Broadwell 2021

Intel Core i7-5600U testing with a LENOVO 20BSCTO1WW (N14ET49W 1.27 BIOS) and Intel HD 5500 3GB on Ubuntu 20.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2101027-HA-BROADWELL87
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:

Audio Encoding 4 Tests
Bioinformatics 2 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 5 Tests
CPU Massive 7 Tests
Creator Workloads 7 Tests
Encoding 4 Tests
HPC - High Performance Computing 4 Tests
Machine Learning 2 Tests
Multi-Core 5 Tests
NVIDIA GPU Compute 5 Tests
Programmer / Developer System Benchmarks 7 Tests
Scientific Computing 2 Tests
Server 4 Tests
Server CPU Tests 2 Tests
Single-Threaded 2 Tests
Vulkan Compute 5 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
Prefer Vertical Bar Graphs

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
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
R1
January 01 2021
  10 Hours, 52 Minutes
2
January 01 2021
  10 Hours, 17 Minutes
3
January 02 2021
  10 Hours, 38 Minutes
Invert Hiding All Results Option
  10 Hours, 36 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):


Broadwell 2021ProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionR123Intel Core i7-5600U @ 3.20GHz (2 Cores / 4 Threads)LENOVO 20BSCTO1WW (N14ET49W 1.27 BIOS)Intel Broadwell-U-OPI8GB128GB SAMSUNG MZNTE128Intel HD 5500 3GB (950MHz)Intel Broadwell-U AudioIntel I218-LM + Intel 7265Ubuntu 20.105.9.1-050901-generic (x86_64)GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.94.6 Mesa 21.0.0-devel (git-bd69765 2021-01-01 groovy-oibaf-ppa)OpenCL 3.01.2.145GCC 10.2.0ext41920x1080OpenBenchmarking.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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,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 Processor Details- Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x2f - Thermald 2.3Security Details- itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable

R123Result OverviewPhoronix Test Suite100%102%104%106%CLOMPTimed MAFFT AlignmentVkResampleSQLite SpeedtestCoremarkNode.js V8 Web Tooling BenchmarkBuild2NCNNoneDNNWarsowTimed Eigen CompilationTimed FFmpeg CompilationBRL-CADVKMarkBetsy GPU CompressorUnpacking FirefoxVkFFTPHPBenchCryptsetupOpus Codec EncodingMonkey Audio EncodingLibplaceboTimed HMMer SearchOgg Audio EncodingWavPack Audio Encodingsimdjson

Broadwell 2021onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUclomp: Static OMP Speedupncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU - blazefaceonednn: IP Shapes 3D - u8s8f32 - CPUncnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - efficientnet-b0vkresample: 2x - Doublemafft: Multiple Sequence Alignment - LSU RNAonednn: IP Shapes 3D - f32 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUncnn: CPU - efficientnet-b0cryptsetup: AES-XTS 256b Decryptionwarsow: 1280 x 1024ncnn: Vulkan GPU - shufflenet-v2ncnn: CPU - mnasnetcryptsetup: AES-XTS 512b Encryptionbuild2: Time To Compileonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUvkmark: 800 x 600cryptsetup: AES-XTS 512b Decryptiononednn: IP Shapes 1D - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUvkmark: 1280 x 1024onednn: Deconvolution Batch shapes_1d - f32 - CPUsqlite-speedtest: Timed Time - Size 1,000onednn: Recurrent Neural Network Inference - u8s8f32 - CPUncnn: Vulkan GPU - alexnetcoremark: CoreMark Size 666 - Iterations Per Secondnode-web-tooling: ncnn: Vulkan GPU-v3-v3 - mobilenet-v3onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUvkmark: 1920 x 1080cryptsetup: AES-XTS 256b Encryptiononednn: Recurrent Neural Network Inference - f32 - CPUncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU-v3-v3 - mobilenet-v3onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUbuild-eigen: Time To Compilencnn: CPU - yolov4-tinyncnn: Vulkan GPU - resnet18vkresample: 2x - Singlencnn: CPU - regnety_400mcryptsetup: Twofish-XTS 256b Encryptionncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: CPU - alexnetncnn: Vulkan GPU - yolov4-tinylibplacebo: av1_grain_lapcryptsetup: Serpent-XTS 512b Decryptionbrl-cad: VGR Performance Metricbuild-ffmpeg: Time To Compilencnn: CPU - shufflenet-v2vkmark: 1024 x 768warsow: 1920 x 1080ncnn: CPU - squeezenet_ssdcryptsetup: Twofish-XTS 512b Encryptiononednn: IP Shapes 1D - f32 - CPUcryptsetup: Serpent-XTS 256b Encryptioncryptsetup: Twofish-XTS 256b Decryptionncnn: CPU - mobilenetbetsy: ETC1 - Highestncnn: Vulkan GPU - squeezenet_ssdbetsy: ETC2 RGB - Highestncnn: CPU - googlenetunpack-firefox: firefox-84.0.source.tar.xzvkfft: ncnn: CPU - blazefacephpbench: PHP Benchmark Suitencnn: Vulkan GPU - resnet50cryptsetup: Serpent-XTS 512b Encryptiononednn: Recurrent Neural Network Training - u8s8f32 - CPUncnn: CPU - resnet18encode-opus: WAV To Opus Encodeonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUncnn: Vulkan GPU - mobilenetncnn: CPU - vgg16onednn: Recurrent Neural Network Training - f32 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUhmmer: Pfam Database Searchcryptsetup: Twofish-XTS 512b Decryptionncnn: CPU - resnet50encode-ape: WAV To APEcryptsetup: PBKDF2-whirlpoollibplacebo: hdr_peakdetectencode-ogg: WAV To Ogglibplacebo: deband_heavycryptsetup: Serpent-XTS 256b Decryptionncnn: Vulkan GPU - vgg16libplacebo: polar_nocomputecryptsetup: PBKDF2-sha512encode-wavpack: WAV To WavPacksimdjson: DistinctUserIDsimdjson: PartialTweetssimdjson: LargeRandsimdjson: KostyaR12310.10941.314.455.657.0195227.8147.3322.46152.68017.00820.908027.796622.501577.470.721.5314.261304.9919.50432.113735.354016761285.113.294733.884069329.149995.58111297.437.0860703.7332897.2412.6911309.14651577.511261.414.5512.6315.4532116.41379.4445.64152.24027.82313.214.5636.8479.35537.11488.714580341.92821.40109645.671.33314.921.0040505.1314.861.4916.14871.2316.18547.1827.46211265.64529100101.55505.220853.045.5110.89420903.361.55153.1820830.526.2014129.006315.0101.1116.84951367232875.8927.95538.65489.0153.0023.73126334818.9460.590.580.330.499.122871.214.135.486.8080927.1346.2621.97155.24417.19620.517227.746322.111586.770.521.3614.051287.5918.46231.695635.637316911289.313.332533.500368529.404594.57711316.936.7060082.6730197.3012.5711320.14611581.411332.214.4412.5615.5448116.86079.0445.41152.37727.72314.314.4737.0278.90534.51490.514562340.63921.34109745.871.03315.721.0875505.4316.061.3816.17770.9716.22147.0127.40411225.62528447101.22506.820912.745.6510.90320853.461.41152.8820852.026.2225129.092315.5101.1716.84351300132911.3327.94238.61488.6152.9823.72126284318.9480.590.580.330.4910.079281.313.985.476.8142327.2546.7022.01156.00817.36920.625127.301022.301559.669.521.2114.161285.9906.24931.969135.803316701273.413.460133.473668929.475295.18711200.436.7660345.6116167.3112.6111214.94631590.411243.514.5312.6515.5638116.03578.8845.73153.24527.64315.214.5436.8179.12534.15487.814504340.20121.30109245.871.26314.421.0772503.4315.961.6116.20871.1816.24447.1627.36411245.63530230101.23506.420847.745.5110.92220858.061.47152.9120812.226.2486129.220315.4101.0116.86851333632916.0827.92438.63488.7152.9323.73126334518.9460.590.580.330.49OpenBenchmarking.org

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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU23R13691215SE +/- 0.02320, N = 3SE +/- 0.13687, N = 4SE +/- 0.04347, N = 39.1228710.0792810.10940MIN: 8.62MIN: 9.14MIN: 9.651. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

CLOMP

CLOMP is the C version of the Livermore OpenMP benchmark developed to measure OpenMP overheads and other performance impacts due to threading in order to influence future system designs. This particular test profile configuration is currently set to look at the OpenMP static schedule speed-up across all available CPU cores using the recommended test configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP Speedup23R10.29250.5850.87751.171.4625SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 121.21.31.31. (CC) gcc options: -fopenmp -O3 -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 20201218Target: Vulkan GPU - Model: mnasnet23R148121620SE +/- 0.47, N = 3SE +/- 0.45, N = 3SE +/- 0.26, N = 314.1313.9814.45MIN: 13.1 / MAX: 14.93MIN: 13.03 / MAX: 14.74MIN: 13.46 / MAX: 16.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazeface23R11.27132.54263.81395.08526.3565SE +/- 0.15, N = 3SE +/- 0.16, N = 3SE +/- 0.01, N = 35.485.475.65MIN: 5.14 / MAX: 5.72MIN: 5.12 / MAX: 5.79MIN: 5.56 / MAX: 7.881. (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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU23R1246810SE +/- 0.09974, N = 3SE +/- 0.10395, N = 15SE +/- 0.06965, N = 156.808096.814237.01952MIN: 6.36MIN: 6.35MIN: 6.611. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 20201218Target: Vulkan GPU - Model: regnety_400m23R1714212835SE +/- 0.60, N = 3SE +/- 0.53, N = 3SE +/- 0.07, N = 327.1327.2527.81MIN: 25.79 / MAX: 37.93MIN: 26 / MAX: 29.97MIN: 26.38 / MAX: 30.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: googlenet23R11122334455SE +/- 0.81, N = 3SE +/- 0.77, N = 3SE +/- 0.16, N = 346.2646.7047.33MIN: 43.6 / MAX: 59.76MIN: 44.45 / MAX: 57.93MIN: 46.92 / MAX: 49.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b023R1510152025SE +/- 0.66, N = 3SE +/- 0.59, N = 3SE +/- 0.30, N = 321.9722.0122.46MIN: 20.55 / MAX: 27.24MIN: 20.61 / MAX: 33.2MIN: 21.52 / MAX: 24.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

VkResample

VkResample is a Vulkan-based image upscaling library based on VkFFT. The sample input file is upscaling a 4K image to 8K using Vulkan-based GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Double23R1306090120150SE +/- 1.45, N = 3SE +/- 0.61, N = 3SE +/- 1.71, N = 3155.24156.01152.681. (CXX) g++ options: -O3 -pthread

Timed MAFFT Alignment

This test performs an alignment of 100 pyruvate decarboxylase sequences. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 7.471Multiple Sequence Alignment - LSU RNA23R148121620SE +/- 0.14, N = 3SE +/- 0.01, N = 3SE +/- 0.11, N = 317.2017.3717.011. (CC) gcc options: -std=c99 -O3 -lm -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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU23R1510152025SE +/- 0.13, N = 3SE +/- 0.05, N = 3SE +/- 0.18, N = 320.5220.6320.91MIN: 20.16MIN: 20.29MIN: 20.391. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU23R1714212835SE +/- 0.09, N = 3SE +/- 0.02, N = 3SE +/- 0.13, N = 327.7527.3027.80MIN: 27.23MIN: 26.98MIN: 27.491. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 20201218Target: CPU - Model: efficientnet-b023R1510152025SE +/- 0.35, N = 3SE +/- 0.39, N = 3SE +/- 0.41, N = 322.1122.3022.50MIN: 20.59 / MAX: 33.72MIN: 21.02 / MAX: 25.56MIN: 20.99 / MAX: 25.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 256b Decryption23R130060090012001500SE +/- 15.21, N = 3SE +/- 11.68, N = 3SE +/- 11.32, N = 31586.71559.61577.4

Warsow

This is a benchmark of Warsow, a popular open-source first-person shooter. This game uses the QFusion engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterWarsow 2.5 BetaResolution: 1280 x 102423R11632486480SE +/- 0.44, N = 3SE +/- 0.63, N = 13SE +/- 0.09, N = 370.569.570.7

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 20201218Target: Vulkan GPU - Model: shufflenet-v223R1510152025SE +/- 0.58, N = 3SE +/- 0.57, N = 3SE +/- 0.37, N = 321.3621.2121.53MIN: 19.98 / MAX: 35.94MIN: 19.96 / MAX: 25.67MIN: 20.55 / MAX: 22.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnet23R148121620SE +/- 0.41, N = 3SE +/- 0.40, N = 3SE +/- 0.37, N = 314.0514.1614.26MIN: 13.01 / MAX: 18.86MIN: 13.06 / MAX: 28.01MIN: 13.24 / MAX: 17.011. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 512b Encryption23R130060090012001500SE +/- 6.69, N = 3SE +/- 2.40, N = 3SE +/- 6.50, N = 31287.51285.91304.9

Build2

This test profile measures the time to bootstrap/install the build2 C++ build toolchain from source. Build2 is a cross-platform build toolchain for C/C++ code and features Cargo-like features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To Compile23R12004006008001000SE +/- 0.55, N = 3SE +/- 0.82, N = 3SE +/- 0.32, N = 3918.46906.25919.50

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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU23R1714212835SE +/- 0.34, N = 3SE +/- 0.27, N = 3SE +/- 0.29, N = 331.7031.9732.11MIN: 30.17MIN: 30.65MIN: 30.061. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU23R1816243240SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 335.6435.8035.35MIN: 33.73MIN: 33.94MIN: 33.411. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

VKMark

VKMark is a collection of Vulkan tests/benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 800 x 60023R1400800120016002000SE +/- 0.88, N = 3SE +/- 10.20, N = 3SE +/- 7.54, N = 31691167016761. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 512b Decryption23R130060090012001500SE +/- 13.32, N = 3SE +/- 6.22, N = 3SE +/- 7.00, N = 31289.31273.41285.1

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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU23R13691215SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 313.3313.4613.29MIN: 13.01MIN: 13.01MIN: 12.991. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU23R1816243240SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.13, N = 333.5033.4733.88MIN: 33.31MIN: 33.25MIN: 33.421. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

VKMark

VKMark is a collection of Vulkan tests/benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 1280 x 102423R1150300450600750SE +/- 2.19, N = 3SE +/- 1.53, N = 3SE +/- 0.67, N = 36856896931. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF

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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU23R1714212835SE +/- 0.29, N = 3SE +/- 0.24, N = 3SE +/- 0.32, N = 329.4029.4829.15MIN: 28MIN: 27.97MIN: 27.381. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

SQLite Speedtest

This is a benchmark of SQLite's speedtest1 benchmark program with an increased problem size of 1,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,00023R120406080100SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.57, N = 394.5895.1995.581. (CC) gcc options: -O2 -ldl -lz -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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU23R12K4K6K8K10KSE +/- 15.94, N = 3SE +/- 18.25, N = 3SE +/- 25.05, N = 311316.911200.411297.4MIN: 11169.5MIN: 11129.9MIN: 11225.51. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 20201218Target: Vulkan GPU - Model: alexnet23R1918273645SE +/- 0.23, N = 3SE +/- 0.12, N = 3SE +/- 0.27, N = 336.7036.7637.08MIN: 35.28 / MAX: 38.63MIN: 35.25 / MAX: 46.5MIN: 34.94 / MAX: 39.371. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per Second23R113K26K39K52K65KSE +/- 445.62, N = 3SE +/- 283.31, N = 3SE +/- 220.80, N = 360082.6760345.6160703.731. (CC) gcc options: -O2 -lrt" -lrt

Node.js V8 Web Tooling Benchmark

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

OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling Benchmark23R1246810SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 37.307.317.241. Nodejs v12.18.2

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 20201218Target: Vulkan GPU-v3-v3 - Model: mobilenet-v323R13691215SE +/- 0.14, N = 3SE +/- 0.14, N = 3SE +/- 0.07, N = 312.5712.6112.69MIN: 12.24 / MAX: 14.08MIN: 12.23 / MAX: 13.54MIN: 12.21 / MAX: 24.641. (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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU23R12K4K6K8K10KSE +/- 44.81, N = 3SE +/- 13.64, N = 3SE +/- 65.97, N = 311320.111214.911309.1MIN: 11234.3MIN: 11171.5MIN: 11154.31. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

VKMark

VKMark is a collection of Vulkan tests/benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 1920 x 108023R1100200300400500SE +/- 0.67, N = 3SE +/- 2.33, N = 3SE +/- 1.76, N = 34614634651. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 256b Encryption23R130060090012001500SE +/- 15.55, N = 3SE +/- 1.02, N = 3SE +/- 19.58, N = 31581.41590.41577.5

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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU23R12K4K6K8K10KSE +/- 15.69, N = 3SE +/- 44.20, N = 3SE +/- 48.62, N = 311332.211243.511261.4MIN: 11227.9MIN: 11158.2MIN: 11146.61. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 20201218Target: CPU-v2-v2 - Model: mobilenet-v223R148121620SE +/- 0.13, N = 3SE +/- 0.18, N = 3SE +/- 0.19, N = 314.4414.5314.55MIN: 14.04 / MAX: 16.1MIN: 13.97 / MAX: 26.42MIN: 14.04 / MAX: 28.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v323R13691215SE +/- 0.13, N = 3SE +/- 0.19, N = 3SE +/- 0.19, N = 312.5612.6512.63MIN: 12.21 / MAX: 15.08MIN: 12.18 / MAX: 14.68MIN: 12.19 / MAX: 14.681. (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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU23R148121620SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 315.5415.5615.45MIN: 14.82MIN: 14.82MIN: 14.571. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Timed Eigen Compilation

This test times how long it takes to build all Eigen examples. The Eigen examples are compiled serially. Eigen is a C++ template library for linear algebra. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To Compile23R1306090120150SE +/- 0.07, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3116.86116.04116.41

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 20201218Target: CPU - Model: yolov4-tiny23R120406080100SE +/- 0.14, N = 3SE +/- 0.14, N = 3SE +/- 0.27, N = 379.0478.8879.44MIN: 78.34 / MAX: 85.24MIN: 78.2 / MAX: 90.69MIN: 78.27 / MAX: 87.181. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet1823R11020304050SE +/- 0.05, N = 3SE +/- 0.29, N = 3SE +/- 0.12, N = 345.4145.7345.64MIN: 45.11 / MAX: 48.07MIN: 45.08 / MAX: 48.32MIN: 45.25 / MAX: 47.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

VkResample

VkResample is a Vulkan-based image upscaling library based on VkFFT. The sample input file is upscaling a 4K image to 8K using Vulkan-based GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Single23R1306090120150SE +/- 1.86, N = 3SE +/- 1.40, N = 3SE +/- 0.49, N = 3152.38153.25152.241. (CXX) g++ options: -O3 -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 20201218Target: CPU - Model: regnety_400m23R1714212835SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 327.7227.6427.82MIN: 27.48 / MAX: 31MIN: 27.49 / MAX: 29.98MIN: 27.57 / MAX: 37.341. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 256b Encryption23R170140210280350SE +/- 0.43, N = 3SE +/- 0.35, N = 3SE +/- 0.84, N = 3314.3315.2313.2

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 20201218Target: Vulkan GPU-v2-v2 - Model: mobilenet-v223R148121620SE +/- 0.15, N = 3SE +/- 0.20, N = 3SE +/- 0.14, N = 314.4714.5414.56MIN: 14.06 / MAX: 28.34MIN: 14.01 / MAX: 17.8MIN: 14.07 / MAX: 17.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnet23R1918273645SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.18, N = 337.0236.8136.84MIN: 35.32 / MAX: 103.85MIN: 34.96 / MAX: 79.9MIN: 35.27 / MAX: 39.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tiny23R120406080100SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 378.9079.1279.35MIN: 78.21 / MAX: 91.76MIN: 78.31 / MAX: 86.73MIN: 78.46 / MAX: 91.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 2.72.2Test: av1_grain_lap23R1120240360480600SE +/- 1.77, N = 3SE +/- 1.17, N = 3SE +/- 0.34, N = 3534.51534.15537.111. (CXX) g++ options: -lm -lglslang -lHLSL -lOGLCompiler -lOSDependent -lSPIRV -lSPVRemapper -lSPIRV-Tools -lSPIRV-Tools-opt -lpthread -pthread -pipe -std=c++11 -fvisibility=hidden -fPIC -MD -MQ -MF

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 512b Decryption23R1110220330440550SE +/- 1.01, N = 3SE +/- 0.51, N = 3SE +/- 1.22, N = 3490.5487.8488.7

BRL-CAD

BRL-CAD 7.28.0 is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.30.8VGR Performance Metric23R13K6K9K12K15K1456214504145801. (CXX) g++ options: -std=c++11 -pipe -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -rdynamic -lSM -lICE -lXi -lGLU -lGL -lGLdispatch -lX11 -lXext -lXrender -lpthread -ldl -luuid -lm

Timed FFmpeg Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.2.2Time To Compile23R170140210280350SE +/- 0.15, N = 3SE +/- 0.31, N = 3SE +/- 0.25, N = 3340.64340.20341.93

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 20201218Target: CPU - Model: shufflenet-v223R1510152025SE +/- 0.63, N = 3SE +/- 0.57, N = 3SE +/- 0.57, N = 321.3421.3021.40MIN: 20 / MAX: 24.44MIN: 20.01 / MAX: 23.24MIN: 20.12 / MAX: 23.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

VKMark

VKMark is a collection of Vulkan tests/benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 1024 x 76823R12004006008001000SE +/- 6.11, N = 3SE +/- 4.18, N = 3SE +/- 6.69, N = 31097109210961. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF

Warsow

This is a benchmark of Warsow, a popular open-source first-person shooter. This game uses the QFusion engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterWarsow 2.5 BetaResolution: 1920 x 108023R11020304050SE +/- 0.13, N = 3SE +/- 0.20, N = 3SE +/- 0.13, N = 345.845.845.6

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 20201218Target: CPU - Model: squeezenet_ssd23R11632486480SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 371.0371.2671.33MIN: 70.28 / MAX: 76.92MIN: 70.52 / MAX: 78.34MIN: 70.87 / MAX: 74.751. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 512b Encryption23R170140210280350SE +/- 0.57, N = 3SE +/- 0.15, N = 3SE +/- 0.73, N = 3315.7314.4314.9

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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU23R1510152025SE +/- 0.27, N = 3SE +/- 0.23, N = 3SE +/- 0.31, N = 321.0921.0821.00MIN: 20.08MIN: 19.85MIN: 19.611. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 256b Encryption23R1110220330440550SE +/- 0.92, N = 3SE +/- 1.12, N = 3SE +/- 0.99, N = 3505.4503.4505.1

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 256b Decryption23R170140210280350SE +/- 0.26, N = 3SE +/- 0.86, N = 3SE +/- 0.48, N = 3316.0315.9314.8

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 20201218Target: CPU - Model: mobilenet23R11428425670SE +/- 0.06, N = 3SE +/- 0.16, N = 3SE +/- 0.08, N = 361.3861.6161.49MIN: 61.05 / MAX: 87.54MIN: 61.11 / MAX: 113.17MIN: 61.05 / MAX: 63.531. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Betsy GPU Compressor

Betsy is an open-source GPU compressor of various GPU compression techniques. Betsy is written in GLSL for Vulkan/OpenGL (compute shader) support for GPU-based texture compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC1 - Quality: Highest23R148121620SE +/- 0.23, N = 4SE +/- 0.27, N = 3SE +/- 0.14, N = 1316.1816.2116.151. (CXX) g++ options: -O3 -O2 -lpthread -ldl

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 20201218Target: Vulkan GPU - Model: squeezenet_ssd23R11632486480SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.10, N = 370.9771.1871.23MIN: 70.58 / MAX: 80.98MIN: 70.63 / MAX: 78.41MIN: 70.8 / MAX: 121.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Betsy GPU Compressor

Betsy is an open-source GPU compressor of various GPU compression techniques. Betsy is written in GLSL for Vulkan/OpenGL (compute shader) support for GPU-based texture compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC2 RGB - Quality: Highest23R148121620SE +/- 0.26, N = 3SE +/- 0.28, N = 3SE +/- 0.20, N = 1316.2216.2416.191. (CXX) g++ options: -O3 -O2 -lpthread -ldl

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 20201218Target: CPU - Model: googlenet23R11122334455SE +/- 0.03, N = 3SE +/- 0.12, N = 3SE +/- 0.09, N = 347.0147.1647.18MIN: 46.74 / MAX: 49.94MIN: 46.7 / MAX: 59.52MIN: 46.85 / MAX: 56.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Unpacking Firefox

This simple test profile measures how long it takes to extract the .tar.xz source package of the Mozilla Firefox Web Browser. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterUnpacking Firefox 84.0Extracting: firefox-84.0.source.tar.xz23R1612182430SE +/- 0.32, N = 6SE +/- 0.16, N = 19SE +/- 0.33, N = 627.4027.3627.46

VkFFT

VkFFT is a Fast Fourier Transform (FFT) Library that is GPU accelerated by means of the Vulkan API. The VkFFT benchmark runs FFT performance differences of many different sizes before returning an overall benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.123R12004006008001000SE +/- 2.00, N = 3SE +/- 1.45, N = 3SE +/- 2.52, N = 31122112411261. (CXX) g++ options: -O3 -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 20201218Target: CPU - Model: blazeface23R11.2692.5383.8075.0766.345SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 35.625.635.64MIN: 5.54 / MAX: 6.01MIN: 5.5 / MAX: 5.84MIN: 5.57 / MAX: 5.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

PHPBench

PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark Suite23R1110K220K330K440K550KSE +/- 388.18, N = 3SE +/- 222.67, N = 3SE +/- 498.65, N = 3528447530230529100

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 20201218Target: Vulkan GPU - Model: resnet5023R120406080100SE +/- 0.21, N = 3SE +/- 0.28, N = 3SE +/- 0.16, N = 3101.22101.23101.55MIN: 100.01 / MAX: 110.27MIN: 100.55 / MAX: 114.34MIN: 100.72 / MAX: 112.981. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 512b Encryption23R1110220330440550SE +/- 0.47, N = 3SE +/- 1.01, N = 3SE +/- 0.87, N = 3506.8506.4505.2

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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU23R14K8K12K16K20KSE +/- 36.91, N = 3SE +/- 5.57, N = 3SE +/- 16.92, N = 320912.720847.720853.0MIN: 20792.3MIN: 20792.3MIN: 20801.41. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 20201218Target: CPU - Model: resnet1823R11020304050SE +/- 0.13, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 345.6545.5145.51MIN: 45.26 / MAX: 56.06MIN: 45.23 / MAX: 48.14MIN: 45.16 / MAX: 47.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Opus Codec Encoding

Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus Encode23R13691215SE +/- 0.02, N = 5SE +/- 0.04, N = 5SE +/- 0.03, N = 510.9010.9210.891. (CXX) g++ options: -fvisibility=hidden -logg -lm

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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU23R14K8K12K16K20KSE +/- 9.04, N = 3SE +/- 3.46, N = 3SE +/- 20.57, N = 320853.420858.020903.3MIN: 20778.6MIN: 20799.8MIN: 20835.51. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 20201218Target: Vulkan GPU - Model: mobilenet23R11428425670SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 361.4161.4761.55MIN: 61.02 / MAX: 63.86MIN: 61.15 / MAX: 64.55MIN: 61.05 / MAX: 64.21. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg1623R1306090120150SE +/- 0.24, N = 3SE +/- 0.19, N = 3SE +/- 0.18, N = 3152.88152.91153.18MIN: 151.28 / MAX: 160.12MIN: 151.95 / MAX: 164.13MIN: 152.21 / MAX: 161.011. (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 initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU23R14K8K12K16K20KSE +/- 51.92, N = 3SE +/- 37.82, N = 3SE +/- 29.30, N = 320852.020812.220830.5MIN: 20691.5MIN: 20694.5MIN: 20740.91. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU23R1612182430SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 326.2226.2526.20MIN: 26.12MIN: 26.08MIN: 26.091. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Timed HMMer Search

This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database Search23R1306090120150SE +/- 0.16, N = 3SE +/- 0.13, N = 3SE +/- 0.14, N = 3129.09129.22129.011. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 512b Decryption23R170140210280350SE +/- 1.09, N = 3SE +/- 0.48, N = 3SE +/- 0.58, N = 3315.5315.4315.0

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 20201218Target: CPU - Model: resnet5023R120406080100SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 3101.17101.01101.11MIN: 100.65 / MAX: 111.1MIN: 100.5 / MAX: 114.21MIN: 100.49 / MAX: 113.371. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Monkey Audio Encoding

This test times how long it takes to encode a sample WAV file to Monkey's Audio APE format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMonkey Audio Encoding 3.99.6WAV To APE23R148121620SE +/- 0.06, N = 5SE +/- 0.04, N = 5SE +/- 0.06, N = 516.8416.8716.851. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Second, More Is BetterCryptsetupPBKDF2-whirlpool23R1110K220K330K440K550KSE +/- 335.33, N = 3SE +/- 335.33, N = 3513001513336513672

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 2.72.2Test: hdr_peakdetect23R17K14K21K28K35KSE +/- 358.35, N = 3SE +/- 401.65, N = 3SE +/- 290.63, N = 332911.3332916.0832875.891. (CXX) g++ options: -lm -lglslang -lHLSL -lOGLCompiler -lOSDependent -lSPIRV -lSPVRemapper -lSPIRV-Tools -lSPIRV-Tools-opt -lpthread -pthread -pipe -std=c++11 -fvisibility=hidden -fPIC -MD -MQ -MF

Ogg Audio Encoding

This test times how long it takes to encode a sample WAV file to Ogg format using the reference Xiph.org tools/libraries. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOgg Audio Encoding 1.3.4WAV To Ogg23R1714212835SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 327.9427.9227.961. (CC) gcc options: -O2 -ffast-math -fsigned-char

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 2.72.2Test: deband_heavy23R1918273645SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 338.6138.6338.651. (CXX) g++ options: -lm -lglslang -lHLSL -lOGLCompiler -lOSDependent -lSPIRV -lSPVRemapper -lSPIRV-Tools -lSPIRV-Tools-opt -lpthread -pthread -pipe -std=c++11 -fvisibility=hidden -fPIC -MD -MQ -MF

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 256b Decryption23R1110220330440550SE +/- 1.59, N = 3SE +/- 1.51, N = 3SE +/- 0.20, N = 3488.6488.7489.0

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 20201218Target: Vulkan GPU - Model: vgg1623R1306090120150SE +/- 0.16, N = 3SE +/- 0.31, N = 3SE +/- 0.07, N = 3152.98152.93153.00MIN: 152.01 / MAX: 164.13MIN: 151.38 / MAX: 166.09MIN: 151.95 / MAX: 164.371. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 2.72.2Test: polar_nocompute23R1612182430SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 323.7223.7323.731. (CXX) g++ options: -lm -lglslang -lHLSL -lOGLCompiler -lOSDependent -lSPIRV -lSPVRemapper -lSPIRV-Tools -lSPIRV-Tools-opt -lpthread -pthread -pipe -std=c++11 -fvisibility=hidden -fPIC -MD -MQ -MF

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Second, More Is BetterCryptsetupPBKDF2-sha51223R1300K600K900K1200K1500KSE +/- 2024.67, N = 3SE +/- 878.73, N = 3SE +/- 1520.33, N = 3126284312633451263348

WavPack Audio Encoding

This test times how long it takes to encode a sample WAV file to WavPack format with very high quality settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPack23R1510152025SE +/- 0.00, N = 5SE +/- 0.00, N = 5SE +/- 0.00, N = 518.9518.9518.951. (CXX) g++ options: -rdynamic

simdjson

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

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: DistinctUserID23R10.13280.26560.39840.53120.664SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.590.590.591. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: PartialTweets23R10.13050.2610.39150.5220.6525SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.580.580.581. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandom23R10.07430.14860.22290.29720.3715SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.330.330.331. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: Kostya23R10.11030.22060.33090.44120.5515SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.490.490.491. (CXX) g++ options: -O3 -pthread

97 Results Shown

oneDNN
CLOMP
NCNN:
  Vulkan GPU - mnasnet
  Vulkan GPU - blazeface
oneDNN
NCNN:
  Vulkan GPU - regnety_400m
  Vulkan GPU - googlenet
  Vulkan GPU - efficientnet-b0
VkResample
Timed MAFFT Alignment
oneDNN:
  IP Shapes 3D - f32 - CPU
  Convolution Batch Shapes Auto - f32 - CPU
NCNN
Cryptsetup
Warsow
NCNN:
  Vulkan GPU - shufflenet-v2
  CPU - mnasnet
Cryptsetup
Build2
oneDNN:
  Deconvolution Batch shapes_1d - u8s8f32 - CPU
  Deconvolution Batch shapes_3d - f32 - CPU
VKMark
Cryptsetup
oneDNN:
  IP Shapes 1D - u8s8f32 - CPU
  Convolution Batch Shapes Auto - u8s8f32 - CPU
VKMark
oneDNN
SQLite Speedtest
oneDNN
NCNN
Coremark
Node.js V8 Web Tooling Benchmark
NCNN
oneDNN
VKMark
Cryptsetup
oneDNN
NCNN:
  CPU-v2-v2 - mobilenet-v2
  CPU-v3-v3 - mobilenet-v3
oneDNN
Timed Eigen Compilation
NCNN:
  CPU - yolov4-tiny
  Vulkan GPU - resnet18
VkResample
NCNN
Cryptsetup
NCNN:
  Vulkan GPU-v2-v2 - mobilenet-v2
  CPU - alexnet
  Vulkan GPU - yolov4-tiny
Libplacebo
Cryptsetup
BRL-CAD
Timed FFmpeg Compilation
NCNN
VKMark
Warsow
NCNN
Cryptsetup
oneDNN
Cryptsetup:
  Serpent-XTS 256b Encryption
  Twofish-XTS 256b Decryption
NCNN
Betsy GPU Compressor
NCNN
Betsy GPU Compressor
NCNN
Unpacking Firefox
VkFFT
NCNN
PHPBench
NCNN
Cryptsetup
oneDNN
NCNN
Opus Codec Encoding
oneDNN
NCNN:
  Vulkan GPU - mobilenet
  CPU - vgg16
oneDNN:
  Recurrent Neural Network Training - f32 - CPU
  Deconvolution Batch shapes_3d - u8s8f32 - CPU
Timed HMMer Search
Cryptsetup
NCNN
Monkey Audio Encoding
Cryptsetup
Libplacebo
Ogg Audio Encoding
Libplacebo
Cryptsetup
NCNN
Libplacebo
Cryptsetup
WavPack Audio Encoding
simdjson:
  DistinctUserID
  PartialTweets
  LargeRand
  Kostya