Core i7 4770K Xmas

Intel Core i7-4770K testing with a Gigabyte Z97-HD3 (F10c BIOS) and Gigabyte Intel HD 4600 2GB 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 2012256-HA-COREI747702
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

Limit displaying results to tests within:

Audio Encoding 3 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 4 Tests
CPU Massive 4 Tests
Creator Workloads 5 Tests
Encoding 3 Tests
HPC - High Performance Computing 3 Tests
Machine Learning 2 Tests
Multi-Core 5 Tests
Programmer / Developer System Benchmarks 6 Tests
Server 3 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
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
1
December 25 2020
  5 Hours, 48 Minutes
2
December 25 2020
  5 Hours, 39 Minutes
3
December 25 2020
  5 Hours, 54 Minutes
Invert Hiding All Results Option
  5 Hours, 47 Minutes

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

1

Compiler Notes: --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 Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x28 - Thermald 2.3
Security Notes: 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: Not affected

2

3

Processor: Intel Core i7-4770K @ 3.90GHz (4 Cores / 8 Threads), Motherboard: Gigabyte Z97-HD3 (F10c BIOS), Chipset: Intel 4th Gen Core DRAM, Memory: 8GB, Disk: 120GB ADATA SU700, Graphics: Gigabyte Intel HD 4600 2GB (1250MHz), Audio: Intel Xeon E3-1200 v3/4th, Monitor: DELL S2409W, Network: Realtek RTL8111/8168/8411

OS: Ubuntu 20.10, Kernel: 5.8.0-31-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.5 Mesa 20.2.1, Vulkan: 1.2.145, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1080

Core i7 4770K XmasProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen Resolution123Intel Core i7-4770K @ 3.90GHz (4 Cores / 8 Threads)Gigabyte Z97-HD3 (F10c BIOS)Intel 4th Gen Core DRAM8GB120GB ADATA SU700Gigabyte Intel HD 4600 2GB (1250MHz)Intel Xeon E3-1200 v3/4thDELL S2409WRealtek RTL8111/8168/8411Ubuntu 20.105.8.0-31-generic (x86_64)GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.94.5 Mesa 20.2.11.2.145GCC 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: 0x28 - Thermald 2.3 Security 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: Not affected

123Result OverviewPhoronix Test Suite100%106%111%117%122%CLOMPoneDNNSQLite SpeedtestNCNNMonkey Audio EncodingBuild2Timed FFmpeg CompilationVKMarkTimed Eigen CompilationNode.js V8 Web Tooling BenchmarkBRL-CADCoremarkOpus Codec EncodingTimed HMMer SearchWavPack Audio Encodingsimdjson

Core i7 4770K Xmasonednn: IP Shapes 3D - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUncnn: CPU-v2-v2 - mobilenet-v2onednn: Recurrent Neural Network Inference - u8s8f32 - CPUncnn: CPU - yolov4-tinyncnn: Vulkan GPU - squeezenet_ssdonednn: Recurrent Neural Network Training - u8s8f32 - CPUncnn: CPU - blazefaceonednn: IP Shapes 1D - f32 - CPUncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: CPU - resnet50ncnn: CPU - mnasnetonednn: Recurrent Neural Network Inference - f32 - CPUncnn: CPU - squeezenet_ssdncnn: Vulkan GPU - yolov4-tinyncnn: CPU - resnet18ncnn: CPU-v3-v3 - mobilenet-v3onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUncnn: CPU - googlenetonednn: Recurrent Neural Network Training - f32 - CPUncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - blazefacencnn: Vulkan GPU - alexnetonednn: Deconvolution Batch shapes_1d - f32 - CPUsqlite-speedtest: Timed Time - Size 1,000ncnn: CPU - regnety_400mncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - mnasnetonednn: Convolution Batch Shapes Auto - f32 - CPUncnn: Vulkan GPU-v3-v3 - mobilenet-v3onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUencode-ape: WAV To APEncnn: CPU - mobilenetncnn: CPU - efficientnet-b0ncnn: Vulkan GPU - mobilenetncnn: Vulkan GPU - efficientnet-b0onednn: IP Shapes 1D - u8s8f32 - CPUbuild2: Time To Compilencnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - shufflenet-v2onednn: Deconvolution Batch shapes_3d - f32 - CPUncnn: Vulkan GPU - resnet18build-ffmpeg: Time To Compilevkmark: 1920 x 1080build-eigen: Time To Compilenode-web-tooling: ncnn: Vulkan GPU - resnet50brl-cad: VGR Performance Metricncnn: CPU - vgg16onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUncnn: CPU - alexnetncnn: CPU - shufflenet-v2coremark: CoreMark Size 666 - Iterations Per Secondencode-opus: WAV To Opus Encodehmmer: Pfam Database Searchencode-wavpack: WAV To WavPacksimdjson: DistinctUserIDsimdjson: PartialTweetssimdjson: LargeRandsimdjson: Kostyaclomp: Static OMP Speedup12315.91428.104354.7261310.425730.7251.5840.2710639.43.2911.173410.5661.858.375784.0440.6151.8330.488.7414.41915907.3010641.529.6510419.229.813.2525.0913.549780.18620.69126.488.5332.11988.777.3295114.05939.3014.2639.2414.405.97793388.39320.7011.5118.926530.47145.94229296.6569.3063.2142373127.4612.956131.026525.1011.41144430.9675179.132137.77415.5480.680.660.40.600.915.65138.097664.6935510.665954.5252.9841.2710420.83.1711.361710.5063.278.425922.9940.3053.0531.428.6614.83075822.1310647.030.4310472.430.533.1825.6213.355781.52420.55128.348.4132.12618.727.3356713.87339.2914.3239.5514.285.94146385.58420.7511.4218.953530.64146.29529297.2899.2963.5542365127.5512.954230.925025.1611.42144054.7592079.133137.93915.5440.680.660.40.601.118.89288.854434.9978611.046041.9054.2342.1010853.53.2811.593310.8964.108.655971.4441.5753.4530.758.9114.56725981.8010925.130.1910683.530.373.2425.1113.292280.10020.90127.978.4332.57808.847.2363113.95339.7714.4339.6414.265.99872388.95320.8811.4919.073730.69146.97829097.2999.2463.3442169128.0212.901930.947425.1711.44144381.8916419.113137.90315.5620.680.660.40.601.1OpenBenchmarking.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: IP Shapes 3D - Data Type: f32 - Engine: CPU123510152025SE +/- 0.19, N = 3SE +/- 0.23, N = 3SE +/- 0.18, N = 915.9115.6518.89MIN: 14.84MIN: 14.56MIN: 17.561. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU123510152025Min: 15.58 / Avg: 15.91 / Max: 16.23Min: 15.2 / Avg: 15.65 / Max: 15.95Min: 18.15 / Avg: 18.89 / Max: 19.841. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU123246810SE +/- 0.02583, N = 3SE +/- 0.00568, N = 3SE +/- 0.08022, N = 108.104358.097668.85443MIN: 7.33MIN: 7.36MIN: 7.411. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU1233691215Min: 8.07 / Avg: 8.1 / Max: 8.15Min: 8.09 / Avg: 8.1 / Max: 8.11Min: 8.14 / Avg: 8.85 / Max: 8.991. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU1231.12452.2493.37354.4985.6225SE +/- 0.00568, N = 3SE +/- 0.01254, N = 3SE +/- 0.00347, N = 34.726134.693554.99786MIN: 4.17MIN: 4.15MIN: 4.391. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU123246810Min: 4.72 / Avg: 4.73 / Max: 4.74Min: 4.67 / Avg: 4.69 / Max: 4.71Min: 4.99 / Avg: 5 / Max: 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: CPU-v2-v2 - Model: mobilenet-v21233691215SE +/- 0.01, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 310.4210.6611.04MIN: 9 / MAX: 26.06MIN: 9.15 / MAX: 22.02MIN: 9.49 / MAX: 23.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v21233691215Min: 10.41 / Avg: 10.42 / Max: 10.44Min: 10.48 / Avg: 10.66 / Max: 10.89Min: 10.92 / Avg: 11.04 / Max: 11.251. (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: u8s8f32 - Engine: CPU12313002600390052006500SE +/- 34.96, N = 3SE +/- 28.84, N = 3SE +/- 103.03, N = 35730.725954.526041.90MIN: 5610.11MIN: 5716.33MIN: 5795.411. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU12310002000300040005000Min: 5669.98 / Avg: 5730.72 / Max: 5791.09Min: 5920.58 / Avg: 5954.52 / Max: 6011.87Min: 5900.82 / Avg: 6041.9 / Max: 6242.51. (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: CPU - Model: yolov4-tiny1231224364860SE +/- 0.60, N = 3SE +/- 0.61, N = 3SE +/- 0.49, N = 351.5852.9854.23MIN: 49.32 / MAX: 73.38MIN: 49.76 / MAX: 62.27MIN: 51.36 / MAX: 69.511. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tiny1231122334455Min: 50.82 / Avg: 51.58 / Max: 52.77Min: 52.02 / Avg: 52.98 / Max: 54.11Min: 53.39 / Avg: 54.23 / Max: 55.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssd1231020304050SE +/- 0.38, N = 3SE +/- 0.71, N = 3SE +/- 0.42, N = 340.2741.2742.10MIN: 38.86 / MAX: 55.4MIN: 38.91 / MAX: 54.13MIN: 39.99 / MAX: 60.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssd123918273645Min: 39.75 / Avg: 40.27 / Max: 41Min: 39.98 / Avg: 41.27 / Max: 42.42Min: 41.57 / Avg: 42.1 / Max: 42.941. (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: u8s8f32 - Engine: CPU1232K4K6K8K10KSE +/- 126.69, N = 3SE +/- 55.37, N = 3SE +/- 89.46, N = 310639.410420.810853.5MIN: 10142.8MIN: 10122.1MIN: 10230.81. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU1232K4K6K8K10KMin: 10508.2 / Avg: 10639.37 / Max: 10892.7Min: 10351.1 / Avg: 10420.83 / Max: 10530.2Min: 10701.9 / Avg: 10853.5 / Max: 11011.61. (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 - Model: blazeface1230.74031.48062.22092.96123.7015SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 33.293.173.28MIN: 2.93 / MAX: 5.59MIN: 2.84 / MAX: 5.58MIN: 2.87 / MAX: 6.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazeface123246810Min: 3.19 / Avg: 3.29 / Max: 3.37Min: 3.14 / Avg: 3.17 / Max: 3.21Min: 3.14 / Avg: 3.28 / Max: 3.361. (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 1D - Data Type: f32 - Engine: CPU1233691215SE +/- 0.05, N = 3SE +/- 0.12, N = 3SE +/- 0.02, N = 311.1711.3611.59MIN: 9.59MIN: 9.64MIN: 9.811. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU1233691215Min: 11.11 / Avg: 11.17 / Max: 11.26Min: 11.17 / Avg: 11.36 / Max: 11.57Min: 11.57 / Avg: 11.59 / Max: 11.631. (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-v2-v2 - Model: mobilenet-v21233691215SE +/- 0.07, N = 3SE +/- 0.13, N = 3SE +/- 0.09, N = 310.5610.5010.89MIN: 9.17 / MAX: 22.12MIN: 9.06 / MAX: 19.24MIN: 9.36 / MAX: 27.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v2-v2 - Model: mobilenet-v21233691215Min: 10.47 / Avg: 10.56 / Max: 10.7Min: 10.28 / Avg: 10.5 / Max: 10.72Min: 10.73 / Avg: 10.89 / Max: 11.041. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet501231428425670SE +/- 0.32, N = 3SE +/- 0.36, N = 3SE +/- 0.63, N = 361.8563.2764.10MIN: 59.67 / MAX: 77.54MIN: 60.14 / MAX: 79.04MIN: 60.8 / MAX: 78.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet501231326395265Min: 61.47 / Avg: 61.85 / Max: 62.48Min: 62.74 / Avg: 63.27 / Max: 63.96Min: 63.35 / Avg: 64.1 / Max: 65.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnet123246810SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.14, N = 38.378.428.65MIN: 7.43 / MAX: 11.53MIN: 7.43 / MAX: 10.7MIN: 7.52 / MAX: 22.411. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnet1233691215Min: 8.32 / Avg: 8.37 / Max: 8.41Min: 8.36 / Avg: 8.42 / Max: 8.54Min: 8.42 / Avg: 8.65 / Max: 8.91. (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: f32 - Engine: CPU12313002600390052006500SE +/- 67.54, N = 3SE +/- 39.11, N = 3SE +/- 76.96, N = 35784.045922.995971.44MIN: 5574.2MIN: 5689.07MIN: 5746.721. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU12310002000300040005000Min: 5655.33 / Avg: 5784.04 / Max: 5883.87Min: 5845.15 / Avg: 5922.99 / Max: 5968.63Min: 5889.27 / Avg: 5971.44 / Max: 6125.231. (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: squeezenet_ssd123918273645SE +/- 0.38, N = 3SE +/- 0.31, N = 3SE +/- 0.08, N = 340.6140.3041.57MIN: 38.69 / MAX: 57.91MIN: 38.88 / MAX: 50.48MIN: 39.79 / MAX: 51.011. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssd123918273645Min: 39.94 / Avg: 40.61 / Max: 41.24Min: 39.88 / Avg: 40.3 / Max: 40.91Min: 41.43 / Avg: 41.57 / Max: 41.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tiny1231224364860SE +/- 0.34, N = 3SE +/- 0.51, N = 3SE +/- 0.37, N = 351.8353.0553.45MIN: 49.56 / MAX: 65.5MIN: 49.64 / MAX: 66.59MIN: 50.95 / MAX: 68.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tiny1231122334455Min: 51.18 / Avg: 51.83 / Max: 52.31Min: 52.18 / Avg: 53.05 / Max: 53.94Min: 52.75 / Avg: 53.45 / Max: 54.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18123714212835SE +/- 0.48, N = 3SE +/- 0.35, N = 3SE +/- 0.49, N = 330.4831.4230.75MIN: 28.51 / MAX: 48.91MIN: 29.38 / MAX: 44.46MIN: 28.65 / MAX: 43.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18123714212835Min: 29.53 / Avg: 30.48 / Max: 30.99Min: 30.74 / Avg: 31.42 / Max: 31.93Min: 29.76 / Avg: 30.75 / Max: 31.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v3123246810SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 38.748.668.91MIN: 7.43 / MAX: 21.86MIN: 7.66 / MAX: 11.93MIN: 7.68 / MAX: 24.141. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v31233691215Min: 8.65 / Avg: 8.74 / Max: 8.83Min: 8.62 / Avg: 8.66 / Max: 8.72Min: 8.83 / Avg: 8.91 / Max: 9.071. (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: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU12348121620SE +/- 0.08, N = 3SE +/- 0.16, N = 7SE +/- 0.19, N = 314.4214.8314.57MIN: 12.59MIN: 12.64MIN: 12.661. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU12348121620Min: 14.27 / Avg: 14.42 / Max: 14.53Min: 14.34 / Avg: 14.83 / Max: 15.42Min: 14.35 / Avg: 14.57 / Max: 14.951. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU12313002600390052006500SE +/- 36.51, N = 3SE +/- 36.61, N = 3SE +/- 45.12, N = 35907.305822.135981.80MIN: 5685.76MIN: 5676.41MIN: 5790.741. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU12310002000300040005000Min: 5838.18 / Avg: 5907.3 / Max: 5962.26Min: 5783.82 / Avg: 5822.13 / Max: 5895.31Min: 5925.2 / Avg: 5981.8 / Max: 6070.971. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU1232K4K6K8K10KSE +/- 84.33, N = 3SE +/- 79.82, N = 3SE +/- 134.60, N = 310641.510647.010925.1MIN: 10074.8MIN: 10136.8MIN: 10351.41. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU1232K4K6K8K10KMin: 10486 / Avg: 10641.5 / Max: 10775.8Min: 10494.1 / Avg: 10647 / Max: 10763.2Min: 10658.7 / Avg: 10925.13 / Max: 11091.71. (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: googlenet123714212835SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.38, N = 329.6530.4330.19MIN: 27.35 / MAX: 44.36MIN: 27.97 / MAX: 44.24MIN: 27.95 / MAX: 43.651. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenet123714212835Min: 29.45 / Avg: 29.65 / Max: 29.86Min: 30.34 / Avg: 30.43 / Max: 30.53Min: 29.69 / Avg: 30.19 / Max: 30.931. (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: CPU1232K4K6K8K10KSE +/- 107.95, N = 3SE +/- 143.78, N = 3SE +/- 103.29, N = 310419.210472.410683.5MIN: 10130.6MIN: 10130.6MIN: 10271.51. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU1232K4K6K8K10KMin: 10206.1 / Avg: 10419.23 / Max: 10555.6Min: 10233.3 / Avg: 10472.4 / Max: 10730.3Min: 10477.4 / Avg: 10683.53 / Max: 10798.31. (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: googlenet123714212835SE +/- 0.08, N = 3SE +/- 0.15, N = 3SE +/- 0.30, N = 329.8130.5330.37MIN: 27.66 / MAX: 43.24MIN: 27.72 / MAX: 49.16MIN: 27.85 / MAX: 41.191. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: googlenet123714212835Min: 29.66 / Avg: 29.81 / Max: 29.91Min: 30.24 / Avg: 30.53 / Max: 30.77Min: 30.07 / Avg: 30.37 / Max: 30.961. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazeface1230.73131.46262.19392.92523.6565SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 33.253.183.24MIN: 2.92 / MAX: 5.82MIN: 2.85 / MAX: 7.23MIN: 2.74 / MAX: 13.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazeface123246810Min: 3.19 / Avg: 3.25 / Max: 3.33Min: 3.14 / Avg: 3.18 / Max: 3.21Min: 3.19 / Avg: 3.24 / Max: 3.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: alexnet123612182430SE +/- 0.03, N = 3SE +/- 0.30, N = 3SE +/- 0.03, N = 325.0925.6225.11MIN: 24.07 / MAX: 34.98MIN: 24.03 / MAX: 37.26MIN: 23.93 / MAX: 31.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: alexnet123612182430Min: 25.03 / Avg: 25.09 / Max: 25.15Min: 25.06 / Avg: 25.62 / Max: 26.07Min: 25.08 / Avg: 25.11 / Max: 25.161. (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: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU1233691215SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 313.5513.3613.29MIN: 11.76MIN: 11.47MIN: 11.341. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU12348121620Min: 13.5 / Avg: 13.55 / Max: 13.64Min: 13.33 / Avg: 13.36 / Max: 13.4Min: 13.25 / Avg: 13.29 / Max: 13.371. (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 1,00012320406080100SE +/- 0.66, N = 3SE +/- 0.42, N = 3SE +/- 0.67, N = 380.1981.5280.101. (CC) gcc options: -O2 -ldl -lz -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,0001231632486480Min: 79.05 / Avg: 80.19 / Max: 81.34Min: 80.86 / Avg: 81.52 / Max: 82.29Min: 79.17 / Avg: 80.1 / Max: 81.411. (CC) gcc options: -O2 -ldl -lz -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: regnety_400m123510152025SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 0.26, N = 320.6920.5520.90MIN: 19.76 / MAX: 32.47MIN: 19.68 / MAX: 41.87MIN: 19.89 / MAX: 33.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400m123510152025Min: 20.5 / Avg: 20.69 / Max: 20.86Min: 20.4 / Avg: 20.55 / Max: 20.76Min: 20.47 / Avg: 20.9 / Max: 21.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg16123306090120150SE +/- 0.12, N = 3SE +/- 0.19, N = 3SE +/- 0.22, N = 3126.48128.34127.97MIN: 123.52 / MAX: 147.25MIN: 124.8 / MAX: 144.28MIN: 124.99 / MAX: 146.711. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg1612320406080100Min: 126.36 / Avg: 126.48 / Max: 126.73Min: 128.06 / Avg: 128.34 / Max: 128.69Min: 127.53 / Avg: 127.97 / Max: 128.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnet123246810SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 38.538.418.43MIN: 7.67 / MAX: 11.27MIN: 7.37 / MAX: 18.13MIN: 7.18 / MAX: 23.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnet1233691215Min: 8.45 / Avg: 8.53 / Max: 8.59Min: 8.31 / Avg: 8.41 / Max: 8.57Min: 8.35 / Avg: 8.43 / Max: 8.551. (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: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU123816243240SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 332.1232.1332.58MIN: 30.77MIN: 30.76MIN: 31.181. (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: CPU123714212835Min: 32.05 / Avg: 32.12 / Max: 32.16Min: 32.1 / Avg: 32.13 / Max: 32.17Min: 32.51 / Avg: 32.58 / Max: 32.621. (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-v3-v3 - Model: mobilenet-v3123246810SE +/- 0.13, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 38.778.728.84MIN: 7.65 / MAX: 16.44MIN: 7.36 / MAX: 21.89MIN: 7.72 / MAX: 20.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v3-v3 - Model: mobilenet-v31233691215Min: 8.6 / Avg: 8.77 / Max: 9.02Min: 8.61 / Avg: 8.72 / Max: 8.83Min: 8.77 / Avg: 8.84 / Max: 8.891. (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: CPU123246810SE +/- 0.01240, N = 3SE +/- 0.02785, N = 3SE +/- 0.02346, N = 37.329517.335677.23631MIN: 6.09MIN: 6.13MIN: 6.081. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU1233691215Min: 7.31 / Avg: 7.33 / Max: 7.35Min: 7.3 / Avg: 7.34 / Max: 7.39Min: 7.19 / Avg: 7.24 / Max: 7.271. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -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 APE12348121620SE +/- 0.09, N = 5SE +/- 0.01, N = 5SE +/- 0.04, N = 514.0613.8713.951. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt
OpenBenchmarking.orgSeconds, Fewer Is BetterMonkey Audio Encoding 3.99.6WAV To APE12348121620Min: 13.83 / Avg: 14.06 / Max: 14.21Min: 13.84 / Avg: 13.87 / Max: 13.91Min: 13.86 / Avg: 13.95 / Max: 14.091. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt

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: mobilenet123918273645SE +/- 0.17, N = 3SE +/- 0.17, N = 3SE +/- 0.05, N = 339.3039.2939.77MIN: 37.67 / MAX: 52.04MIN: 37.69 / MAX: 53.53MIN: 38.23 / MAX: 54.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenet123816243240Min: 38.99 / Avg: 39.3 / Max: 39.56Min: 39.1 / Avg: 39.29 / Max: 39.63Min: 39.69 / Avg: 39.77 / Max: 39.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b012348121620SE +/- 0.06, N = 3SE +/- 0.17, N = 3SE +/- 0.07, N = 314.2614.3214.43MIN: 12.32 / MAX: 58.63MIN: 12.71 / MAX: 21.22MIN: 12.96 / MAX: 26.11. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b012348121620Min: 14.2 / Avg: 14.26 / Max: 14.37Min: 14.12 / Avg: 14.32 / Max: 14.65Min: 14.35 / Avg: 14.43 / Max: 14.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mobilenet123918273645SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 339.2439.5539.64MIN: 37.61 / MAX: 75.08MIN: 37.98 / MAX: 52.81MIN: 38.24 / MAX: 53.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mobilenet123816243240Min: 39.14 / Avg: 39.24 / Max: 39.39Min: 39.44 / Avg: 39.55 / Max: 39.71Min: 39.56 / Avg: 39.64 / Max: 39.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b012348121620SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 314.4014.2814.26MIN: 12.89 / MAX: 32.02MIN: 12.77 / MAX: 26.63MIN: 12.68 / MAX: 28.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b012348121620Min: 14.31 / Avg: 14.4 / Max: 14.51Min: 14.22 / Avg: 14.28 / Max: 14.37Min: 14.09 / Avg: 14.26 / Max: 14.411. (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 1D - Data Type: u8s8f32 - Engine: CPU1231.34972.69944.04915.39886.7485SE +/- 0.00573, N = 3SE +/- 0.01650, N = 3SE +/- 0.02064, N = 35.977935.941465.99872MIN: 5.37MIN: 5.35MIN: 5.391. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU123246810Min: 5.97 / Avg: 5.98 / Max: 5.99Min: 5.92 / Avg: 5.94 / Max: 5.97Min: 5.96 / Avg: 6 / Max: 6.041. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 Compile12380160240320400SE +/- 2.56, N = 3SE +/- 1.77, N = 3SE +/- 1.04, N = 3388.39385.58388.95
OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To Compile12370140210280350Min: 385.78 / Avg: 388.39 / Max: 393.51Min: 382.16 / Avg: 385.58 / Max: 388.08Min: 386.87 / Avg: 388.95 / Max: 390.15

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_400m123510152025SE +/- 0.08, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 320.7020.7520.88MIN: 19.92 / MAX: 33.36MIN: 19.66 / MAX: 33.43MIN: 19.92 / MAX: 33.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: regnety_400m123510152025Min: 20.56 / Avg: 20.7 / Max: 20.84Min: 20.59 / Avg: 20.75 / Max: 20.93Min: 20.67 / Avg: 20.88 / Max: 21.031. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v21233691215SE +/- 0.04, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 311.5111.4211.49MIN: 9.89 / MAX: 25.37MIN: 10.18 / MAX: 14.51MIN: 10.27 / MAX: 24.451. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v21233691215Min: 11.44 / Avg: 11.51 / Max: 11.59Min: 11.24 / Avg: 11.42 / Max: 11.52Min: 11.36 / Avg: 11.49 / Max: 11.611. (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: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU123510152025SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 318.9318.9519.07MIN: 17.94MIN: 17.8MIN: 17.711. (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: CPU123510152025Min: 18.91 / Avg: 18.93 / Max: 18.95Min: 18.92 / Avg: 18.95 / Max: 18.98Min: 19.03 / Avg: 19.07 / Max: 19.161. (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: resnet18123714212835SE +/- 0.34, N = 3SE +/- 0.25, N = 3SE +/- 0.32, N = 330.4730.6430.69MIN: 28.73 / MAX: 39.99MIN: 29 / MAX: 45.72MIN: 29.02 / MAX: 42.891. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet18123714212835Min: 29.81 / Avg: 30.47 / Max: 30.93Min: 30.18 / Avg: 30.64 / Max: 31.03Min: 30.25 / Avg: 30.69 / Max: 31.311. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 Compile123306090120150SE +/- 2.20, N = 3SE +/- 2.25, N = 3SE +/- 1.80, N = 3145.94146.30146.98
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.2.2Time To Compile123306090120150Min: 141.66 / Avg: 145.94 / Max: 148.99Min: 141.81 / Avg: 146.29 / Max: 148.76Min: 143.43 / Avg: 146.98 / Max: 149.31

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 108012360120180240300SE +/- 0.33, N = 3SE +/- 0.67, N = 32922922901. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF
OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 1920 x 108012350100150200250Min: 291 / Avg: 291.67 / Max: 292Min: 291 / Avg: 291.67 / Max: 2931. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF

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 Compile12320406080100SE +/- 0.15, N = 3SE +/- 0.33, N = 3SE +/- 0.28, N = 396.6697.2997.30
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To Compile12320406080100Min: 96.4 / Avg: 96.66 / Max: 96.93Min: 96.89 / Avg: 97.29 / Max: 97.95Min: 96.91 / Avg: 97.3 / Max: 97.84

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 Benchmark1233691215SE +/- 0.08, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 39.309.299.241. Nodejs v12.18.2
OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling Benchmark1233691215Min: 9.14 / Avg: 9.3 / Max: 9.39Min: 9.28 / Avg: 9.29 / Max: 9.3Min: 9.14 / Avg: 9.24 / Max: 9.41. 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 - Model: resnet501231428425670SE +/- 1.40, N = 3SE +/- 0.50, N = 3SE +/- 0.70, N = 363.2163.5563.34MIN: 59.61 / MAX: 81.53MIN: 59.8 / MAX: 83.82MIN: 59.91 / MAX: 77.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet501231224364860Min: 61.7 / Avg: 63.21 / Max: 66Min: 62.56 / Avg: 63.55 / Max: 64.21Min: 62.14 / Avg: 63.34 / Max: 64.571. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 Metric1239K18K27K36K45K4237342365421691. (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

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: vgg16123306090120150SE +/- 0.47, N = 3SE +/- 0.12, N = 3SE +/- 0.28, N = 3127.46127.55128.02MIN: 123.62 / MAX: 155.82MIN: 124.34 / MAX: 142.98MIN: 124.75 / MAX: 141.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg1612320406080100Min: 126.62 / Avg: 127.46 / Max: 128.26Min: 127.3 / Avg: 127.55 / Max: 127.68Min: 127.47 / Avg: 128.02 / Max: 128.381. (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: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU1233691215SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 312.9612.9512.90MIN: 11.96MIN: 11.96MIN: 11.731. (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: CPU12348121620Min: 12.94 / Avg: 12.96 / Max: 12.98Min: 12.92 / Avg: 12.95 / Max: 12.98Min: 12.81 / Avg: 12.9 / Max: 12.981. (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: CPU123714212835SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.25, N = 331.0330.9330.95MIN: 29.17MIN: 29.31MIN: 29.081. (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: CPU123714212835Min: 30.95 / Avg: 31.03 / Max: 31.08Min: 30.87 / Avg: 30.93 / Max: 30.96Min: 30.44 / Avg: 30.95 / Max: 31.241. (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: alexnet123612182430SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 325.1025.1625.17MIN: 23.66 / MAX: 35.12MIN: 23.92 / MAX: 36.39MIN: 23.93 / MAX: 36.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnet123612182430Min: 25.04 / Avg: 25.1 / Max: 25.15Min: 25.1 / Avg: 25.16 / Max: 25.19Min: 25.05 / Avg: 25.17 / Max: 25.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v21233691215SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.10, N = 311.4111.4211.44MIN: 9.61 / MAX: 24.27MIN: 9.78 / MAX: 21.73MIN: 10.2 / MAX: 14.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v21233691215Min: 11.32 / Avg: 11.41 / Max: 11.48Min: 11.24 / Avg: 11.42 / Max: 11.55Min: 11.27 / Avg: 11.44 / Max: 11.631. (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 Second12330K60K90K120K150KSE +/- 201.95, N = 3SE +/- 324.47, N = 3SE +/- 511.87, N = 3144430.97144054.76144381.891. (CC) gcc options: -O2 -lrt" -lrt
OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per Second12330K60K90K120K150KMin: 144066.27 / Avg: 144430.97 / Max: 144763.63Min: 143587.9 / Avg: 144054.76 / Max: 144678.54Min: 143652.36 / Avg: 144381.89 / Max: 145368.651. (CC) gcc options: -O2 -lrt" -lrt

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 Encode1233691215SE +/- 0.019, N = 5SE +/- 0.008, N = 5SE +/- 0.008, N = 59.1329.1339.1131. (CXX) g++ options: -fvisibility=hidden -logg -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus Encode1233691215Min: 9.1 / Avg: 9.13 / Max: 9.2Min: 9.11 / Avg: 9.13 / Max: 9.16Min: 9.09 / Avg: 9.11 / Max: 9.141. (CXX) g++ options: -fvisibility=hidden -logg -lm

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 Search123306090120150SE +/- 0.04, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 3137.77137.94137.901. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database Search123306090120150Min: 137.71 / Avg: 137.77 / Max: 137.85Min: 137.78 / Avg: 137.94 / Max: 138.11Min: 137.87 / Avg: 137.9 / Max: 137.971. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm

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 WavPack12348121620SE +/- 0.03, N = 5SE +/- 0.03, N = 5SE +/- 0.04, N = 515.5515.5415.561. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPack12348121620Min: 15.49 / Avg: 15.55 / Max: 15.63Min: 15.49 / Avg: 15.54 / Max: 15.65Min: 15.5 / Avg: 15.56 / Max: 15.71. (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: DistinctUserID1230.1530.3060.4590.6120.765SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.680.680.681. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: DistinctUserID123246810Min: 0.68 / Avg: 0.68 / Max: 0.68Min: 0.68 / Avg: 0.68 / Max: 0.68Min: 0.68 / Avg: 0.68 / Max: 0.681. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: PartialTweets1230.14850.2970.44550.5940.7425SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.660.660.661. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: PartialTweets123246810Min: 0.66 / Avg: 0.66 / Max: 0.66Min: 0.66 / Avg: 0.66 / Max: 0.66Min: 0.66 / Avg: 0.66 / Max: 0.661. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandom1230.090.180.270.360.45SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.40.40.41. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandom12312345Min: 0.4 / Avg: 0.4 / Max: 0.4Min: 0.4 / Avg: 0.4 / Max: 0.4Min: 0.4 / Avg: 0.4 / Max: 0.41. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: Kostya1230.1350.270.4050.540.675SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.600.600.601. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: Kostya123246810Min: 0.59 / Avg: 0.6 / Max: 0.6Min: 0.59 / Avg: 0.6 / Max: 0.6Min: 0.59 / Avg: 0.6 / Max: 0.61. (CXX) g++ options: -O3 -pthread

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 Speedup1230.24750.4950.74250.991.2375SE +/- 0.04, N = 11SE +/- 0.04, N = 9SE +/- 0.02, N = 120.91.11.11. (CC) gcc options: -fopenmp -O3 -lm
OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP Speedup123246810Min: 0.5 / Avg: 0.9 / Max: 1.1Min: 1 / Avg: 1.13 / Max: 1.3Min: 1 / Avg: 1.14 / Max: 1.21. (CC) gcc options: -fopenmp -O3 -lm

1

Compiler Notes: --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 Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x28 - Thermald 2.3
Security Notes: 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: Not affected

Testing initiated at 25 December 2020 09:56 by user phoronix.

2

Compiler Notes: --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 Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x28 - Thermald 2.3
Security Notes: 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: Not affected

Testing initiated at 25 December 2020 13:24 by user phoronix.

3

Processor: Intel Core i7-4770K @ 3.90GHz (4 Cores / 8 Threads), Motherboard: Gigabyte Z97-HD3 (F10c BIOS), Chipset: Intel 4th Gen Core DRAM, Memory: 8GB, Disk: 120GB ADATA SU700, Graphics: Gigabyte Intel HD 4600 2GB (1250MHz), Audio: Intel Xeon E3-1200 v3/4th, Monitor: DELL S2409W, Network: Realtek RTL8111/8168/8411

OS: Ubuntu 20.10, Kernel: 5.8.0-31-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.5 Mesa 20.2.1, Vulkan: 1.2.145, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1080

Compiler Notes: --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 Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x28 - Thermald 2.3
Security Notes: 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: Not affected

Testing initiated at 25 December 2020 16:32 by user phoronix.