Xeon Platinum 8280 oneDNN 2.0 + More

2 x Intel Xeon Platinum 8280 testing with a GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS) and llvmpipe on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2012110-HA-XEONPLATI68.

Xeon Platinum 8280 oneDNN 2.0 + MoreProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen Resolution122a32 x Intel Xeon Platinum 8280 @ 4.00GHz (56 Cores / 112 Threads)GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS)Intel Sky Lake-E DMI3 Registers378GB280GB INTEL SSDPED1D280GAllvmpipeVE2282 x Intel X722 for 1GbE + 2 x QLogic FastLinQ QL41000 10/25/40/50GbEUbuntu 20.045.4.0-18-generic (x86_64)GNOME Shell 3.36.0X Server 1.20.7modesetting 1.20.73.3 Mesa 20.0.2 (LLVM 9.0.1 256 bits)GCC 9.3.0ext41920x1080OpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Disk Details- 1, 2: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Details- Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002cPython Details- Python 3.8.2Security Details- itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + tsx_async_abort: Mitigation of TSX disabled

Xeon Platinum 8280 oneDNN 2.0 + Moreleveldb: Hot Readleveldb: Fill Syncleveldb: Fill Syncleveldb: Overwriteleveldb: Overwriteleveldb: Rand Fillleveldb: Rand Fillleveldb: Rand Readleveldb: Seek Randleveldb: Rand Deleteleveldb: Seq Fillleveldb: Seq Fillcrafty: Elapsed Timeonednn: IP Shapes 1D - f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUonednn: IP Shapes 1D - bf16bf16bf16 - CPUonednn: IP Shapes 3D - bf16bf16bf16 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_3d - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPUrav1e: 1rav1e: 5rav1e: 6rav1e: 10stockfish: Total Timeasmfish: 1024 Hash Memory, 26 Depthbuild-clash: Time To Compileastcenc: Fastastcenc: Mediumastcenc: Thoroughastcenc: Exhaustiveopenvino: Face Detection 0106 FP16 - CPUopenvino: Face Detection 0106 FP16 - CPUopenvino: Face Detection 0106 FP32 - CPUopenvino: Face Detection 0106 FP32 - CPUopenvino: Person Detection 0106 FP16 - CPUopenvino: Person Detection 0106 FP16 - CPUopenvino: Person Detection 0106 FP32 - CPUopenvino: Person Detection 0106 FP32 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP32 - CPUopenvino: Age Gender Recognition Retail 0013 FP32 - CPUphpbench: PHP Benchmark Suite122a3139.4847.21684.4298.31484.3678.51462.045140.411172.1211483.2668.31493.01978149861.354733.114151.424481.094083.667132.473643.961341.453151.216393.708320.4255680.349150989.801565.554981.4343.192434.383924.44712559.6590.335755979.574565.2680.2715510.8172390.3480.9371.2082.45096447530135035548488.6345.636.617.1652.3315.511784.4815.141814.328.743142.678.683152.6138683.480.6238607.560.6266662478288681.357213.104811.392111.085033.665012.537073.991811.450031.211613.766980.4254700.349469995.326561.3521011.7613.196954.385244.45890560.9730.3370981008.338563.8590.2681510.8232100.3460.9561.2212.48696807159133208525496.0905.606.557.1552.3278363581.355783.112431.405181.095663.664132.687953.942811.462021.217153.704190.4249750.349913997.876563.1241007.0683.202584.381394.44637570.7370.3367531050.412562.3150.2707280.8222540.3460.9491.2172.45595164573132978753492.7835.626.607.1652.3315.531786.6415.191814.558.843107.468.723120.8838792.260.6238565.640.62668619OpenBenchmarking.org

LevelDB

Benchmark: Hot Read

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot Read1306090120150SE +/- 0.23, N = 3139.481. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Fill Sync

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill Sync1246810SE +/- 0.00, N = 37.21. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Fill Sync

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill Sync1400800120016002000SE +/- 8.30, N = 31684.431. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Overwrite

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Overwrite1246810SE +/- 0.03, N = 38.31. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Overwrite

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Overwrite130060090012001500SE +/- 7.25, N = 31484.371. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Random Fill

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random Fill1246810SE +/- 0.03, N = 38.51. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Random Fill

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random Fill130060090012001500SE +/- 3.27, N = 31462.051. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Random Read

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random Read1306090120150SE +/- 1.25, N = 3140.411. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Seek Random

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek Random14080120160200SE +/- 1.28, N = 3172.121. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Random Delete

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random Delete130060090012001500SE +/- 5.02, N = 31483.271. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Sequential Fill

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential Fill1246810SE +/- 0.00, N = 38.31. (CXX) g++ options: -O3 -lsnappy -lpthread

LevelDB

Benchmark: Sequential Fill

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential Fill130060090012001500SE +/- 2.12, N = 31493.021. (CXX) g++ options: -O3 -lsnappy -lpthread

Crafty

Elapsed Time

OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed Time12a32M4M6M8M10MSE +/- 2939.24, N = 3SE +/- 17114.61, N = 3SE +/- 11321.43, N = 37814986782886878363581. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU12a30.30540.61080.91621.22161.527SE +/- 0.00608, N = 3SE +/- 0.00106, N = 3SE +/- 0.00124, N = 31.354731.357211.35578MIN: 1.26MIN: 1.27MIN: 1.271. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU12a30.70071.40142.10212.80283.5035SE +/- 0.00815, N = 3SE +/- 0.00721, N = 3SE +/- 0.00522, N = 33.114153.104813.11243MIN: 3.04MIN: 3.04MIN: 3.041. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU12a30.32050.6410.96151.2821.6025SE +/- 0.01135, N = 15SE +/- 0.01426, N = 3SE +/- 0.02305, N = 31.424481.392111.40518MIN: 1.17MIN: 1.17MIN: 1.151. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU12a30.24650.4930.73950.9861.2325SE +/- 0.00838, N = 15SE +/- 0.00795, N = 3SE +/- 0.00538, N = 31.094081.085031.09566MIN: 0.78MIN: 0.77MIN: 0.781. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU12a30.82511.65022.47533.30044.1255SE +/- 0.00543, N = 3SE +/- 0.00560, N = 3SE +/- 0.00698, N = 33.667133.665013.66413MIN: 3.42MIN: 3.4MIN: 3.421. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU12a30.60481.20961.81442.41923.024SE +/- 0.05729, N = 15SE +/- 0.06186, N = 15SE +/- 0.06290, N = 122.473642.537072.68795MIN: 2.04MIN: 2.05MIN: 2.041. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU12a30.89821.79642.69463.59284.491SE +/- 0.01388, N = 3SE +/- 0.00392, N = 3SE +/- 0.00591, N = 33.961343.991813.94281MIN: 3.85MIN: 3.9MIN: 3.851. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU12a30.3290.6580.9871.3161.645SE +/- 0.01014, N = 3SE +/- 0.01385, N = 3SE +/- 0.01422, N = 31.453151.450031.46202MIN: 1.34MIN: 1.34MIN: 1.341. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU12a30.27390.54780.82171.09561.3695SE +/- 0.00355, N = 3SE +/- 0.00247, N = 3SE +/- 0.00135, N = 31.216391.211611.21715MIN: 1.16MIN: 1.16MIN: 1.151. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU12a30.84761.69522.54283.39044.238SE +/- 0.00882, N = 3SE +/- 0.00174, N = 3SE +/- 0.00878, N = 33.708323.766983.70419MIN: 3.61MIN: 3.66MIN: 3.611. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU12a30.09580.19160.28740.38320.479SE +/- 0.001195, N = 3SE +/- 0.001364, N = 3SE +/- 0.001956, N = 30.4255680.4254700.424975MIN: 0.37MIN: 0.38MIN: 0.381. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU12a30.07870.15740.23610.31480.3935SE +/- 0.002877, N = 3SE +/- 0.003468, N = 3SE +/- 0.002471, N = 30.3491500.3494690.349913MIN: 0.32MIN: 0.33MIN: 0.331. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU12a32004006008001000SE +/- 3.14, N = 3SE +/- 7.10, N = 3SE +/- 2.71, N = 3989.80995.33997.88MIN: 960.49MIN: 974.22MIN: 970.71. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU12a3120240360480600SE +/- 3.54, N = 3SE +/- 4.95, N = 3SE +/- 3.03, N = 3565.55561.35563.12MIN: 550.61MIN: 545.36MIN: 548.721. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU12a32004006008001000SE +/- 1.43, N = 3SE +/- 14.10, N = 3SE +/- 18.50, N = 15981.431011.761007.07MIN: 966.46MIN: 973.13MIN: 963.171. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU12a30.72061.44122.16182.88243.603SE +/- 0.01321, N = 3SE +/- 0.00958, N = 3SE +/- 0.00807, N = 33.192433.196953.20258MIN: 2.96MIN: 2.95MIN: 2.961. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU12a30.98671.97342.96013.94684.9335SE +/- 0.00424, N = 3SE +/- 0.00185, N = 3SE +/- 0.00238, N = 34.383924.385244.38139MIN: 4.24MIN: 4.24MIN: 4.241. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU12a31.00332.00663.00994.01325.0165SE +/- 0.00271, N = 3SE +/- 0.01309, N = 3SE +/- 0.00236, N = 34.447124.458904.44637MIN: 4.39MIN: 4.39MIN: 4.391. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU12a3120240360480600SE +/- 1.36, N = 3SE +/- 1.49, N = 3SE +/- 5.42, N = 13559.66560.97570.74MIN: 549.08MIN: 549.09MIN: 545.081. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU12a30.07580.15160.22740.30320.379SE +/- 0.005221, N = 3SE +/- 0.003594, N = 3SE +/- 0.004807, N = 30.3357550.3370980.336753MIN: 0.3MIN: 0.3MIN: 0.31. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU12a32004006008001000SE +/- 0.37, N = 3SE +/- 13.90, N = 3SE +/- 33.04, N = 15979.571008.341050.41MIN: 967.79MIN: 978.75MIN: 963.151. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU12a3120240360480600SE +/- 4.49, N = 3SE +/- 4.19, N = 3SE +/- 1.65, N = 3565.27563.86562.32MIN: 549.6MIN: 549.3MIN: 549.321. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU12a30.06110.12220.18330.24440.3055SE +/- 0.002247, N = 15SE +/- 0.004018, N = 4SE +/- 0.002990, N = 70.2715510.2681510.270728MIN: 0.24MIN: 0.24MIN: 0.241. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU12a30.18520.37040.55560.74080.926SE +/- 0.002439, N = 3SE +/- 0.002694, N = 3SE +/- 0.001883, N = 30.8172390.8232100.822254MIN: 0.76MIN: 0.76MIN: 0.761. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

rav1e

Speed: 1

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 112a30.07830.15660.23490.31320.3915SE +/- 0.001, N = 3SE +/- 0.002, N = 3SE +/- 0.002, N = 30.3480.3460.346

rav1e

Speed: 5

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 512a30.21510.43020.64530.86041.0755SE +/- 0.001, N = 3SE +/- 0.002, N = 3SE +/- 0.001, N = 30.9370.9560.949

rav1e

Speed: 6

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 612a30.27470.54940.82411.09881.3735SE +/- 0.011, N = 3SE +/- 0.011, N = 3SE +/- 0.013, N = 31.2081.2211.217

rav1e

Speed: 10

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 1012a30.55941.11881.67822.23762.797SE +/- 0.028, N = 3SE +/- 0.033, N = 3SE +/- 0.027, N = 72.4502.4862.455

Stockfish

Total Time

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 12Total Time12a320M40M60M80M100MSE +/- 665117.42, N = 3SE +/- 585345.87, N = 3SE +/- 865487.36, N = 39644753096807159951645731. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver

asmFish

1024 Hash Memory, 26 Depth

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 Depth12a330M60M90M120M150MSE +/- 1433658.44, N = 3SE +/- 1096698.56, N = 3SE +/- 1438873.95, N = 3135035548133208525132978753

Timed Clash Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Clash CompilationTime To Compile12a3110220330440550SE +/- 4.06, N = 3SE +/- 0.83, N = 3SE +/- 1.81, N = 3488.63496.09492.78

ASTC Encoder

Preset: Fast

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Fast12a31.26682.53363.80045.06726.334SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 35.635.605.621. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

ASTC Encoder

Preset: Medium

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Medium12a3246810SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 36.616.556.601. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

ASTC Encoder

Preset: Thorough

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Thorough12a3246810SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 37.167.157.161. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

ASTC Encoder

Preset: Exhaustive

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Exhaustive12a31224364860SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 352.3352.3252.331. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

OpenVINO

Model: Face Detection 0106 FP16 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Face Detection 0106 FP16 - Device: CPU1348121620SE +/- 0.02, N = 3SE +/- 0.02, N = 315.5115.53

OpenVINO

Model: Face Detection 0106 FP16 - Device: CPU

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Face Detection 0106 FP16 - Device: CPU13400800120016002000SE +/- 1.55, N = 3SE +/- 1.08, N = 31784.481786.64

OpenVINO

Model: Face Detection 0106 FP32 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Face Detection 0106 FP32 - Device: CPU1348121620SE +/- 0.09, N = 3SE +/- 0.04, N = 315.1415.19

OpenVINO

Model: Face Detection 0106 FP32 - Device: CPU

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Face Detection 0106 FP32 - Device: CPU13400800120016002000SE +/- 1.30, N = 3SE +/- 1.38, N = 31814.321814.55

OpenVINO

Model: Person Detection 0106 FP16 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Person Detection 0106 FP16 - Device: CPU13246810SE +/- 0.08, N = 10SE +/- 0.04, N = 38.748.84

OpenVINO

Model: Person Detection 0106 FP16 - Device: CPU

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Person Detection 0106 FP16 - Device: CPU137001400210028003500SE +/- 37.41, N = 10SE +/- 19.60, N = 33142.673107.46

OpenVINO

Model: Person Detection 0106 FP32 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Person Detection 0106 FP32 - Device: CPU13246810SE +/- 0.02, N = 3SE +/- 0.03, N = 38.688.72

OpenVINO

Model: Person Detection 0106 FP32 - Device: CPU

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Person Detection 0106 FP32 - Device: CPU137001400210028003500SE +/- 15.31, N = 3SE +/- 3.09, N = 33152.613120.88

OpenVINO

Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU138K16K24K32K40KSE +/- 50.16, N = 3SE +/- 148.89, N = 338683.4838792.26

OpenVINO

Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU130.13950.2790.41850.5580.6975SE +/- 0.00, N = 3SE +/- 0.00, N = 30.620.62

OpenVINO

Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU138K16K24K32K40KSE +/- 80.48, N = 3SE +/- 81.70, N = 338607.5638565.64

OpenVINO

Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU130.13950.2790.41850.5580.6975SE +/- 0.00, N = 3SE +/- 0.00, N = 30.620.62

PHPBench

PHP Benchmark Suite

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark Suite13140K280K420K560K700KSE +/- 1558.00, N = 3SE +/- 351.36, N = 3666624668619

Phoronix Test Suite v10.8.4