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 2207215-NE-TR3990X5306
Jump To Table - Results

Statistics

Remove Outliers Before Calculating Averages

Graph Settings

Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

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

Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Threadripper 3990X
July 20 2022
  21 Hours, 43 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


AMD Ryzen Threadripper 3990XOpenBenchmarking.orgPhoronix Test SuiteAMD 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 11.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionAMD Ryzen Threadripper 3990X BenchmarksSystem Logs- Transparent Huge Pages: madvise- --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 - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301039- OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.22.04.1)- Python 3.10.4- 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 3990Xtscp: AI Chess Performancecompress-lz4: 9 - Compression Speedcompress-lz4: 9 - Decompression Speedcrafty: Elapsed Timecompress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9java-gradle-perf: Reactorstress-ng: Cryptostress-ng: CPU Cachestress-ng: CPU Stressstress-ng: Matrix Mathstress-ng: Vector Mathdacapobench: H2dacapobench: Jythondacapobench: Tradesoapdacapobench: Tradebeansopenscad: Pistolopenscad: Retro Caropenscad: Mini-ITX Caseopenscad: Projector Mount Swivelopenscad: Leonardo Phone Case Slimbasis: UASTC Level 2basis: UASTC Level 3darktable: Boat - CPU-onlydarktable: Masskrug - CPU-onlydarktable: Server Room - CPU-onlytjbench: Decompression Throughputbrl-cad: VGR Performance Metricngspice: C2670ngspice: C7552stargate: 44100 - 512stargate: 96000 - 512stargate: 44100 - 1024stargate: 480000 - 512stargate: 96000 - 1024stargate: 480000 - 1024astcenc: Thoroughastcenc: Exhaustiveetcpak: Multi-Threaded - ETC2etcpak: Single-Threaded - ETC2draco: Liondraco: Church Facadejpegxl-decode: 1jpegxl-decode: Allhelsing: 14 digitjpegxl: PNG - 7jpegxl: PNG - 8jpegxl: JPEG - 7jpegxl: JPEG - 8rawtherapee: Total Benchmark Timewebp2: Defaultwebp2: Quality 75, Compression Effort 7webp2: Quality 95, Compression Effort 7webp2: Quality 100, Compression Effort 5webp: Quality 100, Losslesswebp: Quality 100, Highest Compressionwebp: Quality 100, Lossless, Highest Compressiongegl: Cropgegl: Scalegegl: Color Enhancegegl: Rotate 90 Degreesgimp: resizegimp: rotategimp: auto-levelsgimp: unsharp-maskinkscape: SVG Files To PNGsynthmark: VoiceMark_100securemark: SecureMark-TLSchia-vdf: Square Plain C++chia-vdf: Square Assembly Optimizedquantlib: graph500: 26graph500: 26graph500: 26graph500: 26lczero: BLASlczero: Eigennumpy: cloverleaf: Lagrangian-Eulerian Hydrodynamicsoctave-benchmark: onnx: GPT-2 - CPU - Standardonnx: yolov4 - CPU - Standardonnx: fcn-resnet101-11 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardtensorflow-lite: SqueezeNettensorflow-lite: Inception V4tensorflow-lite: NASNet Mobiletensorflow-lite: Mobilenet Floattensorflow-lite: Mobilenet Quanttensorflow-lite: Inception ResNet V2tnn: CPU - DenseNettnn: CPU - MobileNet v2tnn: CPU - SqueezeNet v2tnn: CPU - SqueezeNet v1.1gromacs: MPI CPU - water_GMX50_barelammps: 20k Atomslammps: Rhodopsin Proteinhpcg: npb: BT.Cnpb: CG.Cnpb: EP.Cnpb: FT.Cnpb: LU.Cnpb: MG.Cnpb: SP.Bnpb: SP.Cnamd: ATPase Simulation - 327,506 Atomsmt-dgemm: Sustained Floating-Point Ratepennant: sedovbigpennant: leblancbigamg: kripke: lulesh: nwchem: C240 Buckyballopenfoam: motorBike - Mesh Timeopenfoam: motorBike - Execution Timeopenfoam: drivaerFastback, Small Mesh Size - Mesh Timeopenfoam: drivaerFastback, Small Mesh Size - Execution Timemocassin: Dust 2D tau100.0minife: Smallincompact3d: input.i3d 129 Cells Per Directionincompact3d: input.i3d 193 Cells Per Directiongpaw: Carbon Nanotubeqe: AUSURF112wireguard: coremark: CoreMark Size 666 - Iterations Per Secondaircrack-ng: build-ffmpeg: Time To Compilebuild-imagemagick: Time To Compilebuild-mplayer: Time To Compileprimesieve: 1e12primesieve: 1e13build-apache: Time To Compilestockfish: Total Timecompress-7zip: Compression Ratingcompress-7zip: Decompression Ratingbuild-llvm: Ninjabuild-llvm: Unix Makefilescompress-zstd: 19 - Compression Speedcompress-zstd: 19 - Decompression Speedcompress-zstd: 19, Long Mode - Compression Speedcompress-zstd: 19, Long Mode - Decompression Speedcompress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compressionasmfish: 1024 Hash Memory, 26 Depthbuild-linux-kernel: defconfigbuild-linux-kernel: allmodconfigkvazaar: Bosphorus 4K - Very Fastkvazaar: Bosphorus 4K - Ultra Fastaom-av1: Speed 6 Realtime - Bosphorus 4Kaom-av1: Speed 8 Realtime - Bosphorus 4Kaom-av1: Speed 9 Realtime - Bosphorus 4Kaom-av1: Speed 10 Realtime - Bosphorus 4Klibgav1: Summer Nature 4Kvpxenc: Speed 5 - Bosphorus 4Ktachyon: Total Timesvt-vp9: VMAF Optimized - Bosphorus 4Ksvt-vp9: PSNR/SSIM Optimized - Bosphorus 4Ksvt-vp9: Visual Quality Optimized - Bosphorus 4Ksvt-av1: Preset 8 - Bosphorus 4Ksvt-av1: Preset 10 - Bosphorus 4Ksvt-av1: Preset 12 - Bosphorus 4Kx265: Bosphorus 4Kx265: Bosphorus 1080pc-ray: Total Time - 4K, 16 Rays Per Pixelv-ray: CPUsvt-hevc: 7 - Bosphorus 4Ksvt-hevc: 10 - Bosphorus 4Kblender: BMW27 - CPU-Onlyblender: Classroom - CPU-Onlyblender: Fishy Cat - CPU-Onlyblender: Barbershop - CPU-Onlyblender: Pabellon Barcelona - CPU-Onlypovray: Trace Timebuild-godot: Time To Compileembree: Pathtracer - Crownembree: Pathtracer ISPC - Crownembree: Pathtracer - Asian Dragonembree: Pathtracer ISPC - Asian Dragonnatron: Spaceshipindigobench: CPU - Bedroomindigobench: CPU - Supercarluxcorerender: DLSC - CPUluxcorerender: Danish Mood - CPUluxcorerender: Orange Juice - CPUluxcorerender: LuxCore Benchmark - CPUospray: particle_volume/ao/real_timeospray: particle_volume/scivis/real_timeospray: particle_volume/pathtracer/real_timeospray: gravity_spheres_volume/dim_512/ao/real_timeospray: gravity_spheres_volume/dim_512/scivis/real_timeospray: gravity_spheres_volume/dim_512/pathtracer/real_timeospray-studio: 1 - 4K - 1 - Path Tracerospray-studio: 2 - 4K - 1 - Path Tracerospray-studio: 3 - 4K - 1 - Path Tracerospray-studio: 1 - 4K - 16 - Path Tracerospray-studio: 1 - 4K - 32 - Path Tracerospray-studio: 2 - 4K - 16 - Path Tracerospray-studio: 2 - 4K - 32 - Path Tracerospray-studio: 3 - 4K - 16 - Path Tracerospray-studio: 3 - 4K - 32 - Path Tracerospray-studio: 1 - 1080p - 1 - Path Tracerospray-studio: 2 - 1080p - 1 - Path Tracerospray-studio: 3 - 1080p - 1 - Path Tracerospray-studio: 1 - 1080p - 16 - Path Tracerospray-studio: 1 - 1080p - 32 - Path Tracerospray-studio: 2 - 1080p - 16 - Path Tracerospray-studio: 2 - 1080p - 32 - Path Tracerospray-studio: 3 - 1080p - 16 - Path Tracerospray-studio: 3 - 1080p - 32 - Path Tracerappleseed: Emilyappleseed: Disney Materialappleseed: Material Testerbuild2: Time To Compilebuild-gem5: Time To Compilebuild-mesa: Time To Compilebuild-nodejs: Time To Compilepyperformance: raytracepyperformance: json_loadspyperformance: crypto_pyaespyperformance: regex_compilepyperformance: python_startupliquid-dsp: 16 - 256 - 57liquid-dsp: 32 - 256 - 57liquid-dsp: 64 - 256 - 57liquid-dsp: 128 - 256 - 57luaradio: Five Back to Back FIR Filtersluaradio: FM Deemphasis Filterluaradio: Hilbert Transformluaradio: Complex Phasesrsran: OFDM_Testsrsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMsrsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMsrsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMsrsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMsrsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMsrsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMsrsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMsrsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMsrsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMsrsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMopenssl: SHA256openssl: RSA4096openssl: RSA4096clickhouse: 100M Rows Web Analytics Dataset, First Run / Cold Cacheclickhouse: 100M Rows Web Analytics Dataset, Second Runclickhouse: 100M Rows Web Analytics Dataset, Third Runcassandra: Writesnode-web-tooling: simdjson: Kostyasimdjson: TopTweetsimdjson: PartialTweetssimdjson: DistinctUserIDpybench: Total For Average Test Timespjsip: INVITEpjsip: OPTIONS, Statefulpjsip: OPTIONS, StatelessThreadripper 3990X132505355.3211528.5836917420.822297.70792388.85150.17123897.45219165.21340961.12453243183800532298.2294.43043.6888.16217.29810.94415.5111.6142.3552.137215.566786714610143.477109.1265.3469823.7464605.7888165.3752344.2700855.6817476.088413.91164515.860225.8505616886169.37539.71100.60510.890.9891.9528.2550.8282.433108.798233.6133.06616.5057.20534.1168.48112.58952.90355.77924.18116.69219.82924.82730.539741.3262188511633671070002790.453556900054378000016124700021568600016591564416.7514.747.220698030215410216181.9056545.051346.94229.253296.1085767.42776.396288.35168.971244.4043.78824.39323.1479.0694668363.638366.744690.7328515.3765067.7526818.0148358.8423124.730.4530917.35457112.315345.5785004556222001059932009413.40572598.839.8772.86174.271369.22379889.269.3721057945.3925336107.872447.69273.8582248635.666496162110.44318.01113.13610.6093.73852.27620.465151075832192724353695161.816212.28081.23371.244.53376.82.11413648753328.878262.43431.1760.6618.7431.8545.1444.0841.7010.1428.4920102.98108.86103.6254.428109.598133.03620.9946.9611.04146971119.00151.1830.1784.3540.06342.54110.3310.05345.10254.314549.256844.250442.28793.09.83121.0817.805.9712.006.5913.412413.1957145.8826.970786.5220510.82202688273631934318391586439869334051334108130673689804109022161911086219751295825752138.13589662.592187153.51178759.300179.02620.351148.32544023.611415714.0980473333178460000028128333333011300000762.4415.7104.3635.8139566667356.7142.8368.7159.7392.7157.0392.9166.8103.758.95923302074813533.3854108.5250.20255.76260.332400229.542.684.293.784.2010173711664640872OpenBenchmarking.org

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 3990X300K600K900K1200K1500KSE +/- 904.73, N = 1213250531. (CC) gcc options: -O3 -march=native

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 - Compression SpeedThreadripper 3990X1224364860SE +/- 0.38, N = 1255.321. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedThreadripper 3990X2K4K6K8K10KSE +/- 20.11, N = 1211528.51. (CC) gcc options: -O3

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 3990X2M4M6M8M10MSE +/- 12538.70, N = 383691741. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm

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 3990X510152025SE +/- 0.05, N = 320.821. (CC) gcc options: -fvisibility=hidden -O2

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 3990X60120180240300SE +/- 5.65, N = 9297.71

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: CryptoThreadripper 3990X20K40K60K80K100KSE +/- 733.42, N = 392388.851. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CPU CacheThreadripper 3990X306090120150SE +/- 2.21, N = 15150.171. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CPU StressThreadripper 3990X30K60K90K120K150KSE +/- 1038.88, N = 3123897.451. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Matrix MathThreadripper 3990X50K100K150K200K250KSE +/- 728.11, N = 3219165.211. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Vector MathThreadripper 3990X70K140K210K280K350KSE +/- 3412.43, N = 3340961.121. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

DaCapo Benchmark

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

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2Threadripper 3990X10002000300040005000SE +/- 25.60, N = 54532

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonThreadripper 3990X9001800270036004500SE +/- 28.07, N = 74318

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradesoapThreadripper 3990X8001600240032004000SE +/- 38.92, N = 53800

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansThreadripper 3990X11002200330044005500SE +/- 44.91, N = 55322

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 3990X20406080100SE +/- 0.34, N = 398.231. OpenSCAD version 2021.01

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Retro CarThreadripper 3990X0.99681.99362.99043.98724.984SE +/- 0.006, N = 84.4301. OpenSCAD version 2021.01

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Mini-ITX CaseThreadripper 3990X1020304050SE +/- 0.19, N = 343.691. OpenSCAD version 2021.01

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Projector Mount SwivelThreadripper 3990X246810SE +/- 0.027, N = 68.1621. OpenSCAD version 2021.01

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenSCADRender: Leonardo Phone Case SlimThreadripper 3990X48121620SE +/- 0.10, N = 317.301. OpenSCAD version 2021.01

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 3990X3691215SE +/- 0.00, N = 510.941. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.13Settings: UASTC Level 3Threadripper 3990X48121620SE +/- 0.04, N = 415.511. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

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 3990X0.36320.72641.08961.45281.816SE +/- 0.006, N = 101.614

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Masskrug - Acceleration: CPU-onlyThreadripper 3990X0.52991.05981.58972.11962.6495SE +/- 0.004, N = 92.355

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Server Room - Acceleration: CPU-onlyThreadripper 3990X0.48080.96161.44241.92322.404SE +/- 0.003, N = 92.137

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 3990X50100150200250SE +/- 0.46, N = 3215.571. (CC) gcc options: -O3 -rdynamic

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 3990X150K300K450K600K750K7146101. (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

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 3990X306090120150SE +/- 1.51, N = 4143.481. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552Threadripper 3990X20406080100SE +/- 0.71, N = 3109.131. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

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 3990X1.20312.40623.60934.81246.0155SE +/- 0.034562, N = 35.3469821. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 96000 - Buffer Size: 512Threadripper 3990X0.8431.6862.5293.3724.215SE +/- 0.082618, N = 153.7464601. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 44100 - Buffer Size: 1024Threadripper 3990X1.30252.6053.90755.216.5125SE +/- 0.007232, N = 35.7888161. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 480000 - Buffer Size: 512Threadripper 3990X1.20942.41883.62824.83766.047SE +/- 0.003417, N = 35.3752341. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 96000 - Buffer Size: 1024Threadripper 3990X0.96081.92162.88243.84324.804SE +/- 0.049478, N = 34.2700851. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 21.10.9Sample Rate: 480000 - Buffer Size: 1024Threadripper 3990X1.27842.55683.83525.11366.392SE +/- 0.015679, N = 35.6817471. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

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 3990X246810SE +/- 0.0086, N = 76.08841. (CXX) g++ options: -O3 -flto -pthread

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: ExhaustiveThreadripper 3990X48121620SE +/- 0.10, N = 313.911. (CXX) g++ options: -O3 -flto -pthread

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 3990X10002000300040005000SE +/- 32.34, N = 84515.861. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 1.0Benchmark: Single-Threaded - Configuration: ETC2Threadripper 3990X50100150200250SE +/- 0.02, N = 3225.851. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

Google Draco

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

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.0Model: LionThreadripper 3990X12002400360048006000SE +/- 5.51, N = 656161. (CXX) g++ options: -O3

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.0Model: Church FacadeThreadripper 3990X2K4K6K8K10KSE +/- 9.97, N = 588611. (CXX) g++ options: -O3

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 3990X1530456075SE +/- 0.07, N = 369.37

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.6.1CPU Threads: AllThreadripper 3990X120240360480600SE +/- 1.46, N = 3539.71

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 3990X20406080100SE +/- 1.36, N = 3100.611. (CC) gcc options: -O2 -pthread

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 3990X3691215SE +/- 0.03, N = 310.891. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: PNG - Encode Speed: 8Threadripper 3990X0.22050.4410.66150.8821.1025SE +/- 0.00, N = 30.981. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: JPEG - Encode Speed: 7Threadripper 3990X20406080100SE +/- 0.13, N = 491.951. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.6.1Input: JPEG - Encode Speed: 8Threadripper 3990X714212835SE +/- 0.05, N = 528.251. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie

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 3990X1122334455SE +/- 0.14, N = 350.831. RawTherapee, version 5.8, command line.

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 3990X0.54741.09481.64222.18962.737SE +/- 0.029, N = 152.4331. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 75, Compression Effort 7Threadripper 3990X20406080100SE +/- 0.83, N = 3108.801. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 95, Compression Effort 7Threadripper 3990X50100150200250SE +/- 0.48, N = 3233.611. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 100, Compression Effort 5Threadripper 3990X0.68991.37982.06972.75963.4495SE +/- 0.004, N = 93.0661. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

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 3990X48121620SE +/- 0.17, N = 316.511. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16 -ltiff

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest CompressionThreadripper 3990X246810SE +/- 0.012, N = 67.2051. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16 -ltiff

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest CompressionThreadripper 3990X816243240SE +/- 0.14, N = 334.121. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16 -ltiff

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 3990X246810SE +/- 0.022, N = 58.481

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: ScaleThreadripper 3990X3691215SE +/- 0.04, N = 412.59

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Color EnhanceThreadripper 3990X1224364860SE +/- 0.03, N = 352.90

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Rotate 90 DegreesThreadripper 3990X1326395265SE +/- 0.46, N = 355.78

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 3990X612182430SE +/- 0.29, N = 424.18

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.30Test: rotateThreadripper 3990X48121620SE +/- 0.11, N = 316.69

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.30Test: auto-levelsThreadripper 3990X510152025SE +/- 0.04, N = 319.83

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.30Test: unsharp-maskThreadripper 3990X612182430SE +/- 0.10, N = 324.83

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 3990X714212835SE +/- 0.03, N = 330.541. Inkscape 1.1.2 (0a00cf5339, 2022-02-04)

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 3990X160320480640800SE +/- 0.84, N = 3741.331. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

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 3990X50K100K150K200K250KSE +/- 671.44, N = 32188511. (CC) gcc options: -pedantic -O3

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 3990X30K60K90K120K150KSE +/- 405.52, N = 31633671. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread

OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.7Test: Square Assembly OptimizedThreadripper 3990X20K40K60K80K100KSE +/- 529.15, N = 31070001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread

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 3990X6001200180024003000SE +/- 30.41, N = 42790.41. (CXX) g++ options: -O3 -march=native -rdynamic

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

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

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

OpenBenchmarking.orgsssp max_TEPS, More Is BetterGraph500 3.0Scale: 26Threadripper 3990X50M100M150M200M250M2156860001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

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: BLASThreadripper 3990X400800120016002000SE +/- 6.17, N = 316591. (CXX) g++ options: -flto -pthread

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: EigenThreadripper 3990X30060090012001500SE +/- 19.50, N = 415641. (CXX) g++ options: -flto -pthread

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 3990X90180270360450SE +/- 1.26, N = 3416.75

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 3990X48121620SE +/- 0.03, N = 414.741. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

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 3990X246810SE +/- 0.017, N = 67.220

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 3990X15003000450060007500SE +/- 31.14, N = 369801. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: StandardThreadripper 3990X70140210280350SE +/- 0.93, N = 33021. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: StandardThreadripper 3990X306090120150SE +/- 1.09, N = 31541. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardThreadripper 3990X2004006008001000SE +/- 2.29, N = 310211. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

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: SqueezeNetThreadripper 3990X13002600390052006500SE +/- 36.36, N = 36181.90

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception V4Threadripper 3990X12K24K36K48K60KSE +/- 547.67, N = 1556545.0

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: NASNet MobileThreadripper 3990X11K22K33K44K55KSE +/- 311.74, N = 351346.9

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet FloatThreadripper 3990X9001800270036004500SE +/- 19.00, N = 34229.25

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet QuantThreadripper 3990X7001400210028003500SE +/- 35.20, N = 33296.10

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception ResNet V2Threadripper 3990X20K40K60K80K100KSE +/- 1222.16, N = 385767.4

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 3990X6001200180024003000SE +/- 6.92, N = 32776.40MIN: 2575.3 / MAX: 3223.811. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v2Threadripper 3990X60120180240300SE +/- 2.20, N = 3288.35MIN: 263.94 / MAX: 386.221. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v2Threadripper 3990X1530456075SE +/- 0.19, N = 768.97MIN: 66.69 / MAX: 72.461. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.1Threadripper 3990X50100150200250SE +/- 0.30, N = 3244.40MIN: 243.4 / MAX: 245.651. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

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 3990X0.85231.70462.55693.40924.2615SE +/- 0.009, N = 33.7881. (CXX) g++ options: -O3

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 3990X612182430SE +/- 0.12, N = 324.391. (CXX) g++ options: -O3 -lm -ldl

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: Rhodopsin ProteinThreadripper 3990X612182430SE +/- 0.17, N = 1523.151. (CXX) g++ options: -O3 -lm -ldl

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 3990X3691215SE +/- 0.00092, N = 39.069461. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

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 3990X15K30K45K60K75KSE +/- 10.73, N = 368363.631. (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, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CThreadripper 3990X2K4K6K8K10KSE +/- 5.53, N = 38366.741. (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, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CThreadripper 3990X10002000300040005000SE +/- 32.72, N = 154690.731. (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, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CThreadripper 3990X6K12K18K24K30KSE +/- 24.04, N = 428515.371. (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, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CThreadripper 3990X14K28K42K56K70KSE +/- 37.32, N = 365067.751. (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, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CThreadripper 3990X6K12K18K24K30KSE +/- 38.09, N = 626818.011. (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, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.BThreadripper 3990X10K20K30K40K50KSE +/- 111.88, N = 648358.841. (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