AMD Ryzen Threadripper 3990X

AMD Ryzen Threadripper 3990X benchmarks by Michael Larabel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2207228-NE-2207215NE88
Jump To Table - Results

View

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

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

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
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Threadripper 3990X
July 20 2022
  21 Hours, 43 Minutes
Threadripper 3960X
July 21 2022
  22 Hours, 27 Minutes
Invert Behavior (Only Show Selected Data)
  22 Hours, 5 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):

Threadripper 3990X

Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads), Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS), Chipset: AMD Starship/Matisse, Memory: 128GB, Disk: Samsung SSD 970 EVO Plus 500GB, Graphics: AMD Radeon RX 5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: DELL P2415Q, Network: Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 22.04, Kernel: 5.19.0-051900rc7-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.47), Vulkan: 1.2.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301039
Java Notes: OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.22.04.1)
Python Notes: Python 3.10.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Threadripper 3960X

Processor: AMD Ryzen Threadripper 3960X 24-Core @ 3.80GHz (24 Cores / 48 Threads), Motherboard: MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS), Chipset: AMD Starship/Matisse, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 1TB, Graphics: Gigabyte AMD Radeon RX 5500 XT 8GB (1900/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: VA2431, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 22.04, Kernel: 5.19.0-051900rc7-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.47), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.22.04.1)
Python Notes: Python 3.10.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

AMD Ryzen Threadripper 3990XProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionThreadripper 3990XThreadripper 3960XAMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads)Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS)AMD Starship/Matisse128GBSamsung SSD 970 EVO Plus 500GBAMD Radeon RX 5700 8GB (1750/875MHz)AMD Navi 10 HDMI AudioDELL P2415QIntel I211 + Intel Wi-Fi 6 AX200Ubuntu 22.045.19.0-051900rc7-generic (x86_64)GNOME Shell 42.2X Server + Wayland4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.47)1.2.204GCC 11.2.0ext43840x2160AMD Ryzen Threadripper 3960X 24-Core @ 3.80GHz (24 Cores / 48 Threads)MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS)32GB1000GB Sabrent Rocket 4.0 1TBGigabyte AMD Radeon RX 5500 XT 8GB (1900/875MHz)VA2431Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX2001.3.2041920x1080OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Threadripper 3990X: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301039- Threadripper 3960X: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301025Java Details- OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.22.04.1)Python Details- Python 3.10.4Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Threadripper 3990X vs. Threadripper 3960X ComparisonPhoronix Test SuiteBaseline+99.5%+99.5%+199%+199%+298.5%+298.5%143.3%143.2%123.7%119.1%72.8%70%43.8%35.1%28.1%26.7%22.2%20.1%20%19.6%19%18.8%15.3%14.9%14.8%14.6%13.5%12.6%12%11.5%9.6%8.7%8.2%7.7%7.7%7.5%7.5%7.3%7.2%6.8%6.8%6.7%6.7%6.6%6.6%6.5%6.2%6%6%5.7%5.7%5.7%5.6%5.5%5.4%5.2%5.1%5%4.9%4.9%4.9%4.8%4.8%4.8%4.7%4.7%4.6%4.5%4.5%4.5%4.5%4.4%4.3%4.2%4.2%4%4%4%3.9%3.8%3.7%3.7%3.7%3.7%3.6%3.5%3.4%3.4%3.4%3.2%3.2%3.2%3.2%3.1%3.1%3.1%3.1%3.1%3%2.9%2.9%2.8%2.8%2.8%2.7%2.6%2.6%2.4%2.4%2.4%2.3%2.2%leblancbig397.8%sedovbig242.2%NASNet MobileSqueezeNetMobilenet FloatI.R.VEP.C106.8%14 digit106.3%Crypto105.7%Total Time105.2%2 - 4K - 16 - Path Tracer101.9%1 - 4K - 16 - Path Tracer99.8%3 - 4K - 16 - Path Tracer99.3%S.F.P.R98.8%CPU Stress98.7%CoreMark Size 666 - I.P.S97.3%SHA25696.8%D.R96.5%Exhaustive96.5%Total Time - 4.1.R.P.P95.7%SP.B95.5%L.E.H92.6%Total Time92.1%1.H.M.2.D91.2%RSA409690.3%2 - 1080p - 32 - Path Tracer89%Vector Math88.8%2 - 4K - 1 - Path Tracer88.8%2 - 1080p - 1 - Path Tracer88.7%3 - 4K - 1 - Path Tracer88.5%3 - 1080p - 1 - Path Tracer88.4%CPU88.3%1 - 1080p - 1 - Path Tracer88.1%3 - 1080p - 32 - Path Tracer88.1%2 - 1080p - 16 - Path Tracer87.3%1e1287.2%1 - 1080p - 32 - Path Tracer87.2%1 - 4K - 1 - Path Tracer87.1%3 - 1080p - 16 - Path Tracer86.7%RSA409685.3%1 - 1080p - 16 - Path Tracer84.9%BMW27 - CPU-Only84.7%2 - 4K - 32 - Path Tracer83.8%3 - 4K - 32 - Path Tracer83.4%Matrix Math83.2%gravity_spheres_volume/dim_512/ao/real_time83.1%DLSC - CPU83.1%CPU - Bedroom82.5%Classroom - CPU-Only82%1 - 4K - 32 - Path Tracer81.9%Barbershop - CPU-Only81.1%gravity_spheres_volume/dim_512/scivis/real_time80.4%V.P.M80.4%Fishy Cat - CPU-Only80.4%128 - 256 - 5780%F.1.0.R.a.m.i.C79.9%79.8%particle_volume/ao/real_time79.1%CPU - Supercar79%gravity_spheres_volume/dim_512/pathtracer/real_time78.5%Danish Mood - CPU78.2%LuxCore Benchmark - CPU77.6%particle_volume/scivis/real_time77%Pathtracer - Crown76.8%Pathtracer ISPC - Crown75.9%Bosphorus 4K - Ultra Fast75.4%Orange Juice - CPU74.2%Inception V4Bosphorus 4K - Very Fast72.2%64 - 256 - 5771.5%Trace Time70.8%SpaceshipPabellon Barcelona - CPU-Only68.6%96000 - 51268.6%ATPase Simulation - 327,506 Atoms68.2%1e1367.7%C240 Buckyball62.8%Eigen58.3%MPI CPU - water_GMX50_bare56.3%allmodconfig55.5%BLAS55.5%UASTC Level 353.5%Disney Material48%i.i.1.C.P.D43.8%ArcFace ResNet-100 - CPU - Standardd.S.M.S - Execution Time43.5%Ninja41.9%20k Atoms41%Pathtracer ISPC - Asian Dragon39.8%Time To Compile37.4%SP.C37.2%Rhodopsin Protein37.1%Time To Compile36.6%particle_volume/pathtracer/real_time35.6%Carbon Nanotube35.5%yolov4 - CPU - Standarddefconfig35%UASTC Level 234.7%Pathtracer - Asian Dragon34.4%i.i.1.C.P.D31.6%Time To Compile31.6%LU.C30.1%Boat - CPU-only28.7%Dust 2D tau100.0Small27.8%Unix Makefiles27.7%CG.CBT.C23.3%Mobilenet QuantThorough20.6%7 - Bosphorus 4K20.4%resizeEmily19.7%GPT-2 - CPU - StandardTime To Compile19.2%CPU CacheTradesoap18.4%d.S.M.S - Mesh Time17.6%32 - 256 - 5715.6%AUSURF112Church FacadeH2Rotate 90 DegreesCompression Rating13.7%Material Tester19 - Compression Speed13.4%LionWrites12.4%CPU - MobileNet v2Time To Compile11.8%11.7%unsharp-maskTime To Compile11.3%Time To Compile11%1.R.W.A.D.F.R.C.C10.9%fcn-resnet101-11 - CPU - Standard10.8%480000 - 51210.8%FT.C10.3%All9.8%auto-levelsScale1.R.W.A.D.T.R8.5%T.B.TVMAF Optimized - Bosphorus 4K8.1%JPEG - 7raytrace7.7%9 - D.S1.R.W.A.D.S.R7.6%crypto_pyaesINVITEPreset 12 - Bosphorus 4KTime To Compile7.1%Preset 10 - Bosphorus 4KSpeed 5 - Bosphorus 4KJythonC7552Croprotate10 - Bosphorus 4KMG.C6.1%KostyaQ.9.C.E.7Speed 6 Realtime - Bosphorus 4K5.9%19, Long Mode - D.SJPEG - 8Summer Nature 4KTime To CompileTradebeans4.P.1.P.M.6.QVoiceMark_100Bosphorus 1080p5.2%PNG - 819 - D.SMulti-Threaded - ETC24.9%Color EnhancemotorBike - Mesh TimeBosphorus 4K4.P.1.P.M.6.QQ.7.C.E.7SecureMark-TLSCPU - SqueezeNet v1.1116 - 256 - 57S.A.OSVG Files To PNGReactorElapsed Time480000 - 10244.4%Retro CarMini-ITX CaseSingle-Threaded - ETC2F.D.FF.B.t.B.F.FHilbert TransformC2670Pistol3.7%P.M.SmotorBike - Execution TimeD.TL.P.C.SP.S.O - Bosphorus 4K3.5%Square Plain C++Q.1.L.H.CQ.1.C.E.54.P.1.P.S.2.Q4.P.1.P.S.2.Q5.P.T.5.P.S.6.QComplex PhaseOPTIONS, StatefulDefaultC.u.1.0.3.s.i.i.C.L.9DistinctUserIDjson_loads5.P.T.5.P.S.6.Q44100 - 10243%Q.1.LQ.1.H.CPartialTweets4.P.1.P.S.6.QTopTweet19, Long Mode - Compression Speed2.8%Speed 10 Realtime - Bosphorus 4K4.P.1.P.M.2.Q4.P.1.P.S.6.Q9 - Compression SpeedA.C.PCPU - DenseNet4.P.1.P.M.2.QCPU - SqueezeNet v2PennantPennantTensorFlow LiteTensorFlow LiteTensorFlow LiteTensorFlow LiteNAS Parallel BenchmarksHelsingStress-NGStockfishOSPRay StudioOSPRay StudioOSPRay StudioACES DGEMMStress-NGCoremarkOpenSSL7-Zip CompressionASTC EncoderC-RayNAS Parallel BenchmarksCloverLeafTachyonasmFishOpenSSLOSPRay StudioStress-NGOSPRay StudioOSPRay StudioOSPRay StudioOSPRay StudioChaos Group V-RAYOSPRay StudioOSPRay StudioOSPRay StudioPrimesieveOSPRay StudioOSPRay StudioOSPRay StudioOpenSSLOSPRay StudioBlenderOSPRay StudioOSPRay StudioStress-NGOSPRayLuxCoreRenderIndigoBenchBlenderOSPRay StudioBlenderOSPRayBRL-CADBlenderLiquid-DSPParallel BZIP2 CompressionAircrack-ngOSPRayIndigoBenchOSPRayLuxCoreRenderLuxCoreRenderOSPRayEmbreeEmbreeKvazaarLuxCoreRenderTensorFlow LiteKvazaarLiquid-DSPPOV-RayNatronBlenderStargate Digital Audio WorkstationNAMDPrimesieveNWChemLeelaChessZeroGROMACSTimed Linux Kernel CompilationLeelaChessZeroBasis UniversalAppleseedXcompact3d Incompact3dONNX RuntimeOpenFOAMTimed LLVM CompilationLAMMPS Molecular Dynamics SimulatorEmbreeTimed MPlayer CompilationNAS Parallel BenchmarksLAMMPS Molecular Dynamics SimulatorTimed Node.js CompilationOSPRayGPAWONNX RuntimeTimed Linux Kernel CompilationBasis UniversalEmbreeXcompact3d Incompact3dTimed FFmpeg CompilationNAS Parallel BenchmarksDarktableMonte Carlo Simulations of Ionised NebulaeminiFETimed LLVM CompilationNAS Parallel BenchmarksNAS Parallel BenchmarksTensorFlow LiteASTC EncoderSVT-HEVCKripkeGIMPAppleseedONNX RuntimeTimed Godot Game Engine CompilationStress-NGDaCapo BenchmarkLULESHOpenFOAMLiquid-DSPQuantum ESPRESSOGoogle DracoDaCapo BenchmarkGEGL7-Zip CompressionAppleseedZstd CompressionGoogle DracoApache CassandraTNNTimed ImageMagick CompilationAlgebraic Multi-Grid BenchmarkGIMPTimed Mesa CompilationTimed Gem5 CompilationClickHouseONNX RuntimeStargate Digital Audio WorkstationNAS Parallel BenchmarksJPEG XL Decoding libjxlGIMPGEGLClickHouseRawTherapeeSVT-VP9JPEG XL libjxlPyPerformanceLZ4 CompressionClickHousePyPerformanceGNU Octave BenchmarkPJSIPSVT-AV1Build2SVT-AV1VP9 libvpx EncodingDaCapo BenchmarkNode.js V8 Web Tooling BenchmarkNgspiceGEGLGIMPSVT-HEVCNAS Parallel BenchmarkssimdjsonWebP2 Image EncodeAOM AV1Zstd CompressionJPEG XL libjxllibgav1Timed Apache CompilationDaCapo BenchmarksrsRANGoogle SynthMarkx265JPEG XL libjxlZstd CompressionEtcpakGEGLOpenFOAMx265srsRANWebP2 Image EncodeSecureMarkTNNJPEG XL Decoding libjxlLiquid-DSPChia Blockchain VDFInkscapeJava Gradle BuildQuantLibCraftyStargate Digital Audio WorkstationNumpy BenchmarkOpenSCADOpenSCADEtcpakLuaRadioLuaRadioLuaRadioNgspiceOpenSCADHigh Performance Conjugate GradientOpenSCADOpenFOAMlibjpeg-turbo tjbenchOpenSCADSVT-VP9Chia Blockchain VDFWebP Image EncodeWebP2 Image EncodesrsRANsrsRANsrsRANLuaRadioPJSIPWebP2 Image EncodeXZ CompressionsimdjsonPyPerformancesrsRANStargate Digital Audio WorkstationWebP Image EncodeWebP Image EncodesimdjsonsrsRANWireGuard + Linux Networking Stack Stress TestsimdjsonZstd CompressionAOM AV1srsRANsrsRANLZ4 CompressionTSCPTNNsrsRANTNNThreadripper 3990XThreadripper 3960X

AMD Ryzen Threadripper 3990Xnwchem: C240 Buckyballjava-gradle-perf: Reactoropenfoam: drivaerFastback, Small Mesh Size - Execution Timeopenfoam: drivaerFastback, Small Mesh Size - Mesh Timelczero: Eigenlczero: BLASblender: Barbershop - CPU-Onlyopenssl: SHA256lammps: 20k Atomsqe: AUSURF112stockfish: Total Timehpcg: build-linux-kernel: allmodconfigospray: particle_volume/pathtracer/real_timeclickhouse: 100M Rows Web Analytics Dataset, Third Runclickhouse: 100M Rows Web Analytics Dataset, Second Runclickhouse: 100M Rows Web Analytics Dataset, First Run / Cold Cachengspice: C7552onnx: fcn-resnet101-11 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardonnx: yolov4 - CPU - Standardjpegxl: PNG - 8securemark: SecureMark-TLSospray: particle_volume/scivis/real_timewireguard: brl-cad: VGR Performance Metricbuild-llvm: Unix Makefilesluaradio: Complex Phaseluaradio: Hilbert Transformluaradio: FM Deemphasis Filterluaradio: Five Back to Back FIR Filtersgraph500: 26graph500: 26graph500: 26graph500: 26webp2: Quality 95, Compression Effort 7numpy: mocassin: Dust 2D tau100.0build-llvm: Ninjatnn: CPU - DenseNetospray: particle_volume/ao/real_timebuild-gem5: Time To Compiletensorflow-lite: Inception V4asmfish: 1024 Hash Memory, 26 Depthbuild-nodejs: Time To Compileprimesieve: 1e13ngspice: C2670ospray-studio: 3 - 4K - 32 - Path Tracerstargate: 96000 - 512helsing: 14 digitstress-ng: CPU Cacheblender: Pabellon Barcelona - CPU-Onlyospray-studio: 2 - 4K - 32 - Path Tracerospray-studio: 3 - 1080p - 16 - Path Tracerospray-studio: 3 - 4K - 1 - Path Tracercompress-lz4: 9 - Decompression Speedcompress-lz4: 9 - Compression Speedospray-studio: 3 - 4K - 16 - Path Tracerospray-studio: 3 - 1080p - 32 - Path Tracerospray-studio: 1 - 4K - 32 - Path Tracercompress-zstd: 19 - Decompression Speedcompress-zstd: 19 - Compression Speedaom-av1: Speed 6 Realtime - Bosphorus 4Kospray-studio: 1 - 4K - 1 - Path Tracerospray-studio: 2 - 1080p - 16 - Path Tracerospray-studio: 2 - 4K - 1 - Path Tracerospray-studio: 1 - 1080p - 16 - Path Tracernatron: Spaceshipgpaw: Carbon Nanotubecassandra: Writesospray-studio: 2 - 4K - 16 - Path Traceronnx: GPT-2 - CPU - Standardospray-studio: 2 - 1080p - 32 - Path Tracerospray-studio: 1 - 4K - 16 - Path Tracerblender: Classroom - CPU-Onlyospray-studio: 1 - 1080p - 32 - Path Tracerospray: gravity_spheres_volume/dim_512/scivis/real_timeospray: gravity_spheres_volume/dim_512/ao/real_timeospray-studio: 3 - 1080p - 1 - Path Tracerospray-studio: 1 - 1080p - 1 - Path Tracerospray-studio: 2 - 1080p - 1 - Path Tracerospray: gravity_spheres_volume/dim_512/pathtracer/real_timewebp2: Quality 75, Compression Effort 7appleseed: Emilyjpegxl: PNG - 7mt-dgemm: Sustained Floating-Point Rateopenscad: Pistolappleseed: Material Testerpyperformance: python_startupstargate: 480000 - 512pyperformance: raytracenode-web-tooling: libgav1: Summer Nature 4Kchia-vdf: Square Assembly Optimizednpb: SP.Cv-ray: CPUsimdjson: PartialTweetssimdjson: DistinctUserIDsimdjson: TopTweetgromacs: MPI CPU - water_GMX50_barepjsip: INVITEpjsip: OPTIONS, Statefulluxcorerender: Orange Juice - CPUluxcorerender: LuxCore Benchmark - CPUluxcorerender: Danish Mood - CPUindigobench: CPU - Bedroomindigobench: CPU - Supercarbuild2: Time To Compileluxcorerender: DLSC - CPUtensorflow-lite: Inception ResNet V2tensorflow-lite: NASNet Mobiletensorflow-lite: Mobilenet Floattensorflow-lite: SqueezeNettensorflow-lite: Mobilenet Quantopenssl: RSA4096openssl: RSA4096build-linux-kernel: defconfigetcpak: Single-Threaded - ETC2vpxenc: Speed 5 - Bosphorus 4Kblender: Fishy Cat - CPU-Onlysimdjson: Kostyaamg: chia-vdf: Square Plain C++incompact3d: input.i3d 193 Cells Per Directiongegl: Rotate 90 Degreesgegl: Color Enhanceappleseed: Disney Materialpjsip: OPTIONS, Statelessrawtherapee: Total Benchmark Timebuild-godot: Time To Compilenpb: BT.Caircrack-ng: namd: ATPase Simulation - 327,506 Atomssvt-vp9: VMAF Optimized - Bosphorus 4Kblender: BMW27 - CPU-Onlycompress-7zip: Decompression Ratingcompress-7zip: Compression Ratingcompress-zstd: 19, Long Mode - Decompression Speedcompress-zstd: 19, Long Mode - Compression Speedopenscad: Mini-ITX Casestargate: 96000 - 1024tachyon: Total Timejpegxl-decode: 1openfoam: motorBike - Execution Timeopenfoam: motorBike - Mesh Timesrsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMsrsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMnpb: LU.Cquantlib: webp: Quality 100, Lossless, Highest Compressionsrsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMsrsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMgimp: resizesrsran: OFDM_Testminife: Smallpyperformance: regex_compilestargate: 44100 - 512stargate: 480000 - 1024lulesh: stargate: 44100 - 1024stress-ng: CPU Stressstress-ng: Cryptostress-ng: Matrix Mathstress-ng: Vector Mathsynthmark: VoiceMark_100pennant: sedovbiginkscape: SVG Files To PNGpennant: leblancbigkripke: x265: Bosphorus 4Kcoremark: CoreMark Size 666 - Iterations Per Secondkvazaar: Bosphorus 4K - Very Fasttjbench: Decompression Throughputastcenc: Exhaustivesvt-hevc: 7 - Bosphorus 4Kcloverleaf: Lagrangian-Eulerian Hydrodynamicsgimp: unsharp-maskpybench: Total For Average Test Timescrafty: Elapsed Timepyperformance: json_loadsbasis: UASTC Level 3srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMsrsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMnpb: FT.Cbuild-mesa: Time To Compilepyperformance: crypto_pyaesbuild-ffmpeg: Time To Compileembree: Pathtracer ISPC - Asian Dragonsrsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMsrsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMcompress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9c-ray: Total Time - 4K, 16 Rays Per Pixelliquid-dsp: 128 - 256 - 57embree: Pathtracer ISPC - Crownliquid-dsp: 64 - 256 - 57aom-av1: Speed 8 Realtime - Bosphorus 4Kliquid-dsp: 32 - 256 - 57liquid-dsp: 16 - 256 - 57build-apache: Time To Compileembree: Pathtracer - Asian Dragontnn: CPU - SqueezeNet v1.1aom-av1: Speed 10 Realtime - Bosphorus 4Kwebp: Quality 100, Losslessbasis: UASTC Level 2tnn: CPU - MobileNet v2aom-av1: Speed 9 Realtime - Bosphorus 4Kgimp: auto-levelssrsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMsrsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMnpb: CG.Cbuild-mplayer: Time To Compilebuild-imagemagick: Time To Compileembree: Pathtracer - Crownnpb: SP.Bincompact3d: input.i3d 129 Cells Per Directionpovray: Trace Timex265: Bosphorus 1080pjpegxl: JPEG - 7dacapobench: Tradebeansjpegxl-decode: Alljpegxl: JPEG - 8openscad: Leonardo Phone Case Slimkvazaar: Bosphorus 4K - Ultra Fastsvt-av1: Preset 8 - Bosphorus 4Kdraco: Church Facadegimp: rotategegl: Scaleopenscad: Projector Mount Swiveldacapobench: Tradesoapsvt-vp9: Visual Quality Optimized - Bosphorus 4Kdacapobench: H2svt-vp9: PSNR/SSIM Optimized - Bosphorus 4Kgegl: Cropcompress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compressionnpb: MG.Cwebp: Quality 100, Highest Compressionastcenc: Thoroughoctave-benchmark: draco: Lionsvt-av1: Preset 10 - Bosphorus 4Ksvt-av1: Preset 12 - Bosphorus 4Kdacapobench: Jythonwebp2: Defaultprimesieve: 1e12svt-hevc: 10 - Bosphorus 4Ktnn: CPU - SqueezeNet v2etcpak: Multi-Threaded - ETC2npb: EP.Copenscad: Retro Cardarktable: Masskrug - CPU-onlydarktable: Server Room - CPU-onlywebp2: Quality 100, Compression Effort 5darktable: Boat - CPU-onlylammps: Rhodopsin Proteintscp: AI Chess PerformanceThreadripper 3990XThreadripper 3960X2598.8297.7071369.2174.2715641659342.545923302074824.393447.691510758329.06946262.434145.882260.33255.76250.20109.12615410213020.9821885113.1957273.858714610212.280635.8104.3415.7762.4535569000215686000543780000161247000233.613416.75237161.8162776.39613.4124179.02656545.0136487533148.32552.276143.4771081303.746460100.605150.17110.339334012958319311528.555.325133425752915863371.281.218.742688110862736109023.0107.872240022439866980219754318384.35216196.522056.9707880467368910.8220108.798138.13589610.8917.35457198.229153.51178714.05.3752344409.5441.7010700023124.73469713.784.204.293.7883711664612.006.595.979.83121.08159.3007.8085767.451346.94229.256181.903296.10854108.513533.328.878225.85010.1440.062.6845562220016336745.392533655.77952.90362.5921874087250.82845.10268363.63162110.4430.45309102.9830.173536951927243376.844.543.6884.27008528.492069.3772.8639.87157.0392.765067.752790.434.116142.8356.724.1811395666679889.261575.3469825.6817479413.40575.788816123897.4592388.85219165.21340961.12741.32612.3153430.5395.57850010599320020.992248635.66649631.17215.56678613.9116119.0014.7424.8271017836917423.615.51158.9103.728515.3720.35111418.01142.2879159.7368.720.82211.041301130000049.2568281283333331.85178460000098047333320.46544.2504244.40444.0816.50510.944288.35145.1419.829166.8392.98366.7410.60913.13654.314548358.849.3721057910.05346.9691.955322539.7128.2517.29860.6654.428886116.69212.5898.1623800103.624532108.868.4812.11426818.017.2056.08847.2205616109.598133.03643182.4333.738151.1868.9714515.8604690.734.4302.3552.1373.0661.61423.14713250534232.1284.8181964.24205.019881067620.493009680859017.298388.17736212508.74452408.134107.567239.85237.69225.58102.36713914684081.032292647.45690266.383396127271.109656.0108.4432.3792.7220.464434.62185229.5822711.9477.48827198.65232727.171399960202.55487.691138.1931982882.221881207.597178.66186.0617151324189601812417.456.66102284484331666033538.671.617.695030207685165201635.1146.1952135088882383504153286291153.52404613.615203.806871515126613006.06147103.823165.38790210.918.72831894.707135.28376813.94.85267047410.1844.0711186716854.32249413.894.334.412.423398268566.893.713.355.38811.77663.5334.2639153.021105.71890.722541.582697.41460980.27110.238.999234.96510.8372.252.8440794600016903359.750363748.66650.42792.6575684040846.96353.75255451.8590173.5390.7619895.2355.711799531695223570.943.341.9404.22073154.719972.6270.2738.02160.6403.150029.412914.732.985150.5373.820.1531418666677737.031605.3382835.4428047952.57835.61954062366.7644920.13119634.42180602.86779.85742.1420929.21327.7689512727630022.011139722.36167618.10223.46906027.335198.8628.3922.2621003874128522.923.80360.7107.025858.5422.65210623.70030.2566163.9379.220.19621.603167270000027.9976164000000031.681543400000102590000019.38332.9238233.43545.2716.03214.740257.48945.6018.096172.4405.510603.8214.57814.68130.728524742.2613.476237117.16644.6699.075043491.6329.8716.69134.5855.048770915.67111.5817.8703198101.723948105.147.9593.80325278.426.9997.34066.7174987117.076142.55440462.3596.999160.6267.5114303.3682268.064.2522.3932.1202.9652.07816.8851356605OpenBenchmarking.org

NWChem

NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 BuckyballThreadripper 3960XThreadripper 3990X90018002700360045004232.12598.81. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz -lcomex -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2
OpenBenchmarking.orgWatts, Fewer Is BetterNWChem 7.0.2CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 46.34 / Avg: 270 / Max: 287.17Min: 57.83 / Avg: 255.42 / Max: 279.97

Java Gradle Build

This test runs Java software project builds using the Gradle build system. It is intended to give developers an idea as to the build performance for development activities and build servers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterJava Gradle BuildGradle Build: ReactorThreadripper 3960XThreadripper 3990X60120180240300SE +/- 7.63, N = 9SE +/- 5.65, N = 9284.82297.71
OpenBenchmarking.orgWatts, Fewer Is BetterJava Gradle BuildCPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 42.65 / Avg: 97.11 / Max: 208.8Min: 52.66 / Avg: 130.86 / Max: 271.04

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 9Input: drivaerFastback, Small Mesh Size - Execution TimeThreadripper 3960XThreadripper 3990X4008001200160020001964.241369.201. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
OpenBenchmarking.orgWatts, Fewer Is BetterOpenFOAM 9CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 43.61 / Avg: 201.23 / Max: 271.06Min: 54.54 / Avg: 260.25 / Max: 280.06

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 9Input: drivaerFastback, Small Mesh Size - Mesh TimeThreadripper 3960XThreadripper 3990X4080120160200205.01174.271. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

LeelaChessZero

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

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: EigenThreadripper 3960XThreadripper 3990X30060090012001500SE +/- 10.14, N = 5SE +/- 19.50, N = 498815641. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.28Backend: EigenThreadripper 3960XThreadripper 3990X2468104.0826.370
OpenBenchmarking.orgWatts, Fewer Is BetterLeelaChessZero 0.28CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 45.13 / Avg: 242.03 / Max: 287.14Min: 61.38 / Avg: 245.53 / Max: 280.55

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: BLASThreadripper 3960XThreadripper 3990X400800120016002000SE +/- 11.76, N = 5SE +/- 6.17, N = 3106716591. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.28Backend: BLASThreadripper 3960XThreadripper 3990X2468104.4296.811
OpenBenchmarking.orgWatts, Fewer Is BetterLeelaChessZero 0.28CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 43.49 / Avg: 240.9 / Max: 286.75Min: 59.68 / Avg: 243.57 / Max: 281.45

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.2Blend File: Barbershop - Compute: CPU-OnlyThreadripper 3960XThreadripper 3990X130260390520650SE +/- 0.45, N = 3SE +/- 2.66, N = 3620.49342.54
OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.2CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 48.3 / Avg: 283.47 / Max: 288.3Min: 57.72 / Avg: 267.26 / Max: 280.34

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.0Algorithm: SHA256Threadripper 3960XThreadripper 3990X13000M26000M39000M52000M65000MSE +/- 118264097.95, N = 3SE +/- 448484476.56, N = 1230096808590592330207481. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
OpenBenchmarking.orgbyte/s Per Watt, More Is BetterOpenSSL 3.0Algorithm: SHA256Threadripper 3960XThreadripper 3990X50M100M150M200M250M118385444.12233130767.15
MinAvgMaxThreadripper 3960X43.4254.2274.1Threadripper 3990X53.0254.1281.4OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.0CPU Power Consumption Monitor70140210280350

LAMMPS Molecular Dynamics Simulator

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

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: 20k AtomsThreadripper 3960XThreadripper 3990X612182430SE +/- 0.06, N = 3SE +/- 0.12, N = 317.3024.391. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgns/day Per Watt, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: 20k AtomsThreadripper 3960XThreadripper 3990X0.02120.04240.06360.08480.1060.0660.094
OpenBenchmarking.orgWatts, Fewer Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 47.37 / Avg: 263.79 / Max: 267.93Min: 60.52 / Avg: 260.1 / Max: 279.79

Quantum ESPRESSO

Quantum ESPRESSO is an integrated suite of Open-Source computer codes for electronic-structure calculations and materials modeling at the nanoscale. It is based on density-functional theory, plane waves, and pseudopotentials. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterQuantum ESPRESSO 7.0Input: AUSURF112Threadripper 3960XThreadripper 3990X100200300400500SE +/- 0.62, N = 3SE +/- 0.50, N = 3388.17447.691. (F9X) gfortran options: -pthread -fopenmp -ldevXlib -lopenblas -lFoX_dom -lFoX_sax -lFoX_wxml -lFoX_common -lFoX_utils -lFoX_fsys -lfftw3_omp -lfftw3 -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
MinAvgMaxThreadripper 3960X25.2265.9288.6Threadripper 3990X60.6253.5280.5OpenBenchmarking.orgWatts, Fewer Is BetterQuantum ESPRESSO 7.0CPU Power Consumption Monitor70140210280350

Stockfish

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

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 15Total TimeThreadripper 3960XThreadripper 3990X30M60M90M120M150MSE +/- 814613.01, N = 15SE +/- 2762000.09, N = 14736212501510758321. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -flto -flto=jobserver
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterStockfish 15Total TimeThreadripper 3960XThreadripper 3990X130K260K390K520K650K270194.72588417.21
MinAvgMaxThreadripper 3960X45.1272.5287.8Threadripper 3990X60.1256.7279.9OpenBenchmarking.orgWatts, Fewer Is BetterStockfish 15CPU Power Consumption Monitor70140210280350

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Threadripper 3960XThreadripper 3990X3691215SE +/- 0.00808, N = 3SE +/- 0.00092, N = 38.744529.069461. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHigh Performance Conjugate Gradient 3.1Threadripper 3960XThreadripper 3990X0.01060.02120.03180.04240.0530.0470.033
MinAvgMaxThreadripper 3960X42.5188.0196.7Threadripper 3990X53.5274.6281.9OpenBenchmarking.orgWatts, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Power Consumption Monitor70140210280350

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.18Build: allmodconfigThreadripper 3960XThreadripper 3990X90180270360450SE +/- 0.46, N = 3SE +/- 0.73, N = 3408.13262.43
OpenBenchmarking.orgWatts, Fewer Is BetterTimed Linux Kernel Compilation 5.18CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 47.2 / Avg: 266.83 / Max: 287.18Min: 56 / Avg: 251.26 / Max: 282.76

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity 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.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: particle_volume/pathtracer/real_timeThreadripper 3960XThreadripper 3990X306090120150SE +/- 0.01, N = 3SE +/- 0.90, N = 3107.57145.88
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.10Benchmark: particle_volume/pathtracer/real_timeThreadripper 3960XThreadripper 3990X0.12760.25520.38280.51040.6380.4310.567
MinAvgMaxThreadripper 3960X45.4249.5288.3Threadripper 3990X58.0257.3279.9OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.10CPU Power Consumption Monitor70140210280350

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all queries performed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse100M Rows Web Analytics Dataset, Third RunThreadripper 3960XThreadripper 3990X60120180240300SE +/- 1.78, N = 15SE +/- 2.52, N = 12239.85260.33MIN: 9.87 / MAX: 15000MIN: 19.03 / MAX: 200001. ClickHouse server version 22.7.1.687 (official build).
OpenBenchmarking.orgQueries Per Minute, Geo Mean Per Watt, More Is BetterClickHouse100M Rows Web Analytics Dataset, Third RunThreadripper 3960XThreadripper 3990X0.36290.72581.08871.45161.81451.6131.148
MinAvgMaxThreadripper 3960X42.2148.7238.9Threadripper 3990X52.8226.7280.3OpenBenchmarking.orgWatts, Fewer Is BetterClickHouseCPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse100M Rows Web Analytics Dataset, Second RunThreadripper 3960XThreadripper 3990X60120180240300SE +/- 1.87, N = 15SE +/- 2.05, N = 12237.69255.76MIN: 16.17 / MAX: 20000MIN: 18.46 / MAX: 200001. ClickHouse server version 22.7.1.687 (official build).

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse100M Rows Web Analytics Dataset, First Run / Cold CacheThreadripper 3960XThreadripper 3990X50100150200250SE +/- 2.70, N = 15SE +/- 4.60, N = 12225.58250.20MIN: 17.1 / MAX: 12000MIN: 18.42 / MAX: 200001. ClickHouse server version 22.7.1.687 (official build).

Ngspice

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

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552Threadripper 3960XThreadripper 3990X20406080100SE +/- 0.73, N = 15SE +/- 0.71, N = 3102.37109.131. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE
OpenBenchmarking.orgWatts, Fewer Is BetterNgspice 34CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X20406080100Min: 40.42 / Avg: 79.61 / Max: 85.23Min: 53.34 / Avg: 107.96 / Max: 112.79

ONNX Runtime

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

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: StandardThreadripper 3960XThreadripper 3990X306090120150SE +/- 5.85, N = 12SE +/- 1.09, N = 31391541. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: StandardThreadripper 3960XThreadripper 3990X0.14870.29740.44610.59480.74350.5620.661
OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 43.42 / Avg: 247.4 / Max: 284.08Min: 55.76 / Avg: 232.84 / Max: 243.95

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardThreadripper 3960XThreadripper 3990X30060090012001500SE +/- 13.51, N = 12SE +/- 2.29, N = 3146810211. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardThreadripper 3960XThreadripper 3990X2468107.2223.811
OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 43.34 / Avg: 203.27 / Max: 211.72Min: 60.57 / Avg: 267.87 / Max: 279.72

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: StandardThreadripper 3960XThreadripper 3990X90180270360450SE +/- 9.99, N = 12SE +/- 0.93, N = 34083021. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: StandardThreadripper 3960XThreadripper 3990X0.48350.9671.45051.9342.41752.1491.116
OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption MonitorThreadripper 3960XThreadripper 3990X50100150200250Min: 44.37 / Avg: 189.85 / Max: 211.51Min: 57 / Avg: 270.53 / Max: 279.98

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: PNG - Encode Speed: 8Threadripper 3960XThreadripper 3990X0.23180.46360.69540.92721.159SE +/- 0.00, N = 3SE +/- 0.00, N = 31.030.981. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie
OpenBenchmarking.orgMP/s Per Watt, More Is BetterJPEG XL libjxl 0.6.1Input: PNG - Encode Speed: 8Threadripper 3960XThreadripper 3990X0.00270.00540.00810.01080.01350.0120.008
MinAvgMaxThreadripper 3960X42.689.3129.6Threadripper 3990X56.0116.9152.5OpenBenchmarking.orgWatts, Fewer Is BetterJPEG XL libjxl 0.6.1CPU Power Consumption Monitor4080120160200

SecureMark

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

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSThreadripper 3960XThreadripper 3990X50K100K150K200K250KSE +/- 104.83, N = 3SE +/- 671.44, N = 32292642188511. (CC) gcc options: -pedantic -O3
OpenBenchmarking.orgmarks Per Watt, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSThreadripper 3960XThreadripper 3990X60012001800240030002770.401953.40
MinAvgMaxThreadripper 3960X42.582.886.2Threadripper 3990X52.0112.0117.7OpenBenchmarking.orgWatts, Fewer Is BetterSecureMark 1.0.4CPU Power Consumption Monitor4080120160200

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity 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.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: particle_volume/scivis/real_timeThreadripper 3960XThreadripper 3990X3691215SE +/- 0.00052, N = 3SE +/- 0.04512, N = 37.4569013.19570
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.10Benchmark: particle_volume/scivis/real_timeThreadripper 3960XThreadripper 3990X0.01170.02340.03510.04680.05850.0320.052
MinAvgMaxThreadripper 3960X44.1229.9273.5Threadripper 3990X57.9254.2279.9OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.10CPU Power Consumption Monitor70140210280350

WireGuard + Linux Networking Stack Stress Test

This is a benchmark of the WireGuard secure VPN tunnel and Linux networking stack stress test. The test runs on the local host but does require root permissions to run. The way it works is it creates three namespaces. ns0 has a loopback device. ns1 and ns2 each have wireguard devices. Those two wireguard devices send traffic through the loopback device of ns0. The end result of this is that tests wind up testing encryption and decryption at the same time -- a pretty CPU and scheduler-heavy workflow. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWireGuard + Linux Networking Stack Stress TestThreadripper 3960XThreadripper 3990X60120180240300SE +/- 0.50, N = 3SE +/- 0.99, N = 3266.38273.86
MinAvgMaxThreadripper 3960X41.9130.9167.6Threadripper 3990X51.1219.3266.0OpenBenchmarking.orgWatts, Fewer Is BetterWireGuard + Linux Networking Stack Stress TestCPU Power Consumption Monitor70140210280350

BRL-CAD

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

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.32.2VGR Performance MetricThreadripper 3960XThreadripper 3990X150K300K450K600K750K3961277146101. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -ldl -lm
OpenBenchmarking.orgVGR Performance Metric Per Watt, More Is BetterBRL-CAD 7.32.2VGR Performance MetricThreadripper 3960XThreadripper 3990X60012001800240030001403.402761.35
MinAvgMaxThreadripper 3960X45.0282.3288.3Threadripper 3990X58.5258.8280.2OpenBenchmarking.orgWatts, Fewer Is BetterBRL-CAD 7.32.2CPU Power Consumption Monitor70140210280350

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 13.0Build System: Unix MakefilesThreadripper 3960XThreadripper 3990X60120180240300SE +/- 3.61, N = 3SE +/- 0.71, N = 3271.11212.28
MinAvgMaxThreadripper 3960X44.8227.6288.3Threadripper 3990X59.0227.1282.6OpenBenchmarking.orgWatts, Fewer Is BetterTimed LLVM Compilation 13.0CPU Power Consumption Monitor70140210280350

LuaRadio

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

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Complex PhaseThreadripper 3960XThreadripper 3990X140280420560700SE +/- 3.20, N = 3SE +/- 0.35, N = 3656.0635.8
OpenBenchmarking.orgMiB/s Per Watt, More Is BetterLuaRadio 0.9.1Test: Complex PhaseThreadripper 3960XThreadripper 3990X2468106.9425.112
MinAvgMaxThreadripper 3960X42.894.5124.1Threadripper 3990X53.3124.4167.5OpenBenchmarking.orgWatts, Fewer Is BetterLuaRadio 0.9.1CPU Power Consumption Monitor50100150200250

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Hilbert TransformThreadripper 3960XThreadripper 3990X20406080100SE +/- 0.03, N = 3SE +/- 0.13, N = 3108.4104.3

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: FM Deemphasis FilterThreadripper 3960XThreadripper 3990X90180270360450SE +/- 0.22, N = 3SE +/- 0.35, N = 3432.3415.7

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Five Back to Back FIR FiltersThreadripper 3960XThreadripper 3990X2004006008001000SE +/- 1.13, N = 3SE +/- 1.05, N = 3792.7762.4

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbfs median_TEPS, More Is BetterGraph500 3.0Scale: 26Threadripper 3990X110M220M330M440M550M5355690001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Scale: 26

Threadripper 3960X: The test quit with a non-zero exit status. E: mpirun noticed that process rank 2 with PID 0 on node phoronix-MS-7C59 exited on signal 9 (Killed).

OpenBenchmarking.orgsssp max_TEPS, More Is BetterGraph500 3.0Scale: 26Threadripper 3990X50M100M150M200M250M2156860001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
OpenBenchmarking.orgsssp max_TEPS Per Watt, More Is BetterGraph500 3.0Scale: 26Threadripper 3990X200K400K600K800K1000K797055.48
OpenBenchmarking.orgWatts, Fewer Is BetterGraph500 3.0CPU Power Consumption MonitorThreadripper 3990X50100150200250Min: 56.12 / Avg: 270.6 / Max: 281.39

Scale: 26

Threadripper 3960X: The test quit with a non-zero exit status. E: mpirun noticed that process rank 2 with PID 0 on node phoronix-MS-7C59 exited on signal 9 (Killed).

OpenBenchmarking.orgbfs max_TEPS, More Is BetterGraph500 3.0Scale: 26Threadripper 3990X120M240M360M480M600M5437800001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Scale: 26

Threadripper 3960X: The test quit with a non-zero exit status. E: mpirun noticed that process rank 2 with PID 0 on node phoronix-MS-7C59 exited on signal 9 (Killed).

OpenBenchmarking.orgsssp median_TEPS, More Is BetterGraph500 3.0Scale: 26Threadripper 3990X30M60M90M120M150M1612470001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Scale: 26

Threadripper 3960X: The test quit with a non-zero exit status. E: mpirun noticed that process rank 2 with PID 0 on node phoronix-MS-7C59 exited on signal 9 (Killed).

WebP2 Image Encode

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

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 95, Compression Effort 7Threadripper 3960XThreadripper 3990X50100150200250SE +/- 0.08, N = 3SE +/- 0.48, N = 3220.46233.611. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
MinAvgMaxThreadripper 3960X43.8230.3249.7Threadripper 3990X57.3233.2277.7OpenBenchmarking.orgWatts, Fewer Is BetterWebP2 Image Encode 20220422CPU Power Consumption Monitor70140210280350

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkThreadripper 3960XThreadripper 3990X90180270360450SE +/- 1.60, N = 3SE +/- 1.26, N = 3434.62416.75
OpenBenchmarking.orgScore Per Watt, More Is BetterNumpy BenchmarkThreadripper 3960XThreadripper 3990X1.13382.26763.40144.53525.6695.0393.585
MinAvgMaxThreadripper 3960X42.186.293.5Threadripper 3990X53.8116.3125.3OpenBenchmarking.orgWatts, Fewer Is BetterNumpy BenchmarkCPU Power Consumption Monitor4080120160200

Monte Carlo Simulations of Ionised Nebulae

Mocassin is the Monte Carlo Simulations of Ionised Nebulae. MOCASSIN is a fully 3D or 2D photoionisation and dust radiative transfer code which employs a Monte Carlo approach to the transfer of radiation through media of arbitrary geometry and density distribution. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMonte Carlo Simulations of Ionised Nebulae 2019-03-24Input: Dust 2D tau100.0Threadripper 3960XThreadripper 3990X50100150200250SE +/- 0.00, N = 3SE +/- 0.33, N = 31852371. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O3 -O2 -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lz
MinAvgMaxThreadripper 3960X45.4206.5243.6Threadripper 3990X52.5260.5280.1OpenBenchmarking.orgWatts, Fewer Is BetterMonte Carlo Simulations of Ionised Nebulae 2019-03-24CPU Power Consumption Monitor70140210280350

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 13.0Build System: NinjaThreadripper 3960XThreadripper 3990X50100150200250SE +/- 0.39, N = 3SE +/- 0.13, N = 3229.58161.82
MinAvgMaxThreadripper 3960X45.5251.2287.4Threadripper 3990X57.1234.3282.2OpenBenchmarking.orgWatts, Fewer Is BetterTimed LLVM Compilation 13.0CPU Power Consumption Monitor70140210280350

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: DenseNetThreadripper 3960XThreadripper 3990X6001200180024003000SE +/- 2.89, N = 3SE +/- 6.92, N = 32711.952776.40MIN: 2655.34 / MAX: 2817.65MIN: 2575.3 / MAX: 3223.811. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
MinAvgMaxThreadripper 3960X42.9124.6130.4Threadripper 3990X58.2211.3231.0OpenBenchmarking.orgWatts, Fewer Is BetterTNN 0.3CPU Power Consumption Monitor60120180240300

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity 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.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: particle_volume/ao/real_timeThreadripper 3960XThreadripper 3990X3691215SE +/- 0.01051, N = 3SE +/- 0.06547, N = 37.4882713.41240
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.10Benchmark: particle_volume/ao/real_timeThreadripper 3960XThreadripper 3990X0.01150.0230.03450.0460.05750.0280.051
MinAvgMaxThreadripper 3960X44.8264.9271.7Threadripper 3990X54.2260.7279.9OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.10CPU Power Consumption Monitor70140210280350

Timed Gem5 Compilation

This test times how long it takes to compile Gem5. Gem5 is a simulator for computer system architecture research. Gem5 is widely used for computer architecture research within the industry, academia, and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 21.2Time To CompileThreadripper 3960XThreadripper 3990X4080120160200SE +/- 1.10, N = 3SE +/- 1.38, N = 3198.65179.03
MinAvgMaxThreadripper 3960X45.0199.1288.3Threadripper 3990X56.5180.1281.2OpenBenchmarking.orgWatts, Fewer Is BetterTimed Gem5 Compilation 21.2CPU Power Consumption Monitor70140210280350

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception V4Threadripper 3960XThreadripper 3990X12K24K36K48K60KSE +/- 214.80, N = 3SE +/- 547.67, N = 1532727.156545.0
MinAvgMaxThreadripper 3960X44.7237.9252.7Threadripper 3990X59.5262.1279.8OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow Lite 2022-05-18CPU Power Consumption Monitor70140210280350

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 3960XThreadripper 3990X30M60M90M120M150MSE +/- 537114.01, N = 3SE +/- 1525861.00, N = 471399960136487533
OpenBenchmarking.orgNodes/second Per Watt, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthThreadripper 3960XThreadripper 3990X110K220K330K440K550K262315.84520308.04
MinAvgMaxThreadripper 3960X47.3272.2287.3Threadripper 3990X60.1262.3283.5OpenBenchmarking.orgWatts, Fewer Is BetterasmFish 2018-07-23CPU Power Consumption Monitor70140210280350

Timed Node.js Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 17.3Time To CompileThreadripper 3960XThreadripper 3990X4080120160200SE +/- 0.33, N = 3SE +/- 0.58, N = 3202.55148.33
MinAvgMaxThreadripper 3960X44.1247.9287.4Threadripper 3990X57.3233.9284.7OpenBenchmarking.orgWatts, Fewer Is BetterTimed Node.js Compilation 17.3CPU Power Consumption Monitor70140210280350

Primesieve

Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 8.0Length: 1e13Threadripper 3960XThreadripper 3990X20406080100SE +/- 0.26, N = 3SE +/- 0.61, N = 1587.6952.281. (CXX) g++ options: -O3
MinAvgMaxThreadripper 3960X44.6275.9287.2Threadripper 3990X57.3242.5279.7OpenBenchmarking.orgWatts, Fewer Is BetterPrimesieve 8.0CPU Power Consumption Monitor70140210280350

Ngspice

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

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670Threadripper 3960XThreadripper 3990X306090120150SE +/- 0.77, N = 3SE +/- 1.51, N = 4138.19143.481. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE
MinAvgMaxThreadripper 3960X41.882.586.9Threadripper 3990X54.6111.9118.4OpenBenchmarking.orgWatts, Fewer Is BetterNgspice 34CPU Power Consumption Monitor4080120160200

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X40K80K120K160K200KSE +/- 357.19, N = 3SE +/- 260.54, N = 31982881081301. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X47.3268.2287.2Threadripper 3990X62.3251.6280.1OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 96000 - Buffer Size: 512Threadripper 3960XThreadripper 3990X0.8431.6862.5293.3724.215SE +/- 0.020150, N = 3SE +/- 0.082618, N = 152.2218813.7464601. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio Per Watt, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 96000 - Buffer Size: 512Threadripper 3960XThreadripper 3990X0.00430.00860.01290.01720.02150.0150.019
MinAvgMaxThreadripper 3960X42.6147.5166.0Threadripper 3990X57.0201.2238.0OpenBenchmarking.orgWatts, Fewer Is BetterStargate Digital Audio Workstation 21.10.9CPU Power Consumption Monitor60120180240300

Helsing

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

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digitThreadripper 3960XThreadripper 3990X50100150200250SE +/- 0.47, N = 3SE +/- 1.36, N = 3207.60100.611. (CC) gcc options: -O2 -pthread
MinAvgMaxThreadripper 3960X41.6282.2288.3Threadripper 3990X54.7266.9279.7OpenBenchmarking.orgWatts, Fewer Is BetterHelsing 1.0-betaCPU Power Consumption Monitor70140210280350

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 0.14Test: CPU CacheThreadripper 3960XThreadripper 3990X4080120160200SE +/- 3.31, N = 15SE +/- 2.21, N = 15178.66150.171. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14Test: CPU CacheThreadripper 3960XThreadripper 3990X0.28370.56740.85111.13481.41851.2610.806
MinAvgMaxThreadripper 3960X42.9141.7180.5Threadripper 3990X55.8186.3252.1OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14CPU Power Consumption Monitor70140210280350

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.2Blend File: Pabellon Barcelona - Compute: CPU-OnlyThreadripper 3960XThreadripper 3990X4080120160200SE +/- 0.39, N = 3SE +/- 1.08, N = 3186.06110.33
MinAvgMaxThreadripper 3960X46.0279.7288.3Threadripper 3990X60.3256.1279.7OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.2CPU Power Consumption Monitor70140210280350

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X40K80K120K160K200KSE +/- 539.62, N = 3SE +/- 115.34, N = 3171513933401. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X45.9265.1287.2Threadripper 3990X58.8248.4279.7OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X5K10K15K20K25KSE +/- 5.51, N = 3SE +/- 16.97, N = 324189129581. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X46.9253.1287.2Threadripper 3990X59.6259.1281.4OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X13002600390052006500SE +/- 6.06, N = 3SE +/- 8.01, N = 3601831931. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X45.7261.1287.2Threadripper 3990X58.1255.3281.4OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedThreadripper 3960XThreadripper 3990X3K6K9K12K15KSE +/- 29.45, N = 3SE +/- 20.11, N = 1212417.411528.51. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s Per Watt, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedThreadripper 3960XThreadripper 3990X306090120150148.40104.79
MinAvgMaxThreadripper 3960X42.583.793.5Threadripper 3990X54.6110.0119.9OpenBenchmarking.orgWatts, Fewer Is BetterLZ4 Compression 1.9.3CPU Power Consumption Monitor4080120160200

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedThreadripper 3960XThreadripper 3990X1326395265SE +/- 0.39, N = 3SE +/- 0.38, N = 1256.6655.321. (CC) gcc options: -O3

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X20K40K60K80K100KSE +/- 157.21, N = 3SE +/- 129.42, N = 3102284513341. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X46.3253.7287.2Threadripper 3990X60.9259.1279.7OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X10K20K30K40K50KSE +/- 88.83, N = 3SE +/- 74.20, N = 348433257521. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X45.2263.0288.3Threadripper 3990X61.9252.4280.2OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X40K80K120K160K200KSE +/- 88.89, N = 3SE +/- 186.79, N = 3166603915861. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X45.6265.1287.2Threadripper 3990X58.5248.0280.8OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression SpeedThreadripper 3960XThreadripper 3990X8001600240032004000SE +/- 12.90, N = 15SE +/- 8.44, N = 33538.63371.21. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s Per Watt, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression SpeedThreadripper 3960XThreadripper 3990X71421283528.5221.39
MinAvgMaxThreadripper 3960X43.7124.1182.3Threadripper 3990X29.4157.6232.7OpenBenchmarking.orgWatts, Fewer Is BetterZstd Compression 1.5.0CPU Power Consumption Monitor60120180240300

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression SpeedThreadripper 3960XThreadripper 3990X20406080100SE +/- 0.48, N = 15SE +/- 0.79, N = 371.681.21. (CC) gcc options: -O3 -pthread -lz -llzma

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.4Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X510152025SE +/- 0.14, N = 9SE +/- 0.30, N = 1517.6918.741. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterAOM AV1 3.4Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.04970.09940.14910.19880.24850.2210.146
MinAvgMaxThreadripper 3960X43.479.9106.9Threadripper 3990X56.3128.7186.3OpenBenchmarking.orgWatts, Fewer Is BetterAOM AV1 3.4CPU Power Consumption Monitor50100150200250

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X11002200330044005500SE +/- 6.36, N = 3SE +/- 5.78, N = 3503026881. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X46.9259.4287.2Threadripper 3990X60.8253.6281.4OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X4K8K12K16K20KSE +/- 24.66, N = 3SE +/- 2.65, N = 320768110861. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X46.1248.3287.2Threadripper 3990X62.3259.8279.9OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X11002200330044005500SE +/- 4.04, N = 3SE +/- 2.91, N = 3516527361. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X46.6259.9287.2Threadripper 3990X60.0254.7281.4OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X4K8K12K16K20KSE +/- 35.18, N = 3SE +/- 37.25, N = 320163109021. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X45.6248.2287.2Threadripper 3990X60.7258.2280.0OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

Natron

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

OpenBenchmarking.orgFPS, More Is BetterNatron 2.4Input: SpaceshipThreadripper 3960XThreadripper 3990X1.14752.2953.44254.595.7375SE +/- 0.09, N = 12SE +/- 0.03, N = 155.13.0
OpenBenchmarking.orgFPS Per Watt, More Is BetterNatron 2.4Input: SpaceshipThreadripper 3960XThreadripper 3990X0.00610.01220.01830.02440.03050.0270.013
MinAvgMaxThreadripper 3960X43.3186.9285.6Threadripper 3990X37.0239.3282.0OpenBenchmarking.orgWatts, Fewer Is BetterNatron 2.4CPU Power Consumption Monitor70140210280350

GPAW

GPAW is a density-functional theory (DFT) Python code based on the projector-augmented wave (PAW) method and the atomic simulation environment (ASE). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGPAW 22.1Input: Carbon NanotubeThreadripper 3960XThreadripper 3990X306090120150SE +/- 0.10, N = 3SE +/- 0.21, N = 3146.20107.871. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi
MinAvgMaxThreadripper 3960X45.6228.0280.6Threadripper 3990X58.7265.9280.6OpenBenchmarking.orgWatts, Fewer Is BetterGPAW 22.1CPU Power Consumption Monitor70140210280350

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 4.0Test: WritesThreadripper 3960XThreadripper 3990X50K100K150K200K250KSE +/- 2367.03, N = 3SE +/- 1231.95, N = 3213508240022
OpenBenchmarking.orgOp/s Per Watt, More Is BetterApache Cassandra 4.0Test: WritesThreadripper 3960XThreadripper 3990X20040060080010001008.661000.17
MinAvgMaxThreadripper 3960X46.2211.7244.6Threadripper 3990X60.2240.0292.0OpenBenchmarking.orgWatts, Fewer Is BetterApache Cassandra 4.0CPU Power Consumption Monitor70140210280350

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X20K40K60K80K100KSE +/- 210.99, N = 3SE +/- 87.48, N = 388823439861. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X45.7248.0287.2Threadripper 3990X58.4257.0279.7OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

ONNX Runtime

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

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: StandardThreadripper 3960XThreadripper 3990X2K4K6K8K10KSE +/- 26.54, N = 3SE +/- 31.14, N = 3835069801. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: StandardThreadripper 3960XThreadripper 3990X122436486052.2536.51
MinAvgMaxThreadripper 3960X44.3159.8164.8Threadripper 3990X58.8191.2197.5OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption Monitor50100150200250

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X9K18K27K36K45KSE +/- 0.88, N = 3SE +/- 30.07, N = 341532219751. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X46.2260.0287.2Threadripper 3990X61.4248.8281.4OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X20K40K60K80K100KSE +/- 39.93, N = 3SE +/- 98.53, N = 386291431831. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X46.3247.8287.2Threadripper 3990X58.7255.3279.9OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.2Blend File: Classroom - Compute: CPU-OnlyThreadripper 3960XThreadripper 3990X306090120150SE +/- 0.14, N = 3SE +/- 0.63, N = 3153.5284.35
MinAvgMaxThreadripper 3960X45.1279.3288.3Threadripper 3990X56.8267.9279.7OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.2CPU Power Consumption Monitor70140210280350

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X9K18K27K36K45KSE +/- 5.24, N = 3SE +/- 51.50, N = 340461216191. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X47.2259.2287.2Threadripper 3990X60.3248.1279.7OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity 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.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeThreadripper 3960XThreadripper 3990X246810SE +/- 0.00186, N = 3SE +/- 0.02450, N = 33.615206.52205
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.10Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeThreadripper 3960XThreadripper 3990X0.00560.01120.01680.02240.0280.0130.025
MinAvgMaxThreadripper 3960X45.8269.7288.3Threadripper 3990X62.6256.9281.8OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.10CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: gravity_spheres_volume/dim_512/ao/real_timeThreadripper 3960XThreadripper 3990X246810SE +/- 0.00671, N = 3SE +/- 0.04916, N = 33.806876.97078
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.10Benchmark: gravity_spheres_volume/dim_512/ao/real_timeThreadripper 3960XThreadripper 3990X0.00610.01220.01830.02440.03050.0150.027
MinAvgMaxThreadripper 3960X44.5261.7281.4Threadripper 3990X58.6254.9279.9OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.10CPU Power Consumption Monitor70140210280350

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X30060090012001500SE +/- 0.88, N = 3SE +/- 1.00, N = 315158041. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X46.4254.8287.2Threadripper 3990X60.0251.5279.7OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X30060090012001500SE +/- 1.53, N = 3SE +/- 0.58, N = 312666731. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X39.5254.3287.2Threadripper 3990X61.1250.7280.3OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path TracerThreadripper 3960XThreadripper 3990X30060090012001500SE +/- 1.53, N = 3SE +/- 2.33, N = 313006891. (CXX) g++ options: -O3 -ldl
MinAvgMaxThreadripper 3960X45.8254.8288.3Threadripper 3990X61.5250.5279.8OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor70140210280350

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity 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.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeThreadripper 3960XThreadripper 3990X3691215SE +/- 0.00466, N = 3SE +/- 0.03914, N = 36.0614710.82200
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.10Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeThreadripper 3960XThreadripper 3990X0.00950.0190.02850.0380.04750.0230.042
MinAvgMaxThreadripper 3960X45.7267.5287.3Threadripper 3990X58.1255.9282.0OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.10CPU Power Consumption Monitor70140210280350

WebP2 Image Encode

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

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 75, Compression Effort 7Threadripper 3960XThreadripper 3990X20406080100SE +/- 0.27, N = 3SE +/- 0.83, N = 3103.82108.801. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
MinAvgMaxThreadripper 3960X42.6224.2249.4Threadripper 3990X56.6238.6276.7OpenBenchmarking.orgWatts, Fewer Is BetterWebP2 Image Encode 20220422CPU Power Consumption Monitor70140210280350

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 3960XThreadripper 3990X4080120160200165.39138.14
MinAvgMaxThreadripper 3960X44.5250.4288.3Threadripper 3990X57.2252.8281.2OpenBenchmarking.orgWatts, Fewer Is BetterAppleseed 2.0 BetaCPU Power Consumption Monitor70140210280350

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: PNG - Encode Speed: 7Threadripper 3960XThreadripper 3990X3691215SE +/- 0.01, N = 3SE +/- 0.03, N = 310.9110.891. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie
OpenBenchmarking.orgMP/s Per Watt, More Is BetterJPEG XL libjxl 0.6.1Input: PNG - Encode Speed: 7Threadripper 3960XThreadripper 3990X0.02340.04680.07020.09360.1170.1040.084
MinAvgMaxThreadripper 3960X43.1104.7138.5Threadripper 3990X56.5129.3161.4OpenBenchmarking.orgWatts, Fewer Is BetterJPEG XL libjxl 0.6.1CPU Power Consumption Monitor50100150200250

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 3960XThreadripper 3990X48121620SE +/- 0.059343, N = 15SE +/- 0.113524, N = 158.72831817.3545711. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateThreadripper 3960XThreadripper 3990X0.01870.03740.05610.07480.09350.0450.083
MinAvgMaxThreadripper 3960X44.5194.2227.6Threadripper 3990X52.2209.8269.6OpenBenchmarking.orgWatts, Fewer Is BetterACES DGEMM 1.0CPU Power Consumption Monitor70140210280350

OpenSCAD

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: PistolThreadripper 3960XThreadripper 3990X20406080100SE +/- 0.41, N = 3SE +/- 0.34, N = 394.7198.231. OpenSCAD version 2021.01
MinAvgMaxThreadripper 3960X41.882.284.6Threadripper 3990X50.9109.9113.1OpenBenchmarking.orgWatts, Fewer Is BetterOpenSCADCPU Power Consumption Monitor4080120160200

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 TesterThreadripper 3960XThreadripper 3990X306090120150135.28153.51
MinAvgMaxThreadripper 3960X44.3247.0265.4Threadripper 3990X58.3262.8279.5OpenBenchmarking.orgWatts, Fewer Is BetterAppleseed 2.0 BetaCPU Power Consumption Monitor70140210280350

PyPerformance

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

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: python_startupThreadripper 3960XThreadripper 3990X48121620SE +/- 0.03, N = 3SE +/- 0.03, N = 313.914.0
MinAvgMaxThreadripper 3960X41.568.373.1Threadripper 3990X54.095.6103.0OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor20406080100

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 480000 - Buffer Size: 512Threadripper 3960XThreadripper 3990X1.20942.41883.62824.83766.047SE +/- 0.097092, N = 12SE +/- 0.003417, N = 34.8526705.3752341. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio Per Watt, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 480000 - Buffer Size: 512Threadripper 3960XThreadripper 3990X0.00610.01220.01830.02440.03050.0270.026
MinAvgMaxThreadripper 3960X43.2179.8215.8Threadripper 3990X57.4207.3242.4OpenBenchmarking.orgWatts, Fewer Is BetterStargate Digital Audio Workstation 21.10.9CPU Power Consumption Monitor60120180240300

PyPerformance

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

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: raytraceThreadripper 3960XThreadripper 3990X100200300400500SE +/- 1.00, N = 3SE +/- 1.20, N = 3474440
MinAvgMaxThreadripper 3960X41.982.485.9Threadripper 3990X56.1113.2117.5OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor4080120160200

Node.js V8 Web Tooling Benchmark

Running the V8 project's Web-Tooling-Benchmark under Node.js. The Web-Tooling-Benchmark stresses JavaScript-related workloads common to web developers like Babel and TypeScript and Babylon. This test profile can test the system's JavaScript performance with Node.js. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling BenchmarkThreadripper 3960XThreadripper 3990X3691215SE +/- 0.09, N = 3SE +/- 0.02, N = 310.189.54
OpenBenchmarking.orgruns/s Per Watt, More Is BetterNode.js V8 Web Tooling BenchmarkThreadripper 3960XThreadripper 3990X0.02590.05180.07770.10360.12950.1150.079
MinAvgMaxThreadripper 3960X42.988.495.8Threadripper 3990X55.9120.4128.6OpenBenchmarking.orgWatts, Fewer Is BetterNode.js V8 Web Tooling BenchmarkCPU Power Consumption Monitor4080120160200

libgav1

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

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.17Video Input: Summer Nature 4KThreadripper 3960XThreadripper 3990X1020304050SE +/- 0.06, N = 3SE +/- 0.06, N = 344.0741.701. (CXX) g++ options: -O3 -lrt
OpenBenchmarking.orgFPS Per Watt, More Is Betterlibgav1 0.17Video Input: Summer Nature 4KThreadripper 3960XThreadripper 3990X0.10150.2030.30450.4060.50750.4510.285
MinAvgMaxThreadripper 3960X42.997.8114.9Threadripper 3990X53.7146.4192.1OpenBenchmarking.orgWatts, Fewer Is Betterlibgav1 0.17CPU Power Consumption Monitor50100150200250

Chia Blockchain VDF

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

OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.7Test: Square Assembly OptimizedThreadripper 3960XThreadripper 3990X20K40K60K80K100KSE +/- 523.87, N = 3SE +/- 529.15, N = 31118671070001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
OpenBenchmarking.orgIPS Per Watt, More Is BetterChia Blockchain VDF 1.0.7Test: Square Assembly OptimizedThreadripper 3960XThreadripper 3990X300600900120015001324.46961.94
MinAvgMaxThreadripper 3960X42.084.587.2Threadripper 3990X54.2111.2116.0OpenBenchmarking.orgWatts, Fewer Is BetterChia Blockchain VDF 1.0.7CPU Power Consumption Monitor4080120160200

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.CThreadripper 3960XThreadripper 3990X5K10K15K20K25KSE +/- 87.77, N = 3SE +/- 3.03, N = 316854.3223124.731. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CThreadripper 3960XThreadripper 3990X20406080100103.1788.85
MinAvgMaxThreadripper 3960X40.8163.4171.5Threadripper 3990X57.8260.3281.9OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor70140210280350

Chaos Group V-RAY

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

OpenBenchmarking.orgvsamples, More Is BetterChaos Group V-RAY 5.02Mode: CPUThreadripper 3960XThreadripper 3990X10K20K30K40K50KSE +/- 302.89, N = 3SE +/- 412.35, N = 32494146971
OpenBenchmarking.orgvsamples Per Watt, More Is BetterChaos Group V-RAY 5.02Mode: CPUThreadripper 3960XThreadripper 3990X4080120160200103.72192.63
MinAvgMaxThreadripper 3960X46.6240.5287.2Threadripper 3990X56.3243.5281.5OpenBenchmarking.orgWatts, Fewer Is BetterChaos Group V-RAY 5.02CPU Power Consumption Monitor70140210280350

simdjson

This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: PartialTweetsThreadripper 3960XThreadripper 3990X0.87531.75062.62593.50124.3765SE +/- 0.02, N = 3SE +/- 0.00, N = 33.893.781. (CXX) g++ options: -O3
OpenBenchmarking.orgGB/s Per Watt, More Is Bettersimdjson 2.0Throughput Test: PartialTweetsThreadripper 3960XThreadripper 3990X0.01060.02120.03180.04240.0530.0470.035
MinAvgMaxThreadripper 3960X42.282.985.6Threadripper 3990X54.4109.4112.6OpenBenchmarking.orgWatts, Fewer Is Bettersimdjson 2.0CPU Power Consumption Monitor4080120160200

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: DistinctUserIDThreadripper 3960XThreadripper 3990X0.97431.94862.92293.89724.8715SE +/- 0.01, N = 3SE +/- 0.01, N = 34.334.201. (CXX) g++ options: -O3
OpenBenchmarking.orgGB/s Per Watt, More Is Bettersimdjson 2.0Throughput Test: DistinctUserIDThreadripper 3960XThreadripper 3990X0.01170.02340.03510.04680.05850.0520.039
MinAvgMaxThreadripper 3960X41.682.985.6Threadripper 3990X53.1109.1112.4OpenBenchmarking.orgWatts, Fewer Is Bettersimdjson 2.0CPU Power Consumption Monitor4080120160200

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: TopTweetThreadripper 3960XThreadripper 3990X0.99231.98462.97693.96924.9615SE +/- 0.02, N = 3SE +/- 0.01, N = 34.414.291. (CXX) g++ options: -O3
OpenBenchmarking.orgGB/s Per Watt, More Is Bettersimdjson 2.0Throughput Test: TopTweetThreadripper 3960XThreadripper 3990X0.01190.02380.03570.04760.05950.0530.039
MinAvgMaxThreadripper 3960X42.383.285.6Threadripper 3990X53.7109.5113.0OpenBenchmarking.orgWatts, Fewer Is Bettersimdjson 2.0CPU Power Consumption Monitor4080120160200

GROMACS

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

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2022.1Implementation: MPI CPU - Input: water_GMX50_bareThreadripper 3960XThreadripper 3990X0.85231.70462.55693.40924.2615SE +/- 0.003, N = 3SE +/- 0.009, N = 32.4233.7881. (CXX) g++ options: -O3
OpenBenchmarking.orgNs Per Day Per Watt, More Is BetterGROMACS 2022.1Implementation: MPI CPU - Input: water_GMX50_bareThreadripper 3960XThreadripper 3990X0.00340.00680.01020.01360.0170.0100.015
MinAvgMaxThreadripper 3960X45.9231.9278.9Threadripper 3990X33.3253.7281.2OpenBenchmarking.orgWatts, Fewer Is BetterGROMACS 2022.1CPU Power Consumption Monitor70140210280350

PJSIP

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

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: INVITEThreadripper 3960XThreadripper 3990X9001800270036004500SE +/- 6.49, N = 3SE +/- 13.69, N = 3398237111. (CC) gcc options: -lstdc++ -lopus -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
OpenBenchmarking.orgResponses Per Second Per Watt, More Is BetterPJSIP 2.11Method: INVITEThreadripper 3960XThreadripper 3990X102030405044.2031.78
MinAvgMaxThreadripper 3960X43.390.1129.6Threadripper 3990X53.8116.8163.3OpenBenchmarking.orgWatts, Fewer Is BetterPJSIP 2.11CPU Power Consumption Monitor50100150200250

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatefulThreadripper 3960XThreadripper 3990X15003000450060007500SE +/- 7.22, N = 3SE +/- 9.67, N = 3685666461. (CC) gcc options: -lstdc++ -lopus -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
OpenBenchmarking.orgResponses Per Second Per Watt, More Is BetterPJSIP 2.11Method: OPTIONS, StatefulThreadripper 3960XThreadripper 3990X2040608010097.1967.61
MinAvgMaxThreadripper 3960X41.570.5112.9Threadripper 3990X52.698.3139.5OpenBenchmarking.orgWatts, Fewer Is BetterPJSIP 2.11CPU Power Consumption Monitor4080120160200

LuxCoreRender

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

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: CPUThreadripper 3960XThreadripper 3990X3691215SE +/- 0.01, N = 3SE +/- 0.04, N = 36.8912.00MIN: 6.28 / MAX: 7.12MIN: 10.24 / MAX: 13.72
OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: CPUThreadripper 3960XThreadripper 3990X0.01040.02080.03120.04160.0520.0260.046
MinAvgMaxThreadripper 3960X47.5265.5288.3Threadripper 3990X59.1262.4281.6OpenBenchmarking.orgWatts, Fewer Is BetterLuxCoreRender 2.6CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: CPUThreadripper 3960XThreadripper 3990X246810SE +/- 0.01, N = 3SE +/- 0.03, N = 33.716.59MIN: 1.36 / MAX: 4.33MIN: 2.54 / MAX: 7.72
OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: CPUThreadripper 3960XThreadripper 3990X0.00560.01120.01680.02240.0280.0140.025
MinAvgMaxThreadripper 3960X48.0263.6287.2Threadripper 3990X58.7260.9281.8OpenBenchmarking.orgWatts, Fewer Is BetterLuxCoreRender 2.6CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: CPUThreadripper 3960XThreadripper 3990X1.34332.68664.02995.37326.7165SE +/- 0.03, N = 3SE +/- 0.06, N = 33.355.97MIN: 1.26 / MAX: 3.89MIN: 2.44 / MAX: 7.13
OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: CPUThreadripper 3960XThreadripper 3990X0.00520.01040.01560.02080.0260.0130.023
MinAvgMaxThreadripper 3960X45.7263.6287.3Threadripper 3990X57.8260.3281.6OpenBenchmarking.orgWatts, Fewer Is BetterLuxCoreRender 2.6CPU Power Consumption Monitor70140210280350

IndigoBench

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

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: BedroomThreadripper 3960XThreadripper 3990X3691215SE +/- 0.010, N = 3SE +/- 0.043, N = 35.3889.831
OpenBenchmarking.orgM samples/s Per Watt, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: BedroomThreadripper 3960XThreadripper 3990X0.00830.01660.02490.03320.04150.0200.037
MinAvgMaxThreadripper 3960X46.6266.4287.2Threadripper 3990X60.6262.7279.9OpenBenchmarking.orgWatts, Fewer Is BetterIndigoBench 4.4CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: SupercarThreadripper 3960XThreadripper 3990X510152025SE +/- 0.01, N = 3SE +/- 0.18, N = 311.7821.08
OpenBenchmarking.orgM samples/s Per Watt, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: SupercarThreadripper 3960XThreadripper 3990X0.01820.03640.05460.07280.0910.0440.081
MinAvgMaxThreadripper 3960X47.7267.1287.4Threadripper 3990X63.2261.2280.2OpenBenchmarking.orgWatts, Fewer Is BetterIndigoBench 4.4CPU Power Consumption Monitor70140210280350

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 0.13Time To CompileThreadripper 3960XThreadripper 3990X1428425670SE +/- 0.15, N = 3SE +/- 0.08, N = 363.5359.30
MinAvgMaxThreadripper 3960X44.4192.4287.4Threadripper 3990X54.2184.1280.4OpenBenchmarking.orgWatts, Fewer Is BetterBuild2 0.13CPU Power Consumption Monitor70140210280350

LuxCoreRender

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

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: CPUThreadripper 3960XThreadripper 3990X246810SE +/- 0.02, N = 3SE +/- 0.06, N = 34.267.80MIN: 4.13 / MAX: 4.55MIN: 7.54 / MAX: 8.54
OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: CPUThreadripper 3960XThreadripper 3990X0.00680.01360.02040.02720.0340.0160.030
MinAvgMaxThreadripper 3960X42.6268.1287.8Threadripper 3990X54.6264.3281.6OpenBenchmarking.orgWatts, Fewer Is BetterLuxCoreRender 2.6CPU Power Consumption Monitor70140210280350

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception ResNet V2Threadripper 3960XThreadripper 3990X20K40K60K80K100KSE +/- 120.04, N = 3SE +/- 1222.16, N = 339153.085767.4
MinAvgMaxThreadripper 3960X45.2210.6222.3Threadripper 3990X63.2261.6280.6OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow Lite 2022-05-18CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: NASNet MobileThreadripper 3960XThreadripper 3990X11K22K33K44K55KSE +/- 145.17, N = 3SE +/- 311.74, N = 321105.751346.9
MinAvgMaxThreadripper 3960X44.4154.0163.3Threadripper 3990X62.0261.2281.4OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow Lite 2022-05-18CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet FloatThreadripper 3960XThreadripper 3990X9001800270036004500SE +/- 6.65, N = 3SE +/- 19.00, N = 31890.724229.25
MinAvgMaxThreadripper 3960X44.3216.7229.5Threadripper 3990X61.9240.0267.0OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow Lite 2022-05-18CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: SqueezeNetThreadripper 3960XThreadripper 3990X13002600390052006500SE +/- 2.61, N = 3SE +/- 36.36, N = 32541.586181.90
MinAvgMaxThreadripper 3960X45.5215.8229.0Threadripper 3990X58.8258.0279.7OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow Lite 2022-05-18CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet QuantThreadripper 3960XThreadripper 3990X7001400210028003500SE +/- 26.42, N = 3SE +/- 35.20, N = 32697.413296.10
MinAvgMaxThreadripper 3960X46.0251.0267.0Threadripper 3990X58.7255.9279.7OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow Lite 2022-05-18CPU Power Consumption Monitor70140210280350

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096Threadripper 3960XThreadripper 3990X200K400K600K800K1000KSE +/- 382.82, N = 3SE +/- 4577.03, N = 3460980.2854108.51. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
OpenBenchmarking.orgverify/s Per Watt, More Is BetterOpenSSL 3.0Algorithm: RSA4096Threadripper 3960XThreadripper 3990X80016002400320040001932.673513.65
MinAvgMaxThreadripper 3960X46.6238.5255.6Threadripper 3990X61.8243.1279.7OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.0CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096Threadripper 3960XThreadripper 3990X3K6K9K12K15KSE +/- 13.05, N = 3SE +/- 189.66, N = 37110.213533.31. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.18Build: defconfigThreadripper 3960XThreadripper 3990X918273645SE +/- 0.43, N = 4SE +/- 0.27, N = 739.0028.88
MinAvgMaxThreadripper 3960X44.7196.8287.2Threadripper 3990X55.1186.8280.5OpenBenchmarking.orgWatts, Fewer Is BetterTimed Linux Kernel Compilation 5.18CPU Power Consumption Monitor70140210280350

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. The test profile uses a 8K x 8K game texture as a sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 1.0Benchmark: Single-Threaded - Configuration: ETC2Threadripper 3960XThreadripper 3990X50100150200250SE +/- 0.12, N = 3SE +/- 0.02, N = 3234.97225.851. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s Per Watt, More Is BetterEtcpak 1.0Benchmark: Single-Threaded - Configuration: ETC2Threadripper 3960XThreadripper 3990X0.651.31.952.63.252.8892.067
MinAvgMaxThreadripper 3960X40.381.384.4Threadripper 3990X54.2109.3113.2OpenBenchmarking.orgWatts, Fewer Is BetterEtcpak 1.0CPU Power Consumption Monitor4080120160200

VP9 libvpx Encoding

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

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X3691215SE +/- 0.07, N = 3SE +/- 0.03, N = 310.8310.141. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.02340.04680.07020.09360.1170.1040.079
MinAvgMaxThreadripper 3960X43.7104.3113.6Threadripper 3990X55.7129.1141.4OpenBenchmarking.orgWatts, Fewer Is BetterVP9 libvpx Encoding 1.10.0CPU Power Consumption Monitor4080120160200

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.2Blend File: Fishy Cat - Compute: CPU-OnlyThreadripper 3960XThreadripper 3990X1632486480SE +/- 0.04, N = 3SE +/- 0.17, N = 372.2540.06
MinAvgMaxThreadripper 3960X46.0269.2288.3Threadripper 3990X58.3253.3279.7OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.2CPU Power Consumption Monitor70140210280350

simdjson

This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: KostyaThreadripper 3960XThreadripper 3990X0.6391.2781.9172.5563.195SE +/- 0.03, N = 3SE +/- 0.01, N = 32.842.681. (CXX) g++ options: -O3
OpenBenchmarking.orgGB/s Per Watt, More Is Bettersimdjson 2.0Throughput Test: KostyaThreadripper 3960XThreadripper 3990X0.00740.01480.02220.02960.0370.0330.024
MinAvgMaxThreadripper 3960X42.085.889.4Threadripper 3990X54.9113.4118.0OpenBenchmarking.orgWatts, Fewer Is Bettersimdjson 2.0CPU Power Consumption Monitor4080120160200

Algebraic Multi-Grid Benchmark

AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2Threadripper 3960XThreadripper 3990X100M200M300M400M500MSE +/- 273233.64, N = 3SE +/- 201832.70, N = 34079460004556222001. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi
OpenBenchmarking.orgFigure Of Merit Per Watt, More Is BetterAlgebraic Multi-Grid Benchmark 1.2Threadripper 3960XThreadripper 3990X500K1000K1500K2000K2500K2373215.061716094.62
MinAvgMaxThreadripper 3960X44.2171.9216.0Threadripper 3990X60.4265.5280.7OpenBenchmarking.orgWatts, Fewer Is BetterAlgebraic Multi-Grid Benchmark 1.2CPU Power Consumption Monitor70140210280350

Chia Blockchain VDF

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

OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.7Test: Square Plain C++Threadripper 3960XThreadripper 3990X40K80K120K160K200KSE +/- 66.67, N = 3SE +/- 405.52, N = 31690331633671. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
OpenBenchmarking.orgIPS Per Watt, More Is BetterChia Blockchain VDF 1.0.7Test: Square Plain C++Threadripper 3960XThreadripper 3990X4008001200160020002095.361532.21
MinAvgMaxThreadripper 3960X41.980.783.7Threadripper 3990X52.4106.6110.7OpenBenchmarking.orgWatts, Fewer Is BetterChia Blockchain VDF 1.0.7CPU Power Consumption Monitor20406080100

Xcompact3d Incompact3d

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

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionThreadripper 3960XThreadripper 3990X1326395265SE +/- 0.24, N = 3SE +/- 0.23, N = 359.7545.391. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
MinAvgMaxThreadripper 3960X46.3191.2209.2Threadripper 3990X59.0256.3280.4OpenBenchmarking.orgWatts, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Power Consumption Monitor70140210280350

GEGL

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

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Rotate 90 DegreesThreadripper 3960XThreadripper 3990X1326395265SE +/- 0.23, N = 3SE +/- 0.46, N = 348.6755.78
MinAvgMaxThreadripper 3960X41.798.5141.5Threadripper 3990X53.8149.2228.3OpenBenchmarking.orgWatts, Fewer Is BetterGEGLCPU Power Consumption Monitor60120180240300

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Color EnhanceThreadripper 3960XThreadripper 3990X1224364860SE +/- 0.09, N = 3SE +/- 0.03, N = 350.4352.90
MinAvgMaxThreadripper 3960X42.184.6101.3Threadripper 3990X54.4116.9163.7OpenBenchmarking.orgWatts, Fewer Is BetterGEGLCPU Power Consumption Monitor50100150200250

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 3960XThreadripper 3990X2040608010092.6662.59
MinAvgMaxThreadripper 3960X46.5267.9282.4Threadripper 3990X58.4262.5276.3OpenBenchmarking.orgWatts, Fewer Is BetterAppleseed 2.0 BetaCPU Power Consumption Monitor70140210280350

PJSIP

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

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatelessThreadripper 3960XThreadripper 3990X9K18K27K36K45KSE +/- 103.50, N = 3SE +/- 54.01, N = 340408408721. (CC) gcc options: -lstdc++ -lopus -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
OpenBenchmarking.orgResponses Per Second Per Watt, More Is BetterPJSIP 2.11Method: OPTIONS, StatelessThreadripper 3960XThreadripper 3990X70140210280350342.99268.54
MinAvgMaxThreadripper 3960X42.6117.5128.2Threadripper 3990X50.8152.2167.3OpenBenchmarking.orgWatts, Fewer Is BetterPJSIP 2.11CPU Power Consumption Monitor50100150200250

RawTherapee

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

OpenBenchmarking.orgSeconds, Fewer Is BetterRawTherapeeTotal Benchmark TimeThreadripper 3960XThreadripper 3990X1122334455SE +/- 0.09, N = 3SE +/- 0.14, N = 346.9650.831. RawTherapee, version 5.8, command line.
MinAvgMaxThreadripper 3960X41.3134.0181.9Threadripper 3990X54.2192.7237.1OpenBenchmarking.orgWatts, Fewer Is BetterRawTherapeeCPU Power Consumption Monitor60120180240300

Timed Godot Game Engine Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3Time To CompileThreadripper 3960XThreadripper 3990X1224364860SE +/- 0.13, N = 3SE +/- 0.15, N = 353.7545.10
MinAvgMaxThreadripper 3960X43.4201.5287.5Threadripper 3990X55.9200.1281.2OpenBenchmarking.orgWatts, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3CPU Power Consumption Monitor70140210280350

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 3960XThreadripper 3990X15K30K45K60K75KSE +/- 137.19, N = 3SE +/- 10.73, N = 355451.8568363.631. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CThreadripper 3960XThreadripper 3990X70140210280350300.84272.97
MinAvgMaxThreadripper 3960X44.0184.3199.8Threadripper 3990X60.0250.4280.2OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor70140210280350

Aircrack-ng

Aircrack-ng is a tool for assessing WiFi/WLAN network security. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgk/s, More Is BetterAircrack-ng 1.7Threadripper 3960XThreadripper 3990X30K60K90K120K150KSE +/- 475.48, N = 3SE +/- 1637.62, N = 690173.54162110.441. (CXX) g++ options: -std=gnu++17 -O3 -fvisibility=hidden -fcommon -rdynamic -lnl-3 -lnl-genl-3 -lpcre -lpthread -lz -lssl -lcrypto -lhwloc -ldl -lm -pthread
OpenBenchmarking.orgk/s Per Watt, More Is BetterAircrack-ng 1.7Threadripper 3960XThreadripper 3990X140280420560700350.22646.77
MinAvgMaxThreadripper 3960X44.4257.5287.2Threadripper 3990X56.1250.6279.7OpenBenchmarking.orgWatts, Fewer Is BetterAircrack-ng 1.7CPU Power Consumption Monitor70140210280350

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 2.14ATPase Simulation - 327,506 AtomsThreadripper 3960XThreadripper 3990X0.17140.34280.51420.68560.857SE +/- 0.00226, N = 3SE +/- 0.00110, N = 30.761980.45309
MinAvgMaxThreadripper 3960X43.9258.1287.2Threadripper 3990X58.0248.3280.2OpenBenchmarking.orgWatts, Fewer Is BetterNAMD 2.14CPU Power Consumption Monitor70140210280350

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X20406080100SE +/- 2.23, N = 15SE +/- 1.07, N = 1595.23102.981. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.14940.29880.44820.59760.7470.6640.617
MinAvgMaxThreadripper 3960X44.4143.4236.1Threadripper 3990X54.1167.0277.8OpenBenchmarking.orgWatts, Fewer Is BetterSVT-VP9 0.3CPU Power Consumption Monitor70140210280350

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.2Blend File: BMW27 - Compute: CPU-OnlyThreadripper 3960XThreadripper 3990X1326395265SE +/- 0.28, N = 3SE +/- 0.34, N = 355.7130.17
MinAvgMaxThreadripper 3960X40.0266.2288.3Threadripper 3990X53.9247.1281.3OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.2CPU Power Consumption Monitor70140210280350

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.00Test: Decompression RatingThreadripper 3960XThreadripper 3990X80K160K240K320K400KSE +/- 496.44, N = 3SE +/- 2744.18, N = 31799533536951. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
OpenBenchmarking.orgMIPS Per Watt, More Is Better7-Zip Compression 22.00Test: Decompression RatingThreadripper 3960XThreadripper 3990X30060090012001500854.601391.03
MinAvgMaxThreadripper 3960X45.4210.6258.7Threadripper 3990X57.2254.3283.9OpenBenchmarking.orgWatts, Fewer Is Better7-Zip Compression 22.00CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.00Test: Compression RatingThreadripper 3960XThreadripper 3990X40K80K120K160K200KSE +/- 599.92, N = 3SE +/- 294.26, N = 31695221927241. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Decompression SpeedThreadripper 3960XThreadripper 3990X8001600240032004000SE +/- 8.76, N = 3SE +/- 3.38, N = 33570.93376.81. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s Per Watt, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Decompression SpeedThreadripper 3960XThreadripper 3990X81624324032.6925.04
MinAvgMaxThreadripper 3960X43.9109.3148.2Threadripper 3990X56.6134.9175.0OpenBenchmarking.orgWatts, Fewer Is BetterZstd Compression 1.5.0CPU Power Consumption Monitor50100150200250

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression SpeedThreadripper 3960XThreadripper 3990X1020304050SE +/- 0.15, N = 3SE +/- 0.03, N = 343.344.51. (CC) gcc options: -O3 -pthread -lz -llzma

OpenSCAD

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Mini-ITX CaseThreadripper 3960XThreadripper 3990X1020304050SE +/- 0.13, N = 3SE +/- 0.19, N = 341.9443.691. OpenSCAD version 2021.01
MinAvgMaxThreadripper 3960X41.280.584.5Threadripper 3990X53.9107.1112.6OpenBenchmarking.orgWatts, Fewer Is BetterOpenSCADCPU Power Consumption Monitor4080120160200

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 96000 - Buffer Size: 1024Threadripper 3960XThreadripper 3990X0.96081.92162.88243.84324.804SE +/- 0.014569, N = 3SE +/- 0.049478, N = 34.2207314.2700851. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio Per Watt, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 96000 - Buffer Size: 1024Threadripper 3960XThreadripper 3990X0.00520.01040.01560.02080.0260.0230.021
MinAvgMaxThreadripper 3960X43.8182.9217.0Threadripper 3990X55.0207.5242.9OpenBenchmarking.orgWatts, Fewer Is BetterStargate Digital Audio Workstation 21.10.9CPU Power Consumption Monitor60120180240300

Tachyon

This is a test of the threaded Tachyon, a parallel ray-tracing system, measuring the time to ray-trace a sample scene. The sample scene used is the Teapot scene ray-traced to 8K x 8K with 32 samples. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.99.2Total TimeThreadripper 3960XThreadripper 3990X1224364860SE +/- 0.25, N = 3SE +/- 0.31, N = 354.7228.491. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread
MinAvgMaxThreadripper 3960X43.1269.4287.2Threadripper 3990X53.2251.8279.7OpenBenchmarking.orgWatts, Fewer Is BetterTachyon 0.99.2CPU Power Consumption Monitor70140210280350

JPEG XL Decoding libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.6.1CPU Threads: 1Threadripper 3960XThreadripper 3990X1632486480SE +/- 0.02, N = 3SE +/- 0.07, N = 372.6269.37
OpenBenchmarking.orgMP/s Per Watt, More Is BetterJPEG XL Decoding libjxl 0.6.1CPU Threads: 1Threadripper 3960XThreadripper 3990X0.19760.39520.59280.79040.9880.8780.637
MinAvgMaxThreadripper 3960X42.482.787.6Threadripper 3990X53.8109.0115.0OpenBenchmarking.orgWatts, Fewer Is BetterJPEG XL Decoding libjxl 0.6.1CPU Power Consumption Monitor4080120160200

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 9Input: motorBike - Execution TimeThreadripper 3960XThreadripper 3990X163248648070.2772.861. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
MinAvgMaxThreadripper 3960X45.1131.0139.8Threadripper 3990X60.3160.1169.9OpenBenchmarking.orgWatts, Fewer Is BetterOpenFOAM 9CPU Power Consumption Monitor50100150200250

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 9Input: motorBike - Mesh TimeThreadripper 3960XThreadripper 3990X91827364538.0239.871. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

srsRAN

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

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAMThreadripper 3960XThreadripper 3990X4080120160200SE +/- 0.35, N = 3SE +/- 0.46, N = 3160.6157.01. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f
OpenBenchmarking.orgUE Mb/s Per Watt, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAMThreadripper 3960XThreadripper 3990X0.44420.88841.33261.77682.2211.9741.469
MinAvgMaxThreadripper 3960X43.181.385.6Threadripper 3990X52.8106.8112.5OpenBenchmarking.orgWatts, Fewer Is BettersrsRAN 21.10CPU Power Consumption Monitor4080120160200

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAMThreadripper 3960XThreadripper 3990X90180270360450SE +/- 1.08, N = 3SE +/- 1.77, N = 3403.1392.71. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f

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 3960XThreadripper 3990X14K28K42K56K70KSE +/- 39.75, N = 3SE +/- 37.32, N = 350029.4165067.751. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CThreadripper 3960XThreadripper 3990X70140210280350301.30257.65
MinAvgMaxThreadripper 3960X42.3166.0181.7Threadripper 3990X58.1252.5280.3OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor70140210280350

QuantLib

QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21Threadripper 3960XThreadripper 3990X6001200180024003000SE +/- 28.82, N = 3SE +/- 30.41, N = 42914.72790.41. (CXX) g++ options: -O3 -march=native -rdynamic
OpenBenchmarking.orgMFLOPS Per Watt, More Is BetterQuantLib 1.21Threadripper 3960XThreadripper 3990X81624324036.0625.50
MinAvgMaxThreadripper 3960X40.280.886.9Threadripper 3990X54.9109.4118.2OpenBenchmarking.orgWatts, Fewer Is BetterQuantLib 1.21CPU Power Consumption Monitor4080120160200

WebP Image Encode

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

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest CompressionThreadripper 3960XThreadripper 3990X816243240SE +/- 0.05, N = 3SE +/- 0.14, N = 332.9934.121. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16 -ltiff
MinAvgMaxThreadripper 3960X42.686.195.4Threadripper 3990X52.1111.7122.1OpenBenchmarking.orgWatts, Fewer Is BetterWebP Image Encode 1.1CPU Power Consumption Monitor4080120160200

srsRAN

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

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAMThreadripper 3960XThreadripper 3990X306090120150SE +/- 0.38, N = 3SE +/- 0.15, N = 3150.5142.81. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f
OpenBenchmarking.orgUE Mb/s Per Watt, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAMThreadripper 3960XThreadripper 3990X0.42030.84061.26091.68122.10151.8681.342
MinAvgMaxThreadripper 3960X42.980.587.6Threadripper 3990X54.8106.4113.9OpenBenchmarking.orgWatts, Fewer Is BettersrsRAN 21.10CPU Power Consumption Monitor4080120160200

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAMThreadripper 3960XThreadripper 3990X80160240320400SE +/- 2.37, N = 3SE +/- 1.52, N = 3373.8356.71. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f

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 2.10.30Test: resizeThreadripper 3960XThreadripper 3990X612182430SE +/- 0.21, N = 5SE +/- 0.29, N = 420.1524.18
MinAvgMaxThreadripper 3960X40.799.7135.3OpenBenchmarking.orgWatts, Fewer Is BetterGIMP 2.10.30CPU Power Consumption Monitor4080120160200

srsRAN

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

OpenBenchmarking.orgSamples / Second, More Is BettersrsRAN 21.10Test: OFDM_TestThreadripper 3960XThreadripper 3990X30M60M90M120M150MSE +/- 1946221.41, N = 3SE +/- 1105038.96, N = 31418666671395666671. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f
OpenBenchmarking.orgSamples / Second Per Watt, More Is BettersrsRAN 21.10Test: OFDM_TestThreadripper 3960XThreadripper 3990X400K800K1200K1600K2000K1679732.171271784.63
MinAvgMaxThreadripper 3960X43.484.590.3Threadripper 3990X54.2109.7117.0OpenBenchmarking.orgWatts, Fewer Is BettersrsRAN 21.10CPU Power Consumption Monitor4080120160200

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 3960XThreadripper 3990X2K4K6K8K10KSE +/- 21.65, N = 3SE +/- 10.86, N = 37737.039889.261. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi
OpenBenchmarking.orgCG Mflops Per Watt, More Is BetterminiFE 2.2Problem Size: SmallThreadripper 3960XThreadripper 3990X122436486051.2742.42
MinAvgMaxThreadripper 3960X45.0150.9263.0Threadripper 3990X60.2233.1281.7OpenBenchmarking.orgWatts, Fewer Is BetterminiFE 2.2CPU Power Consumption Monitor70140210280350

PyPerformance

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

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: regex_compileThreadripper 3960XThreadripper 3990X4080120160200SE +/- 0.00, N = 3SE +/- 0.00, N = 3160157
MinAvgMaxThreadripper 3960X41.679.585.3Threadripper 3990X56.7107.3114.6OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor4080120160200

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 44100 - Buffer Size: 512Threadripper 3960XThreadripper 3990X1.20312.40623.60934.81246.0155SE +/- 0.032958, N = 3SE +/- 0.034562, N = 35.3382835.3469821. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio Per Watt, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 44100 - Buffer Size: 512Threadripper 3960XThreadripper 3990X0.00650.0130.01950.0260.03250.0290.027
MinAvgMaxThreadripper 3960X46.2184.9216.8Threadripper 3990X58.9200.3243.9OpenBenchmarking.orgWatts, Fewer Is BetterStargate Digital Audio Workstation 21.10.9CPU Power Consumption Monitor60120180240300

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 480000 - Buffer Size: 1024Threadripper 3960XThreadripper 3990X1.27842.55683.83525.11366.392SE +/- 0.037269, N = 3SE +/- 0.015679, N = 35.4428045.6817471. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio Per Watt, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 480000 - Buffer Size: 1024Threadripper 3960XThreadripper 3990X0.00650.0130.01950.0260.03250.0290.027
MinAvgMaxThreadripper 3960X43.7185.4219.5Threadripper 3990X58.6208.4246.4OpenBenchmarking.orgWatts, Fewer Is BetterStargate Digital Audio Workstation 21.10.9CPU Power Consumption Monitor60120180240300

LULESH

LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3Threadripper 3960XThreadripper 3990X2K4K6K8K10KSE +/- 8.27, N = 6SE +/- 7.48, N = 37952.589413.411. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi
OpenBenchmarking.orgz/s Per Watt, More Is BetterLULESH 2.0.3Threadripper 3960XThreadripper 3990X2040608010077.5837.18
MinAvgMaxThreadripper 3960X43.0102.5152.9Threadripper 3990X56.7253.2282.2OpenBenchmarking.orgWatts, Fewer Is BetterLULESH 2.0.3CPU Power Consumption Monitor70140210280350

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 44100 - Buffer Size: 1024Threadripper 3960XThreadripper 3990X1.30252.6053.90755.216.5125SE +/- 0.074227, N = 3SE +/- 0.007232, N = 35.6195405.7888161. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio Per Watt, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 44100 - Buffer Size: 1024Threadripper 3960XThreadripper 3990X0.00680.01360.02040.02720.0340.0300.028
MinAvgMaxThreadripper 3960X42.2186.0222.0Threadripper 3990X54.6207.0246.1OpenBenchmarking.orgWatts, Fewer Is BetterStargate Digital Audio Workstation 21.10.9CPU Power Consumption Monitor60120180240300

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 0.14Test: CPU StressThreadripper 3960XThreadripper 3990X30K60K90K120K150KSE +/- 186.10, N = 3SE +/- 1038.88, N = 362366.76123897.451. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14Test: CPU StressThreadripper 3960XThreadripper 3990X110220330440550268.72493.10
MinAvgMaxThreadripper 3960X43.3232.1259.2Threadripper 3990X56.1251.3281.4OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CryptoThreadripper 3960XThreadripper 3990X20K40K60K80K100KSE +/- 172.69, N = 3SE +/- 733.42, N = 344920.1392388.851. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14Test: CryptoThreadripper 3960XThreadripper 3990X80160240320400191.90366.73
MinAvgMaxThreadripper 3960X42.7234.1260.3Threadripper 3990X55.3251.9279.7OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Matrix MathThreadripper 3960XThreadripper 3990X50K100K150K200K250KSE +/- 154.06, N = 3SE +/- 728.11, N = 3119634.42219165.211. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14Test: Matrix MathThreadripper 3960XThreadripper 3990X2004006008001000464.40865.69
MinAvgMaxThreadripper 3960X45.9257.6287.2Threadripper 3990X55.5253.2279.7OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Vector MathThreadripper 3960XThreadripper 3990X70K140K210K280K350KSE +/- 294.28, N = 3SE +/- 3412.43, N = 3180602.86340961.121. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14Test: Vector MathThreadripper 3960XThreadripper 3990X30060090012001500712.221375.61
MinAvgMaxThreadripper 3960X44.4253.6285.7Threadripper 3990X56.9247.9279.7OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14CPU Power Consumption Monitor70140210280350

Google SynthMark

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

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100Threadripper 3960XThreadripper 3990X2004006008001000SE +/- 1.52, N = 3SE +/- 0.84, N = 3779.86741.331. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast
OpenBenchmarking.orgVoices Per Watt, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100Threadripper 3960XThreadripper 3990X369121510.3917.109
MinAvgMaxThreadripper 3960X42.075.180.3Threadripper 3990X55.0104.3112.0OpenBenchmarking.orgWatts, Fewer Is BetterGoogle SynthMark 20201109CPU Power Consumption Monitor4080120160200

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: sedovbigThreadripper 3960XThreadripper 3990X1020304050SE +/- 0.22, N = 3SE +/- 0.11, N = 442.1412.321. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi
MinAvgMaxThreadripper 3960X43.9196.7219.1Threadripper 3990X59.6224.5279.7OpenBenchmarking.orgWatts, Fewer Is BetterPennant 1.0.1CPU Power Consumption Monitor70140210280350

Inkscape

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

OpenBenchmarking.orgSeconds, Fewer Is BetterInkscapeOperation: SVG Files To PNGThreadripper 3960XThreadripper 3990X714212835SE +/- 0.13, N = 3SE +/- 0.03, N = 329.2130.541. Inkscape 1.1.2 (0a00cf5339, 2022-02-04)
MinAvgMaxThreadripper 3960X41.678.084.5Threadripper 3990X55.0114.3124.4OpenBenchmarking.orgWatts, Fewer Is BetterInkscapeCPU Power Consumption Monitor4080120160200

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: leblancbigThreadripper 3960XThreadripper 3990X714212835SE +/- 0.151231, N = 3SE +/- 0.064494, N = 1527.7689505.5785001. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi
MinAvgMaxThreadripper 3960X45.5194.4223.5Threadripper 3990X56.0188.0280.0OpenBenchmarking.orgWatts, Fewer Is BetterPennant 1.0.1CPU Power Consumption Monitor70140210280350

Kripke

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

OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.4Threadripper 3960XThreadripper 3990X30M60M90M120M150MSE +/- 675029.72, N = 3SE +/- 1456890.22, N = 31272763001059932001. (CXX) g++ options: -O3 -fopenmp
OpenBenchmarking.orgThroughput FoM Per Watt, More Is BetterKripke 1.2.4Threadripper 3960XThreadripper 3990X160K320K480K640K800K757026.41435217.70
MinAvgMaxThreadripper 3960X44.1168.1202.6Threadripper 3990X58.8243.5280.0OpenBenchmarking.orgWatts, Fewer Is BetterKripke 1.2.4CPU Power Consumption Monitor70140210280350

x265

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

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X510152025SE +/- 0.08, N = 3SE +/- 0.08, N = 322.0120.991. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFrames Per Second Per Watt, More Is Betterx265 3.4Video Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.02880.05760.08640.11520.1440.1280.094
MinAvgMaxThreadripper 3960X42.4171.9216.5Threadripper 3990X54.2223.1270.6OpenBenchmarking.orgWatts, Fewer Is Betterx265 3.4CPU Power Consumption Monitor70140210280350

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 3960XThreadripper 3990X500K1000K1500K2000K2500KSE +/- 3023.03, N = 3SE +/- 19224.58, N = 31139722.362248635.671. (CC) gcc options: -O2 -lrt" -lrt
OpenBenchmarking.orgIterations/Sec Per Watt, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondThreadripper 3960XThreadripper 3990X2K4K6K8K10K5285.589853.11
MinAvgMaxThreadripper 3960X46.3215.6262.3Threadripper 3990X60.4228.2281.4OpenBenchmarking.orgWatts, Fewer Is BetterCoremark 1.0CPU Power Consumption Monitor70140210280350

Kvazaar

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

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.1Video Input: Bosphorus 4K - Video Preset: Very FastThreadripper 3960XThreadripper 3990X714212835SE +/- 0.01, N = 3SE +/- 0.03, N = 318.1031.171. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterKvazaar 2.1Video Input: Bosphorus 4K - Video Preset: Very FastThreadripper 3960XThreadripper 3990X0.03310.06620.09930.13240.16550.1170.147
MinAvgMaxThreadripper 3960X45.0154.8180.0Threadripper 3990X55.2212.0266.3OpenBenchmarking.orgWatts, Fewer Is BetterKvazaar 2.1CPU Power Consumption Monitor70140210280350

libjpeg-turbo tjbench

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

OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.1.0Test: Decompression ThroughputThreadripper 3960XThreadripper 3990X50100150200250SE +/- 0.48, N = 3SE +/- 0.46, N = 3223.47215.571. (CC) gcc options: -O3 -rdynamic
OpenBenchmarking.orgMegapixels/sec Per Watt, More Is Betterlibjpeg-turbo tjbench 2.1.0Test: Decompression ThroughputThreadripper 3960XThreadripper 3990X0.64061.28121.92182.56243.2032.8471.997
MinAvgMaxThreadripper 3960X42.078.585.9Threadripper 3990X58.1108.0116.8OpenBenchmarking.orgWatts, Fewer Is Betterlibjpeg-turbo tjbench 2.1.0CPU Power Consumption Monitor4080120160200

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: ExhaustiveThreadripper 3960XThreadripper 3990X612182430SE +/- 0.06, N = 3SE +/- 0.10, N = 327.3413.911. (CXX) g++ options: -O3 -flto -pthread
MinAvgMaxThreadripper 3960X45.3230.5282.2Threadripper 3990X57.2207.8279.7OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 3.2CPU Power Consumption Monitor70140210280350

SVT-HEVC

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X306090120150SE +/- 0.85, N = 15SE +/- 0.22, N = 798.86119.001. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.15370.30740.46110.61480.76850.5560.683
MinAvgMaxThreadripper 3960X45.4177.8287.1Threadripper 3990X56.3174.2280.1OpenBenchmarking.orgWatts, Fewer Is BetterSVT-HEVC 1.5.0CPU Power Consumption Monitor70140210280350

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm.in input file (Problem 5). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsThreadripper 3960XThreadripper 3990X714212835SE +/- 0.05, N = 3SE +/- 0.03, N = 428.3914.741. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
MinAvgMaxThreadripper 3960X45.3182.1204.5Threadripper 3990X57.2234.1279.7OpenBenchmarking.orgWatts, Fewer Is BetterCloverLeafCPU Power Consumption Monitor70140210280350

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 2.10.30Test: unsharp-maskThreadripper 3960XThreadripper 3990X612182430SE +/- 0.04, N = 3SE +/- 0.10, N = 322.2624.83
MinAvgMaxThreadripper 3960X41.197.2125.5OpenBenchmarking.orgWatts, Fewer Is BetterGIMP 2.10.30CPU Power Consumption Monitor4080120160200

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 TimesThreadripper 3960XThreadripper 3990X2004006008001000SE +/- 4.16, N = 3SE +/- 2.52, N = 310031017
MinAvgMaxThreadripper 3960X43.180.587.6Threadripper 3990X57.4111.0120.9OpenBenchmarking.orgWatts, Fewer Is BetterPyBench 2018-02-16CPU Power Consumption Monitor4080120160200

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 TimeThreadripper 3960XThreadripper 3990X2M4M6M8M10MSE +/- 26321.35, N = 3SE +/- 12538.70, N = 3874128583691741. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterCrafty 25.2Elapsed TimeThreadripper 3960XThreadripper 3990X20K40K60K80K100K111595.0980747.19
MinAvgMaxThreadripper 3960X42.878.385.8Threadripper 3990X54.9103.6112.7OpenBenchmarking.orgWatts, Fewer Is BetterCrafty 25.2CPU Power Consumption Monitor4080120160200

PyPerformance

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

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: json_loadsThreadripper 3960XThreadripper 3990X612182430SE +/- 0.03, N = 3SE +/- 0.00, N = 322.923.6
MinAvgMaxThreadripper 3960X42.077.084.5Threadripper 3990X55.8105.4114.6OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor4080120160200

Basis Universal

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: UASTC Level 3Threadripper 3960XThreadripper 3990X612182430SE +/- 0.04, N = 3SE +/- 0.04, N = 423.8015.511. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
MinAvgMaxThreadripper 3960X43.7201.1282.7Threadripper 3990X55.2180.1279.7OpenBenchmarking.orgWatts, Fewer Is BetterBasis Universal 1.13CPU Power Consumption Monitor70140210280350

srsRAN

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

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.10Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMThreadripper 3960XThreadripper 3990X1428425670SE +/- 0.06, N = 3SE +/- 0.07, N = 360.758.91. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f
OpenBenchmarking.orgUE Mb/s Per Watt, More Is BettersrsRAN 21.10Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMThreadripper 3960XThreadripper 3990X0.17660.35320.52980.70640.8830.7850.579
MinAvgMaxThreadripper 3960X44.277.483.3Threadripper 3990X54.2101.7110.5OpenBenchmarking.orgWatts, Fewer Is BettersrsRAN 21.10CPU Power Consumption Monitor20406080100

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.10Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMThreadripper 3960XThreadripper 3990X20406080100SE +/- 0.20, N = 3SE +/- 0.23, N = 3107.0103.71. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f

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.CThreadripper 3960XThreadripper 3990X6K12K18K24K30KSE +/- 26.71, N = 4SE +/- 24.04, N = 425858.5428515.371. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CThreadripper 3960XThreadripper 3990X4080120160200165.05125.43
MinAvgMaxThreadripper 3960X43.6156.7194.2Threadripper 3990X57.8227.3280.7OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor70140210280350

Timed Mesa Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Mesa Compilation 21.0Time To CompileThreadripper 3960XThreadripper 3990X510152025SE +/- 0.02, N = 3SE +/- 0.10, N = 322.6520.35
MinAvgMaxThreadripper 3960X44.5180.9287.2Threadripper 3990X57.7176.2283.7OpenBenchmarking.orgWatts, Fewer Is BetterTimed Mesa Compilation 21.0CPU Power Consumption Monitor70140210280350

PyPerformance

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

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: crypto_pyaesThreadripper 3960XThreadripper 3990X306090120150SE +/- 0.00, N = 3SE +/- 0.00, N = 3106114
MinAvgMaxThreadripper 3960X42.076.985.0Threadripper 3990X53.5104.7114.1OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor4080120160200

Timed FFmpeg Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.4Time To CompileThreadripper 3960XThreadripper 3990X612182430SE +/- 0.08, N = 3SE +/- 0.02, N = 323.7018.01
MinAvgMaxThreadripper 3960X42.4174.6287.2Threadripper 3990X56.6167.7279.8OpenBenchmarking.orgWatts, Fewer Is BetterTimed FFmpeg Compilation 4.4CPU Power Consumption Monitor70140210280350

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian DragonThreadripper 3960XThreadripper 3990X1020304050SE +/- 0.17, N = 3SE +/- 0.05, N = 430.2642.29MIN: 29.74 / MAX: 30.94MIN: 41.39 / MAX: 44.16
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian DragonThreadripper 3960XThreadripper 3990X0.04280.08560.12840.17120.2140.1280.190
MinAvgMaxThreadripper 3960X45.8235.5278.1Threadripper 3990X56.7222.7279.8OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 3.13CPU Power Consumption Monitor70140210280350

srsRAN

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

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB SISO 64-QAMThreadripper 3960XThreadripper 3990X4080120160200SE +/- 0.23, N = 4SE +/- 0.09, N = 4163.9159.71. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f
OpenBenchmarking.orgUE Mb/s Per Watt, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB SISO 64-QAMThreadripper 3960XThreadripper 3990X0.48490.96981.45471.93962.42452.1551.585
MinAvgMaxThreadripper 3960X42.376.185.1Threadripper 3990X54.1100.8112.1OpenBenchmarking.orgWatts, Fewer Is BettersrsRAN 21.10CPU Power Consumption Monitor4080120160200

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB SISO 64-QAMThreadripper 3960XThreadripper 3990X80160240320400SE +/- 0.77, N = 4SE +/- 0.53, N = 4379.2368.71. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f

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 9Threadripper 3960XThreadripper 3990X510152025SE +/- 0.06, N = 3SE +/- 0.05, N = 320.2020.821. (CC) gcc options: -fvisibility=hidden -O2
MinAvgMaxThreadripper 3960X42.3130.7167.3Threadripper 3990X53.3157.4193.8OpenBenchmarking.orgWatts, Fewer Is BetterXZ Compression 5.2.4CPU Power Consumption Monitor50100150200250

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 3960XThreadripper 3990X510152025SE +/- 0.04, N = 3SE +/- 0.06, N = 521.6011.041. (CC) gcc options: -lm -lpthread -O3
MinAvgMaxThreadripper 3960X44.3235.4278.3Threadripper 3990X56.9217.4281.4OpenBenchmarking.orgWatts, Fewer Is BetterC-Ray 1.1CPU Power Consumption Monitor70140210280350

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 57Threadripper 3960XThreadripper 3990X600M1200M1800M2400M3000MSE +/- 2165640.78, N = 3SE +/- 9335059.36, N = 3167270000030113000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 57Threadripper 3960XThreadripper 3990X3M6M9M12M15M6811625.1712418251.26
MinAvgMaxThreadripper 3960X45.0245.6287.2Threadripper 3990X57.4242.5279.7OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 2021.01.31CPU Power Consumption Monitor70140210280350

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: CrownThreadripper 3960XThreadripper 3990X1122334455SE +/- 0.08, N = 3SE +/- 0.13, N = 428.0049.26MIN: 27.53 / MAX: 28.55MIN: 48.23 / MAX: 51.75
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: CrownThreadripper 3960XThreadripper 3990X0.04910.09820.14730.19640.24550.1130.218
MinAvgMaxThreadripper 3960X45.0246.7287.2Threadripper 3990X56.8225.6279.8OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 3.13CPU Power Consumption Monitor70140210280350

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 64 - Buffer Length: 256 - Filter Length: 57Threadripper 3960XThreadripper 3990X600M1200M1800M2400M3000MSE +/- 1792577.29, N = 3SE +/- 12081988.43, N = 3164000000028128333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 2021.01.31Threads: 64 - Buffer Length: 256 - Filter Length: 57Threadripper 3960XThreadripper 3990X2M4M6M8M10M6645557.8711633287.63
MinAvgMaxThreadripper 3960X44.4246.8288.3Threadripper 3990X31.2241.8279.7OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 2021.01.31CPU Power Consumption Monitor70140210280350

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.4Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X714212835SE +/- 0.10, N = 3SE +/- 0.07, N = 331.6831.851. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterAOM AV1 3.4Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.09250.1850.27750.370.46250.4110.250
MinAvgMaxThreadripper 3960X43.077.085.4Threadripper 3990X56.4127.3188.8OpenBenchmarking.orgWatts, Fewer Is BetterAOM AV1 3.4CPU Power Consumption Monitor50100150200250

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 32 - Buffer Length: 256 - Filter Length: 57Threadripper 3960XThreadripper 3990X400M800M1200M1600M2000MSE +/- 3467467.86, N = 3SE +/- 6018582.34, N = 3154340000017846000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 2021.01.31Threads: 32 - Buffer Length: 256 - Filter Length: 57Threadripper 3960XThreadripper 3990X1.6M3.2M4.8M6.4M8M6306123.187333859.58
MinAvgMaxThreadripper 3960X43.3244.6287.2Threadripper 3990X57.2243.3279.7OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 2021.01.31CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57Threadripper 3960XThreadripper 3990X200M400M600M800M1000MSE +/- 2818391.98, N = 3SE +/- 4403878.34, N = 310259000009804733331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57Threadripper 3960XThreadripper 3990X1.1M2.2M3.3M4.4M5.5M4943928.604453239.19
MinAvgMaxThreadripper 3960X42.1207.5242.8Threadripper 3990X56.6220.2267.6OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 2021.01.31CPU Power Consumption Monitor70140210280350

Timed Apache Compilation

This test times how long it takes to build the Apache HTTPD web server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.41Time To CompileThreadripper 3960XThreadripper 3990X510152025SE +/- 0.05, N = 3SE +/- 0.10, N = 319.3820.47
MinAvgMaxThreadripper 3960X42.493.5185.6Threadripper 3990X54.6128.4210.6OpenBenchmarking.orgWatts, Fewer Is BetterTimed Apache Compilation 2.4.41CPU Power Consumption Monitor60120180240300

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian DragonThreadripper 3960XThreadripper 3990X1020304050SE +/- 0.26, N = 3SE +/- 0.04, N = 432.9244.25MIN: 32.28 / MAX: 33.92MIN: 43.39 / MAX: 46.47
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian DragonThreadripper 3960XThreadripper 3990X0.0470.0940.1410.1880.2350.1430.209
MinAvgMaxThreadripper 3960X47.5230.3273.9Threadripper 3990X57.8211.3279.8OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 3.13CPU Power Consumption Monitor70140210280350

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.1Threadripper 3960XThreadripper 3990X50100150200250SE +/- 0.30, N = 4SE +/- 0.30, N = 3233.44244.40MIN: 232.23 / MAX: 246.28MIN: 243.4 / MAX: 245.651. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
MinAvgMaxThreadripper 3960X41.575.183.2Threadripper 3990X54.0102.2113.0OpenBenchmarking.orgWatts, Fewer Is BetterTNN 0.3CPU Power Consumption Monitor4080120160200

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.4Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X1020304050SE +/- 0.23, N = 4SE +/- 0.13, N = 445.2744.081. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterAOM AV1 3.4Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.11570.23140.34710.46280.57850.5140.384
MinAvgMaxThreadripper 3960X42.588.0129.6Threadripper 3990X29.4114.9164.2OpenBenchmarking.orgWatts, Fewer Is BetterAOM AV1 3.4CPU Power Consumption Monitor50100150200250

WebP Image Encode

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

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessThreadripper 3960XThreadripper 3990X48121620SE +/- 0.15, N = 4SE +/- 0.17, N = 316.0316.511. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16 -ltiff
MinAvgMaxThreadripper 3960X42.575.686.1Threadripper 3990X54.4100.5113.5OpenBenchmarking.orgWatts, Fewer Is BetterWebP Image Encode 1.1CPU Power Consumption Monitor4080120160200

Basis Universal

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: UASTC Level 2Threadripper 3960XThreadripper 3990X48121620SE +/- 0.01, N = 4SE +/- 0.00, N = 514.7410.941. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
MinAvgMaxThreadripper 3960X42.3165.1283.4Threadripper 3990X57.8149.8279.7OpenBenchmarking.orgWatts, Fewer Is BetterBasis Universal 1.13CPU Power Consumption Monitor70140210280350

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v2Threadripper 3960XThreadripper 3990X60120180240300SE +/- 0.30, N = 3SE +/- 2.20, N = 3257.49288.35MIN: 246.92 / MAX: 318.98MIN: 263.94 / MAX: 386.221. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
MinAvgMaxThreadripper 3960X42.397.8111.0Threadripper 3990X58.4171.1198.3OpenBenchmarking.orgWatts, Fewer Is BetterTNN 0.3CPU Power Consumption Monitor50100150200250

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.4Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X1020304050SE +/- 0.11, N = 4SE +/- 0.08, N = 445.6045.141. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterAOM AV1 3.4Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.12020.24040.36060.48080.6010.5340.360
MinAvgMaxThreadripper 3960X42.185.3127.3Threadripper 3990X53.4125.3185.4OpenBenchmarking.orgWatts, Fewer Is BetterAOM AV1 3.4CPU Power Consumption Monitor50100150200250

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 2.10.30Test: auto-levelsThreadripper 3960XThreadripper 3990X510152025SE +/- 0.08, N = 3SE +/- 0.04, N = 318.1019.83
MinAvgMaxThreadripper 3960X42.587.7106.4OpenBenchmarking.orgWatts, Fewer Is BetterGIMP 2.10.30CPU Power Consumption Monitor20406080100

srsRAN

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

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB SISO 256-QAMThreadripper 3960XThreadripper 3990X4080120160200SE +/- 0.12, N = 3SE +/- 0.06, N = 3172.4166.81. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f
OpenBenchmarking.orgUE Mb/s Per Watt, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB SISO 256-QAMThreadripper 3960XThreadripper 3990X0.49860.99721.49581.99442.4932.2161.636
MinAvgMaxThreadripper 3960X42.977.884.7Threadripper 3990X54.2102.0111.9OpenBenchmarking.orgWatts, Fewer Is BettersrsRAN 21.10CPU Power Consumption Monitor4080120160200

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 21.10Test: 4G PHY_DL_Test 100 PRB SISO 256-QAMThreadripper 3960XThreadripper 3990X90180270360450SE +/- 0.06, N = 3SE +/- 0.44, N = 3405.5392.91. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lpthread -lm -lfftw3f

NAS Parallel Benchmarks

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

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CThreadripper 3960XThreadripper 3990X2K4K6K8K10KSE +/- 22.97, N = 4SE +/- 5.53, N = 310603.828366.741. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CThreadripper 3960XThreadripper 3990X163248648069.6635.57
MinAvgMaxThreadripper 3960X44.3152.2187.8Threadripper 3990X56.6235.2280.0OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor70140210280350

Timed MPlayer Compilation

This test times how long it takes to build the MPlayer open-source media player program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MPlayer Compilation 1.5Time To CompileThreadripper 3960XThreadripper 3990X48121620SE +/- 0.04, N = 4SE +/- 0.03, N = 514.5810.61
MinAvgMaxThreadripper 3960X44.6179.3287.7Threadripper 3990X56.9172.9282.2OpenBenchmarking.orgWatts, Fewer Is BetterTimed MPlayer Compilation 1.5CPU Power Consumption Monitor70140210280350

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 3960XThreadripper 3990X48121620SE +/- 0.06, N = 4SE +/- 0.05, N = 414.6813.14
MinAvgMaxThreadripper 3960X43.9127.3287.2Threadripper 3990X55.5139.9279.6OpenBenchmarking.orgWatts, Fewer Is BetterTimed ImageMagick Compilation 6.9.0CPU Power Consumption Monitor70140210280350

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: CrownThreadripper 3960XThreadripper 3990X1224364860SE +/- 0.12, N = 3SE +/- 0.19, N = 430.7354.31MIN: 30.15 / MAX: 31.5MIN: 53.05 / MAX: 57
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 3.13Binary: Pathtracer - Model: CrownThreadripper 3960XThreadripper 3990X0.05560.11120.16680.22240.2780.1260.247
MinAvgMaxThreadripper 3960X42.3243.9287.2Threadripper 3990X54.5219.5279.8OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 3.13CPU Power Consumption Monitor70140210280350

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 3960XThreadripper 3990X10K20K30K40K50KSE +/- 82.61, N = 4SE +/- 111.88, N = 624742.2648358.841. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.BThreadripper 3960XThreadripper 3990X50100150200250169.03239.66
MinAvgMaxThreadripper 3960X42.9146.4177.3Threadripper 3990X53.9201.8280.2OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor70140210280350

Xcompact3d Incompact3d

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

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per DirectionThreadripper 3960XThreadripper 3990X3691215SE +/- 0.03602470, N = 4SE +/- 0.01563652, N = 513.476237109.372105791. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
MinAvgMaxThreadripper 3960X45.0157.3199.5Threadripper 3990X60.4211.9280.3OpenBenchmarking.orgWatts, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Power Consumption Monitor70140210280350

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 3960XThreadripper 3990X48121620SE +/- 0.03, N = 3SE +/- 0.06, N = 417.1710.051. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -R/usr/lib -lXpm -lSM -lICE -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system
MinAvgMaxThreadripper 3960X44.5216.4287.2Threadripper 3990X55.5193.3279.8OpenBenchmarking.orgWatts, Fewer Is BetterPOV-Ray 3.7.0.7CPU Power Consumption Monitor70140210280350

x265

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

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pThreadripper 3960XThreadripper 3990X1122334455SE +/- 0.08, N = 4SE +/- 0.11, N = 444.6646.961. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFrames Per Second Per Watt, More Is Betterx265 3.4Video Input: Bosphorus 1080pThreadripper 3960XThreadripper 3990X0.09270.18540.27810.37080.46350.4120.350
MinAvgMaxThreadripper 3960X45.1108.3140.4Threadripper 3990X56.5134.1196.9OpenBenchmarking.orgWatts, Fewer Is Betterx265 3.4CPU Power Consumption Monitor50100150200250

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: JPEG - Encode Speed: 7Threadripper 3960XThreadripper 3990X20406080100SE +/- 0.75, N = 5SE +/- 0.13, N = 499.0791.951. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie
OpenBenchmarking.orgMP/s Per Watt, More Is BetterJPEG XL libjxl 0.6.1Input: JPEG - Encode Speed: 7Threadripper 3960XThreadripper 3990X0.24950.4990.74850.9981.24751.1090.713
MinAvgMaxThreadripper 3960X43.589.3111.3Threadripper 3990X55.5129.0161.0OpenBenchmarking.orgWatts, Fewer Is BetterJPEG XL libjxl 0.6.1CPU Power Consumption Monitor50100150200250

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 3960XThreadripper 3990X11002200330044005500SE +/- 33.30, N = 5SE +/- 44.91, N = 550435322
MinAvgMaxThreadripper 3960X44.185.6146.9Threadripper 3990X55.5118.3213.2OpenBenchmarking.orgWatts, Fewer Is BetterDaCapo Benchmark 9.12-MR1CPU Power Consumption Monitor60120180240300

JPEG XL Decoding libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.6.1CPU Threads: AllThreadripper 3960XThreadripper 3990X120240360480600SE +/- 3.04, N = 3SE +/- 1.46, N = 3491.63539.71
OpenBenchmarking.orgMP/s Per Watt, More Is BetterJPEG XL Decoding libjxl 0.6.1CPU Threads: AllThreadripper 3960XThreadripper 3990X0.96411.92822.89233.85644.82054.2853.945
MinAvgMaxThreadripper 3960X43.1114.7170.7Threadripper 3990X54.5136.8207.4OpenBenchmarking.orgWatts, Fewer Is BetterJPEG XL Decoding libjxl 0.6.1CPU Power Consumption Monitor60120180240300

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: JPEG - Encode Speed: 8Threadripper 3960XThreadripper 3990X714212835SE +/- 0.11, N = 5SE +/- 0.05, N = 529.8728.251. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie
OpenBenchmarking.orgMP/s Per Watt, More Is BetterJPEG XL libjxl 0.6.1Input: JPEG - Encode Speed: 8Threadripper 3960XThreadripper 3990X0.0880.1760.2640.3520.440.3910.277
MinAvgMaxThreadripper 3960X43.976.492.7Threadripper 3990X55.0102.1123.8OpenBenchmarking.orgWatts, Fewer Is BetterJPEG XL libjxl 0.6.1CPU Power Consumption Monitor4080120160200

OpenSCAD

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Leonardo Phone Case SlimThreadripper 3960XThreadripper 3990X48121620SE +/- 0.02, N = 3SE +/- 0.10, N = 316.6917.301. OpenSCAD version 2021.01
MinAvgMaxThreadripper 3960X41.375.784.3Threadripper 3990X54.0101.0111.9OpenBenchmarking.orgWatts, Fewer Is BetterOpenSCADCPU Power Consumption Monitor4080120160200

Kvazaar

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

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.1Video Input: Bosphorus 4K - Video Preset: Ultra FastThreadripper 3960XThreadripper 3990X1428425670SE +/- 0.13, N = 3SE +/- 0.08, N = 534.5860.661. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterKvazaar 2.1Video Input: Bosphorus 4K - Video Preset: Ultra FastThreadripper 3960XThreadripper 3990X0.06370.12740.19110.25480.31850.2240.283
MinAvgMaxThreadripper 3960X44.4154.7189.7Threadripper 3990X58.8214.1280.0OpenBenchmarking.orgWatts, Fewer Is BetterKvazaar 2.1CPU Power Consumption Monitor70140210280350

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 8 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X1224364860SE +/- 0.29, N = 4SE +/- 0.17, N = 455.0554.431. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 8 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.06980.13960.20940.27920.3490.3100.278
MinAvgMaxThreadripper 3960X46.4177.6286.7Threadripper 3990X56.9195.6279.0OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 1.0CPU Power Consumption Monitor70140210280350

Google Draco

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

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.0Model: Church FacadeThreadripper 3960XThreadripper 3990X2K4K6K8K10KSE +/- 11.99, N = 5SE +/- 9.97, N = 5770988611. (CXX) g++ options: -O3
MinAvgMaxThreadripper 3960X41.469.082.9Threadripper 3990X54.194.5110.6OpenBenchmarking.orgWatts, Fewer Is BetterGoogle Draco 1.5.0CPU Power Consumption Monitor20406080100

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 2.10.30Test: rotateThreadripper 3960XThreadripper 3990X48121620SE +/- 0.13, N = 3SE +/- 0.11, N = 315.6716.69
MinAvgMaxThreadripper 3960X41.976.396.6OpenBenchmarking.orgWatts, Fewer Is BetterGIMP 2.10.30CPU Power Consumption Monitor20406080100

GEGL

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

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: ScaleThreadripper 3960XThreadripper 3990X3691215SE +/- 0.10, N = 4SE +/- 0.04, N = 411.5812.59
MinAvgMaxThreadripper 3960X41.992.2118.6Threadripper 3990X55.0132.5176.6OpenBenchmarking.orgWatts, Fewer Is BetterGEGLCPU Power Consumption Monitor50100150200250

OpenSCAD

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Projector Mount SwivelThreadripper 3960XThreadripper 3990X246810SE +/- 0.019, N = 6SE +/- 0.027, N = 67.8708.1621. OpenSCAD version 2021.01
MinAvgMaxThreadripper 3960X41.970.184.3Threadripper 3990X51.794.1112.3OpenBenchmarking.orgWatts, Fewer Is BetterOpenSCADCPU Power Consumption Monitor4080120160200

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 3960XThreadripper 3990X8001600240032004000SE +/- 15.71, N = 6SE +/- 38.92, N = 531983800
MinAvgMaxThreadripper 3960X22.9105.2237.6Threadripper 3990X55.5137.4279.0OpenBenchmarking.orgWatts, Fewer Is BetterDaCapo Benchmark 9.12-MR1CPU Power Consumption Monitor70140210280350

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X20406080100SE +/- 0.21, N = 6SE +/- 0.30, N = 5101.72103.621. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.14060.28120.42180.56240.7030.6250.623
MinAvgMaxThreadripper 3960X44.9162.7275.8Threadripper 3990X54.1166.3279.7OpenBenchmarking.orgWatts, Fewer Is BetterSVT-VP9 0.3CPU Power Consumption Monitor70140210280350

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 3960XThreadripper 3990X10002000300040005000SE +/- 29.30, N = 6SE +/- 25.60, N = 539484532
MinAvgMaxThreadripper 3960X44.179.5123.8Threadripper 3990X54.7113.9175.3OpenBenchmarking.orgWatts, Fewer Is BetterDaCapo Benchmark 9.12-MR1CPU Power Consumption Monitor50100150200250

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X20406080100SE +/- 0.18, N = 6SE +/- 0.32, N = 5105.14108.861. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.16810.33620.50430.67240.84050.7470.669
MinAvgMaxThreadripper 3960X43.9140.8235.4Threadripper 3990X55.6162.6278.1OpenBenchmarking.orgWatts, Fewer Is BetterSVT-VP9 0.3CPU Power Consumption Monitor70140210280350

GEGL

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

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: CropThreadripper 3960XThreadripper 3990X246810SE +/- 0.052, N = 6SE +/- 0.022, N = 57.9598.481
MinAvgMaxThreadripper 3960X40.189.5119.8Threadripper 3990X53.8135.5190.6OpenBenchmarking.orgWatts, Fewer Is BetterGEGLCPU Power Consumption Monitor50100150200250

Parallel BZIP2 Compression

This test measures the time needed to compress a file (FreeBSD-13.0-RELEASE-amd64-memstick.img) using Parallel BZIP2 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.13FreeBSD-13.0-RELEASE-amd64-memstick.img CompressionThreadripper 3960XThreadripper 3990X0.85571.71142.56713.42284.2785SE +/- 0.043, N = 15SE +/- 0.022, N = 153.8032.1141. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread
MinAvgMaxThreadripper 3960X42.9130.2287.2Threadripper 3990X55.5123.0280.1OpenBenchmarking.orgWatts, Fewer Is BetterParallel BZIP2 Compression 1.1.13CPU Power Consumption Monitor70140210280350

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 3960XThreadripper 3990X6K12K18K24K30KSE +/- 3.74, N = 6SE +/- 38.09, N = 625278.4226818.011. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CThreadripper 3960XThreadripper 3990X50100150200250210.62136.79
MinAvgMaxThreadripper 3960X43.3120.0167.2Threadripper 3990X60.0196.1280.5OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor70140210280350

WebP Image Encode

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

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest CompressionThreadripper 3960XThreadripper 3990X246810SE +/- 0.010, N = 6SE +/- 0.012, N = 66.9997.2051. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16 -ltiff
MinAvgMaxThreadripper 3960X40.470.285.6Threadripper 3990X54.492.8113.3OpenBenchmarking.orgWatts, Fewer Is BetterWebP Image Encode 1.1CPU Power Consumption Monitor4080120160200

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: ThoroughThreadripper 3960XThreadripper 3990X246810SE +/- 0.0116, N = 6SE +/- 0.0086, N = 77.34066.08841. (CXX) g++ options: -O3 -flto -pthread
MinAvgMaxThreadripper 3960X43.5122.9284.5Threadripper 3990X58.2123.4279.4OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 3.2CPU Power Consumption Monitor70140210280350

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 6.4.0Threadripper 3960XThreadripper 3990X246810SE +/- 0.029, N = 6SE +/- 0.017, N = 66.7177.220
MinAvgMaxThreadripper 3960X42.576.7118.5OpenBenchmarking.orgWatts, Fewer Is BetterGNU Octave Benchmark 6.4.0CPU Power Consumption Monitor4080120160200

Google Draco

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

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.0Model: LionThreadripper 3960XThreadripper 3990X12002400360048006000SE +/- 18.19, N = 6SE +/- 5.51, N = 6498756161. (CXX) g++ options: -O3
MinAvgMaxThreadripper 3960X41.668.084.0Threadripper 3990X55.092.7112.9OpenBenchmarking.orgWatts, Fewer Is BetterGoogle Draco 1.5.0CPU Power Consumption Monitor4080120160200

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 10 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X306090120150SE +/- 0.55, N = 6SE +/- 0.36, N = 6117.08109.601. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 10 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.19490.38980.58470.77960.97450.8660.732
MinAvgMaxThreadripper 3960X44.5135.3255.1Threadripper 3990X57.1149.7260.4OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 1.0CPU Power Consumption Monitor70140210280350

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 12 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X306090120150SE +/- 1.06, N = 7SE +/- 0.42, N = 7142.55133.041. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 12 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.27680.55360.83041.10721.3841.2301.014
MinAvgMaxThreadripper 3960X44.0115.9217.2Threadripper 3990X54.1131.2233.5OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 1.0CPU Power Consumption Monitor60120180240300

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 3960XThreadripper 3990X9001800270036004500SE +/- 11.63, N = 7SE +/- 28.07, N = 740464318
MinAvgMaxThreadripper 3960X42.981.5154.4Threadripper 3990X56.8104.9193.0OpenBenchmarking.orgWatts, Fewer Is BetterDaCapo Benchmark 9.12-MR1CPU Power Consumption Monitor50100150200250

WebP2 Image Encode

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

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: DefaultThreadripper 3960XThreadripper 3990X0.54741.09481.64222.18962.737SE +/- 0.020, N = 15SE +/- 0.029, N = 152.3592.4331. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
MinAvgMaxThreadripper 3960X42.499.8243.3Threadripper 3990X53.4120.9268.0OpenBenchmarking.orgWatts, Fewer Is BetterWebP2 Image Encode 20220422CPU Power Consumption Monitor70140210280350

Primesieve

Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 8.0Length: 1e12Threadripper 3960XThreadripper 3990X246810SE +/- 0.008, N = 6SE +/- 0.009, N = 96.9993.7381. (CXX) g++ options: -O3
MinAvgMaxThreadripper 3960X43.5196.8287.2Threadripper 3990X55.3166.0280.0OpenBenchmarking.orgWatts, Fewer Is BetterPrimesieve 8.0CPU Power Consumption Monitor70140210280350

SVT-HEVC

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X4080120160200SE +/- 0.25, N = 8SE +/- 0.74, N = 7160.62151.181. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 4KThreadripper 3960XThreadripper 3990X0.25920.51840.77761.03681.2961.1520.950
MinAvgMaxThreadripper 3960X43.8139.5264.2Threadripper 3990X53.1159.1279.9OpenBenchmarking.orgWatts, Fewer Is BetterSVT-HEVC 1.5.0CPU Power Consumption Monitor70140210280350

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v2Threadripper 3960XThreadripper 3990X1530456075SE +/- 0.25, N = 8SE +/- 0.19, N = 767.5168.97MIN: 66.25 / MAX: 77.32MIN: 66.69 / MAX: 72.461. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
MinAvgMaxThreadripper 3960X42.666.584.4Threadripper 3990X56.090.5116.0OpenBenchmarking.orgWatts, Fewer Is BetterTNN 0.3CPU Power Consumption Monitor4080120160200

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. The test profile uses a 8K x 8K game texture as a sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 1.0Benchmark: Multi-Threaded - Configuration: ETC2Threadripper 3960XThreadripper 3990X10002000300040005000SE +/- 15.41, N = 7SE +/- 32.34, N = 84303.374515.861. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s Per Watt, More Is BetterEtcpak 1.0Benchmark: Multi-Threaded - Configuration: ETC2Threadripper 3960XThreadripper 3990X91827364536.1638.86
MinAvgMaxThreadripper 3960X21.2119.0214.8Threadripper 3990X54.7116.2213.0OpenBenchmarking.orgWatts, Fewer Is BetterEtcpak 1.0CPU Power Consumption Monitor60120180240300

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.CThreadripper 3960XThreadripper 3990X10002000300040005000SE +/- 1.87, N = 8SE +/- 32.72, N = 152268.064690.731. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CThreadripper 3960XThreadripper 3990X81624324017.1934.96
MinAvgMaxThreadripper 3960X43.6131.9228.8Threadripper 3990X57.7134.2279.8OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor70140210280350

OpenSCAD

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Retro CarThreadripper 3960XThreadripper 3990X0.99681.99362.99043.98724.984SE +/- 0.013, N = 8SE +/- 0.006, N = 84.2524.4301. OpenSCAD version 2021.01
MinAvgMaxThreadripper 3960X41.464.584.3Threadripper 3990X51.886.3111.7OpenBenchmarking.orgWatts, Fewer Is BetterOpenSCADCPU Power Consumption Monitor4080120160200

Darktable

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

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Masskrug - Acceleration: CPU-onlyThreadripper 3960XThreadripper 3990X0.53841.07681.61522.15362.692SE +/- 0.003, N = 9SE +/- 0.004, N = 92.3932.355
MinAvgMaxThreadripper 3960X40.994.3207.3OpenBenchmarking.orgWatts, Fewer Is BetterDarktable 3.8.1CPU Power Consumption Monitor60120180240300

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Server Room - Acceleration: CPU-onlyThreadripper 3960XThreadripper 3990X0.48080.96161.44241.92322.404SE +/- 0.004, N = 9SE +/- 0.003, N = 92.1202.137
MinAvgMaxThreadripper 3960X41.288.7205.6OpenBenchmarking.orgWatts, Fewer Is BetterDarktable 3.8.1CPU Power Consumption Monitor60120180240300

WebP2 Image Encode

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

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 100, Compression Effort 5Threadripper 3960XThreadripper 3990X0.68991.37982.06972.75963.4495SE +/- 0.008, N = 9SE +/- 0.004, N = 92.9653.0661. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
MinAvgMaxThreadripper 3960X43.3121.9245.0Threadripper 3990X53.4147.3271.2OpenBenchmarking.orgWatts, Fewer Is BetterWebP2 Image Encode 20220422CPU Power Consumption Monitor70140210280350

Darktable

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

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Boat - Acceleration: CPU-onlyThreadripper 3960XThreadripper 3990X0.46760.93521.40281.87042.338SE +/- 0.010, N = 10SE +/- 0.006, N = 102.0781.614
MinAvgMaxThreadripper 3960X42.9107.8255.7OpenBenchmarking.orgWatts, Fewer Is BetterDarktable 3.8.1CPU Power Consumption Monitor70140210280350

LAMMPS Molecular Dynamics Simulator

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

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: Rhodopsin ProteinThreadripper 3960XThreadripper 3990X612182430SE +/- 0.10, N = 11SE +/- 0.17, N = 1516.8923.151. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgns/day Per Watt, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: Rhodopsin ProteinThreadripper 3960XThreadripper 3990X0.04430.08860.13290.17720.22150.1690.197
MinAvgMaxThreadripper 3960X43.599.9263.9Threadripper 3990X55.4117.6279.6OpenBenchmarking.orgWatts, Fewer Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022CPU Power Consumption Monitor70140210280350

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 PerformanceThreadripper 3960XThreadripper 3990X300K600K900K1200K1500KSE +/- 1085.85, N = 12SE +/- 904.73, N = 12135660513250531. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterTSCP 1.81AI Chess PerformanceThreadripper 3960XThreadripper 3990X5K10K15K20K25K24597.9618985.72
MinAvgMaxThreadripper 3960X42.755.281.3Threadripper 3990X54.469.8108.1OpenBenchmarking.orgWatts, Fewer Is BetterTSCP 1.81CPU Power Consumption Monitor20406080100

Meta Performance Per Watts

OpenBenchmarking.orgPerformance Per Watts, More Is BetterMeta Performance Per WattsPerformance Per WattsThreadripper 3960X20040060080010001068.78

CPU Power Consumption Monitor

OpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System MonitoringThreadripper 3960XThreadripper 3990X70140210280350Min: 21.21 / Avg: 193.06 / Max: 404.16Min: 27.63 / Avg: 202.59 / Max: 292.03

257 Results Shown

NWChem
Java Gradle Build
OpenFOAM:
  drivaerFastback, Small Mesh Size - Execution Time
  drivaerFastback, Small Mesh Size - Mesh Time
LeelaChessZero:
  Eigen
  BLAS
Blender
OpenSSL
LAMMPS Molecular Dynamics Simulator
Quantum ESPRESSO
Stockfish
High Performance Conjugate Gradient
Timed Linux Kernel Compilation
OSPRay
ClickHouse:
  100M Rows Web Analytics Dataset, Third Run
  100M Rows Web Analytics Dataset, Second Run
  100M Rows Web Analytics Dataset, First Run / Cold Cache
Ngspice
ONNX Runtime:
  fcn-resnet101-11 - CPU - Standard
  ArcFace ResNet-100 - CPU - Standard
  yolov4 - CPU - Standard
JPEG XL libjxl
SecureMark
OSPRay
WireGuard + Linux Networking Stack Stress Test
BRL-CAD
Timed LLVM Compilation
LuaRadio:
  Complex Phase
  Hilbert Transform
  FM Deemphasis Filter
  Five Back to Back FIR Filters
Graph500:
  26:
    bfs median_TEPS
    sssp max_TEPS
    bfs max_TEPS
    sssp median_TEPS
WebP2 Image Encode
Numpy Benchmark
Monte Carlo Simulations of Ionised Nebulae
Timed LLVM Compilation
TNN
OSPRay
Timed Gem5 Compilation
TensorFlow Lite
asmFish
Timed Node.js Compilation
Primesieve
Ngspice
OSPRay Studio
Stargate Digital Audio Workstation
Helsing
Stress-NG
Blender
OSPRay Studio:
  2 - 4K - 32 - Path Tracer
  3 - 1080p - 16 - Path Tracer
  3 - 4K - 1 - Path Tracer
LZ4 Compression:
  9 - Decompression Speed
  9 - Compression Speed
OSPRay Studio:
  3 - 4K - 16 - Path Tracer
  3 - 1080p - 32 - Path Tracer
  1 - 4K - 32 - Path Tracer
Zstd Compression:
  19 - Decompression Speed
  19 - Compression Speed
AOM AV1
OSPRay Studio:
  1 - 4K - 1 - Path Tracer
  2 - 1080p - 16 - Path Tracer
  2 - 4K - 1 - Path Tracer
  1 - 1080p - 16 - Path Tracer
Natron
GPAW
Apache Cassandra
OSPRay Studio
ONNX Runtime
OSPRay Studio:
  2 - 1080p - 32 - Path Tracer
  1 - 4K - 16 - Path Tracer
Blender
OSPRay Studio
OSPRay:
  gravity_spheres_volume/dim_512/scivis/real_time
  gravity_spheres_volume/dim_512/ao/real_time
OSPRay Studio:
  3 - 1080p - 1 - Path Tracer
  1 - 1080p - 1 - Path Tracer
  2 - 1080p - 1 - Path Tracer
OSPRay
WebP2 Image Encode
Appleseed
JPEG XL libjxl
ACES DGEMM
OpenSCAD
Appleseed
PyPerformance
Stargate Digital Audio Workstation
PyPerformance
Node.js V8 Web Tooling Benchmark
libgav1
Chia Blockchain VDF
NAS Parallel Benchmarks
Chaos Group V-RAY
simdjson:
  PartialTweets
  DistinctUserID
  TopTweet
GROMACS
PJSIP:
  INVITE
  OPTIONS, Stateful
LuxCoreRender:
  Orange Juice - CPU
  LuxCore Benchmark - CPU
  Danish Mood - CPU
IndigoBench:
  CPU - Bedroom
  CPU - Supercar
Build2
LuxCoreRender
TensorFlow Lite:
  Inception ResNet V2
  NASNet Mobile
  Mobilenet Float
  SqueezeNet
  Mobilenet Quant
OpenSSL:
  RSA4096:
    verify/s
    sign/s
Timed Linux Kernel Compilation
Etcpak
VP9 libvpx Encoding
Blender
simdjson
Algebraic Multi-Grid Benchmark
Chia Blockchain VDF
Xcompact3d Incompact3d
GEGL:
  Rotate 90 Degrees
  Color Enhance
Appleseed
PJSIP
RawTherapee
Timed Godot Game Engine Compilation
NAS Parallel Benchmarks
Aircrack-ng
NAMD
SVT-VP9
Blender
7-Zip Compression:
  Decompression Rating
  Compression Rating
Zstd Compression:
  19, Long Mode - Decompression Speed
  19, Long Mode - Compression Speed
OpenSCAD
Stargate Digital Audio Workstation
Tachyon
JPEG XL Decoding libjxl
OpenFOAM:
  motorBike - Execution Time
  motorBike - Mesh Time
srsRAN:
  4G PHY_DL_Test 100 PRB MIMO 256-QAM:
    UE Mb/s
    eNb Mb/s
NAS Parallel Benchmarks
QuantLib
WebP Image Encode
srsRAN:
  4G PHY_DL_Test 100 PRB MIMO 64-QAM:
    UE Mb/s
    eNb Mb/s
GIMP
srsRAN
miniFE
PyPerformance
Stargate Digital Audio Workstation:
  44100 - 512
  480000 - 1024
LULESH
Stargate Digital Audio Workstation
Stress-NG:
  CPU Stress
  Crypto
  Matrix Math
  Vector Math
Google SynthMark
Pennant
Inkscape
Pennant
Kripke
x265
Coremark
Kvazaar
libjpeg-turbo tjbench
ASTC Encoder
SVT-HEVC
CloverLeaf
GIMP
PyBench
Crafty
PyPerformance
Basis Universal
srsRAN:
  5G PHY_DL_NR Test 52 PRB SISO 64-QAM:
    UE Mb/s
    eNb Mb/s
NAS Parallel Benchmarks
Timed Mesa Compilation
PyPerformance
Timed FFmpeg Compilation
Embree
srsRAN:
  4G PHY_DL_Test 100 PRB SISO 64-QAM:
    UE Mb/s
    eNb Mb/s
XZ Compression
C-Ray
Liquid-DSP
Embree
Liquid-DSP
AOM AV1
Liquid-DSP:
  32 - 256 - 57
  16 - 256 - 57
Timed Apache Compilation
Embree
TNN
AOM AV1
WebP Image Encode
Basis Universal
TNN
AOM AV1
GIMP
srsRAN:
  4G PHY_DL_Test 100 PRB SISO 256-QAM:
    UE Mb/s
    eNb Mb/s
NAS Parallel Benchmarks
Timed MPlayer Compilation
Timed ImageMagick Compilation
Embree
NAS Parallel Benchmarks
Xcompact3d Incompact3d
POV-Ray
x265
JPEG XL libjxl
DaCapo Benchmark
JPEG XL Decoding libjxl
JPEG XL libjxl
OpenSCAD
Kvazaar
SVT-AV1
Google Draco
GIMP
GEGL
OpenSCAD
DaCapo Benchmark
SVT-VP9
DaCapo Benchmark
SVT-VP9
GEGL
Parallel BZIP2 Compression
NAS Parallel Benchmarks
WebP Image Encode
ASTC Encoder
GNU Octave Benchmark
Google Draco
SVT-AV1:
  Preset 10 - Bosphorus 4K
  Preset 12 - Bosphorus 4K
DaCapo Benchmark
WebP2 Image Encode
Primesieve
SVT-HEVC
TNN
Etcpak
NAS Parallel Benchmarks
OpenSCAD
Darktable:
  Masskrug - CPU-only
  Server Room - CPU-only
WebP2 Image Encode
Darktable
LAMMPS Molecular Dynamics Simulator
TSCP
Meta Performance Per Watts:
  Performance Per Watts
  Phoronix Test Suite System Monitoring

Threadripper 3990X

Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads), Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS), Chipset: AMD Starship/Matisse, Memory: 128GB, Disk: Samsung SSD 970 EVO Plus 500GB, Graphics: AMD Radeon RX 5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: DELL P2415Q, Network: Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 22.04, Kernel: 5.19.0-051900rc7-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.47), Vulkan: 1.2.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301039
Java Notes: OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.22.04.1)
Python Notes: Python 3.10.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 20 July 2022 18:44 by user phoronix.

Threadripper 3960X

Processor: AMD Ryzen Threadripper 3960X 24-Core @ 3.80GHz (24 Cores / 48 Threads), Motherboard: MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS), Chipset: AMD Starship/Matisse, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 1TB, Graphics: Gigabyte AMD Radeon RX 5500 XT 8GB (1900/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: VA2431, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 22.04, Kernel: 5.19.0-051900rc7-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.47), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.22.04.1)
Python Notes: Python 3.10.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 21 July 2022 20:17 by user phoronix.