hackbench
Redis:
  SET
  GET
Go Benchmarks
Hackbench
Apache Benchmark
Coremark
glibc bench
dav1d
ASKAP:
  tConvolve OpenMP - Degridding
  tConvolve OpenMP - Gridding
Stress-NG:
  CPU Stress
  Semaphores
  Crypto
  Forking
  Vector Math
  Matrix Math
  Socket Activity
  Glibc Qsort Data Sorting
CppPerformanceBenchmarks
C-Ray
ACES DGEMM
CppPerformanceBenchmarks
Hackbench
TTSIOD 3D Renderer
Embree
Renaissance
Tungsten Renderer
Embree
SciMark
MKL-DNN DNNL
Tesseract OCR
Embree
NAS Parallel Benchmarks
Tensorflow
GIMP
Go Benchmarks
Tungsten Renderer
Crafty
Embree
Selenium
PyBench
MKL-DNN DNNL
OSPray
glibc bench:
  cos
  sincos
  sin
XZ Compression
FFTW
Selenium
Hackbench
Selenium
RAMspeed SMP
GIMP
Timed ImageMagick Compilation
RAMspeed SMP
FFTW
RAMspeed SMP
SQLite
RAMspeed SMP:
  Scale - Floating Point
  Add - Floating Point
  Average - Floating Point
  Add - Integer
  Copy - Integer
  Scale - Integer
  Triad - Integer
ParaView:
  Many Spheres - 2560 x 1440:
    MiPolys / Sec
    Frames / Sec
dav1d
DaCapo Benchmark
OSPray
MKL-DNN DNNL
Xsbench
Selenium
FFTW
FLAC Audio Encoding
GIMP
PHPBench
SVT-AV1
NeatBench
glibc bench
CppPerformanceBenchmarks
Go Benchmarks
Rodinia
Tungsten Renderer
DaCapo Benchmark
Zstd Compression
DaCapo Benchmark
Opus Codec Encoding
x265
Rust Prime Benchmark
MBW
LeelaChessZero
NAS Parallel Benchmarks
FFTW
Rodinia
glibc bench:
  atanh
  sqrt
  modf
  log2
  tanh
  pthread_once
  asinh
  sinh
  ffsll
GIMP
SVT-VP9
NAS Parallel Benchmarks
Parboil
Go Benchmarks
Darktable
dav1d
FFTW:
  Float + SSE - 2D FFT Size 64
  Float + SSE - 1D FFT Size 64
LAME MP3 Encoding
FFTW:
  Stock - 2D FFT Size 32
  Float + SSE - 2D FFT Size 32
N-Queens
Sunflow Rendering System
Hackbench
DaCapo Benchmark
FFTW
OSPray
Bork File Encrypter
FFTW
SVT-AV1
x264
ParaView:
  Wavelet Volume - 2560 x 1440:
    MiVoxels / Sec
    Frames / Sec
Tungsten Renderer
MBW
FFTW
ParaView:
  Wavelet Contour - 2560 x 1440:
    MiPolys / Sec
    Frames / Sec
OSBench
Intel Open Image Denoise
Git
OSBench:
  Memory Allocations
  Create Processes
  Launch Programs
  Create Threads
FFTW:
  Stock - 1D FFT Size 4096
  Float + SSE - 1D FFT Size 1024
  Stock - 2D FFT Size 64
Darktable
Smallpt
FFTW:
  Float + SSE - 1D FFT Size 512
  Float + SSE - 2D FFT Size 256
MBW
FFTW
Bullet Physics Engine
Hackbench
FFTW:
  Stock - 2D FFT Size 256
  Stock - 1D FFT Size 2048
  Stock - 2D FFT Size 128
  Float + SSE - 2D FFT Size 128
Hackbench
FFTW
Darktable
FFTW:
  Stock - 1D FFT Size 256
  Float + SSE - 1D FFT Size 32
  Stock - 1D FFT Size 32
MKL-DNN DNNL
FFTW
Optcarrot
Hackbench
FFTW
SVT-VP9
FFTW
SVT-VP9
Darktable
Tachyon
TSCP
MBW
Parboil
MBW
ctx_clock
Bullet Physics Engine:
  Convex Trimesh
  Prim Trimesh
  136 Ragdolls
  1000 Convex
  1000 Stack
  3000 Fall
HPC Challenge:
  Max Ping Pong Bandwidth
  Rand Ring Bandwidth
  Rand Ring Latency
  G-Rand Access
  EP-STREAM Triad
  G-Ptrans
  EP-DGEMM
  G-Ffte
  G-Ffte
Stream:
  Add
  Triad
  Scale