RTX 3070 Compute AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (3402 BIOS) and NVIDIA GeForce RTX 3070 8GB on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2104078-IB-RTX3070CO18&grt&sor .
RTX 3070 Compute Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL OpenCL Vulkan Compiler File-System Screen Resolution 1 2 3 AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads) ASUS ROG CROSSHAIR VIII HERO (3402 BIOS) AMD Starship/Matisse 16GB 1000GB Sabrent Rocket 4.0 Plus + 2000GB NVIDIA GeForce RTX 3070 8GB NVIDIA Device 228b ASUS VP28U Realtek RTL8125 2.5GbE + Intel I211 Ubuntu 20.04 5.8.0-48-generic (x86_64) GNOME Shell 3.36.7 X Server 1.20.9 NVIDIA 460.67 4.6.0 OpenCL 1.2 CUDA 11.2.162 1.2.155 GCC 9.3.0 + CUDA 11.2 ext4 3840x2160 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/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: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009 OpenCL Details - GPU Compute Cores: 5888 Python Details - 1: Python 3.8.5 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
RTX 3070 Compute arrayfire: Conjugate Gradient OpenCL betsy: ETC1 - Highest betsy: ETC2 RGB - Highest blender: BMW27 - CUDA blender: Classroom - CUDA blender: Fishy Cat - CUDA blender: Barbershop - CUDA blender: BMW27 - NVIDIA OptiX blender: Classroom - NVIDIA OptiX blender: Fishy Cat - NVIDIA OptiX blender: Barbershop - NVIDIA OptiX blender: Pabellon Barcelona - CUDA blender: Pabellon Barcelona - NVIDIA OptiX v-ray: NVIDIA RTX GPU v-ray: NVIDIA CUDA GPU cl-mem: Copy cl-mem: Read cl-mem: Write clpeak: Integer Compute INT clpeak: Single-Precision Float clpeak: Double-Precision Double clpeak: Global Memory Bandwidth fahbench: financebench: Black-Scholes OpenCL gromacs-gpu: Water Benchmark hashcat: MD5 hashcat: SHA1 hashcat: 7-Zip hashcat: SHA-512 hashcat: TrueCrypt RIPEMD160 + XTS indigobench: OpenCL GPU - Bedroom indigobench: OpenCL GPU - Supercar lczero: OpenCL luxcorerender-cl: DLSC luxcorerender-cl: Food luxcorerender-cl: LuxCore Benchmark luxcorerender-cl: Rainbow Colors and Prism mandelgpu: GPU mixbench: OpenCL - Integer mixbench: OpenCL - Double Precision mixbench: OpenCL - Single Precision namd-cuda: ATPase Simulation - 327,506 Atoms ncnn: Vulkan GPU - mobilenet ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 ncnn: Vulkan GPU - shufflenet-v2 ncnn: Vulkan GPU - mnasnet ncnn: Vulkan GPU - efficientnet-b0 ncnn: Vulkan GPU - blazeface ncnn: Vulkan GPU - googlenet ncnn: Vulkan GPU - vgg16 ncnn: Vulkan GPU - resnet18 ncnn: Vulkan GPU - alexnet ncnn: Vulkan GPU - resnet50 ncnn: Vulkan GPU - yolov4-tiny ncnn: Vulkan GPU - squeezenet_ssd ncnn: Vulkan GPU - regnety_400m octanebench: Total Score realsr-ncnn: 4x - No realsr-ncnn: 4x - Yes redshift: rodinia: OpenCL Particle Filter shoc: OpenCL - S3D shoc: OpenCL - Triad shoc: OpenCL - FFT SP shoc: OpenCL - MD5 Hash shoc: OpenCL - Reduction shoc: OpenCL - GEMM SGEMM_N shoc: OpenCL - Max SP Flops shoc: OpenCL - Bus Speed Download shoc: OpenCL - Bus Speed Readback shoc: OpenCL - Texture Read Bandwidth viennacl: CPU BLAS - sCOPY viennacl: CPU BLAS - sAXPY viennacl: CPU BLAS - sDOT viennacl: CPU BLAS - dCOPY viennacl: CPU BLAS - dAXPY viennacl: CPU BLAS - dDOT viennacl: CPU BLAS - dGEMV-N viennacl: CPU BLAS - dGEMV-T viennacl: CPU BLAS - dGEMM-NN viennacl: CPU BLAS - dGEMM-NT viennacl: CPU BLAS - dGEMM-TN viennacl: CPU BLAS - dGEMM-TT viennacl: OpenCL BLAS - sCOPY viennacl: OpenCL BLAS - sAXPY viennacl: OpenCL BLAS - sDOT viennacl: OpenCL BLAS - dCOPY viennacl: OpenCL BLAS - dAXPY viennacl: OpenCL BLAS - dDOT viennacl: OpenCL BLAS - dGEMV-N viennacl: OpenCL BLAS - dGEMV-T viennacl: OpenCL BLAS - dGEMM-NN viennacl: OpenCL BLAS - dGEMM-NT viennacl: OpenCL BLAS - dGEMM-TN vkfft: vkresample: 2x - Double vkresample: 2x - Single waifu2x-ncnn: 2x - 3 - Yes 1 2 3 2.086 4.309 6.039 29.07 76.10 54.20 509.33 16.11 48.17 35.70 465.14 190.29 76.77 1713 1342 297.6 393.6 380.2 10264.28 20099.75 360.90 389.57 267.0901 10.483 8.138 38778833333 13120133333 686733 1664800000 501033 12.917 37.257 28914 7.97 3.39 6.51 19.18 319970658.9 11433.74 295.11 22081.79 0.13273 12.85 4.39 4.22 4.85 4.05 5.58 1.84 13.27 55.71 13.95 11.08 24.47 22.20 14.49 16.75 410.514043 8.438 50.058 228 5.958 218.440 24.6721 1134.99 25.4712 325.671 3806.57 23117.8 26.2814 26.3950 2120.56 62.0 94.0 141 23.2 34.5 44.3 76.5 81.3 54.7 53.4 57.0 55.3 293 358 325 365 396 397 222 334 342 343 342 32004 220.452 17.489 4.297 2.094 4.300 6.051 29.03 76.49 54.44 508.38 16.17 48.26 35.84 464.01 190.98 77.16 1710 1337 296.9 393.2 379.9 10202.39 19991.73 364.99 389.61 265.8324 10.496 8.083 38839033333 13144266667 686700 1669566667 502833 12.882 37.163 29117 7.94 3.39 6.50 19.19 319688588.0 11336.97 299.65 21965.66 0.13388 13.56 4.35 4.24 4.92 4.10 5.62 1.90 13.17 55.89 14.03 11.22 24.75 23.73 15.45 16.87 411.111675 8.427 49.994 228 6.016 218.605 24.6714 1133.82 25.4940 325.801 3769.31 23179.0 26.3097 26.3909 2131.28 63.4 92.8 139 22.6 33.4 43.5 77.0 81.9 53.5 52.7 55.8 54.5 293 357 324 364 395 396 220 332 338 340 336 32323 221.037 17.532 4.303 8.437 50.181 4.321 OpenBenchmarking.org
ArrayFire Test: Conjugate Gradient OpenCL OpenBenchmarking.org ms, Fewer Is Better ArrayFire 3.7 Test: Conjugate Gradient OpenCL 1 2 0.4712 0.9424 1.4136 1.8848 2.356 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 2.086 2.094 1. (CXX) g++ options: -rdynamic
Betsy GPU Compressor Codec: ETC1 - Quality: Highest OpenBenchmarking.org Seconds, Fewer Is Better Betsy GPU Compressor 1.1 Beta Codec: ETC1 - Quality: Highest 2 1 0.9695 1.939 2.9085 3.878 4.8475 SE +/- 0.019, N = 3 SE +/- 0.010, N = 3 4.300 4.309 1. (CXX) g++ options: -O3 -O2 -lpthread -ldl
Betsy GPU Compressor Codec: ETC2 RGB - Quality: Highest OpenBenchmarking.org Seconds, Fewer Is Better Betsy GPU Compressor 1.1 Beta Codec: ETC2 RGB - Quality: Highest 1 2 2 4 6 8 10 SE +/- 0.025, N = 3 SE +/- 0.015, N = 3 6.039 6.051 1. (CXX) g++ options: -O3 -O2 -lpthread -ldl
Blender Blend File: BMW27 - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: BMW27 - Compute: CUDA 2 1 7 14 21 28 35 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 29.03 29.07
Blender Blend File: Classroom - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Classroom - Compute: CUDA 1 2 20 40 60 80 100 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 76.10 76.49
Blender Blend File: Fishy Cat - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Fishy Cat - Compute: CUDA 1 2 12 24 36 48 60 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 54.20 54.44
Blender Blend File: Barbershop - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Barbershop - Compute: CUDA 2 1 110 220 330 440 550 SE +/- 0.25, N = 3 SE +/- 0.43, N = 3 508.38 509.33
Blender Blend File: BMW27 - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: BMW27 - Compute: NVIDIA OptiX 1 2 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 16.11 16.17
Blender Blend File: Classroom - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Classroom - Compute: NVIDIA OptiX 1 2 11 22 33 44 55 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 48.17 48.26
Blender Blend File: Fishy Cat - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Fishy Cat - Compute: NVIDIA OptiX 1 2 8 16 24 32 40 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 35.70 35.84
Blender Blend File: Barbershop - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Barbershop - Compute: NVIDIA OptiX 2 1 100 200 300 400 500 SE +/- 1.26, N = 3 SE +/- 1.92, N = 3 464.01 465.14
Blender Blend File: Pabellon Barcelona - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Pabellon Barcelona - Compute: CUDA 1 2 40 80 120 160 200 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 190.29 190.98
Blender Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX 1 2 20 40 60 80 100 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 76.77 77.16
Chaos Group V-RAY Mode: NVIDIA RTX GPU OpenBenchmarking.org vrays, More Is Better Chaos Group V-RAY 5 Mode: NVIDIA RTX GPU 1 2 400 800 1200 1600 2000 SE +/- 2.00, N = 3 1713 1710
Chaos Group V-RAY Mode: NVIDIA CUDA GPU OpenBenchmarking.org vpaths, More Is Better Chaos Group V-RAY 5 Mode: NVIDIA CUDA GPU 1 2 300 600 900 1200 1500 SE +/- 1.20, N = 3 SE +/- 0.67, N = 3 1342 1337
cl-mem Benchmark: Copy OpenBenchmarking.org GB/s, More Is Better cl-mem 2017-01-13 Benchmark: Copy 1 2 60 120 180 240 300 SE +/- 0.17, N = 3 SE +/- 0.20, N = 3 297.6 296.9 1. (CC) gcc options: -O2 -flto -lOpenCL
cl-mem Benchmark: Read OpenBenchmarking.org GB/s, More Is Better cl-mem 2017-01-13 Benchmark: Read 1 2 90 180 270 360 450 SE +/- 0.07, N = 3 SE +/- 0.36, N = 3 393.6 393.2 1. (CC) gcc options: -O2 -flto -lOpenCL
cl-mem Benchmark: Write OpenBenchmarking.org GB/s, More Is Better cl-mem 2017-01-13 Benchmark: Write 1 2 80 160 240 320 400 SE +/- 0.06, N = 3 SE +/- 0.13, N = 3 380.2 379.9 1. (CC) gcc options: -O2 -flto -lOpenCL
clpeak OpenCL Test: Integer Compute INT OpenBenchmarking.org GIOPS, More Is Better clpeak OpenCL Test: Integer Compute INT 1 2 2K 4K 6K 8K 10K SE +/- 105.22, N = 5 SE +/- 86.71, N = 3 10264.28 10202.39 1. (CXX) g++ options: -O3 -rdynamic -lOpenCL
clpeak OpenCL Test: Single-Precision Float OpenBenchmarking.org GFLOPS, More Is Better clpeak OpenCL Test: Single-Precision Float 1 2 4K 8K 12K 16K 20K SE +/- 2.13, N = 3 SE +/- 109.12, N = 3 20099.75 19991.73 1. (CXX) g++ options: -O3 -rdynamic -lOpenCL
clpeak OpenCL Test: Double-Precision Double OpenBenchmarking.org GFLOPS, More Is Better clpeak OpenCL Test: Double-Precision Double 2 1 80 160 240 320 400 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 364.99 360.90 1. (CXX) g++ options: -O3 -rdynamic -lOpenCL
clpeak OpenCL Test: Global Memory Bandwidth OpenBenchmarking.org GBPS, More Is Better clpeak OpenCL Test: Global Memory Bandwidth 2 1 80 160 240 320 400 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 389.61 389.57 1. (CXX) g++ options: -O3 -rdynamic -lOpenCL
FAHBench OpenBenchmarking.org Ns Per Day, More Is Better FAHBench 2.3.2 1 2 60 120 180 240 300 SE +/- 0.04, N = 3 SE +/- 0.14, N = 3 267.09 265.83
FinanceBench Benchmark: Black-Scholes OpenCL OpenBenchmarking.org ms, Fewer Is Better FinanceBench 2016-07-25 Benchmark: Black-Scholes OpenCL 1 2 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 10.48 10.50 1. (CXX) g++ options: -O3 -march=native -fopenmp
GROMACS Water Benchmark OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2020.3 Water Benchmark 1 2 2 4 6 8 10 SE +/- 0.024, N = 3 SE +/- 0.016, N = 3 8.138 8.083 1. (CXX) g++ options: -O3 -lpthread -ldl -lrt -lm
Hashcat Benchmark: MD5 OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: MD5 2 1 8000M 16000M 24000M 32000M 40000M SE +/- 42362968.63, N = 3 SE +/- 49139065.70, N = 3 38839033333 38778833333
Hashcat Benchmark: SHA1 OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: SHA1 2 1 3000M 6000M 9000M 12000M 15000M SE +/- 5394235.61, N = 3 SE +/- 21817526.08, N = 3 13144266667 13120133333
Hashcat Benchmark: 7-Zip OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: 7-Zip 1 2 150K 300K 450K 600K 750K SE +/- 1386.04, N = 3 SE +/- 1069.27, N = 3 686733 686700
Hashcat Benchmark: SHA-512 OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: SHA-512 2 1 400M 800M 1200M 1600M 2000M SE +/- 866666.67, N = 3 SE +/- 723417.81, N = 3 1669566667 1664800000
Hashcat Benchmark: TrueCrypt RIPEMD160 + XTS OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: TrueCrypt RIPEMD160 + XTS 2 1 110K 220K 330K 440K 550K SE +/- 1197.68, N = 3 SE +/- 1550.63, N = 3 502833 501033
IndigoBench Acceleration: OpenCL GPU - Scene: Bedroom OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: OpenCL GPU - Scene: Bedroom 1 2 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 12.92 12.88
IndigoBench Acceleration: OpenCL GPU - Scene: Supercar OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: OpenCL GPU - Scene: Supercar 1 2 9 18 27 36 45 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 37.26 37.16
LeelaChessZero Backend: OpenCL OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: OpenCL 2 1 6K 12K 18K 24K 30K SE +/- 234.03, N = 3 SE +/- 74.99, N = 3 29117 28914 1. (CXX) g++ options: -flto -pthread
LuxCoreRender OpenCL Scene: DLSC OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender OpenCL 2.3 Scene: DLSC 1 2 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 7.97 7.94 MIN: 7.86 / MAX: 8.17 MIN: 7.82 / MAX: 8.14
LuxCoreRender OpenCL Scene: Food OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender OpenCL 2.3 Scene: Food 2 1 0.7628 1.5256 2.2884 3.0512 3.814 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 3.39 3.39 MIN: 0.26 / MAX: 4.22 MIN: 0.23 / MAX: 4.24
LuxCoreRender OpenCL Scene: LuxCore Benchmark OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender OpenCL 2.3 Scene: LuxCore Benchmark 1 2 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 6.51 6.50 MIN: 0.27 / MAX: 7.46 MIN: 0.32 / MAX: 7.45
LuxCoreRender OpenCL Scene: Rainbow Colors and Prism OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender OpenCL 2.3 Scene: Rainbow Colors and Prism 2 1 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 19.19 19.18 MIN: 17.89 / MAX: 20.09 MIN: 17.88 / MAX: 20.08
MandelGPU OpenCL Device: GPU OpenBenchmarking.org Samples/sec, More Is Better MandelGPU 1.3pts1 OpenCL Device: GPU 1 2 70M 140M 210M 280M 350M SE +/- 682590.08, N = 3 SE +/- 1129798.58, N = 3 319970658.9 319688588.0 1. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
Mixbench Backend: OpenCL - Benchmark: Integer OpenBenchmarking.org GIOPS, More Is Better Mixbench 2020-06-23 Backend: OpenCL - Benchmark: Integer 1 2 2K 4K 6K 8K 10K SE +/- 3.19, N = 3 SE +/- 53.09, N = 3 11433.74 11336.97 1. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2
Mixbench Backend: OpenCL - Benchmark: Double Precision OpenBenchmarking.org GFLOPS, More Is Better Mixbench 2020-06-23 Backend: OpenCL - Benchmark: Double Precision 2 1 70 140 210 280 350 SE +/- 0.76, N = 3 SE +/- 1.86, N = 3 299.65 295.11 1. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2
Mixbench Backend: OpenCL - Benchmark: Single Precision OpenBenchmarking.org GFLOPS, More Is Better Mixbench 2020-06-23 Backend: OpenCL - Benchmark: Single Precision 1 2 5K 10K 15K 20K 25K SE +/- 8.94, N = 3 SE +/- 109.50, N = 3 22081.79 21965.66 1. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2
NAMD CUDA ATPase Simulation - 327,506 Atoms OpenBenchmarking.org days/ns, Fewer Is Better NAMD CUDA 2.14 ATPase Simulation - 327,506 Atoms 1 2 0.0301 0.0602 0.0903 0.1204 0.1505 SE +/- 0.00056, N = 3 SE +/- 0.00147, N = 3 0.13273 0.13388
NCNN Target: Vulkan GPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: mobilenet 1 2 3 6 9 12 15 SE +/- 0.08, N = 3 SE +/- 0.01, N = 3 12.85 13.56 MIN: 11.95 / MAX: 34.82 MIN: 12.13 / MAX: 52.71 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 2 1 0.9878 1.9756 2.9634 3.9512 4.939 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 4.35 4.39 MIN: 3.99 / MAX: 6.3 MIN: 4.12 / MAX: 5.78 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 1 2 0.954 1.908 2.862 3.816 4.77 SE +/- 0.06, N = 3 SE +/- 0.15, N = 3 4.22 4.24 MIN: 3.9 / MAX: 30.64 MIN: 3.8 / MAX: 24.56 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: shufflenet-v2 1 2 1.107 2.214 3.321 4.428 5.535 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 4.85 4.92 MIN: 4.59 / MAX: 6.05 MIN: 4.48 / MAX: 25.29 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: mnasnet 1 2 0.9225 1.845 2.7675 3.69 4.6125 SE +/- 0.11, N = 3 SE +/- 0.01, N = 3 4.05 4.10 MIN: 3.65 / MAX: 20.14 MIN: 3.66 / MAX: 26.14 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: efficientnet-b0 1 2 1.2645 2.529 3.7935 5.058 6.3225 SE +/- 0.10, N = 3 SE +/- 0.12, N = 3 5.58 5.62 MIN: 5.12 / MAX: 20.98 MIN: 5.1 / MAX: 21.88 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: blazeface 1 2 0.4275 0.855 1.2825 1.71 2.1375 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 1.84 1.90 MIN: 1.75 / MAX: 3.08 MIN: 1.73 / MAX: 11.77 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: googlenet 2 1 3 6 9 12 15 SE +/- 0.27, N = 3 SE +/- 0.08, N = 3 13.17 13.27 MIN: 11.88 / MAX: 33.16 MIN: 12.01 / MAX: 35.62 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: vgg16 1 2 13 26 39 52 65 SE +/- 0.18, N = 3 SE +/- 0.11, N = 3 55.71 55.89 MIN: 51.94 / MAX: 108.04 MIN: 52.71 / MAX: 91.56 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: resnet18 1 2 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 13.95 14.03 MIN: 13.09 / MAX: 42.15 MIN: 12.85 / MAX: 39.19 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: alexnet 1 2 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 11.08 11.22 MIN: 10.26 / MAX: 26.8 MIN: 10.16 / MAX: 47.07 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: resnet50 1 2 6 12 18 24 30 SE +/- 0.33, N = 3 SE +/- 0.37, N = 3 24.47 24.75 MIN: 22.75 / MAX: 54.75 MIN: 22.79 / MAX: 62.12 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: yolov4-tiny 1 2 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 1.20, N = 3 22.20 23.73 MIN: 21.01 / MAX: 44.66 MIN: 21.14 / MAX: 142.5 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: squeezenet_ssd 1 2 4 8 12 16 20 SE +/- 0.15, N = 3 SE +/- 0.12, N = 3 14.49 15.45 MIN: 13.53 / MAX: 35.44 MIN: 13.94 / MAX: 74.44 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: regnety_400m 1 2 4 8 12 16 20 SE +/- 0.25, N = 3 SE +/- 0.32, N = 3 16.75 16.87 MIN: 15.59 / MAX: 33.2 MIN: 15.42 / MAX: 40.88 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OctaneBench Total Score OpenBenchmarking.org Score, More Is Better OctaneBench 2020.1 Total Score 2 1 90 180 270 360 450 411.11 410.51
RealSR-NCNN Scale: 4x - TAA: No OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No 2 3 1 2 4 6 8 10 SE +/- 0.093, N = 3 SE +/- 0.096, N = 3 SE +/- 0.085, N = 3 8.427 8.437 8.438
RealSR-NCNN Scale: 4x - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: Yes 2 1 3 11 22 33 44 55 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 49.99 50.06 50.18
RedShift Demo OpenBenchmarking.org Seconds, Fewer Is Better RedShift Demo 3.0 1 2 50 100 150 200 250 SE +/- 0.88, N = 3 SE +/- 1.00, N = 3 228 228
Rodinia Test: OpenCL Particle Filter OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenCL Particle Filter 1 2 2 4 6 8 10 SE +/- 0.014, N = 3 SE +/- 0.023, N = 3 5.958 6.016 1. (CXX) g++ options: -m64 -lm -lcuda -lcudart -lcudadevrt -lcudart_static -lrt -lpthread -ldl
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: S3D OpenBenchmarking.org GFLOPS, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: S3D 2 1 50 100 150 200 250 SE +/- 0.28, N = 3 SE +/- 0.27, N = 3 218.61 218.44 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Triad OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Triad 1 2 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 24.67 24.67 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: FFT SP OpenBenchmarking.org GFLOPS, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: FFT SP 1 2 200 400 600 800 1000 SE +/- 0.71, N = 3 SE +/- 0.99, N = 3 1134.99 1133.82 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: MD5 Hash OpenBenchmarking.org GHash/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: MD5 Hash 2 1 6 12 18 24 30 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 25.49 25.47 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Reduction OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Reduction 2 1 70 140 210 280 350 SE +/- 0.46, N = 3 SE +/- 0.56, N = 3 325.80 325.67 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: GEMM SGEMM_N OpenBenchmarking.org GFLOPS, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: GEMM SGEMM_N 1 2 800 1600 2400 3200 4000 SE +/- 10.00, N = 3 SE +/- 22.78, N = 3 3806.57 3769.31 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Max SP Flops OpenBenchmarking.org GFLOPS, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Max SP Flops 2 1 5K 10K 15K 20K 25K SE +/- 49.54, N = 3 SE +/- 31.09, N = 3 23179.0 23117.8 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Bus Speed Download OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Bus Speed Download 2 1 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 26.31 26.28 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Bus Speed Readback OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Bus Speed Readback 1 2 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 26.40 26.39 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Texture Read Bandwidth OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Texture Read Bandwidth 2 1 500 1000 1500 2000 2500 SE +/- 5.26, N = 3 SE +/- 7.35, N = 3 2131.28 2120.56 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
ViennaCL Test: CPU BLAS - sCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sCOPY 2 1 14 28 42 56 70 SE +/- 0.59, N = 3 SE +/- 0.84, N = 3 63.4 62.0 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - sAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sAXPY 1 2 20 40 60 80 100 SE +/- 0.87, N = 3 SE +/- 2.27, N = 3 94.0 92.8 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - sDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sDOT 1 2 30 60 90 120 150 SE +/- 1.15, N = 3 SE +/- 3.53, N = 3 141 139 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dCOPY 1 2 6 12 18 24 30 SE +/- 0.10, N = 3 SE +/- 0.74, N = 3 23.2 22.6 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dAXPY 1 2 8 16 24 32 40 SE +/- 0.22, N = 3 SE +/- 0.65, N = 3 34.5 33.4 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dDOT 1 2 10 20 30 40 50 SE +/- 0.42, N = 3 SE +/- 0.62, N = 3 44.3 43.5 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dGEMV-N OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-N 2 1 20 40 60 80 100 SE +/- 0.49, N = 3 SE +/- 0.45, N = 3 77.0 76.5 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dGEMV-T OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-T 2 1 20 40 60 80 100 SE +/- 0.57, N = 3 SE +/- 1.43, N = 3 81.9 81.3 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dGEMM-NN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NN 1 2 12 24 36 48 60 SE +/- 0.10, N = 3 SE +/- 1.00, N = 3 54.7 53.5 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dGEMM-NT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NT 1 2 12 24 36 48 60 SE +/- 0.15, N = 3 SE +/- 0.70, N = 3 53.4 52.7 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dGEMM-TN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TN 1 2 13 26 39 52 65 SE +/- 0.12, N = 3 SE +/- 1.22, N = 3 57.0 55.8 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dGEMM-TT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TT 1 2 12 24 36 48 60 SE +/- 0.15, N = 3 SE +/- 0.97, N = 3 55.3 54.5 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - sCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - sCOPY 2 1 60 120 180 240 300 SE +/- 0.58, N = 3 293 293 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - sAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - sAXPY 1 2 80 160 240 320 400 358 357 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - sDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - sDOT 1 2 70 140 210 280 350 325 324 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dCOPY 1 2 80 160 240 320 400 SE +/- 0.33, N = 3 365 364 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dAXPY 1 2 90 180 270 360 450 396 395 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dDOT 1 2 90 180 270 360 450 397 396 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMV-N OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMV-N 1 2 50 100 150 200 250 222 220 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMV-T OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMV-T 1 2 70 140 210 280 350 334 332 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMM-NN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMM-NN 1 2 70 140 210 280 350 SE +/- 1.00, N = 3 SE +/- 0.67, N = 3 342 338 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMM-NT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMM-NT 1 2 70 140 210 280 350 SE +/- 1.50, N = 2 343 340 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMM-TN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMM-TN 1 2 70 140 210 280 350 342 336 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
VkFFT OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.1.1 2 1 7K 14K 21K 28K 35K SE +/- 422.26, N = 3 SE +/- 238.49, N = 3 32323 32004 1. (CXX) g++ options: -O3 -pthread
VkResample Upscale: 2x - Precision: Double OpenBenchmarking.org ms, Fewer Is Better VkResample 1.0 Upscale: 2x - Precision: Double 1 2 50 100 150 200 250 SE +/- 0.07, N = 3 SE +/- 0.07, N = 3 220.45 221.04 1. (CXX) g++ options: -O3 -pthread
VkResample Upscale: 2x - Precision: Single OpenBenchmarking.org ms, Fewer Is Better VkResample 1.0 Upscale: 2x - Precision: Single 1 2 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 17.49 17.53 1. (CXX) g++ options: -O3 -pthread
Waifu2x-NCNN Vulkan Scale: 2x - Denoise: 3 - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better Waifu2x-NCNN Vulkan 20200818 Scale: 2x - Denoise: 3 - TAA: Yes 1 2 3 0.9722 1.9444 2.9166 3.8888 4.861 SE +/- 0.003, N = 3 SE +/- 0.004, N = 3 SE +/- 0.001, N = 3 4.297 4.303 4.321
Phoronix Test Suite v10.8.4