7773x

tests for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2305086-NE-7773X808085
Jump To Table - Results

View

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

Limit displaying results to tests within:

C++ Boost Tests 2 Tests
Timed Code Compilation 2 Tests
C/C++ Compiler Tests 7 Tests
Compression Tests 2 Tests
CPU Massive 9 Tests
Creator Workloads 10 Tests
Encoding 3 Tests
Fortran Tests 4 Tests
Game Development 3 Tests
HPC - High Performance Computing 13 Tests
Machine Learning 5 Tests
Molecular Dynamics 6 Tests
MPI Benchmarks 4 Tests
Multi-Core 17 Tests
NVIDIA GPU Compute 4 Tests
Intel oneAPI 4 Tests
OpenMPI Tests 8 Tests
Programmer / Developer System Benchmarks 4 Tests
Python Tests 5 Tests
Scientific Computing 7 Tests
Server CPU Tests 7 Tests
Video Encoding 3 Tests

Statistics

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

Graph Settings

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

Additional Graphs

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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
a
May 02 2023
  3 Hours, 1 Minute
b
May 02 2023
  4 Hours, 10 Minutes
5 a
May 04 2023
  4 Hours, 53 Minutes
5 b
May 04 2023
  59 Minutes
5 2p a
May 04 2023
  4 Hours, 1 Minute
5 2p b
May 04 2023
  5 Hours, 22 Minutes
7373x
May 08 2023
  3 Hours, 34 Minutes
2 x AMD EPYC 7373X 16-Core
May 08 2023
  4 Hours, 5 Minutes
7373X 2P
May 08 2023
  3 Hours, 57 Minutes
Invert Hiding All Results Option
  3 Hours, 47 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


7773x ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerVulkanCompilerFile-SystemScreen Resolutionab5 a5 b5 2p a5 2p b7373x2 x AMD EPYC 7373X 16-Core7373X 2PAMD EPYC 7773X 64-Core @ 2.20GHz (64 Cores / 128 Threads)AMD DAYTONA_X (RYM1009B BIOS)AMD Starship/Matisse256GB3841GB Micron_9300_MTFDHAL3T8TDPASPEEDVE2282 x Mellanox MT27710Ubuntu 22.045.15.0-47-generic (x86_64)GNOME Shell 42.4X Server 1.21.1.31.2.204GCC 11.2.0ext41920x1080AMD EPYC 7573X 32-Core @ 2.80GHz (32 Cores / 64 Threads)2 x AMD EPYC 7573X 32-Core @ 2.80GHz (64 Cores / 128 Threads)512GBAMD EPYC 7373X 16-Core @ 3.05GHz (16 Cores / 32 Threads)256GB2 x AMD EPYC 7373X 16-Core @ 3.05GHz (32 Cores / 64 Threads)512GBOpenBenchmarking.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- Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001229Python Details- Python 3.10.6Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

ab5 a5 b5 2p a5 2p b7373x2 x AMD EPYC 7373X 16-Core7373X 2PResult OverviewPhoronix Test Suite100%174%248%322%396%ACES DGEMMOpenFOAMSPECFEM3DEmbreeJohn The RipperTimed LLVM CompilationOpenVKLTimed FFmpeg CompilationSVT-AV1VVenCFFmpegQuantLib

7773x ncnn: CPU - googlenetncnn: CPU - mobilenetncnn: CPU - resnet50ncnn: CPU - efficientnet-b0ncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU - resnet18ncnn: CPU - regnety_400mncnn: CPU - FastestDetlczero: Eigenncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - shufflenet-v2ncnn: CPU - alexnetopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timelczero: BLASopenvino: Person Vehicle Bike Detection FP16 - CPUlulesh: openvino: Vehicle Detection FP16 - CPUmt-dgemm: Sustained Floating-Point Ratespecfem3d: Mount St. Helensopenvino: Vehicle Detection FP16-INT8 - CPUjohn-the-ripper: bcryptjohn-the-ripper: Blowfishjohn-the-ripper: WPA PSKblender: Classroom - CPU-Onlyopenvino: Weld Porosity Detection FP16-INT8 - CPUspecfem3d: Layered Halfspaceblender: Barbershop - CPU-Onlyaskap: tConvolve MPI - Degriddingopenvino: Face Detection FP16-INT8 - CPUspecfem3d: Homogeneous Halfspacecompress-7zip: Decompression Ratingblender: Pabellon Barcelona - CPU-Onlyopenvino: Weld Porosity Detection FP16 - CPUblender: BMW27 - CPU-Onlyjohn-the-ripper: MD5opencv: Coreembree: Pathtracer - Crownblender: Fishy Cat - CPU-Onlyembree: Pathtracer ISPC - Crownpennant: leblancbigspecfem3d: Water-layered Halfspaceopenvino: Person Detection FP32 - CPUopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Person Detection FP16 - CPUopenvino: Face Detection FP16 - CPUaskap: tConvolve MPI - Griddingembree: Pathtracer ISPC - Asian Dragon Objopencv: Object Detectionspecfem3d: Tomographic Modelembree: Pathtracer - Asian Dragonembree: Pathtracer ISPC - Asian Dragonpennant: sedovbigembree: Pathtracer - Asian Dragon Objncnn: CPU - squeezenet_ssdonednn: Convolution Batch Shapes Auto - f32 - CPUbuild-llvm: Ninjagromacs: MPI CPU - water_GMX50_bareincompact3d: input.i3d 129 Cells Per Directioncloverleaf: Lagrangian-Eulerian Hydrodynamicsopenvino: Age Gender Recognition Retail 0013 FP16 - CPUcompress-7zip: Compression Ratingncnn: CPU - yolov4-tinyopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenfoam: drivaerFastback, Small Mesh Size - Execution Timeincompact3d: input.i3d 193 Cells Per Directiononednn: Deconvolution Batch shapes_1d - f32 - CPUpetsc: Streamsjohn-the-ripper: HMAC-SHA512onednn: IP Shapes 1D - f32 - CPUopenvkl: vklBenchmark ISPCopencv: DNN - Deep Neural Networkonednn: IP Shapes 3D - f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUopencv: Graph APIopenfoam: drivaerFastback, Medium Mesh Size - Execution Timencnn: CPU - mnasnetncnn: CPU - blazefaceaskap: Hogbom Clean OpenMPonednn: Recurrent Neural Network Inference - f32 - CPUbuild-llvm: Unix Makefilesbuild-ffmpeg: Time To Compileaskap: tConvolve OpenMP - Griddingaskap: tConvolve MT - Griddingopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUncnn: CPU - vgg16openvino: Person Detection FP32 - CPUopenvino: Person Detection FP16 - CPUopencv: Stitchingopenvino: Face Detection FP16 - CPUopenvino: Weld Porosity Detection FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUaskap: tConvolve MT - Degriddingsvt-av1: Preset 8 - Bosphorus 4Ksvt-av1: Preset 12 - Bosphorus 4Konednn: Recurrent Neural Network Training - f32 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUaskap: tConvolve OpenMP - Degriddingopenvino: Vehicle Detection FP16 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUsvt-av1: Preset 13 - Bosphorus 4Kopenvino: Person Vehicle Bike Detection FP16 - CPUvvenc: Bosphorus 1080p - Fastersvt-av1: Preset 12 - Bosphorus 1080pclickhouse: 100M Rows Hits Dataset, First Run / Cold Cachevvenc: Bosphorus 4K - Fasterffmpeg: libx265 - Liveffmpeg: libx265 - Livesvt-av1: Preset 4 - Bosphorus 4Kclickhouse: 100M Rows Hits Dataset, Second Runclickhouse: 100M Rows Hits Dataset, Third Runsvt-av1: Preset 8 - Bosphorus 1080popenfoam: drivaerFastback, Small Mesh Size - Mesh Timecompress-zstd: 8, Long Mode - Compression Speedcompress-zstd: 8 - Compression Speedcompress-zstd: 3, Long Mode - Compression Speedvvenc: Bosphorus 1080p - Fastsvt-av1: Preset 13 - Bosphorus 1080pvvenc: Bosphorus 4K - Fastncnn: CPU - vision_transformercompress-zstd: 12 - Compression Speedcompress-zstd: 19, Long Mode - Decompression Speedcompress-zstd: 8 - Decompression Speedcompress-zstd: 19 - Decompression Speedcompress-zstd: 19 - Compression Speedcompress-zstd: 3, Long Mode - Decompression Speedcompress-zstd: 8, Long Mode - Decompression Speedcompress-zstd: 19, Long Mode - Compression Speedcompress-zstd: 3 - Decompression Speedcompress-zstd: 12 - Decompression Speedsvt-av1: Preset 4 - Bosphorus 1080pquantlib: draco: Church Facadeffmpeg: libx265 - Video On Demandffmpeg: libx265 - Video On Demandcompress-zstd: 3 - Compression Speeddraco: Lionffmpeg: libx265 - Platformffmpeg: libx265 - Platformespeak: Text-To-Speech Synthesisffmpeg: libx265 - Uploadffmpeg: libx265 - Uploadz3: 1.smt2z3: 2.smt2ab5 a5 b5 2p a5 2p b7373x2 x AMD EPYC 7373X 16-Core7373X 2P25.34858329.38349611.678414616862308646020138872.4530.518912661260.8317.20934908288.7627.61553400069.107734.7763.371128.08154095064.511713.86220230078.475173.844670.5238165.7077.29040.45056813679600047040.481652246.95716.92873.442222.837195.92430.006604.84211.367105.7847.744.863109.98424.72995716.552552.2416.28610.2952746.55793173.4743.67486743.73173.2321.59116.9319.0917.8419.5815.4411.310.8953.5219.08514210.2915.67.13120.3933755541888.222257.1371205.1129.0456911.753321461889.05912768736020295772.022674.2930.115899704259.8732799.527.0316.81708024641121388.731198.4527.655430007706134.6163.36475.10407827.4675402738.04138.848.0512.1636827.564.64093191013.53017939473.8289.5636119.320.827622164.4497.2994.4380412111.3435737.8239692823.9238091.4640.24442817.21184736.9214556185.75041351650001.26107469397561.246183.12487235810363.735911.718.12806.452749.622248.48117.04220481.25493.351.621.7424.0439273911.972011602608.3426.68230.278308.3367.548214.9331183.6323.921175.0119018.326.5416.93198.13516.9330.114601.984428.5111.364106.9847.2067748894.826439.07437.35109.71125.596085744.5988.5752.616.374549.8346.128133.39270.81190.813901276.817.61300.81434.29.31277.31445.910.2342692.45833172.53393950643.902819488543.90172.5695669428.15721.62116.8091581414.1113.714.89.036.468.1226.349.2512386.197.735.48117.437814191252.0520920.487841.4917.874518.862199071244.866019260325130848113.311767.1349.136149469408.7920991.717.9224.987445867244220136.58822.1942.5236210006834343.949252.6840.24087.97402245.709427135.4792.615.488.2324990.139.19412773919.08966485748.522844.911814.0011143.958914.851.14656230.3785.0225.0675640112.0925209.6227191819.9827340.9650.74590419.63050084.4956531992.6068960590001.37996340399970.625932.76234206970428.65926.083.82869.565679.645290.53920.37817750.46599.681.161.2621.082886.122881.041840321935.619.45172.539611.0565.337220.8131182.6118.1886.4119018.31912.84199.6912.7731.837629.3437.2011.961110.4945.714.743457.81456.30107.38222.337396792.11067.8854.216.846571.0846.406126.26289.11229.11438.81304.218.21361.51460.19.541326.51478.110.762756.95718161.5446.892856.6476046.70162.1927.82622.95110.01117.604518.640319.355553092601526034513107248.73867645425.089622797360500043.978740.119944.33457433539.223619.63009634248.507445.105444.1884231.09150.69975294924000339427.71276297.22320.6470.856226.733196.06331.795622.73212.02110.2945.794.76108.36422.70639917.253560.5656.3910.8282820.422.93110.1493.2977.7780.8135.1227.738.94100.6540.45828621.2633.2718.699.38414582842523.5742030.0381632.8131.3445159.6094538682504.3611850211857925941358.553501.6925.221006745213.641983.334.8314.49141946945399671.641594.1422.59684500081.450827.9673.33014.29473123.93606599710.64175.4810.5415.8649980.164.409111.22476781376.806773.96127.65636968.692237.170.674336138.7178.2852.4865429415.6739455.8640373938.133.00107910.624130210.5221790020002.678014520.9118061.85167205.4639238.4914.65436.6811212.84219.40315.45512102.57738.591.3733.552946.972975.152013.5220.06182.112506.762.836169.1121390.3618.26914.241664119.5812.76169.65812.6624.492540.729445.189.57492.5254.584.784463.68461.53107.122.563946702.81024808.514.812525.1545.548144.31283.61222.91437.51315.6181332.31454.79.571306.71482.710.412760.95682168.45694118644.972783.7473444.31170.94271146327.86822.38112.819760537112.43115.68120.2760.7938.6751.8128.3553.25781628.8837.8928.5599.9326879912524.0542551.8981623.8331.7608769.594769352498.7411754211777125845758.573504.1525.472190291213.4641983.334.9114.41593115844074071.731592.7122.51668300023644581.091327.973.26024.41801223.26803264110.64176.7610.6315.7651199.164.04418855311.79752558576.628773.77797.79332568.702235.810.669349138.5658.2222.4838690815.5138800.3439684445.8940352.4832.97431610.762441610.368574130.1789732600002.88451452871330.9172261.89859419702203.6511849.4624.56434.7831305.96221.78915.34412678.97176.411.331.432.812952.762948.392874922020.2120.07180.9111068.869.536176.1071385.3318.25912.4215662.119.6912.79172.57312.6624.268565.057440.569.906107.2947.074.842467.46448.03104.16923.335275702.21013.4741.515.044546.0345.768145.49282.81227.11441.11312181336.51445.19.51309.51498.910.4952850.85729173.05225728343.772775.9481843.64173.5600295828.13722.10114.26177556211.7612.2313.256.974.927.0517.67.3517144.55.864.76149.003721955598.410188.248400.078.02433637.795953129652.03315733165170803212.97964.1991.272692415768.94116629.7151.264616343128235252.02454.978.75195900023.58995.9621.385114.5090978.6002645433.1652.663.174.741543520.104535.50584661924.907323.568423.8794723.358712.711.93416391.7922.9856.8646898318.2614994.7615587918.2316278.2578.74933425.21746066.82625596800001.874462130.6097463.77803596.969164.522.4729.927906.609433.97129.97913312.88783.70.971.0620.472496.522493.981684.8617.58151.7310352.649.204152.1941760.4916.58821.361479219.9912.26147.35413.3631.274487.076351.0111.224114.6344.063.993386.60384.9692.98426.386727727913.5865.415.801506.5175.719140.06305.11312.91523.91389.619.214131541.710.11281.11557.210.0232854.15406162.3346.662661456346.67162.29602734526.50722.83110.60550232727.6367.1525.3846.6728.8513.6286.2221.36145024.1931.889.78114.4461216761181.7527032.903786.1714.48164718.4066297691274.046236062244136260108.31914.2248.717394203393.8721866.319.3123.273738774232623127.96892.6740.6137350008562144.879749.2740.36047.46537343.7801354736.198.396.178.8629155.135.24083185018.03326655742.831240.506213.2864838.777922.581.03456211.4895.2483.932473930.6729028.2524180331.9831155.5976.08765213.7996965.78798672130002.05818309656570.7508532.70057263067334.8952437.6218.4458.7161177.81277.06119.5611576.39627.341.011.0932.772595.652547.352329771782.0617.91162.511891.458.355163.3281377.3916.7826.7713312.820.3312.54166.53913.5224.314525.637420.2910.127111.0745.4673073194.644438.36429.56100.58724.265467680.91022.3854.914.787510.7875.877145.33288.91289.91523.61375.818.61415.41559.810.11384.1156310.1952894.25498165.87146396645.672842.3464244.56169.98578945326.46722.21113.67282212621.21870.28645.4179.8230.7146.6933.1615.1484.8839.48135729.5231.6510.29116.27444170927107.14314.70415317.8203903956255162302139127108.0649.831882811396.821420.123.898547572236136127.540.4737390009873844.21649.4439.99927.41973943.85974441929155.134.85663218718.33177041242.7440.181213.3102638.425623.891.04348212.525.1934.7539529828.5924386233.7474.59384813.7177015.87951685300002.17496306823600.9481432.73357363745332.0343238.9818.75458.7161349.92271.82919.69712102.59605.6329.8222812312025.758.697162.7841370.6113312.8159.26124.216522.16420.9410.08399.8250.5908738714.676433.56436.29101.82124.672394671.6992.4813.614.936490.9865.774144.552951292.11518.11394.318.91414.11538.410.11362.4155210.4342907.35485170.67963610544.382801.4461445.72165.70013276126.51422.35112.96105218521.71770.51OpenBenchmarking.org

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: googlenet2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb30609012015027.6393.29112.4314.1145.4111.7619.09MIN: 23.37 / MAX: 37.04MIN: 52.46 / MAX: 137.19MIN: 29.99 / MAX: 148.59MIN: 28.42 / MAX: 186.92MIN: 18.77 / MAX: 25.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: mobilenet2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb30609012015067.1577.77115.6813.7079.8212.2317.84MIN: 21.26 / MAX: 78.89MIN: 67.11 / MAX: 156MIN: 64.45 / MAX: 159.93MIN: 65.63 / MAX: 95.23MIN: 17.55 / MAX: 25.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: resnet502 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb30609012015025.3880.81120.2714.8030.7113.2519.58MIN: 22.8 / MAX: 34.84MIN: 62.51 / MAX: 112.38MIN: 42.99 / MAX: 192.41MIN: 27.2 / MAX: 36.97MIN: 19.15 / MAX: 44.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: efficientnet-b02 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb142842567046.6735.1260.799.0346.696.9715.44MIN: 36.84 / MAX: 52.67MIN: 33.83 / MAX: 41.96MIN: 47.67 / MAX: 141.85MIN: 8.93 / MAX: 11.07MIN: 36.92 / MAX: 81.01MIN: 6.9 / MAX: 7.4MIN: 13.63 / MAX: 18.671. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU-v2-v2 - Model: mobilenet-v22 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb91827364528.8527.7038.676.4633.164.9211.30MIN: 18.28 / MAX: 40.64MIN: 23.18 / MAX: 43.54MIN: 28.02 / MAX: 119.64MIN: 6.35 / MAX: 8.73MIN: 24.4 / MAX: 44MIN: 4.84 / MAX: 5.41MIN: 9.74 / MAX: 14.961. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: resnet182 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb122436486013.6238.9451.808.1215.147.0510.89MIN: 13.43 / MAX: 16.6MIN: 16.05 / MAX: 124.98MIN: 16.09 / MAX: 93.56MIN: 8.01 / MAX: 10.04MIN: 13.62 / MAX: 18.75MIN: 6.95 / MAX: 7.65MIN: 10.68 / MAX: 11.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: regnety_400m2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb30609012015086.22100.65128.3526.3484.8817.6053.52MIN: 82.8 / MAX: 122.51MIN: 97.81 / MAX: 136.12MIN: 111.43 / MAX: 240.27MIN: 25.99 / MAX: 28.26MIN: 80.66 / MAX: 90.33MIN: 17.32 / MAX: 18.74MIN: 50.93 / MAX: 71.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: FastestDet2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb122436486021.3640.4553.259.2539.487.3519.08MIN: 19.71 / MAX: 29.28MIN: 27.4 / MAX: 462.07MIN: 30.07 / MAX: 66.77MIN: 9.12 / MAX: 9.81MIN: 21.82 / MAX: 52.04MIN: 7.23 / MAX: 7.88MIN: 13.65 / MAX: 21.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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: Eigen2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb2K4K6K8K10K14508286781612381357171451421. (CXX) g++ options: -flto -pthread

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU-v3-v3 - Model: mobilenet-v32 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb71421283524.1921.2628.886.1929.524.5010.29MIN: 19.28 / MAX: 34.5MIN: 20.79 / MAX: 28.25MIN: 23.66 / MAX: 168.43MIN: 6.05 / MAX: 6.93MIN: 25.64 / MAX: 34.13MIN: 4.43 / MAX: 5.26MIN: 9.51 / MAX: 11.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: shufflenet-v22 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb91827364531.8833.2737.897.7331.655.8615.60MIN: 27.33 / MAX: 35.67MIN: 29.37 / MAX: 96.69MIN: 34.63 / MAX: 113.32MIN: 7.58 / MAX: 9.71MIN: 28.59 / MAX: 37.25MIN: 5.76 / MAX: 6.62MIN: 12.95 / MAX: 19.731. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: alexnet2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb7142128359.7818.6028.555.4810.294.767.13MIN: 9.12 / MAX: 10.95MIN: 17.59 / MAX: 34.64MIN: 12.6 / MAX: 62.18MIN: 5.36 / MAX: 6.34MIN: 9.46 / MAX: 14.07MIN: 4.68 / MAX: 5.71MIN: 6.97 / MAX: 7.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 10Input: drivaerFastback, Medium Mesh Size - Mesh Time2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab306090120150114.4599.3899.93117.44117.60116.27149.0025.35120.391. (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: BLAS2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb2K4K6K8K10K16768284799114191709195555541. (CXX) g++ options: -flto -pthread

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb50010001500200025001181.752523.572524.051252.05598.401888.201. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

LULESH

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

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.32 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb9K18K27K36K45K27032.9042030.0442551.9020920.4927107.1410188.2522257.141. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb400800120016002000786.171632.811623.83841.49400.071205.111. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

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 Rate2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab714212835SE +/- 0.247911, N = 1514.48164731.34451531.76087617.87450018.64030014.7041538.02433629.38349629.0456901. (CC) gcc options: -O3 -march=native -fopenmp

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Mount St. Helens2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab918273645SE +/- 0.038048290, N = 318.4066297699.6094538689.59476935018.86219907019.35555309217.82039039537.79595312911.67841461611.7533214601. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb50010001500200025001274.042504.362498.741244.86652.031889.051. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcrypt2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab30K60K90K120K150KSE +/- 99.80, N = 3623601185021175426019260152625513157386230912761. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: Blowfish2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab30K60K90K120K150KSE +/- 185.66, N = 3622441185791177716032560345623023165186460873601. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: WPA PSK2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab60K120K180K240K300KSE +/- 297.59, N = 3136260259413258457130848131072139127708032013882029571. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: Classroom - Compute: CPU-Only2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab50100150200250SE +/- 0.17, N = 3108.3058.5558.57113.31108.06212.9772.4572.02

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb80016002400320040001914.223501.693504.151767.13964.192674.291. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Layered Halfspace2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab20406080100SE +/- 0.29, N = 348.7225.2225.4749.1448.7449.8391.2730.5230.121. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: Barbershop - Compute: CPU-Only2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab170340510680850SE +/- 0.09, N = 3393.87213.60213.46408.79396.80768.94260.83259.87

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - Degridding2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb9K18K27K36K45K21866.341983.341983.320991.721420.111662.032799.51. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb81624324019.3134.8334.9117.929.7127.031. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Homogeneous Halfspace2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab1224364860SE +/- 0.20, N = 323.2714.4914.4224.9925.0923.9051.2617.2116.821. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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 22.01Test: Decompression Rating2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb100K200K300K400K500K2326234539964407402442202361361282354112131. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: Pabellon Barcelona - Compute: CPU-Only2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab60120180240300SE +/- 0.10, N = 3127.9671.6471.73136.58127.50252.0288.7688.73

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb30060090012001500892.671594.141592.71822.19454.901198.451. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: BMW27 - Compute: CPU-Only2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab20406080100SE +/- 0.03, N = 340.6122.5922.5142.5240.4778.7527.6127.60

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: MD52 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab1.5M3M4.5M6M7.5MSE +/- 1000.00, N = 33735000684500066830003621000360500037390001959000553400055430001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Core2 x AMD EPYC 7373X 16-Core5 2p b5 a7373X 2Pb50K100K150K200K250K856212364456834398738770611. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer - Model: Crown2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xa20406080100SE +/- 0.05, N = 344.8881.4581.0943.9543.9844.2223.5969.11MIN: 44.32 / MAX: 45.69MIN: 80.49 / MAX: 83.06MIN: 80.03 / MAX: 82.56MIN: 43.49 / MAX: 44.42MIN: 43.29 / MAX: 44.67MIN: 43.54 / MAX: 44.81MIN: 23.47 / MAX: 23.76MIN: 68.18 / MAX: 71.75

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: Fishy Cat - Compute: CPU-Only2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab20406080100SE +/- 0.04, N = 349.2727.9627.9052.6849.4495.9634.7734.61

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer ISPC - Model: Crown2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab1632486480SE +/- 0.11, N = 340.3673.3373.2640.2440.1240.0021.3963.3763.36MIN: 39.78 / MAX: 41.03MIN: 71.99 / MAX: 75.06MIN: 72.13 / MAX: 74.82MIN: 39.64 / MAX: 40.95MIN: 39.64 / MAX: 40.56MIN: 39.56 / MAX: 40.59MIN: 21.25 / MAX: 21.62MIN: 62.18 / MAX: 66.58MIN: 62.44 / MAX: 66.52

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: leblancbig2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb481216207.4653734.2947314.4180127.9740227.41973914.5090905.1040781. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Water-layered Halfspace2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab20406080100SE +/- 0.13, N = 343.7823.9423.2745.7144.3343.8678.6028.0827.471. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb36912156.1010.6410.645.473.168.041. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Machine Translation EN To DE FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb408012016020098.39175.48176.7692.6152.66138.841. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb36912156.1710.5410.635.483.178.051. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb481216208.8615.8615.768.234.7412.161. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - Gridding2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb11K22K33K44K55K29155.149980.151199.124990.129155.115435.036827.51. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer ISPC - Model: Asian Dragon Obj2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab1428425670SE +/- 0.03, N = 335.2464.4164.0439.1939.2234.8620.1064.5164.64MIN: 35.01 / MAX: 35.91MIN: 63.78 / MAX: 65.49MIN: 63.29 / MAX: 64.92MIN: 38.97 / MAX: 39.59MIN: 39 / MAX: 39.63MIN: 34.63 / MAX: 35.47MIN: 20.01 / MAX: 21.52MIN: 63.96 / MAX: 66.18MIN: 64.13 / MAX: 65.46

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Object Detection2 x AMD EPYC 7373X 16-Core5 2p b5 a7373X 2Pb20K40K60K80K100K31850885532773932187319101. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Tomographic Model2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab816243240SE +/- 0.15, N = 318.0311.2211.8019.0919.6318.3335.5113.8613.531. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer - Model: Asian Dragon2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xa20406080100SE +/- 0.14, N = 342.8376.8176.6348.5248.5142.7424.9178.48MIN: 42.56 / MAX: 43.84MIN: 76.07 / MAX: 77.73MIN: 75.97 / MAX: 78.17MIN: 48.32 / MAX: 48.91MIN: 48.29 / MAX: 48.79MIN: 42.44 / MAX: 44.03MIN: 24.47 / MAX: 26.52MIN: 77.64 / MAX: 80.46

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer ISPC - Model: Asian Dragon2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab1632486480SE +/- 0.01, N = 340.5173.9673.7844.9145.1140.1823.5773.8473.83MIN: 40.25 / MAX: 41.33MIN: 73.12 / MAX: 75.27MIN: 72.91 / MAX: 74.82MIN: 44.68 / MAX: 45.26MIN: 44.88 / MAX: 45.48MIN: 39.87 / MAX: 41.09MIN: 23.24 / MAX: 24.98MIN: 73.31 / MAX: 75.9MIN: 73.34 / MAX: 74.66

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: sedovbig2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb61218243013.2864807.6563697.79332514.00111013.31026023.8794709.5636101. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer - Model: Asian Dragon Obj2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xa1632486480SE +/- 0.08, N = 338.7868.6968.7043.9644.1938.4323.3670.52MIN: 38.54 / MAX: 39.74MIN: 68.03 / MAX: 70.15MIN: 67.96 / MAX: 69.64MIN: 43.73 / MAX: 44.3MIN: 43.94 / MAX: 44.51MIN: 38.17 / MAX: 39.69MIN: 22.36 / MAX: 24.19MIN: 69.67 / MAX: 71.94

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: squeezenet_ssd2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb91827364522.5837.1735.8114.8523.8912.7119.32MIN: 21.5 / MAX: 36.98MIN: 30.62 / MAX: 51.54MIN: 29.14 / MAX: 58.45MIN: 14.49 / MAX: 25.32MIN: 22.35 / MAX: 46.47MIN: 12.37 / MAX: 13.22MIN: 18.86 / MAX: 22.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb0.43520.87041.30561.74082.1761.0345600.6743360.6693491.1465601.0434801.9341600.827622MIN: 0.98MIN: 0.62MIN: 0.6MIN: 1.09MIN: 0.98MIN: 1.89MIN: 0.781. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Ninja2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab80160240320400SE +/- 0.08, N = 3211.49138.72138.57230.38231.09212.52391.79165.71164.45

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 2023Implementation: MPI CPU - Input: water_GMX50_bare2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab246810SE +/- 0.026, N = 35.2488.2858.2225.0225.1932.9857.2907.2991. (CXX) g++ options: -O3

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 Direction2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb2468103.932473902.486542942.483869085.067564014.753952986.864689834.438041211. (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

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 Hydrodynamics2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb71421283530.6715.6715.5112.0928.5918.2611.341. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb8K16K24K32K40K29028.2539455.8638800.3425209.6214994.7635737.821. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

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 22.01Test: Compression Rating2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb90K180K270K360K450K2418034037393968442719182438621558793969281. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: yolov4-tiny2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb102030405031.9838.1045.8919.9833.7418.2323.92MIN: 26.72 / MAX: 48.39MIN: 29.08 / MAX: 101.85MIN: 34.08 / MAX: 64.66MIN: 19.53 / MAX: 23.19MIN: 28.02 / MAX: 43.18MIN: 17.97 / MAX: 23.77MIN: 23.17 / MAX: 30.481. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p b5 a7373xb9K18K27K36K45K31155.5940352.4827340.9616278.2538091.461. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

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 10Input: drivaerFastback, Small Mesh Size - Execution Time2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab2040608010076.0933.0032.9750.7550.7074.5978.7540.4540.241. (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

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 Direction2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb61218243013.8010.6210.7619.6313.7225.2217.211. (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

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb36912155.7879810.5221010.368504.495655.879516.826256.92145MIN: 4.87MIN: 8.88MIN: 8.54MIN: 3.9MIN: 4.97MIN: 5.9MIN: 6.321. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

PETSc

PETSc, the Portable, Extensible Toolkit for Scientific Computation, is for the scalable (parallel) solution of scientific applications modeled by partial differential equations. This test profile runs the PETSc "make streams" benchmark and records the throughput rate when all available cores are utilized for the MPI Streams build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterPETSc 3.19Test: Streams5 2p b5 ab16K32K48K64K80K74130.1831992.6156185.751. (CC) gcc options: -fPIC -O3 -O2 -lpthread -ludev -lpciaccess -lm

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: HMAC-SHA5122 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab30M60M90M120M150MSE +/- 266743.32, N = 3672130007900200073260000960590009492400068530000596800001367960001351650001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb0.6491.2981.9472.5963.2452.058182.678012.884511.379962.174961.874461.26107MIN: 1.58MIN: 1.89MIN: 1.79MIN: 1.25MIN: 1.61MIN: 1.72MIN: 1.061. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

OpenVKL

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

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 1.3.1Benchmark: vklBenchmark ISPC2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab100200300400500SE +/- 0.33, N = 3309452452340339306213470469MIN: 54 / MAX: 1845MIN: 98 / MAX: 1875MIN: 99 / MAX: 2013MIN: 55 / MAX: 2309MIN: 54 / MAX: 2307MIN: 54 / MAX: 1694MIN: 29 / MAX: 1843MIN: 84 / MAX: 2616MIN: 84 / MAX: 2565

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: DNN - Deep Neural Network2 x AMD EPYC 7373X 16-Core5 2p b5 a7373X 2Pb20K40K60K80K100K65657871333999782360397561. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb0.28040.56080.84121.12161.4020.7508530.9118060.9172260.6259300.9481430.6097461.246180MIN: 0.57MIN: 0.78MIN: 0.77MIN: 0.57MIN: 0.62MIN: 0.57MIN: 1.131. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb0.85011.70022.55033.40044.25052.700571.851671.898592.762342.733573.778033.12487MIN: 2.52MIN: 1.61MIN: 1.58MIN: 2.59MIN: 2.6MIN: 3.69MIN: 2.061. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Graph API2 x AMD EPYC 7373X 16-Core5 2p b5 a7373X 2Pb90K180K270K360K450K2630674197022069703637452358101. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

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 10Input: drivaerFastback, Medium Mesh Size - Execution Time2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab130260390520650334.90205.46203.65428.66427.71332.03596.9740.48363.741. (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

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: mnasnet2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb112233445537.6238.4949.466.0838.984.5211.71MIN: 29.31 / MAX: 44.7MIN: 25.51 / MAX: 75.24MIN: 36.42 / MAX: 176.46MIN: 6.01 / MAX: 6.54MIN: 32.96 / MAX: 44.92MIN: 9.43 / MAX: 20.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: blazeface2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb61218243018.4014.6524.563.8218.752.408.12MIN: 16.72 / MAX: 22.27MIN: 11.34 / MAX: 73.36MIN: 20.64 / MAX: 91.2MIN: 3.45 / MAX: 79.74MIN: 15.74 / MAX: 22.59MIN: 6.94 / MAX: 11.11. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Second, More Is BetterASKAP 1.0Test: Hogbom Clean OpenMP2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb2004006008001000458.72436.68434.78869.57458.72729.93806.451. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb300600900120015001177.811212.841305.96679.651349.92906.61749.62MIN: 1030.68MIN: 1070.52MIN: 1060.61MIN: 664.64MIN: 1150.78MIN: 898.73MIN: 731.751. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Unix Makefiles2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab90180270360450SE +/- 0.96, N = 3277.06219.40221.79290.54297.22271.83433.97246.96248.48

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 6.0Time To Compile2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab714212835SE +/- 0.03, N = 319.5615.4615.3420.3820.6419.7029.9816.9317.04

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - Gridding2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb4K8K12K16K20K11576.312102.512678.917750.412102.513312.820481.21. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - Gridding2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb2K4K6K8K10K9627.347738.597176.416599.689605.638783.705493.351. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p b5 a7373xb0.36450.7291.09351.4581.82251.011.331.160.971.62MIN: 0.6 / MAX: 20.54MIN: 0.64 / MAX: 26.34MIN: 0.66 / MAX: 12.23MIN: 0.58 / MAX: 12.03MIN: 0.69 / MAX: 13.341. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb0.39150.7831.17451.5661.95751.091.371.401.261.061.74MIN: 0.64 / MAX: 18.03MIN: 0.67 / MAX: 28.86MIN: 0.68 / MAX: 42.1MIN: 0.69 / MAX: 12.96MIN: 0.62 / MAX: 11.46MIN: 0.84 / MAX: 14.571. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: vgg162 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb81624324032.7733.5532.8121.0829.8220.4724.04MIN: 29.33 / MAX: 38.66MIN: 28.65 / MAX: 42.53MIN: 30.06 / MAX: 44.76MIN: 20.78 / MAX: 24.35MIN: 27.79 / MAX: 34.74MIN: 20.22 / MAX: 21.34MIN: 23.49 / MAX: 30.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb80016002400320040002595.652946.972952.762886.122496.523927.00MIN: 2193.47 / MAX: 2987.27MIN: 2004.15 / MAX: 3534.38MIN: 2193.59 / MAX: 3652.32MIN: 1694.68 / MAX: 3104.67MIN: 1381.65 / MAX: 2622.64MIN: 3402.58 / MAX: 4474.471. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb80016002400320040002547.352975.152948.392881.042493.983911.97MIN: 1516.1 / MAX: 2961.68MIN: 2241.37 / MAX: 3616.82MIN: 1547.54 / MAX: 3537.3MIN: 1536.21 / MAX: 3142.35MIN: 2275.06 / MAX: 2587.16MIN: 3337.4 / MAX: 4451.631. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Stitching2 x AMD EPYC 7373X 16-Core5 2p b5 a7373X 2Pb60K120K180K240K300K2329772874921840322281232011601. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb60012001800240030001782.062013.522020.211935.601684.862608.34MIN: 1650.91 / MAX: 2051.03MIN: 1890.96 / MAX: 2802.82MIN: 1823.96 / MAX: 3111.51MIN: 1852.05 / MAX: 1974.19MIN: 1636.65 / MAX: 1721.46MIN: 2421.38 / MAX: 2754.891. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb61218243017.9120.0620.0719.4517.5826.68MIN: 9.87 / MAX: 67.23MIN: 11.41 / MAX: 47.48MIN: 13.26 / MAX: 75.83MIN: 11.98 / MAX: 28.4MIN: 15.76 / MAX: 27.27MIN: 15.32 / MAX: 46.541. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Machine Translation EN To DE FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb50100150200250162.50182.10180.91172.53151.73230.27MIN: 133.61 / MAX: 274.15MIN: 117.14 / MAX: 548.18MIN: 124.49 / MAX: 288.01MIN: 81.35 / MAX: 207.95MIN: 131.11 / MAX: 163.85MIN: 166.99 / MAX: 311.891. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - Degridding2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb3K6K9K12K15K11891.4012506.7011068.809611.0512025.7010352.608308.331. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 8 - Input: Bosphorus 4K2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab1632486480SE +/- 0.56, N = 1258.3662.8469.5465.3470.8658.7049.2073.4467.551. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 12 - Input: Bosphorus 4K2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab50100150200250SE +/- 0.73, N = 3163.33169.11176.11220.81226.73162.78152.19222.84214.931. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb4008001200160020001377.391390.361385.331182.611370.611760.491183.63MIN: 1316MIN: 1287.43MIN: 1314.76MIN: 1161.83MIN: 1312.47MIN: 1746.27MIN: 1162.731. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb61218243016.7018.2618.2518.1016.5823.92MIN: 9.01 / MAX: 41.25MIN: 10.53 / MAX: 60.19MIN: 8.84 / MAX: 40.73MIN: 9.23 / MAX: 28.11MIN: 12.62 / MAX: 25.96MIN: 15.02 / MAX: 35.741. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb30060090012001500826.77914.24912.42886.41821.361175.01MIN: 796.91 / MAX: 888.72MIN: 797.83 / MAX: 966.84MIN: 878.63 / MAX: 988.74MIN: 851.09 / MAX: 900.44MIN: 798.88 / MAX: 834.55MIN: 982.69 / MAX: 1202.211. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - Degridding2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb4K8K12K16K20K13312.816641.015662.119018.313312.814792.019018.31. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb61218243020.3319.5819.6919.0019.9926.54MIN: 14.4 / MAX: 46.38MIN: 11.25 / MAX: 73.87MIN: 11.03 / MAX: 75.83MIN: 13.62 / MAX: 32.75MIN: 12.45 / MAX: 25.77MIN: 14.19 / MAX: 63.191. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb4812162012.5412.7612.7912.8412.2616.93MIN: 9.03 / MAX: 28.23MIN: 7.65 / MAX: 43.58MIN: 7.68 / MAX: 43.18MIN: 6.96 / MAX: 23.28MIN: 7.79 / MAX: 21.63MIN: 10.73 / MAX: 31.241. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 13 - Input: Bosphorus 4K2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab4080120160200SE +/- 0.70, N = 3166.54169.66172.57199.69196.06159.26147.35195.92198.141. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPU2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373xb4812162013.5212.6612.6612.7713.3616.93MIN: 10.48 / MAX: 44.97MIN: 7.58 / MAX: 53.45MIN: 8.38 / MAX: 48.88MIN: 7.39 / MAX: 23.62MIN: 8.79 / MAX: 23.44MIN: 13.88 / MAX: 33.111. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.8Video Input: Bosphorus 1080p - Video Preset: Faster2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab714212835SE +/- 0.03, N = 324.3124.4924.2731.8431.8024.2231.2730.0130.111. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 12 - Input: Bosphorus 1080p2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab140280420560700SE +/- 2.97, N = 3525.64540.73565.06629.30622.73522.16487.08604.84601.981. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold Cache2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb100200300400500420.29445.18440.56437.20420.94351.01428.51MIN: 24.7 / MAX: 5454.55MIN: 41.18 / MAX: 3157.89MIN: 40.98 / MAX: 4000MIN: 24.3 / MAX: 6000MIN: 24.79 / MAX: 3750MIN: 13.78 / MAX: 6666.67MIN: 34.8 / MAX: 5454.55

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.8Video Input: Bosphorus 4K - Video Preset: Faster2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab3691215SE +/- 0.022, N = 310.1279.5749.90611.96112.02010.08311.22411.36711.3641. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Live2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab306090120150SE +/- 0.29, N = 3111.0792.52107.29110.49110.2999.82114.63105.78106.981. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Live2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab1224364860SE +/- 0.13, N = 345.4754.5847.0745.7145.7950.5944.0647.7447.211. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 4 - Input: Bosphorus 4K2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab1.09422.18843.28264.37685.471SE +/- 0.021, N = 34.6444.7844.8424.7434.7604.6763.9934.8634.8261. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second Run2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb100200300400500438.36463.68467.46457.81433.56386.60439.07MIN: 25.08 / MAX: 4285.71MIN: 41.64 / MAX: 4615.38MIN: 40.6 / MAX: 4615.38MIN: 24.13 / MAX: 5454.55MIN: 24.58 / MAX: 4285.71MIN: 13.7 / MAX: 6666.67MIN: 35.82 / MAX: 6000

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third Run2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb100200300400500429.56461.53448.03456.30436.29384.96437.35MIN: 25.01 / MAX: 4615.38MIN: 41.49 / MAX: 3000MIN: 41.38 / MAX: 2608.7MIN: 24.65 / MAX: 5454.55MIN: 24.91 / MAX: 3750MIN: 13.8 / MAX: 6666.67MIN: 35.59 / MAX: 5454.55

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 8 - Input: Bosphorus 1080p2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab20406080100SE +/- 1.23, N = 5100.59107.10104.17107.38108.36101.8292.98109.98109.711. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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 10Input: drivaerFastback, Small Mesh Size - Mesh Time2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab61218243024.2722.5623.3422.3422.7124.6726.3924.7325.601. (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

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8, Long Mode - Compression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb2004006008001000680.9702.8702.2792.1671.6727.0744.51. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8 - Compression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb20040060080010001022.31024.01013.41067.8992.4913.5988.51. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3, Long Mode - Compression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb2004006008001000854.9808.5741.5854.2813.6865.4752.61. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.8Video Input: Bosphorus 1080p - Video Preset: Fast2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab48121620SE +/- 0.10, N = 314.7914.8115.0416.8517.2514.9415.8016.5516.371. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 13 - Input: Bosphorus 1080p2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab120240360480600SE +/- 5.77, N = 3510.79525.15546.03571.08560.57490.99506.52552.24549.831. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.8Video Input: Bosphorus 4K - Video Preset: Fast2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab246810SE +/- 0.010, N = 35.8775.5485.7686.4066.3905.7745.7196.2866.1281. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: vision_transformer2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb306090120150145.33144.31145.49126.26144.55140.06133.39MIN: 142.5 / MAX: 157.84MIN: 140.5 / MAX: 157.63MIN: 141.27 / MAX: 245.34MIN: 125.42 / MAX: 132.03MIN: 140.93 / MAX: 165.57MIN: 137.05 / MAX: 145.39MIN: 129.7 / MAX: 252.771. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 12 - Compression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb70140210280350288.9283.6282.8289.1295.0305.1270.81. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Decompression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb300600900120015001289.91222.91227.11229.11292.11312.91190.81. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8 - Decompression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb300600900120015001523.61437.51441.11438.81518.11523.91390.01. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Decompression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb300600900120015001375.81315.61312.01304.21394.31389.61276.81. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Compression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb51015202518.618.018.018.218.919.217.61. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3, Long Mode - Decompression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb300600900120015001415.41332.31336.51361.51414.11413.01300.81. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8, Long Mode - Decompression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb300600900120015001559.81454.71445.11460.11538.41541.71434.21. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Compression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb369121510.109.579.509.5410.1010.109.301. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3 - Decompression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb300600900120015001384.11306.71309.51326.51362.41281.11277.31. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 12 - Decompression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb300600900120015001563.01482.71498.91478.11552.01557.21445.91. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 4 - Input: Bosphorus 1080p2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab3691215SE +/- 0.02, N = 310.2010.4110.5010.7610.8310.4310.0210.3010.231. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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 1.302 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab6001200180024003000SE +/- 25.52, N = 32894.22760.92850.82756.92820.42907.32854.12746.52692.41. (CXX) g++ options: -O3 -march=native -fPIE -pie

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 1.5.6Model: Church Facade2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab13002600390052006500SE +/- 3.21, N = 3549856825729571854855406579358331. (CXX) g++ options: -O3

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On Demand2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab4080120160200SE +/- 0.15, N = 3165.87168.46173.05161.54170.68162.33173.47172.531. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On Demand2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab1122334455SE +/- 0.04, N = 345.6744.9743.7746.8944.3846.6643.6743.901. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3 - Compression Speed2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb60012001800240030002842.32783.72775.92856.62801.42661.02819.01. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

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 1.5.6Model: Lion2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab10002000300040005000SE +/- 7.02, N = 3464247344818476046144563486748851. (CXX) g++ options: -O3

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Platform2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab1122334455SE +/- 0.02, N = 344.5644.3143.6446.7045.7246.6743.7343.901. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Platform2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xab4080120160200SE +/- 0.08, N = 3169.99170.94173.56162.19165.70162.30173.23172.571. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

eSpeak-NG Speech Engine

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

OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech Synthesis2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a7373X 2P7373xb71421283526.4727.8728.1427.8326.5126.5128.161. (CC) gcc options: -O2 -std=c99

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Upload2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab510152025SE +/- 0.01, N = 322.2122.3822.1022.9522.9322.3522.8321.5921.621. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Upload2 x AMD EPYC 7373X 16-Core5 2p a5 2p b5 a5 b7373X 2P7373xab306090120150SE +/- 0.05, N = 3113.67112.82114.26110.01110.14112.96110.61116.93116.811. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Z3 Theorem Prover

The Z3 Theorem Prover / SMT solver is developed by Microsoft Research under the MIT license. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterZ3 Theorem Prover 4.12.1SMT File: 1.smt22 x AMD EPYC 7373X 16-Core7373X 2P51015202521.2221.721. (CXX) g++ options: -lpthread -std=c++17 -fvisibility=hidden -mfpmath=sse -msse -msse2 -O3 -fPIC

OpenBenchmarking.orgSeconds, Fewer Is BetterZ3 Theorem Prover 4.12.1SMT File: 2.smt22 x AMD EPYC 7373X 16-Core7373X 2P163248648070.2970.511. (CXX) g++ options: -lpthread -std=c++17 -fvisibility=hidden -mfpmath=sse -msse -msse2 -O3 -fPIC

145 Results Shown

NCNN:
  CPU - googlenet
  CPU - mobilenet
  CPU - resnet50
  CPU - efficientnet-b0
  CPU-v2-v2 - mobilenet-v2
  CPU - resnet18
  CPU - regnety_400m
  CPU - FastestDet
LeelaChessZero
NCNN:
  CPU-v3-v3 - mobilenet-v3
  CPU - shufflenet-v2
  CPU - alexnet
OpenFOAM
LeelaChessZero
OpenVINO
LULESH
OpenVINO
ACES DGEMM
SPECFEM3D
OpenVINO
John The Ripper:
  bcrypt
  Blowfish
  WPA PSK
Blender
OpenVINO
SPECFEM3D
Blender
ASKAP
OpenVINO
SPECFEM3D
7-Zip Compression
Blender
OpenVINO
Blender
John The Ripper
OpenCV
Embree
Blender
Embree
Pennant
SPECFEM3D
OpenVINO:
  Person Detection FP32 - CPU
  Machine Translation EN To DE FP16 - CPU
  Person Detection FP16 - CPU
  Face Detection FP16 - CPU
ASKAP
Embree
OpenCV
SPECFEM3D
Embree:
  Pathtracer - Asian Dragon
  Pathtracer ISPC - Asian Dragon
Pennant
Embree
NCNN
oneDNN
Timed LLVM Compilation
GROMACS
Xcompact3d Incompact3d
CloverLeaf
OpenVINO
7-Zip Compression
NCNN
OpenVINO
OpenFOAM
Xcompact3d Incompact3d
oneDNN
PETSc
John The Ripper
oneDNN
OpenVKL
OpenCV
oneDNN:
  IP Shapes 3D - f32 - CPU
  Deconvolution Batch shapes_3d - f32 - CPU
OpenCV
OpenFOAM
NCNN:
  CPU - mnasnet
  CPU - blazeface
ASKAP
oneDNN
Timed LLVM Compilation
Timed FFmpeg Compilation
ASKAP:
  tConvolve OpenMP - Gridding
  tConvolve MT - Gridding
OpenVINO:
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU
  Age Gender Recognition Retail 0013 FP16 - CPU
NCNN
OpenVINO:
  Person Detection FP32 - CPU
  Person Detection FP16 - CPU
OpenCV
OpenVINO:
  Face Detection FP16 - CPU
  Weld Porosity Detection FP16 - CPU
  Machine Translation EN To DE FP16 - CPU
ASKAP
SVT-AV1:
  Preset 8 - Bosphorus 4K
  Preset 12 - Bosphorus 4K
oneDNN
OpenVINO:
  Weld Porosity Detection FP16-INT8 - CPU
  Face Detection FP16-INT8 - CPU
ASKAP
OpenVINO:
  Vehicle Detection FP16 - CPU
  Vehicle Detection FP16-INT8 - CPU
SVT-AV1
OpenVINO
VVenC
SVT-AV1
ClickHouse
VVenC
FFmpeg:
  libx265 - Live:
    FPS
    Seconds
SVT-AV1
ClickHouse:
  100M Rows Hits Dataset, Second Run
  100M Rows Hits Dataset, Third Run
SVT-AV1
OpenFOAM
Zstd Compression:
  8, Long Mode - Compression Speed
  8 - Compression Speed
  3, Long Mode - Compression Speed
VVenC
SVT-AV1
VVenC
NCNN
Zstd Compression:
  12 - Compression Speed
  19, Long Mode - Decompression Speed
  8 - Decompression Speed
  19 - Decompression Speed
  19 - Compression Speed
  3, Long Mode - Decompression Speed
  8, Long Mode - Decompression Speed
  19, Long Mode - Compression Speed
  3 - Decompression Speed
  12 - Decompression Speed
SVT-AV1
QuantLib
Google Draco
FFmpeg:
  libx265 - Video On Demand:
    Seconds
    FPS
Zstd Compression
Google Draco
FFmpeg:
  libx265 - Platform:
    FPS
    Seconds
eSpeak-NG Speech Engine
FFmpeg:
  libx265 - Upload:
    FPS
    Seconds
Z3 Theorem Prover:
  1.smt2
  2.smt2