Core i3 7100 Xmas Eve

Intel Core i3-7100 testing with a Gigabyte B250M-DS3H-CF (F9 BIOS) and Gigabyte Intel HD 630 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 2012250-HA-COREI371034
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
Bioinformatics 2 Tests
Chess Test Suite 3 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 5 Tests
CPU Massive 9 Tests
Creator Workloads 6 Tests
Encoding 4 Tests
HPC - High Performance Computing 5 Tests
Machine Learning 2 Tests
Multi-Core 8 Tests
NVIDIA GPU Compute 3 Tests
Programmer / Developer System Benchmarks 6 Tests
Scientific Computing 3 Tests
Server 4 Tests
Server CPU Tests 4 Tests
Vulkan Compute 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 24 2020
  9 Hours, 21 Minutes
2
December 24 2020
  9 Hours, 17 Minutes
3
December 25 2020
  9 Hours, 16 Minutes
Invert Hiding All Results Option
  9 Hours, 18 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):


Core i3 7100 Xmas EveProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen Resolution123Intel Core i3-7100 @ 3.90GHz (2 Cores / 4 Threads)Gigabyte B250M-DS3H-CF (F9 BIOS)Intel Xeon E3-1200 v6/7th + B2508GB250GB Western Digital WDS250G1B0A-Gigabyte Intel HD 630 3GB (1100MHz)Realtek ALC887-VDVA2431Realtek RTL8111/8168/8411Ubuntu 20.105.8.0-28-generic (x86_64)GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.94.6 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 Disk Details- MQ-DEADLINE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Details- Scaling Governor: intel_pstate powersave - CPU Microcode: 0xde - 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%101%103%104%105%SQLiteStockfishVKMarkasmFishHPC ChallengeNode.js V8 Web Tooling BenchmarkTimed MAFFT AlignmentoneDNNCoremarkVkResampleBuild2Craftyrav1eASTC EncoderPHPBenchMonkey Audio EncodingVkFFTTimed Eigen CompilationWavPack Audio EncodingNCNNOpus Codec EncodingTimed FFmpeg CompilationTimed HMMer SearchsimdjsonCLOMP

Core i3 7100 Xmas Eveonednn: IP Shapes 3D - f32 - CPUhpcc: Rand Ring Bandwidthsqlite: 1stockfish: Total Timeonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUvkresample: 2x - Singlevkmark: 1280 x 1024onednn: IP Shapes 3D - u8s8f32 - CPUrav1e: 1hpcc: Max Ping Pong Bandwidthasmfish: 1024 Hash Memory, 26 Depthhpcc: G-HPLastcenc: Exhaustivevkmark: 1920 x 1080hpcc: EP-DGEMMnode-web-tooling: mafft: Multiple Sequence Alignment - LSU RNAonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: IP Shapes 1D - f32 - CPUhpcc: G-Rand Accesscoremark: CoreMark Size 666 - Iterations Per Secondhpcc: EP-STREAM Triadonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUhpcc: Rand Ring Latencyonednn: Convolution Batch Shapes Auto - f32 - CPUbuild2: Time To Compilencnn: CPU - efficientnet-b0crafty: Elapsed Timencnn: CPU - mnasnetncnn: Vulkan GPU - regnety_400mhpcc: G-Ffteonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUncnn: Vulkan GPU - mobilenetvkresample: 2x - Doublencnn: Vulkan GPU-v3-v3 - mobilenet-v3rav1e: 5ncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU - squeezenet_ssdonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUncnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - mnasnetonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUrav1e: 6phpbench: PHP Benchmark Suitencnn: CPU - mobilenetncnn: CPU - resnet18ncnn: Vulkan GPU - blazefacencnn: CPU - blazefacencnn: CPU - vgg16encode-ape: WAV To APEvkfft: ncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - vgg16build-eigen: Time To Compileonednn: Deconvolution Batch shapes_3d - f32 - CPUncnn: CPU - yolov4-tinyncnn: Vulkan GPU - squeezenet_ssdncnn: CPU - shufflenet-v2ncnn: Vulkan GPU - resnet18astcenc: Fastncnn: Vulkan GPU - alexnetbuild-ffmpeg: Time To Compilencnn: CPU - resnet50ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - regnety_400mncnn: Vulkan GPU - resnet50onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUastcenc: Thoroughencode-wavpack: WAV To WavPackrav1e: 10ncnn: CPU - alexnetncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - shufflenet-v2encode-opus: WAV To Opus Encodencnn: CPU - googlenethmmer: Pfam Database Searchastcenc: Mediumsimdjson: DistinctUserIDsimdjson: PartialTweetssimdjson: LargeRandsimdjson: Kostyaclomp: Static OMP Speeduphpcc: G-Ptrans12315.53615.1470130.315423295312989.97184.0413022.37192.687188.4912994.0523.0349054.694110.2538711.352614400986.839171087.5861447.154439.4414.00020.042915.79950.0127077661.5579969.8747920.06426.977300.2103520.7537644.78915.4473322079.8618.601.7552324.019943.071011.3209.190.77910.6210.6347.116.0388915.329.8413.36411.03565788743.1230.823.863.88111.3712.781133231.55111.2992.84724.575757.1747.1915.3130.757.9724.93232.08568.59.1818.6068.4312.3520135.2916.6932.40424.9557.1515.339.59331.51114.82620.420.620.60.370.491.41.3112414.83655.1846231.916404387113484.97447.3013494.47437.087424.1013403.5532.9579064.816950.2488813.097625743985.294871104.7061447.504039.3413.86719.968415.65800.0126777631.8269199.8881920.02476.970300.2099620.6337648.38015.4173697769.9118.671.7621023.943943.201010.7409.150.77910.6410.6547.286.0311215.369.8713.34471.03865812643.2430.753.873.87111.6312.752133531.53111.3292.73024.623657.2347.2415.3330.767.9724.93232.00268.429.1818.6268.3612.3525135.4216.6962.40624.9657.1815.329.59731.51114.83420.420.620.60.370.491.41.4607616.18174.9025531.702410198013005.47201.5113017.57207.997201.0913012.2538.2868814.709810.2518644.793624041286.813801088.4660647.694609.3414.00419.861215.74340.0127778242.2261169.8205219.92897.015590.2090820.7444645.34915.3673470219.8618.581.7538624.053143.261006.8909.150.77610.6610.6747.156.0527115.379.8613.32391.03665629543.2430.743.873.88111.5912.768133231.48111.0992.92024.603157.2647.2615.3130.727.9624.90231.80668.429.1918.6068.4212.3401135.3116.7092.40624.9457.1915.329.59931.52114.82620.420.620.600.370.491.41.46079OpenBenchmarking.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: CPU32148121620SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 316.1814.8415.54MIN: 15.82MIN: 14.45MIN: 15.011. (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: CPU32148121620Min: 16.15 / Avg: 16.18 / Max: 16.21Min: 14.81 / Avg: 14.84 / Max: 14.88Min: 15.46 / Avg: 15.54 / Max: 15.641. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Random Ring Bandwidth3211.16652.3333.49954.6665.8325SE +/- 0.08913, N = 3SE +/- 0.08185, N = 3SE +/- 0.09024, N = 34.902555.184625.147011. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Random Ring Bandwidth321246810Min: 4.72 / Avg: 4.9 / Max: 5Min: 5.02 / Avg: 5.18 / Max: 5.29Min: 4.97 / Avg: 5.15 / Max: 5.261. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

SQLite

This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.30.1Threads / Copies: 1321714212835SE +/- 0.36, N = 3SE +/- 0.04, N = 3SE +/- 0.36, N = 1531.7031.9230.321. (CC) gcc options: -O2 -lz -lm -ldl -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.30.1Threads / Copies: 1321714212835Min: 31.16 / Avg: 31.7 / Max: 32.39Min: 31.84 / Avg: 31.92 / Max: 31.96Min: 28.89 / Avg: 30.32 / Max: 33.371. (CC) gcc options: -O2 -lz -lm -ldl -lpthread

Stockfish

This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 12Total Time321900K1800K2700K3600K4500KSE +/- 14449.66, N = 3SE +/- 24874.87, N = 3SE +/- 40004.50, N = 34101980404387142329531. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver
OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 12Total Time321700K1400K2100K2800K3500KMin: 4079624 / Avg: 4101980.33 / Max: 4129018Min: 4000273 / Avg: 4043871.33 / Max: 4086423Min: 4180251 / Avg: 4232953 / Max: 43114381. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver

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: CPU3213K6K9K12K15KSE +/- 5.12, N = 3SE +/- 28.75, N = 3SE +/- 2.60, N = 313005.413484.912989.9MIN: 12979.2MIN: 13078.8MIN: 12963.91. (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: CPU3212K4K6K8K10KMin: 12996.3 / Avg: 13005.43 / Max: 13014Min: 13427.9 / Avg: 13484.87 / Max: 13520.1Min: 12984.7 / Avg: 12989.87 / Max: 12992.91. (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: CPU32116003200480064008000SE +/- 5.32, N = 3SE +/- 7.22, N = 3SE +/- 4.85, N = 37201.517447.307184.04MIN: 7179.7MIN: 7277.81MIN: 7162.141. (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: CPU32113002600390052006500Min: 7193.22 / Avg: 7201.51 / Max: 7211.42Min: 7433.73 / Avg: 7447.3 / Max: 7458.35Min: 7174.93 / Avg: 7184.04 / Max: 7191.471. (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: CPU3213K6K9K12K15KSE +/- 6.69, N = 3SE +/- 5.06, N = 3SE +/- 34.32, N = 313017.513494.413022.3MIN: 12988.3MIN: 13309.7MIN: 12970.11. (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: CPU3212K4K6K8K10KMin: 13006.5 / Avg: 13017.53 / Max: 13029.6Min: 13485.1 / Avg: 13494.43 / Max: 13502.5Min: 12986.9 / Avg: 13022.27 / Max: 13090.91. (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: CPU32116003200480064008000SE +/- 7.86, N = 3SE +/- 4.96, N = 3SE +/- 5.80, N = 37207.997437.087192.68MIN: 7180.31MIN: 7270.53MIN: 7171.231. (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: CPU32113002600390052006500Min: 7192.49 / Avg: 7207.99 / Max: 7218.07Min: 7427.38 / Avg: 7437.08 / Max: 7443.76Min: 7185.07 / Avg: 7192.68 / Max: 7204.071. (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: CPU32116003200480064008000SE +/- 2.75, N = 3SE +/- 4.88, N = 3SE +/- 4.08, N = 37201.097424.107188.49MIN: 7180.72MIN: 7267.22MIN: 7171.821. (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: CPU32113002600390052006500Min: 7196.37 / Avg: 7201.09 / Max: 7205.9Min: 7417.09 / Avg: 7424.1 / Max: 7433.48Min: 7183.03 / Avg: 7188.49 / Max: 7196.481. (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: CPU3213K6K9K12K15KSE +/- 7.57, N = 3SE +/- 76.51, N = 3SE +/- 4.10, N = 313012.213403.512994.0MIN: 12979.3MIN: 13048.6MIN: 12966.61. (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: CPU3212K4K6K8K10KMin: 12997.1 / Avg: 13012.2 / Max: 13020.6Min: 13250.5 / Avg: 13403.5 / Max: 13482.5Min: 12988.2 / Avg: 12993.97 / Max: 13001.91. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -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: Single321120240360480600SE +/- 4.21, N = 3SE +/- 4.10, N = 3SE +/- 3.39, N = 3538.29532.96523.031. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Single321100200300400500Min: 529.89 / Avg: 538.29 / Max: 542.91Min: 528.55 / Avg: 532.96 / Max: 541.15Min: 517.19 / Avg: 523.03 / Max: 528.931. (CXX) g++ options: -O3 -pthread

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 10243212004006008001000SE +/- 3.28, N = 3SE +/- 0.58, N = 3SE +/- 1.00, N = 38819069051. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF
OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 1280 x 1024321160320480640800Min: 875 / Avg: 881.33 / Max: 886Min: 905 / Avg: 906 / Max: 907Min: 903 / Avg: 905 / Max: 9061. (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: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU3211.08382.16763.25144.33525.419SE +/- 0.00674, N = 3SE +/- 0.00803, N = 3SE +/- 0.05317, N = 34.709814.816954.69411MIN: 4.56MIN: 4.39MIN: 4.481. (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: CPU321246810Min: 4.7 / Avg: 4.71 / Max: 4.72Min: 4.81 / Avg: 4.82 / Max: 4.83Min: 4.64 / Avg: 4.69 / Max: 4.81. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 13210.05690.11380.17070.22760.2845SE +/- 0.002, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 30.2510.2480.253
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 132112345Min: 0.25 / Avg: 0.25 / Max: 0.25Min: 0.25 / Avg: 0.25 / Max: 0.25Min: 0.25 / Avg: 0.25 / Max: 0.25

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Max Ping Pong Bandwidth3212K4K6K8K10KSE +/- 80.85, N = 3SE +/- 66.91, N = 3SE +/- 98.36, N = 38644.798813.108711.351. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgMB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Max Ping Pong Bandwidth32115003000450060007500Min: 8483.51 / Avg: 8644.79 / Max: 8735.48Min: 8679.92 / Avg: 8813.1 / Max: 8891Min: 8515.57 / Avg: 8711.35 / Max: 8825.741. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 Depth3211.3M2.6M3.9M5.2M6.5MSE +/- 17217.84, N = 3SE +/- 82994.07, N = 3SE +/- 29250.15, N = 3624041262574396144009
OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 Depth3211.1M2.2M3.3M4.4M5.5MMin: 6206939 / Avg: 6240412.33 / Max: 6264150Min: 6129742 / Avg: 6257438.67 / Max: 6413125Min: 6087624 / Avg: 6144009 / Max: 6185702

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-HPL32120406080100SE +/- 0.15, N = 3SE +/- 0.24, N = 3SE +/- 0.04, N = 386.8185.2986.841. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-HPL3211632486480Min: 86.51 / Avg: 86.81 / Max: 86.99Min: 84.97 / Avg: 85.29 / Max: 85.76Min: 86.75 / Avg: 86.84 / Max: 86.91. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Exhaustive3212004006008001000SE +/- 0.17, N = 3SE +/- 8.12, N = 3SE +/- 0.37, N = 31088.461104.701087.581. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Exhaustive3212004006008001000Min: 1088.17 / Avg: 1088.46 / Max: 1088.76Min: 1088.51 / Avg: 1104.7 / Max: 1113.92Min: 1087.02 / Avg: 1087.58 / Max: 1088.291. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -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 1080321130260390520650SE +/- 7.51, N = 3SE +/- 0.58, N = 3SE +/- 1.20, N = 36066146141. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF
OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 1920 x 1080321110220330440550Min: 591 / Avg: 606 / Max: 614Min: 613 / Avg: 614 / Max: 615Min: 612 / Avg: 614.33 / Max: 6161. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: EP-DGEMM3211122334455SE +/- 0.10, N = 3SE +/- 0.13, N = 3SE +/- 0.52, N = 347.6947.5047.151. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: EP-DGEMM3211020304050Min: 47.5 / Avg: 47.69 / Max: 47.8Min: 47.25 / Avg: 47.5 / Max: 47.7Min: 46.12 / Avg: 47.15 / Max: 47.781. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

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 Benchmark3213691215SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 39.349.349.441. Nodejs v12.18.2
OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling Benchmark3213691215Min: 9.31 / Avg: 9.34 / Max: 9.4Min: 9.26 / Avg: 9.34 / Max: 9.42Min: 9.26 / Avg: 9.44 / Max: 9.61. Nodejs v12.18.2

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 RNA32148121620SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 314.0013.8714.001. (CC) gcc options: -std=c99 -O3 -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 7.471Multiple Sequence Alignment - LSU RNA32148121620Min: 13.82 / Avg: 14 / Max: 14.2Min: 13.83 / Avg: 13.87 / Max: 13.95Min: 13.85 / Avg: 14 / Max: 14.121. (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: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU321510152025SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 319.8619.9720.04MIN: 19.64MIN: 19.66MIN: 19.641. (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: CPU321510152025Min: 19.84 / Avg: 19.86 / Max: 19.87Min: 19.91 / Avg: 19.97 / Max: 20.08Min: 19.85 / Avg: 20.04 / Max: 20.231. (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: CPU32148121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 315.7415.6615.80MIN: 14.72MIN: 14.75MIN: 14.731. (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: CPU32148121620Min: 15.73 / Avg: 15.74 / Max: 15.77Min: 15.63 / Avg: 15.66 / Max: 15.67Min: 15.74 / Avg: 15.8 / Max: 15.831. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGUP/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-Random Access3210.00290.00580.00870.01160.0145SE +/- 0.00004, N = 3SE +/- 0.00011, N = 3SE +/- 0.00008, N = 30.012770.012670.012701. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGUP/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-Random Access32112345Min: 0.01 / Avg: 0.01 / Max: 0.01Min: 0.01 / Avg: 0.01 / Max: 0.01Min: 0.01 / Avg: 0.01 / Max: 0.011. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

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 Second32120K40K60K80K100KSE +/- 631.76, N = 3SE +/- 652.47, N = 3SE +/- 672.82, N = 378242.2377631.8377661.561. (CC) gcc options: -O2 -lrt" -lrt
OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per Second32114K28K42K56K70KMin: 77008.23 / Avg: 78242.23 / Max: 79094.37Min: 76687.12 / Avg: 77631.83 / Max: 78883.79Min: 76317.67 / Avg: 77661.56 / Max: 78392.941. (CC) gcc options: -O2 -lrt" -lrt

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: EP-STREAM Triad3213691215SE +/- 0.01214, N = 3SE +/- 0.00967, N = 3SE +/- 0.01832, N = 39.820529.888199.874791. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: EP-STREAM Triad3213691215Min: 9.8 / Avg: 9.82 / Max: 9.84Min: 9.87 / Avg: 9.89 / Max: 9.9Min: 9.84 / Avg: 9.87 / Max: 9.91. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

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: CPU321510152025SE +/- 0.04, N = 3SE +/- 0.26, N = 3SE +/- 0.24, N = 319.9320.0220.06MIN: 19.62MIN: 19.56MIN: 19.521. (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: CPU321510152025Min: 19.87 / Avg: 19.93 / Max: 20.01Min: 19.73 / Avg: 20.02 / Max: 20.53Min: 19.73 / Avg: 20.06 / Max: 20.521. (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: CPU321246810SE +/- 0.03839, N = 3SE +/- 0.01316, N = 3SE +/- 0.01584, N = 37.015596.970306.97730MIN: 6.87MIN: 6.87MIN: 6.861. (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: CPU3213691215Min: 6.96 / Avg: 7.02 / Max: 7.09Min: 6.95 / Avg: 6.97 / Max: 6.99Min: 6.95 / Avg: 6.98 / Max: 71. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgusecs, Fewer Is BetterHPC Challenge 1.5.0Test / Class: Random Ring Latency3210.04730.09460.14190.18920.2365SE +/- 0.00069, N = 3SE +/- 0.00200, N = 3SE +/- 0.00110, N = 30.209080.209960.210351. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgusecs, Fewer Is BetterHPC Challenge 1.5.0Test / Class: Random Ring Latency32112345Min: 0.21 / Avg: 0.21 / Max: 0.21Min: 0.21 / Avg: 0.21 / Max: 0.21Min: 0.21 / Avg: 0.21 / Max: 0.211. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

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: CPU321510152025SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 320.7420.6320.75MIN: 20.47MIN: 20.39MIN: 20.421. (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: CPU321510152025Min: 20.71 / Avg: 20.74 / Max: 20.79Min: 20.55 / Avg: 20.63 / Max: 20.7Min: 20.71 / Avg: 20.75 / Max: 20.781. (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 Compile321140280420560700SE +/- 0.41, N = 3SE +/- 0.56, N = 3SE +/- 0.27, N = 3645.35648.38644.79
OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To Compile321110220330440550Min: 644.86 / Avg: 645.35 / Max: 646.17Min: 647.28 / Avg: 648.38 / Max: 649.08Min: 644.31 / Avg: 644.79 / Max: 645.24

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-b032148121620SE +/- 0.01, N = 3SE +/- 0.05, N = 2SE +/- 0.04, N = 215.3615.4115.44MIN: 15.2 / MAX: 16.15MIN: 15.24 / MAX: 25.3MIN: 15.2 / MAX: 25.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b032148121620Min: 15.34 / Avg: 15.36 / Max: 15.37Min: 15.36 / Avg: 15.41 / Max: 15.46Min: 15.39 / Avg: 15.44 / Max: 15.481. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Crafty

This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed Time3211.6M3.2M4.8M6.4M8MSE +/- 3975.85, N = 3SE +/- 11879.59, N = 3SE +/- 19203.96, N = 37347021736977673322071. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed Time3211.3M2.6M3.9M5.2M6.5MMin: 7339539 / Avg: 7347020.67 / Max: 7353094Min: 7355287 / Avg: 7369776.33 / Max: 7393328Min: 7301846 / Avg: 7332207.33 / Max: 73677601. (CC) gcc options: -pthread -lstdc++ -fprofile-use -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: mnasnet3213691215SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 39.869.919.86MIN: 9.74 / MAX: 10.66MIN: 9.73 / MAX: 22MIN: 9.73 / MAX: 20.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnet3213691215Min: 9.84 / Avg: 9.86 / Max: 9.89Min: 9.85 / Avg: 9.91 / Max: 10.02Min: 9.84 / Avg: 9.86 / Max: 9.891. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: regnety_400m321510152025SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 318.5818.6718.60MIN: 18.48 / MAX: 28.89MIN: 18.47 / MAX: 29.58MIN: 18.4 / MAX: 28.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: regnety_400m321510152025Min: 18.55 / Avg: 18.58 / Max: 18.61Min: 18.66 / Avg: 18.67 / Max: 18.69Min: 18.53 / Avg: 18.6 / Max: 18.671. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-Ffte3210.39650.7931.18951.5861.9825SE +/- 0.00359, N = 3SE +/- 0.00193, N = 3SE +/- 0.00362, N = 31.753861.762101.755231. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-Ffte321246810Min: 1.75 / Avg: 1.75 / Max: 1.76Min: 1.76 / Avg: 1.76 / Max: 1.77Min: 1.75 / Avg: 1.76 / Max: 1.761. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

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: u8s8f32 - Engine: CPU321612182430SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 324.0523.9424.02MIN: 23.81MIN: 23.7MIN: 23.781. (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: CPU321612182430Min: 24.03 / Avg: 24.05 / Max: 24.1Min: 23.91 / Avg: 23.94 / Max: 23.99Min: 23.99 / Avg: 24.02 / Max: 24.051. (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: mobilenet3211020304050SE +/- 0.09, N = 3SE +/- 0.11, N = 3SE +/- 0.01, N = 343.2643.2043.07MIN: 42.86 / MAX: 53.83MIN: 42.87 / MAX: 53.52MIN: 42.83 / MAX: 53.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mobilenet321918273645Min: 43.09 / Avg: 43.26 / Max: 43.36Min: 43.09 / Avg: 43.2 / Max: 43.42Min: 43.06 / Avg: 43.07 / Max: 43.081. (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: Double3212004006008001000SE +/- 2.44, N = 3SE +/- 4.98, N = 3SE +/- 5.17, N = 31006.891010.741011.321. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Double3212004006008001000Min: 1003.33 / Avg: 1006.89 / Max: 1011.56Min: 1001.91 / Avg: 1010.74 / Max: 1019.16Min: 1001.88 / Avg: 1011.32 / Max: 1019.711. (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: Vulkan GPU-v3-v3 - Model: mobilenet-v33213691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 39.159.159.19MIN: 9.01 / MAX: 10.75MIN: 9.02 / MAX: 10.69MIN: 9.01 / MAX: 19.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v3-v3 - Model: mobilenet-v33213691215Min: 9.14 / Avg: 9.15 / Max: 9.17Min: 9.14 / Avg: 9.15 / Max: 9.17Min: 9.16 / Avg: 9.19 / Max: 9.211. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 53210.17530.35060.52590.70120.8765SE +/- 0.001, N = 3SE +/- 0.003, N = 3SE +/- 0.003, N = 30.7760.7790.779
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 5321246810Min: 0.77 / Avg: 0.78 / Max: 0.78Min: 0.77 / Avg: 0.78 / Max: 0.78Min: 0.77 / Avg: 0.78 / Max: 0.78

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-v23213691215SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 310.6610.6410.62MIN: 10.45 / MAX: 21.04MIN: 10.46 / MAX: 20.48MIN: 10.45 / MAX: 12.131. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v2-v2 - Model: mobilenet-v23213691215Min: 10.64 / Avg: 10.66 / Max: 10.69Min: 10.62 / Avg: 10.64 / Max: 10.68Min: 10.61 / Avg: 10.62 / Max: 10.631. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v23213691215SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 310.6710.6510.63MIN: 10.49 / MAX: 25.76MIN: 10.45 / MAX: 20.53MIN: 10.46 / MAX: 12.771. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v23213691215Min: 10.64 / Avg: 10.67 / Max: 10.71Min: 10.61 / Avg: 10.65 / Max: 10.71Min: 10.63 / Avg: 10.63 / Max: 10.641. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssd3211122334455SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 347.1547.2847.11MIN: 46.78 / MAX: 57.9MIN: 46.79 / MAX: 58.17MIN: 46.79 / MAX: 57.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssd3211020304050Min: 47.12 / Avg: 47.15 / Max: 47.19Min: 47.18 / Avg: 47.28 / Max: 47.41Min: 47.03 / Avg: 47.11 / Max: 47.241. (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: f32 - Engine: CPU321246810SE +/- 0.00775, N = 3SE +/- 0.00495, N = 3SE +/- 0.00365, N = 36.052716.031126.03889MIN: 5.92MIN: 5.89MIN: 5.921. (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: CPU321246810Min: 6.04 / Avg: 6.05 / Max: 6.06Min: 6.02 / Avg: 6.03 / Max: 6.04Min: 6.03 / Avg: 6.04 / Max: 6.051. (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: efficientnet-b032148121620SE +/- 0.03, N = 2SE +/- 0.03, N = 3SE +/- 0.00, N = 315.3715.3615.32MIN: 15.21 / MAX: 25.63MIN: 15.2 / MAX: 25.21MIN: 15.19 / MAX: 161. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b032148121620Min: 15.34 / Avg: 15.37 / Max: 15.4Min: 15.31 / Avg: 15.36 / Max: 15.4Min: 15.32 / Avg: 15.32 / Max: 15.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnet3213691215SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 39.869.879.84MIN: 9.72 / MAX: 20.18MIN: 9.75 / MAX: 19.76MIN: 9.75 / MAX: 10.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnet3213691215Min: 9.83 / Avg: 9.86 / Max: 9.9Min: 9.86 / Avg: 9.87 / Max: 9.89Min: 9.84 / Avg: 9.84 / Max: 9.851. (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: CPU3213691215SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 313.3213.3413.36MIN: 13.18MIN: 13.14MIN: 13.21. (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: CPU32148121620Min: 13.32 / Avg: 13.32 / Max: 13.33Min: 13.25 / Avg: 13.34 / Max: 13.4Min: 13.32 / Avg: 13.36 / Max: 13.41. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 63210.23360.46720.70080.93441.168SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 31.0361.0381.035
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 6321246810Min: 1.04 / Avg: 1.04 / Max: 1.04Min: 1.04 / Avg: 1.04 / Max: 1.04Min: 1.03 / Avg: 1.04 / Max: 1.04

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 Suite321140K280K420K560K700KSE +/- 1397.20, N = 3SE +/- 1397.02, N = 3SE +/- 1373.01, N = 3656295658126657887
OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark Suite321110K220K330K440K550KMin: 653528 / Avg: 656295.33 / Max: 658015Min: 655487 / Avg: 658126.33 / Max: 660240Min: 655159 / Avg: 657887.33 / Max: 659521

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: mobilenet3211020304050SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 343.2443.2443.12MIN: 42.96 / MAX: 53.6MIN: 42.94 / MAX: 54.26MIN: 42.84 / MAX: 53.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenet321918273645Min: 43.23 / Avg: 43.24 / Max: 43.26Min: 43.19 / Avg: 43.24 / Max: 43.31Min: 43.08 / Avg: 43.12 / Max: 43.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18321714212835SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 330.7430.7530.82MIN: 30.51 / MAX: 41.31MIN: 30.52 / MAX: 41.61MIN: 30.55 / MAX: 41.041. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18321714212835Min: 30.66 / Avg: 30.74 / Max: 30.79Min: 30.66 / Avg: 30.75 / Max: 30.82Min: 30.74 / Avg: 30.82 / Max: 30.911. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazeface3210.87081.74162.61243.48324.354SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 33.873.873.86MIN: 3.81 / MAX: 4.04MIN: 3.82 / MAX: 4.01MIN: 3.81 / MAX: 4.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazeface321246810Min: 3.86 / Avg: 3.87 / Max: 3.87Min: 3.85 / Avg: 3.87 / Max: 3.89Min: 3.86 / Avg: 3.86 / Max: 3.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazeface3210.8731.7462.6193.4924.365SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 33.883.873.88MIN: 3.82 / MAX: 4.55MIN: 3.81 / MAX: 4.48MIN: 3.82 / MAX: 4.531. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazeface321246810Min: 3.88 / Avg: 3.88 / Max: 3.88Min: 3.86 / Avg: 3.87 / Max: 3.88Min: 3.87 / Avg: 3.88 / Max: 3.881. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg1632120406080100SE +/- 0.22, N = 3SE +/- 0.25, N = 3SE +/- 0.11, N = 3111.59111.63111.37MIN: 110.86 / MAX: 122.85MIN: 110.94 / MAX: 123.3MIN: 110.74 / MAX: 121.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg1632120406080100Min: 111.23 / Avg: 111.59 / Max: 111.99Min: 111.36 / Avg: 111.63 / Max: 112.12Min: 111.15 / Avg: 111.37 / Max: 111.481. (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 APE3213691215SE +/- 0.07, N = 5SE +/- 0.07, N = 5SE +/- 0.04, N = 512.7712.7512.781. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt
OpenBenchmarking.orgSeconds, Fewer Is BetterMonkey Audio Encoding 3.99.6WAV To APE32148121620Min: 12.6 / Avg: 12.77 / Max: 12.98Min: 12.57 / Avg: 12.75 / Max: 12.99Min: 12.65 / Avg: 12.78 / Max: 12.871. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt

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 1.1.132130060090012001500SE +/- 1.20, N = 3SE +/- 1.67, N = 31332133513321. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.13212004006008001000Min: 1330 / Avg: 1332.33 / Max: 1334Min: 1333 / Avg: 1334.67 / Max: 13381. (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: Vulkan GPU - Model: googlenet321714212835SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 331.4831.5331.55MIN: 31.29 / MAX: 41.94MIN: 31.34 / MAX: 41.26MIN: 31.32 / MAX: 41.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: googlenet321714212835Min: 31.43 / Avg: 31.48 / Max: 31.53Min: 31.5 / Avg: 31.53 / Max: 31.57Min: 31.52 / Avg: 31.55 / Max: 31.571. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg1632120406080100SE +/- 0.08, N = 3SE +/- 0.11, N = 3SE +/- 0.11, N = 3111.09111.32111.29MIN: 110.52 / MAX: 121.85MIN: 110.69 / MAX: 121.89MIN: 110.65 / MAX: 122.031. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg1632120406080100Min: 110.95 / Avg: 111.09 / Max: 111.22Min: 111.11 / Avg: 111.32 / Max: 111.46Min: 111.1 / Avg: 111.29 / Max: 111.471. (CXX) g++ options: -O3 -rdynamic -lgomp -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 Compile32120406080100SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 392.9292.7392.85
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To Compile32120406080100Min: 92.89 / Avg: 92.92 / Max: 92.95Min: 92.67 / Avg: 92.73 / Max: 92.78Min: 92.81 / Avg: 92.85 / Max: 92.9

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: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU321612182430SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 324.6024.6224.58MIN: 24.43MIN: 24.37MIN: 24.351. (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: CPU321612182430Min: 24.59 / Avg: 24.6 / Max: 24.62Min: 24.55 / Avg: 24.62 / Max: 24.68Min: 24.44 / Avg: 24.58 / Max: 24.671. (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-tiny3211326395265SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 357.2657.2357.17MIN: 56.9 / MAX: 67.71MIN: 56.88 / MAX: 68MIN: 56.85 / MAX: 67.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tiny3211122334455Min: 57.21 / Avg: 57.26 / Max: 57.29Min: 57.17 / Avg: 57.23 / Max: 57.26Min: 57.1 / Avg: 57.17 / Max: 57.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssd3211122334455SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 347.2647.2447.19MIN: 46.76 / MAX: 57.51MIN: 46.79 / MAX: 55.85MIN: 46.77 / MAX: 98.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssd3211020304050Min: 47.05 / Avg: 47.26 / Max: 47.4Min: 47.04 / Avg: 47.24 / Max: 47.45Min: 47.05 / Avg: 47.19 / Max: 47.41. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v232148121620SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 315.3115.3315.31MIN: 15.14 / MAX: 25.64MIN: 15.14 / MAX: 26.25MIN: 15.18 / MAX: 15.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v232148121620Min: 15.28 / Avg: 15.31 / Max: 15.33Min: 15.25 / Avg: 15.33 / Max: 15.38Min: 15.3 / Avg: 15.31 / Max: 15.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet18321714212835SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 330.7230.7630.75MIN: 30.5 / MAX: 40.97MIN: 30.56 / MAX: 41.08MIN: 30.52 / MAX: 40.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet18321714212835Min: 30.67 / Avg: 30.72 / Max: 30.75Min: 30.74 / Avg: 30.76 / Max: 30.79Min: 30.67 / Avg: 30.75 / Max: 30.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Fast321246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 37.967.977.971. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Fast3213691215Min: 7.96 / Avg: 7.96 / Max: 7.97Min: 7.96 / Avg: 7.97 / Max: 7.97Min: 7.96 / Avg: 7.97 / Max: 7.981. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -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: alexnet321612182430SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 324.9024.9324.93MIN: 24.72 / MAX: 35.18MIN: 24.67 / MAX: 34.83MIN: 24.74 / MAX: 351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: alexnet321612182430Min: 24.86 / Avg: 24.9 / Max: 24.95Min: 24.88 / Avg: 24.93 / Max: 24.97Min: 24.88 / Avg: 24.93 / Max: 24.951. (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 Compile32150100150200250SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.36, N = 3231.81232.00232.09
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.2.2Time To Compile3214080120160200Min: 231.63 / Avg: 231.81 / Max: 231.93Min: 231.96 / Avg: 232 / Max: 232.08Min: 231.41 / Avg: 232.09 / Max: 232.62

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: resnet503211530456075SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 368.4268.4268.50MIN: 68.07 / MAX: 79.07MIN: 68.09 / MAX: 79.66MIN: 68.08 / MAX: 122.191. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet503211326395265Min: 68.36 / Avg: 68.42 / Max: 68.46Min: 68.39 / Avg: 68.42 / Max: 68.48Min: 68.4 / Avg: 68.47 / Max: 68.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v33213691215SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 39.199.189.18MIN: 9.02 / MAX: 19.6MIN: 9.03 / MAX: 10.57MIN: 9.02 / MAX: 10.671. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v33213691215Min: 9.16 / Avg: 9.19 / Max: 9.21Min: 9.16 / Avg: 9.18 / Max: 9.22Min: 9.17 / Avg: 9.18 / Max: 9.181. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400m321510152025SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 318.6018.6218.60MIN: 18.41 / MAX: 29MIN: 18.42 / MAX: 26.27MIN: 18.49 / MAX: 28.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400m321510152025Min: 18.55 / Avg: 18.6 / Max: 18.63Min: 18.57 / Avg: 18.62 / Max: 18.66Min: 18.57 / Avg: 18.6 / Max: 18.631. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet503211530456075SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 368.4268.3668.43MIN: 68.07 / MAX: 79.4MIN: 68.07 / MAX: 79.76MIN: 68.13 / MAX: 78.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet503211326395265Min: 68.3 / Avg: 68.42 / Max: 68.5Min: 68.31 / Avg: 68.36 / Max: 68.39Min: 68.35 / Avg: 68.43 / Max: 68.521. (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: CPU3213691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 312.3412.3512.35MIN: 12.1MIN: 12.09MIN: 12.131. (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: CPU32148121620Min: 12.33 / Avg: 12.34 / Max: 12.36Min: 12.34 / Avg: 12.35 / Max: 12.36Min: 12.33 / Avg: 12.35 / Max: 12.371. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Thorough321306090120150SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3135.31135.42135.291. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Thorough321306090120150Min: 135.26 / Avg: 135.31 / Max: 135.4Min: 135.39 / Avg: 135.42 / Max: 135.45Min: 135.26 / Avg: 135.29 / Max: 135.321. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

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 WavPack32148121620SE +/- 0.02, N = 5SE +/- 0.00, N = 5SE +/- 0.00, N = 516.7116.7016.691. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPack32148121620Min: 16.69 / Avg: 16.71 / Max: 16.78Min: 16.69 / Avg: 16.7 / Max: 16.7Min: 16.69 / Avg: 16.69 / Max: 16.71. (CXX) g++ options: -rdynamic

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 103210.54141.08281.62422.16562.707SE +/- 0.003, N = 3SE +/- 0.004, N = 3SE +/- 0.008, N = 32.4062.4062.404
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 10321246810Min: 2.4 / Avg: 2.41 / Max: 2.41Min: 2.4 / Avg: 2.41 / Max: 2.41Min: 2.39 / Avg: 2.4 / Max: 2.41

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: alexnet321612182430SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 324.9424.9624.95MIN: 24.72 / MAX: 35.4MIN: 24.78 / MAX: 34.8MIN: 24.75 / MAX: 34.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnet321612182430Min: 24.85 / Avg: 24.94 / Max: 25.04Min: 24.94 / Avg: 24.96 / Max: 25.01Min: 24.93 / Avg: 24.95 / Max: 24.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tiny3211326395265SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 357.1957.1857.15MIN: 56.86 / MAX: 67.49MIN: 56.79 / MAX: 67.52MIN: 56.82 / MAX: 67.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tiny3211122334455Min: 57.18 / Avg: 57.19 / Max: 57.21Min: 57.13 / Avg: 57.18 / Max: 57.27Min: 57.14 / Avg: 57.15 / Max: 57.161. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v232148121620SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 315.3215.3215.33MIN: 15.15 / MAX: 25.61MIN: 15.18 / MAX: 27.16MIN: 15.18 / MAX: 25.51. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v232148121620Min: 15.29 / Avg: 15.32 / Max: 15.34Min: 15.27 / Avg: 15.32 / Max: 15.37Min: 15.29 / Avg: 15.33 / Max: 15.351. (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 Encode3213691215SE +/- 0.022, N = 5SE +/- 0.022, N = 5SE +/- 0.021, N = 59.5999.5979.5931. (CXX) g++ options: -fvisibility=hidden -logg -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus Encode3213691215Min: 9.57 / Avg: 9.6 / Max: 9.69Min: 9.56 / Avg: 9.6 / Max: 9.68Min: 9.56 / Avg: 9.59 / Max: 9.671. (CXX) g++ options: -fvisibility=hidden -logg -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: googlenet321714212835SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 331.5231.5131.51MIN: 31.29 / MAX: 41.86MIN: 31.31 / MAX: 41.35MIN: 31.3 / MAX: 41.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenet321714212835Min: 31.51 / Avg: 31.52 / Max: 31.55Min: 31.48 / Avg: 31.51 / Max: 31.53Min: 31.48 / Avg: 31.51 / Max: 31.561. (CXX) g++ options: -O3 -rdynamic -lgomp -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 Search321306090120150SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3114.83114.83114.831. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database Search32120406080100Min: 114.76 / Avg: 114.83 / Max: 114.91Min: 114.75 / Avg: 114.83 / Max: 114.96Min: 114.75 / Avg: 114.83 / Max: 114.931. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Medium321510152025SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 320.4220.4220.421. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: Medium321510152025Min: 20.41 / Avg: 20.42 / Max: 20.43Min: 20.41 / Avg: 20.42 / Max: 20.42Min: 20.41 / Avg: 20.42 / Max: 20.431. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

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: DistinctUserID3210.13950.2790.41850.5580.6975SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.620.620.621. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: DistinctUserID321246810Min: 0.62 / Avg: 0.62 / Max: 0.62Min: 0.62 / Avg: 0.62 / Max: 0.62Min: 0.62 / Avg: 0.62 / Max: 0.621. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: PartialTweets3210.1350.270.4050.540.675SE +/- 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: PartialTweets321246810Min: 0.6 / Avg: 0.6 / Max: 0.61Min: 0.6 / Avg: 0.6 / Max: 0.6Min: 0.6 / Avg: 0.6 / Max: 0.61. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandom3210.08330.16660.24990.33320.4165SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.370.370.371. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandom32112345Min: 0.37 / Avg: 0.37 / Max: 0.37Min: 0.37 / Avg: 0.37 / Max: 0.37Min: 0.37 / Avg: 0.37 / Max: 0.371. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: Kostya3210.11030.22060.33090.44120.5515SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.490.490.491. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: Kostya321246810Min: 0.49 / Avg: 0.49 / Max: 0.49Min: 0.49 / Avg: 0.49 / Max: 0.49Min: 0.49 / Avg: 0.49 / Max: 0.491. (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 Speedup3210.3150.630.9451.261.575SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.41.41.41. (CC) gcc options: -fopenmp -O3 -lm
OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP Speedup321246810Min: 1.4 / Avg: 1.4 / Max: 1.4Min: 1.4 / Avg: 1.4 / Max: 1.4Min: 1.4 / Avg: 1.4 / Max: 1.41. (CC) gcc options: -fopenmp -O3 -lm

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-Ptrans3210.32870.65740.98611.31481.6435SE +/- 0.00604, N = 3SE +/- 0.00835, N = 3SE +/- 0.15709, N = 31.460791.460761.311241. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-Ptrans321246810Min: 1.45 / Avg: 1.46 / Max: 1.47Min: 1.44 / Avg: 1.46 / Max: 1.47Min: 1 / Avg: 1.31 / Max: 1.471. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

90 Results Shown

oneDNN
HPC Challenge
SQLite
Stockfish
oneDNN:
  Recurrent Neural Network Training - bf16bf16bf16 - CPU
  Recurrent Neural Network Inference - bf16bf16bf16 - CPU
  Recurrent Neural Network Training - u8s8f32 - CPU
  Recurrent Neural Network Inference - u8s8f32 - CPU
  Recurrent Neural Network Inference - f32 - CPU
  Recurrent Neural Network Training - f32 - CPU
VkResample
VKMark
oneDNN
rav1e
HPC Challenge
asmFish
HPC Challenge
ASTC Encoder
VKMark
HPC Challenge
Node.js V8 Web Tooling Benchmark
Timed MAFFT Alignment
oneDNN:
  Deconvolution Batch shapes_1d - f32 - CPU
  IP Shapes 1D - f32 - CPU
HPC Challenge
Coremark
HPC Challenge
oneDNN:
  Deconvolution Batch shapes_1d - u8s8f32 - CPU
  IP Shapes 1D - u8s8f32 - CPU
HPC Challenge
oneDNN
Build2
NCNN
Crafty
NCNN:
  CPU - mnasnet
  Vulkan GPU - regnety_400m
HPC Challenge
oneDNN
NCNN
VkResample
NCNN
rav1e
NCNN:
  Vulkan GPU-v2-v2 - mobilenet-v2
  CPU-v2-v2 - mobilenet-v2
  CPU - squeezenet_ssd
oneDNN
NCNN:
  Vulkan GPU - efficientnet-b0
  Vulkan GPU - mnasnet
oneDNN
rav1e
PHPBench
NCNN:
  CPU - mobilenet
  CPU - resnet18
  Vulkan GPU - blazeface
  CPU - blazeface
  CPU - vgg16
Monkey Audio Encoding
VkFFT
NCNN:
  Vulkan GPU - googlenet
  Vulkan GPU - vgg16
Timed Eigen Compilation
oneDNN
NCNN:
  CPU - yolov4-tiny
  Vulkan GPU - squeezenet_ssd
  CPU - shufflenet-v2
  Vulkan GPU - resnet18
ASTC Encoder
NCNN
Timed FFmpeg Compilation
NCNN:
  CPU - resnet50
  CPU-v3-v3 - mobilenet-v3
  CPU - regnety_400m
  Vulkan GPU - resnet50
oneDNN
ASTC Encoder
WavPack Audio Encoding
rav1e
NCNN:
  CPU - alexnet
  Vulkan GPU - yolov4-tiny
  Vulkan GPU - shufflenet-v2
Opus Codec Encoding
NCNN
Timed HMMer Search
ASTC Encoder
simdjson:
  DistinctUserID
  PartialTweets
  LargeRand
  Kostya
CLOMP
HPC Challenge