AMD 3D V-Cache Comparison

Tests for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2204297-NE-CC929132156&grs&sro.

AMD 3D V-Cache ComparisonProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionDisplay DriverRyzen 9 5950XRyzen 7 5800X3DRyzen 7 5800XRyzen 9 5900XCore i9 12900KAMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VIII HERO (WI-FI) (4006 BIOS)AMD Starship/Matisse32GB1000GB Sabrent Rocket 4.0 1TBAMD Radeon RX 6800 16GB (2475/1000MHz)AMD Navi 21 HDMI AudioASUS MG28URealtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 22.045.17.4-051704-generic (x86_64)GNOME Shell 42.0X Server + Wayland4.6 Mesa 22.2.0-devel (git-092ac67 2022-04-21 jammy-oibaf-ppa) (LLVM 14.0.0 DRM 3.44)1.3.211GCC 11.2.0ext43840x2160AMD Ryzen 7 5800X3D 8-Core @ 3.40GHz (8 Cores / 16 Threads)ASRock X570 Pro4 (P4.30 BIOS)16GBAMD Radeon RX 6800 XT 16GB (2575/1000MHz)ASUS VP28UIntel I211AMD Ryzen 7 5800X 8-Core @ 3.80GHz (8 Cores / 16 Threads)AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads)ASUS ROG CROSSHAIR VIII HERO (3904 BIOS)NVIDIA NV134 8GBNVIDIA GP104 HD AudioASUS MG28URealtek RTL8125 2.5GbE + Intel I211nouveau4.3 Mesa 22.2.0-devel (git-092ac67 2022-04-21 jammy-oibaf-ppa)Intel Core i9-12900K @ 5.20GHz (16 Cores / 24 Threads)ASUS ROG STRIX Z690-E GAMING WIFI (1003 BIOS)Intel Device 7aa732GBAMD Radeon RX 6800 XT 16GB (2575/1000MHz)Intel Device 7ad0ASUS VP28UIntel I225-V + Intel Wi-Fi 6 AX210/AX211/AX4114.6 Mesa 22.2.0-devel (git-092ac67 2022-04-21 jammy-oibaf-ppa) (LLVM 14.0.0 DRM 3.44)OpenBenchmarking.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- Ryzen 9 5950X: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201016- Ryzen 7 5800X3D: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201205- Ryzen 7 5800X: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201016- Ryzen 9 5900X: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201016- Core i9 12900K: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x18 - Thermald 2.4.9Python Details- Python 3.10.4Security Details- Ryzen 9 5950X: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected- Ryzen 7 5800X3D: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected- Ryzen 7 5800X: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected- Ryzen 9 5900X: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected- Core i9 12900K: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

AMD 3D V-Cache Comparisononnx: ArcFace ResNet-100 - CPU - Parallelaskap: tConvolve OpenMP - Griddingecp-candle: P3B2askap: tConvolve MT - Griddinglczero: BLASlczero: Eigenonednn: IP Shapes 3D - u8s8f32 - CPUecp-candle: P3B1onednn: Convolution Batch Shapes Auto - u8s8f32 - CPUaskap: tConvolve MT - Degriddingopenfoam: Motorbike 60Mincompact3d: input.i3d 193 Cells Per Directionaskap: tConvolve OpenMP - Degriddingonednn: IP Shapes 3D - f32 - CPUincompact3d: input.i3d 129 Cells Per Directionaskap: Hogbom Clean OpenMPmnn: mobilenet-v1-1.0onnx: yolov4 - CPU - Parallelopenfoam: Motorbike 30Mmnn: mobilenetV3askap: tConvolve MPI - Griddingwebp2: Quality 100, Lossless Compressionmlpack: scikit_qdawebp2: Quality 100, Compression Effort 5tnn: CPU - SqueezeNet v1.1webp2: Quality 75, Compression Effort 7webp2: Quality 95, Compression Effort 7askap: tConvolve MPI - Degriddingmlpack: scikit_svmonnx: bertsquad-12 - CPU - Parallelonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUmnn: MobileNetV2_224onednn: Deconvolution Batch shapes_3d - f32 - CPUwebp2: Defaultonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUtnn: CPU - DenseNetecp-candle: P1B2mnn: squeezenetv1.1onnx: fcn-resnet101-11 - CPU - Paralleltnn: CPU - SqueezeNet v2onnx: GPT-2 - CPU - Standardtnn: CPU - MobileNet v2onnx: super-resolution-10 - CPU - Parallelncnn: CPU - alexnetonednn: IP Shapes 1D - f32 - CPUonnx: GPT-2 - CPU - Parallelncnn: CPU - resnet50caffe: GoogleNet - CPU - 200opm-git: Flow MPI Norne - 8opm-git: Flow MPI Norne-4C MSW - 8caffe: AlexNet - CPU - 100caffe: GoogleNet - CPU - 100caffe: AlexNet - CPU - 200numpy: ncnn: CPU - squeezenet_ssdonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUopm-git: Flow MPI Extra - 8opm-git: Flow MPI Extra - 4opm-git: Flow MPI Norne-4C MSW - 4mlpack: scikit_icaopm-git: Flow MPI Norne-4C MSW - 2opm-git: Flow MPI Norne-4C MSW - 1opm-git: Flow MPI Norne - 1opm-git: Flow MPI Norne - 4opm-git: Flow MPI Norne - 2opm-git: Flow MPI Extra - 2opm-git: Flow MPI Extra - 1onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUmnn: inception-v3onnx: ArcFace ResNet-100 - CPU - Standardonnx: bertsquad-12 - CPU - Standardonnx: super-resolution-10 - CPU - Standardonnx: fcn-resnet101-11 - CPU - Standardonnx: yolov4 - CPU - Standardmlpack: scikit_linearridgeregressionmnn: SqueezeNetV1.0mnn: resnet-v2-50ncnn: CPU - regnety_400mncnn: CPU - yolov4-tinyncnn: CPU - resnet18ncnn: CPU - vgg16ncnn: CPU - googlenetncnn: CPU - blazefacencnn: CPU - efficientnet-b0ncnn: CPU - mnasnetncnn: CPU - shufflenet-v2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU - mobilenetonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPURyzen 9 5950XRyzen 7 5800X3DRyzen 7 5800XRyzen 9 5900XCore i9 12900K15402755.72654.353784.5076686840.4760971309.59418.03211346.571382.54156.1625263214.587.7965233.9481214220.0362.63929698.442.3916728.09536.76755.063.249213.323110.284234.9276643.6416.565541.069263.2753.626092.1651.618990.8230092425.59431.3923.9888850.9407062224.264643611.083.89647553424.40193710357.86575.13366589662473352591.5414.551814.131783.851820.40808.05690.62377.0341.77441.07569.14270.81232.61203.89712.621074.332750.182703.332745.4425.72616688006169984871.695.23120.5159.6120.4914.2856.5512.921.805.233.874.163.774.2912.274.7553316.705112236946.11553.745930.574125411600.6047801023.31510.38001594.951090.21126.8610338741.66.5576527.1814976527.7231.67630880.291.0828746.52864.93538.275.169222.332178.013376.9566453.2216.175181.776411.8315.571093.1702.483931.237852609.84329.9922.5907253.2218826233.17946069.622.89717691918.13162022361.91581.46302228086260494603.0912.441382.161387.861385.79781.98643.96357.3334.88403.35476.91224.10223.97187.49672.48997.342691.242683.162691.6723.193193982241071075721.634.21316.3565.1814.8010.2742.607.311.063.012.012.111.902.137.627.2762312.622711071709.25390.269854.7078678541.457521158.67416.35341471.581270.72141.2109543367.678.9355237.2105344221.7321.816273177.601.1564256.05974.59152.575.875266.187199.645420.0323976.3020.294802.036001.9426.370233.6192.839131.407283016.47934.8412.8056763.6056832272.037410311.833.29136562120.22179146447.63733.91339058982567661491.5616.761859.141864.541847.701027.17829.49454.3439.16511.31601.31281.47275.40234.63799.401154.093065.263053.303055.8425.69210425573628494311.824.53618.4425.9319.7813.0255.9710.221.223.602.242.352.282.6111.638.3420318.753514212936.85667.282837.9449548860.5197331174.46916.73821541.671277.55144.4737503732.727.5484332.1536293238.6674.07229996.161.8388258.02614.81251.393.709213.776129.733268.6127668.0516.155401.300492.9754.387732.4391.997421.0069412563.75730.2413.3948550.8527862224.199559110.013.44548568721.21178725361.84581.35343298949268676605.1713.511785.051742.061776.19810.33691.38378.3339.37441.14572.60273.89233.63203.98713.341073.712904.332873.452908.2823.692151994975451155391.584.75824.4998.4320.6312.5050.7511.441.634.773.443.883.463.9111.425.3868916.13983624631.821503.2642721.15220121610.812047429.616.002143872.03487.8055.60967137793.773.4464314.5530450540.5412.89162984.471.1744472.73476.45127.312.936133.970101.568215.5824198.5110.549191.352122.415.253942.0622.223361.052001792.97220.9412.40511138.91511082170.58345167.582.63611808816.84134045515.71825.56255906783351713665.3413.271613.811616.061617.10891.78671.56451.0032.93475.95537.38247.89280.85223.20717.041053.382881.322881.722881.1024.301197598847471016931.604.15123.0857.3915.869.6628.249.941.465.343.103.102.903.4111.118.737505.87536OpenBenchmarking.org

ONNX Runtime

Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: ParallelCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X30060090012001500SE +/- 0.17, N = 3SE +/- 4.25, N = 3SE +/- 3.51, N = 3SE +/- 1.17, N = 3SE +/- 8.09, N = 336211071223142115401. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ASKAP

Test: tConvolve OpenMP - Gridding

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - GriddingCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X15003000450060007500SE +/- 34.51, N = 6SE +/- 14.53, N = 15SE +/- 78.12, N = 15SE +/- 10.74, N = 6SE +/- 27.21, N = 64631.821709.256946.112936.852755.721. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

ECP-CANDLE

Benchmark: P3B2

OpenBenchmarking.orgSeconds, Fewer Is BetterECP-CANDLE 0.4Benchmark: P3B2Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X300600900120015001503.26390.27553.75667.28654.35

ASKAP

Test: tConvolve MT - Gridding

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - GriddingCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X6001200180024003000SE +/- 2.70, N = 3SE +/- 2.02, N = 3SE +/- 2.64, N = 3SE +/- 0.90, N = 3SE +/- 1.65, N = 32721.15854.71930.57837.94784.511. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

LeelaChessZero

Backend: BLAS

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: BLASCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X5001000150020002500SE +/- 22.18, N = 3SE +/- 6.36, N = 3SE +/- 6.08, N = 3SE +/- 11.02, N = 3SE +/- 2.73, N = 3220186712549546681. (CXX) g++ options: -flto -pthread

LeelaChessZero

Backend: Eigen

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: EigenCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X5001000150020002500SE +/- 19.84, N = 7SE +/- 8.85, N = 9SE +/- 8.89, N = 3SE +/- 7.53, N = 9SE +/- 7.32, N = 9216185411608866841. (CXX) g++ options: -flto -pthread

oneDNN

Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.32790.65580.98371.31161.6395SE +/- 0.002976, N = 5SE +/- 0.002722, N = 5SE +/- 0.002108, N = 5SE +/- 0.003942, N = 5SE +/- 0.002527, N = 50.8120471.4575200.6047800.5197330.476097MIN: 0.79-lpthread - MIN: 1.35-lpthread - MIN: 0.58-lpthread - MIN: 0.47-lpthread - MIN: 0.421. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

ECP-CANDLE

Benchmark: P3B1

OpenBenchmarking.orgSeconds, Fewer Is BetterECP-CANDLE 0.4Benchmark: P3B1Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X30060090012001500429.611158.671023.321174.471309.59

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X48121620SE +/- 0.00249, N = 7SE +/- 0.22593, N = 15SE +/- 0.04088, N = 7SE +/- 0.04249, N = 7SE +/- 0.02368, N = 76.0021416.3534010.3800016.7382018.03210MIN: 5.9-lpthread - MIN: 14.78-lpthread - MIN: 9.78-lpthread - MIN: 16.08-lpthread - MIN: 17.61. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

ASKAP

Test: tConvolve MT - Degridding

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - DegriddingCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X8001600240032004000SE +/- 2.07, N = 3SE +/- 5.57, N = 3SE +/- 1.83, N = 3SE +/- 3.45, N = 3SE +/- 0.92, N = 33872.031471.581594.951541.671346.571. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenFOAM

Input: Motorbike 60M

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 60MCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X30060090012001500SE +/- 0.14, N = 3SE +/- 0.23, N = 3SE +/- 1.12, N = 3SE +/- 0.77, N = 3SE +/- 0.26, N = 3487.801270.721090.211277.551382.54-lfoamToVTK -llagrangian -lfileFormats1. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -lgenericPatchFields -lOpenFOAM -ldl -lm

Xcompact3d Incompact3d

Input: input.i3d 193 Cells Per Direction

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X306090120150SE +/- 0.58, N = 3SE +/- 1.17, N = 9SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.16, N = 355.61141.21126.86144.47156.161. (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

ASKAP

Test: tConvolve OpenMP - Degridding

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - DegriddingCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2K4K6K8K10KSE +/- 37.29, N = 6SE +/- 6.53, N = 15SE +/- 38.17, N = 15SE +/- 10.98, N = 6SE +/- 11.91, N = 67793.773367.678741.603732.723214.581. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

oneDNN

Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X246810SE +/- 0.00441, N = 5SE +/- 0.01310, N = 5SE +/- 0.01065, N = 5SE +/- 0.02525, N = 5SE +/- 0.05069, N = 53.446438.935526.557657.548437.79652MIN: 3.4-lpthread - MIN: 8.33-lpthread - MIN: 6.32-lpthread - MIN: 7.29-lpthread - MIN: 7.291. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

Xcompact3d Incompact3d

Input: input.i3d 129 Cells Per Direction

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per DirectionCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X918273645SE +/- 0.01, N = 4SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.32, N = 3SE +/- 0.05, N = 314.5537.2127.1832.1533.951. (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

ASKAP

Test: Hogbom Clean OpenMP

OpenBenchmarking.orgIterations Per Second, More Is BetterASKAP 1.0Test: Hogbom Clean OpenMPCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X120240360480600SE +/- 0.00, N = 5SE +/- 0.57, N = 3SE +/- 1.80, N = 4SE +/- 0.52, N = 4SE +/- 1.00, N = 4540.54221.73527.72238.67220.041. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Mobile Neural Network

Model: mobilenet-v1-1.0

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenet-v1-1.0Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.91621.83242.74863.66484.581SE +/- 0.008, N = 3SE +/- 0.011, N = 3SE +/- 0.010, N = 3SE +/- 0.037, N = 3SE +/- 0.067, N = 32.8911.8161.6764.0722.639MIN: 2.85 / MAX: 8.54MIN: 1.79 / MAX: 3.08MIN: 1.63 / MAX: 2.97MIN: 3.95 / MAX: 4.31MIN: 2.52 / MAX: 11.271. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

ONNX Runtime

Model: yolov4 - Device: CPU - Executor: Parallel

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: ParallelCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X140280420560700SE +/- 1.09, N = 3SE +/- 0.44, N = 3SE +/- 0.60, N = 3SE +/- 0.33, N = 3SE +/- 0.44, N = 36292733082992961. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

OpenFOAM

Input: Motorbike 30M

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 30MCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X4080120160200SE +/- 0.18, N = 3SE +/- 0.25, N = 3SE +/- 0.32, N = 3SE +/- 0.15, N = 3SE +/- 0.24, N = 384.47177.6080.2996.1698.44-lfoamToVTK -llagrangian -lfileFormats1. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -lgenericPatchFields -lOpenFOAM -ldl -lm

Mobile Neural Network

Model: mobilenetV3

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenetV3Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.5381.0761.6142.1522.69SE +/- 0.007, N = 3SE +/- 0.005, N = 3SE +/- 0.006, N = 3SE +/- 0.018, N = 3SE +/- 0.003, N = 31.1741.1561.0821.8382.391MIN: 1.15 / MAX: 2.06MIN: 1.14 / MAX: 1.73MIN: 1.06 / MAX: 2.25MIN: 1.79 / MAX: 2.07MIN: 1.87 / MAX: 3.851. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

ASKAP

Test: tConvolve MPI - Gridding

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2K4K6K8K10KSE +/- 12.67, N = 3SE +/- 45.52, N = 3SE +/- 0.00, N = 3SE +/- 58.16, N = 34472.734256.058746.528258.026728.091. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

WebP2 Image Encode

Encode Settings: Quality 100, Lossless Compression

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 100, Lossless CompressionCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2004006008001000SE +/- 0.96, N = 3SE +/- 1.21, N = 3SE +/- 2.17, N = 3SE +/- 4.10, N = 3SE +/- 2.59, N = 3476.45974.59864.94614.81536.771. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

Mlpack Benchmark

Benchmark: scikit_qda

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_qdaCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X1224364860SE +/- 0.13, N = 3SE +/- 0.10, N = 3SE +/- 0.08, N = 3SE +/- 0.34, N = 3SE +/- 0.20, N = 327.3152.5738.2751.3955.06

WebP2 Image Encode

Encode Settings: Quality 100, Compression Effort 5

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 100, Compression Effort 5Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X1.32192.64383.96575.28766.6095SE +/- 0.003, N = 9SE +/- 0.004, N = 7SE +/- 0.003, N = 7SE +/- 0.003, N = 8SE +/- 0.006, N = 92.9365.8755.1693.7093.2491. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

TNN

Target: CPU - Model: SqueezeNet v1.1

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.1Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X60120180240300SE +/- 0.06, N = 5SE +/- 0.18, N = 3SE +/- 0.03, N = 4SE +/- 1.66, N = 4SE +/- 1.16, N = 4133.97266.19222.33213.78213.32MIN: 133.46 / MAX: 134.81MIN: 265.87 / MAX: 266.68MIN: 222.15 / MAX: 222.66MIN: 210.63 / MAX: 219.97MIN: 209.34 / MAX: 215.211. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

WebP2 Image Encode

Encode Settings: Quality 75, Compression Effort 7

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 75, Compression Effort 7Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X4080120160200SE +/- 0.23, N = 3SE +/- 0.74, N = 3SE +/- 0.62, N = 3SE +/- 0.51, N = 3SE +/- 1.48, N = 3101.57199.65178.01129.73110.281. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

WebP2 Image Encode

Encode Settings: Quality 95, Compression Effort 7

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 95, Compression Effort 7Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X90180270360450SE +/- 0.51, N = 3SE +/- 1.68, N = 3SE +/- 0.77, N = 3SE +/- 0.76, N = 3SE +/- 1.17, N = 3215.58420.03376.96268.61234.931. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

ASKAP

Test: tConvolve MPI - Degridding

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X16003200480064008000SE +/- 19.39, N = 3SE +/- 34.79, N = 3SE +/- 53.33, N = 3SE +/- 49.47, N = 3SE +/- 48.56, N = 34198.513976.306453.227668.056643.641. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Mlpack Benchmark

Benchmark: scikit_svm

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_svmCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X510152025SE +/- 0.01, N = 4SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 310.5420.2916.1716.1516.56

ONNX Runtime

Model: bertsquad-12 - Device: CPU - Executor: Parallel

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: ParallelCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2004006008001000SE +/- 11.29, N = 3SE +/- 0.29, N = 3SE +/- 1.17, N = 3SE +/- 0.60, N = 3SE +/- 0.29, N = 39194805185405541. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.45810.91621.37431.83242.2905SE +/- 0.00773, N = 3SE +/- 0.00040, N = 3SE +/- 0.00274, N = 3SE +/- 0.00489, N = 3SE +/- 0.00299, N = 31.352122.036001.776411.300491.06926MIN: 1.28-lpthread - MIN: 2.01-lpthread - MIN: 1.74-lpthread - MIN: 1.2-lpthread - MIN: 0.961. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

Mobile Neural Network

Model: MobileNetV2_224

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: MobileNetV2_224Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.73691.47382.21072.94763.6845SE +/- 0.020, N = 3SE +/- 0.029, N = 3SE +/- 0.011, N = 3SE +/- 0.045, N = 3SE +/- 0.014, N = 32.4101.9421.8312.9753.275MIN: 2.36 / MAX: 3.86MIN: 1.9 / MAX: 3.6MIN: 1.79 / MAX: 2.92MIN: 2.88 / MAX: 3.44MIN: 3.21 / MAX: 10.861. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X246810SE +/- 0.00117, N = 9SE +/- 0.00140, N = 9SE +/- 0.00483, N = 9SE +/- 0.00484, N = 9SE +/- 0.00224, N = 95.253946.370235.571094.387733.62609MIN: 5.16-lpthread - MIN: 6.32-lpthread - MIN: 5.45-lpthread - MIN: 4.17-lpthread - MIN: 3.431. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

WebP2 Image Encode

Encode Settings: Default

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: DefaultCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.81431.62862.44293.25724.0715SE +/- 0.010, N = 11SE +/- 0.006, N = 8SE +/- 0.007, N = 9SE +/- 0.008, N = 10SE +/- 0.013, N = 102.0623.6193.1702.4392.1651. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.63881.27761.91642.55523.194SE +/- 0.00121, N = 9SE +/- 0.00474, N = 9SE +/- 0.00503, N = 9SE +/- 0.00298, N = 9SE +/- 0.00702, N = 92.223362.839132.483931.997421.61899MIN: 2.2-lpthread - MIN: 2.8-lpthread - MIN: 2.41-lpthread - MIN: 1.82-lpthread - MIN: 1.451. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

oneDNN

Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.31660.63320.94981.26641.583SE +/- 0.008609, N = 4SE +/- 0.000645, N = 4SE +/- 0.001247, N = 4SE +/- 0.007658, N = 4SE +/- 0.006926, N = 151.0520001.4072801.2378501.0069410.823009MIN: 1.01-lpthread - MIN: 1.39-lpthread - MIN: 1.21-lpthread - MIN: 0.94-lpthread - MIN: 0.71. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

TNN

Target: CPU - Model: DenseNet

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: DenseNetCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X6001200180024003000SE +/- 1.88, N = 3SE +/- 0.87, N = 3SE +/- 1.54, N = 3SE +/- 3.11, N = 3SE +/- 10.08, N = 31792.973016.482609.842563.762425.59MIN: 1751.55 / MAX: 1868.31MIN: 2943.1 / MAX: 3087.97MIN: 2559.5 / MAX: 2657.73MIN: 2481.34 / MAX: 2640.2MIN: 2339.59 / MAX: 2518.621. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

ECP-CANDLE

Benchmark: P1B2

OpenBenchmarking.orgSeconds, Fewer Is BetterECP-CANDLE 0.4Benchmark: P1B2Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X81624324020.9434.8429.9930.2431.39

Mobile Neural Network

Model: squeezenetv1.1

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: squeezenetv1.1Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.89731.79462.69193.58924.4865SE +/- 0.049, N = 3SE +/- 0.017, N = 3SE +/- 0.006, N = 3SE +/- 0.098, N = 3SE +/- 0.098, N = 32.4052.8052.5903.3943.988MIN: 2.33 / MAX: 3.56MIN: 2.76 / MAX: 10.42MIN: 2.55 / MAX: 4.48MIN: 3.15 / MAX: 4.21MIN: 3.72 / MAX: 4.791. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

ONNX Runtime

Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: ParallelCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X20406080100SE +/- 0.44, N = 3SE +/- 0.29, N = 3SE +/- 0.17, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3111677285881. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

TNN

Target: CPU - Model: SqueezeNet v2

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v2Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X1428425670SE +/- 0.08, N = 10SE +/- 0.12, N = 8SE +/- 0.13, N = 9SE +/- 0.22, N = 9SE +/- 0.08, N = 938.9263.6153.2250.8550.94MIN: 38.37 / MAX: 39.92MIN: 62.79 / MAX: 64.28MIN: 52.43 / MAX: 54.16MIN: 49.98 / MAX: 52.59MIN: 50.34 / MAX: 52.461. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

ONNX Runtime

Model: GPT-2 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: StandardCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2K4K6K8K10KSE +/- 20.71, N = 3SE +/- 104.18, N = 12SE +/- 14.95, N = 3SE +/- 10.85, N = 3SE +/- 58.87, N = 81108268328826786270621. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

TNN

Target: CPU - Model: MobileNet v2

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v2Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X60120180240300SE +/- 0.31, N = 4SE +/- 0.18, N = 3SE +/- 0.14, N = 4SE +/- 0.47, N = 4SE +/- 0.71, N = 4170.58272.04233.18224.20224.26MIN: 157.87 / MAX: 209.34MIN: 270.94 / MAX: 276.3MIN: 232.19 / MAX: 237.22MIN: 218.68 / MAX: 249.25MIN: 219.36 / MAX: 242.571. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

ONNX Runtime

Model: super-resolution-10 - Device: CPU - Executor: Parallel

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: ParallelCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X14002800420056007000SE +/- 46.82, N = 4SE +/- 10.11, N = 3SE +/- 15.67, N = 3SE +/- 5.61, N = 3SE +/- 22.98, N = 3451641034606559164361. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

NCNN

Target: CPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: alexnetCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X3691215SE +/- 0.05, N = 15SE +/- 0.01, N = 15SE +/- 0.03, N = 15SE +/- 0.01, N = 3SE +/- 0.13, N = 47.5811.839.6210.0111.08MIN: 7.18 / MAX: 9.34MIN: 11.63 / MAX: 18.46MIN: 8.9 / MAX: 11.14MIN: 9.92 / MAX: 12.23MIN: 10.7 / MAX: 12.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

oneDNN

Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.87671.75342.63013.50684.3835SE +/- 0.00329, N = 4SE +/- 0.00733, N = 4SE +/- 0.00793, N = 4SE +/- 0.02871, N = 4SE +/- 0.01755, N = 42.636113.291362.897173.445483.89647MIN: 2.5-lpthread - MIN: 3.12-lpthread - MIN: 2.81-lpthread - MIN: 3.05-lpthread - MIN: 3.661. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

ONNX Runtime

Model: GPT-2 - Device: CPU - Executor: Parallel

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: ParallelCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2K4K6K8K10KSE +/- 20.34, N = 3SE +/- 8.85, N = 3SE +/- 5.78, N = 3SE +/- 11.00, N = 3SE +/- 11.18, N = 3808856216919568755341. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

NCNN

Target: CPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: resnet50Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X612182430SE +/- 0.07, N = 15SE +/- 0.07, N = 15SE +/- 0.09, N = 15SE +/- 0.04, N = 3SE +/- 0.34, N = 416.8420.2218.1321.2124.40MIN: 16.32 / MAX: 21.73MIN: 19.74 / MAX: 28.22MIN: 17.64 / MAX: 24.72MIN: 20.94 / MAX: 23.21MIN: 23.71 / MAX: 27.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Caffe

Model: GoogleNet - Acceleration: CPU - Iterations: 200

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: GoogleNet - Acceleration: CPU - Iterations: 200Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X40K80K120K160K200KSE +/- 116.45, N = 3SE +/- 39.00, N = 3SE +/- 248.50, N = 3SE +/- 46.44, N = 3SE +/- 215.28, N = 31340451791461620221787251937101. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

Open Porous Media Git

OPM Benchmark: Flow MPI Norne - Threads: 8

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Norne - Threads: 8Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X110220330440550SE +/- 0.26, N = 3SE +/- 0.04, N = 3SE +/- 0.25, N = 3SE +/- 0.23, N = 3SE +/- 0.16, N = 3515.71447.63361.91361.84357.861. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Open Porous Media Git

OPM Benchmark: Flow MPI Norne-4C MSW - Threads: 8

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Norne-4C MSW - Threads: 8Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2004006008001000SE +/- 0.65, N = 3SE +/- 0.32, N = 3SE +/- 0.28, N = 3SE +/- 0.16, N = 3SE +/- 0.32, N = 3825.56733.91581.46581.35575.131. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Caffe

Model: AlexNet - Acceleration: CPU - Iterations: 100

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: AlexNet - Acceleration: CPU - Iterations: 100Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X8K16K24K32K40KSE +/- 17.32, N = 3SE +/- 7.06, N = 3SE +/- 24.84, N = 3SE +/- 38.11, N = 3SE +/- 30.55, N = 325590339053022234329366581. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

Caffe

Model: GoogleNet - Acceleration: CPU - Iterations: 100

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: GoogleNet - Acceleration: CPU - Iterations: 100Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X20K40K60K80K100KSE +/- 741.46, N = 4SE +/- 34.64, N = 3SE +/- 158.22, N = 3SE +/- 260.55, N = 3SE +/- 34.42, N = 367833898258086289492966241. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

Caffe

Model: AlexNet - Acceleration: CPU - Iterations: 200

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: AlexNet - Acceleration: CPU - Iterations: 200Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X16K32K48K64K80KSE +/- 323.04, N = 3SE +/- 49.12, N = 3SE +/- 94.00, N = 3SE +/- 57.49, N = 3SE +/- 178.17, N = 351713676616049468676733521. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

Numpy Benchmark

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X140280420560700SE +/- 0.68, N = 3SE +/- 0.43, N = 3SE +/- 1.21, N = 3SE +/- 8.26, N = 3SE +/- 0.86, N = 3665.34491.56603.09605.17591.54

NCNN

Target: CPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: squeezenet_ssdCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X48121620SE +/- 0.17, N = 15SE +/- 0.06, N = 15SE +/- 0.05, N = 15SE +/- 0.02, N = 3SE +/- 0.09, N = 413.2716.7612.4413.5114.55MIN: 12.19 / MAX: 43.4MIN: 16.11 / MAX: 23.01MIN: 12.03 / MAX: 14.14MIN: 13.16 / MAX: 20.71MIN: 13.65 / MAX: 21.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X400800120016002000SE +/- 0.28, N = 3SE +/- 5.60, N = 3SE +/- 2.31, N = 3SE +/- 8.68, N = 3SE +/- 8.84, N = 31613.811859.141382.161785.051814.13MIN: 1608.23-lpthread - MIN: 1846.97-lpthread - MIN: 1372.35-lpthread - MIN: 1762.05-lpthread - MIN: 1783.131. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X400800120016002000SE +/- 1.95, N = 3SE +/- 4.89, N = 3SE +/- 0.61, N = 3SE +/- 13.72, N = 3SE +/- 22.20, N = 31616.061864.541387.861742.061783.85MIN: 1608.68-lpthread - MIN: 1851.66-lpthread - MIN: 1380.9-lpthread - MIN: 1715.13-lpthread - MIN: 1730.181. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X400800120016002000SE +/- 2.95, N = 3SE +/- 9.79, N = 3SE +/- 1.81, N = 3SE +/- 5.06, N = 3SE +/- 19.76, N = 51617.101847.701385.791776.191820.40MIN: 1608.6-lpthread - MIN: 1820.32-lpthread - MIN: 1375.42-lpthread - MIN: 1756.45-lpthread - MIN: 1761.071. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

Open Porous Media Git

OPM Benchmark: Flow MPI Extra - Threads: 8

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Extra - Threads: 8Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2004006008001000SE +/- 0.86, N = 3SE +/- 0.32, N = 3SE +/- 1.20, N = 3SE +/- 0.19, N = 3SE +/- 0.30, N = 3891.781027.17781.98810.33808.051. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Open Porous Media Git

OPM Benchmark: Flow MPI Extra - Threads: 4

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Extra - Threads: 4Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2004006008001000SE +/- 0.81, N = 3SE +/- 0.80, N = 3SE +/- 0.34, N = 3SE +/- 0.23, N = 3SE +/- 0.65, N = 3671.56829.49643.96691.38690.621. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Open Porous Media Git

OPM Benchmark: Flow MPI Norne-4C MSW - Threads: 4

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Norne-4C MSW - Threads: 4Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X100200300400500SE +/- 0.53, N = 3SE +/- 0.49, N = 3SE +/- 0.21, N = 3SE +/- 0.78, N = 3SE +/- 0.39, N = 3451.00454.34357.33378.33377.031. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Mlpack Benchmark

Benchmark: scikit_ica

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_icaCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X1020304050SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.24, N = 3SE +/- 0.12, N = 332.9339.1634.8839.3741.77

Open Porous Media Git

OPM Benchmark: Flow MPI Norne-4C MSW - Threads: 2

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Norne-4C MSW - Threads: 2Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X110220330440550SE +/- 0.98, N = 3SE +/- 0.29, N = 3SE +/- 0.45, N = 3SE +/- 0.35, N = 3SE +/- 0.66, N = 3475.95511.31403.35441.14441.071. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Open Porous Media Git

OPM Benchmark: Flow MPI Norne-4C MSW - Threads: 1

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Norne-4C MSW - Threads: 1Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X130260390520650SE +/- 1.75, N = 3SE +/- 1.44, N = 3SE +/- 0.27, N = 3SE +/- 0.96, N = 3SE +/- 2.73, N = 3537.38601.31476.91572.60569.141. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Open Porous Media Git

OPM Benchmark: Flow MPI Norne - Threads: 1

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Norne - Threads: 1Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X60120180240300SE +/- 0.07, N = 3SE +/- 0.17, N = 3SE +/- 0.10, N = 3SE +/- 0.23, N = 3SE +/- 1.36, N = 3247.89281.47224.10273.89270.811. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Open Porous Media Git

OPM Benchmark: Flow MPI Norne - Threads: 4

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Norne - Threads: 4Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X60120180240300SE +/- 0.28, N = 3SE +/- 0.16, N = 3SE +/- 0.14, N = 3SE +/- 0.18, N = 3SE +/- 0.31, N = 3280.85275.40223.97233.63232.611. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Open Porous Media Git

OPM Benchmark: Flow MPI Norne - Threads: 2

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Norne - Threads: 2Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X50100150200250SE +/- 0.42, N = 3SE +/- 0.53, N = 3SE +/- 0.14, N = 3SE +/- 0.16, N = 3SE +/- 0.42, N = 3223.20234.63187.49203.98203.891. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Open Porous Media Git

OPM Benchmark: Flow MPI Extra - Threads: 2

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Extra - Threads: 2Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2004006008001000SE +/- 0.67, N = 3SE +/- 2.38, N = 3SE +/- 0.69, N = 3SE +/- 1.56, N = 3SE +/- 1.57, N = 3717.04799.40672.48713.34712.621. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

Open Porous Media Git

OPM Benchmark: Flow MPI Extra - Threads: 1

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Flow MPI Extra - Threads: 1Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2004006008001000SE +/- 2.12, N = 3SE +/- 5.46, N = 3SE +/- 2.03, N = 3SE +/- 4.75, N = 3SE +/- 3.75, N = 31053.381154.09997.341073.711074.331. (CXX) g++ options: -pipe -pthread -fopenmp -O3 -mtune=native -UNDEBUG -lm -ldl -lrt2. Core i9 12900K: Build Time Thu Apr 28 06:45:36 PM EDT 20223. Ryzen 7 5800X: Build Time Mon Apr 25 06:10:54 PM EDT 20224. Ryzen 7 5800X3D: Build Time Mon Apr 25 06:10:54 PM EDT 20225. Ryzen 9 5900X: Build Time Mon Apr 25 06:10:54 PM EDT 20226. Ryzen 9 5950X: Build Time Mon Apr 25 06:10:54 PM EDT 2022

oneDNN

Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X7001400210028003500SE +/- 0.76, N = 3SE +/- 1.56, N = 3SE +/- 2.34, N = 3SE +/- 32.41, N = 3SE +/- 26.41, N = 32881.323065.262691.242904.332750.18MIN: 2872.06-lpthread - MIN: 3058.73-lpthread - MIN: 2678.39-lpthread - MIN: 2837.45-lpthread - MIN: 2685.911. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

oneDNN

Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X7001400210028003500SE +/- 0.27, N = 3SE +/- 2.27, N = 3SE +/- 5.12, N = 3SE +/- 23.67, N = 3SE +/- 10.92, N = 32881.723053.302683.162873.452703.33MIN: 2874.65-lpthread - MIN: 3045.76-lpthread - MIN: 2665.66-lpthread - MIN: 2830.55-lpthread - MIN: 2665.51. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

oneDNN

Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X7001400210028003500SE +/- 1.61, N = 3SE +/- 0.89, N = 3SE +/- 1.77, N = 3SE +/- 15.10, N = 3SE +/- 23.91, N = 32881.103055.842691.672908.282745.44MIN: 2869.73-lpthread - MIN: 3050.91-lpthread - MIN: 2680.17-lpthread - MIN: 2862.45-lpthread - MIN: 2684.631. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

Mobile Neural Network

Model: inception-v3

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: inception-v3Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X612182430SE +/- 0.66, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.37, N = 3SE +/- 0.13, N = 324.3025.6923.1923.6925.73MIN: 22.9 / MAX: 36.09MIN: 25.45 / MAX: 31.44MIN: 22.93 / MAX: 31MIN: 22.99 / MAX: 31.36MIN: 25.09 / MAX: 33.941. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

ONNX Runtime

Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X400800120016002000SE +/- 0.67, N = 3SE +/- 1.33, N = 3SE +/- 3.42, N = 3SE +/- 45.67, N = 12SE +/- 46.80, N = 9197510421939151916681. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ONNX Runtime

Model: bertsquad-12 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: StandardCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X2004006008001000SE +/- 0.76, N = 3SE +/- 0.17, N = 3SE +/- 57.02, N = 12SE +/- 4.80, N = 3SE +/- 1.20, N = 39885578229498001. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ONNX Runtime

Model: super-resolution-10 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: StandardCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X16003200480064008000SE +/- 10.33, N = 3SE +/- 10.27, N = 3SE +/- 11.77, N = 3SE +/- 318.33, N = 12SE +/- 21.53, N = 3474736284107754561691. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ONNX Runtime

Model: fcn-resnet101-11 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: StandardCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X306090120150SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.17, N = 3SE +/- 0.44, N = 3SE +/- 4.73, N = 1210149107115981. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ONNX Runtime

Model: yolov4 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: StandardCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X150300450600750SE +/- 1.42, N = 3SE +/- 28.87, N = 12SE +/- 42.62, N = 12SE +/- 17.59, N = 12SE +/- 17.40, N = 96934315725394871. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

Mlpack Benchmark

Benchmark: scikit_linearridgeregression

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_linearridgeregressionCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.40950.8191.22851.6382.0475SE +/- 0.05, N = 12SE +/- 0.02, N = 7SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 31.601.821.631.581.69

Mobile Neural Network

Model: SqueezeNetV1.0

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: SqueezeNetV1.0Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X1.1772.3543.5314.7085.885SE +/- 0.177, N = 3SE +/- 0.032, N = 3SE +/- 0.010, N = 3SE +/- 0.058, N = 3SE +/- 0.113, N = 34.1514.5364.2134.7585.231MIN: 3.91 / MAX: 6.32MIN: 4.47 / MAX: 5.71MIN: 4.16 / MAX: 5.46MIN: 4.59 / MAX: 12.51MIN: 4.95 / MAX: 6.511. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

Mobile Neural Network

Model: resnet-v2-50

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: resnet-v2-50Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X612182430SE +/- 1.14, N = 3SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.14, N = 3SE +/- 0.14, N = 323.0918.4416.3624.5020.52MIN: 21.7 / MAX: 30.17MIN: 18.25 / MAX: 25.86MIN: 16 / MAX: 24.17MIN: 23.86 / MAX: 52.44MIN: 19.84 / MAX: 24.051. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

NCNN

Target: CPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: regnety_400mCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X3691215SE +/- 0.22, N = 15SE +/- 0.01, N = 15SE +/- 0.01, N = 15SE +/- 0.01, N = 3SE +/- 0.08, N = 47.395.935.188.439.61MIN: 6.17 / MAX: 27.59MIN: 5.83 / MAX: 7.68MIN: 5.07 / MAX: 12.17MIN: 8.36 / MAX: 8.75MIN: 9.37 / MAX: 11.031. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: yolov4-tinyCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X510152025SE +/- 0.33, N = 15SE +/- 0.21, N = 15SE +/- 0.19, N = 15SE +/- 0.40, N = 3SE +/- 0.32, N = 415.8619.7814.8020.6320.49MIN: 14.24 / MAX: 21MIN: 18.64 / MAX: 21.1MIN: 14 / MAX: 16.98MIN: 19.48 / MAX: 21.6MIN: 19.6 / MAX: 21.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: resnet18Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X48121620SE +/- 0.15, N = 15SE +/- 0.03, N = 15SE +/- 0.05, N = 15SE +/- 0.05, N = 3SE +/- 0.17, N = 49.6613.0210.2712.5014.28MIN: 7.55 / MAX: 14.7MIN: 12.77 / MAX: 32.91MIN: 9.71 / MAX: 12.09MIN: 12.28 / MAX: 12.82MIN: 13.94 / MAX: 16.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: vgg16Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X1326395265SE +/- 0.48, N = 15SE +/- 0.07, N = 15SE +/- 0.13, N = 15SE +/- 0.09, N = 3SE +/- 0.08, N = 428.2455.9742.6050.7556.55MIN: 25.72 / MAX: 45.6MIN: 54.91 / MAX: 64.62MIN: 41.52 / MAX: 50.92MIN: 49.97 / MAX: 60.2MIN: 55.53 / MAX: 62.981. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: googlenetCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X3691215SE +/- 0.21, N = 15SE +/- 0.02, N = 15SE +/- 0.05, N = 15SE +/- 0.02, N = 3SE +/- 0.28, N = 49.9410.227.3111.4412.92MIN: 7.91 / MAX: 14.3MIN: 9.79 / MAX: 18.21MIN: 7.05 / MAX: 14.4MIN: 11.28 / MAX: 11.82MIN: 12.09 / MAX: 15.321. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: blazefaceCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.4050.811.2151.622.025SE +/- 0.05, N = 15SE +/- 0.00, N = 15SE +/- 0.00, N = 15SE +/- 0.00, N = 3SE +/- 0.02, N = 41.461.221.061.631.80MIN: 1.15 / MAX: 2.96MIN: 1.19 / MAX: 2.1MIN: 1.04 / MAX: 4.31MIN: 1.61 / MAX: 1.81MIN: 1.75 / MAX: 2.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: efficientnet-b0Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X1.20152.4033.60454.8066.0075SE +/- 0.11, N = 15SE +/- 0.01, N = 15SE +/- 0.01, N = 15SE +/- 0.01, N = 3SE +/- 0.06, N = 45.343.603.014.775.23MIN: 4.35 / MAX: 9.28MIN: 3.53 / MAX: 5.22MIN: 2.93 / MAX: 13.07MIN: 4.7 / MAX: 5MIN: 5.09 / MAX: 6.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mnasnetCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.87081.74162.61243.48324.354SE +/- 0.07, N = 15SE +/- 0.00, N = 15SE +/- 0.00, N = 15SE +/- 0.01, N = 3SE +/- 0.05, N = 43.102.242.013.443.87MIN: 2.66 / MAX: 4.79MIN: 2.2 / MAX: 3.68MIN: 1.97 / MAX: 2.76MIN: 3.39 / MAX: 3.72MIN: 3.76 / MAX: 10.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: shufflenet-v2Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.9361.8722.8083.7444.68SE +/- 0.08, N = 14SE +/- 0.00, N = 15SE +/- 0.01, N = 15SE +/- 0.01, N = 3SE +/- 0.01, N = 43.102.352.113.884.16MIN: 2.68 / MAX: 4.51MIN: 2.31 / MAX: 3.74MIN: 2.07 / MAX: 3MIN: 3.82 / MAX: 4.06MIN: 4.05 / MAX: 4.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU-v3-v3 - Model: mobilenet-v3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU-v3-v3 - Model: mobilenet-v3Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.84831.69662.54493.39324.2415SE +/- 0.06, N = 15SE +/- 0.00, N = 15SE +/- 0.00, N = 15SE +/- 0.01, N = 3SE +/- 0.00, N = 42.902.281.903.463.77MIN: 2.53 / MAX: 4.55MIN: 2.22 / MAX: 3.92MIN: 1.84 / MAX: 2.38MIN: 3.39 / MAX: 3.66MIN: 3.7 / MAX: 4.731. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU-v2-v2 - Model: mobilenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU-v2-v2 - Model: mobilenet-v2Core i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X0.96531.93062.89593.86124.8265SE +/- 0.12, N = 15SE +/- 0.01, N = 15SE +/- 0.00, N = 15SE +/- 0.01, N = 3SE +/- 0.01, N = 43.412.612.133.914.29MIN: 2.72 / MAX: 5.86MIN: 2.54 / MAX: 3.77MIN: 2.03 / MAX: 2.88MIN: 3.82 / MAX: 4.11MIN: 4.15 / MAX: 7.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mobilenetCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X3691215SE +/- 0.26, N = 15SE +/- 0.13, N = 15SE +/- 0.08, N = 15SE +/- 0.01, N = 3SE +/- 0.15, N = 411.1111.637.6211.4212.27MIN: 8.87 / MAX: 13.96MIN: 11.06 / MAX: 12.93MIN: 7.25 / MAX: 9.92MIN: 11.16 / MAX: 18.48MIN: 11.71 / MAX: 18.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X246810SE +/- 0.16860, N = 12SE +/- 0.11645, N = 15SE +/- 0.06264, N = 3SE +/- 0.07639, N = 3SE +/- 0.22726, N = 158.737508.342037.276235.386894.75533MIN: 4.15-lpthread - MIN: 5.88-lpthread - MIN: 5.11-lpthread - MIN: 4-lpthread - MIN: 3.381. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUCore i9 12900KRyzen 7 5800XRyzen 7 5800X3DRyzen 9 5900XRyzen 9 5950X510152025SE +/- 0.00220, N = 7SE +/- 0.12899, N = 7SE +/- 0.01917, N = 7SE +/- 0.29208, N = 15SE +/- 0.00896, N = 75.8753618.7535012.6227016.1398016.70510MIN: 5.78-lpthread - MIN: 18.34-lpthread - MIN: 12.28-lpthread - MIN: 15.35-lpthread - MIN: 16.311. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl

Phoronix Test Suite v10.8.5