RTX 3090 Compute AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (3801 BIOS) and NVIDIA GeForce RTX 3090 24GB on Ubuntu 21.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2108296-TJ-RTX3090CO87&grs .
RTX 3090 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 RTX 3090 2 3 AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads) ASUS ROG CROSSHAIR VIII HERO (3801 BIOS) AMD Starship/Matisse 16GB 1000GB Western Digital WD_BLACK SN850 1TB + 2000GB NVIDIA GeForce RTX 3090 24GB NVIDIA GA102 HD Audio ASUS VP28U Realtek RTL8125 2.5GbE + Intel I211 Ubuntu 21.10 5.13.0-14-generic (x86_64) GNOME Shell 40.2 X Server 1.20.11 NVIDIA 470.63.01 4.6.0 OpenCL 3.0 CUDA 11.4.112 1.2.175 GCC 11.2.0 ext4 3840x2160 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-M6DaQn/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-M6DaQn/gcc-11-11.2.0/debian/tmp-gcn/usr --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-build-config=bootstrap-lto-lean --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: 0xa201016 OpenCL Details - GPU Compute Cores: 10496 Python Details - RTX 3090: Python 3.9.6 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 3090 Compute ncnn: Vulkan GPU - googlenet ncnn: Vulkan GPU - yolov4-tiny ncnn: Vulkan GPU - squeezenet_ssd viennacl: CPU BLAS - sAXPY viennacl: CPU BLAS - sCOPY luxcorerender: Danish Mood - GPU viennacl: OpenCL BLAS - dCOPY viennacl: CPU BLAS - dGEMM-NN viennacl: CPU BLAS - sDOT vkfft: clpeak: Integer Compute INT shoc: OpenCL - GEMM SGEMM_N ncnn: Vulkan GPU - alexnet viennacl: CPU BLAS - dAXPY viennacl: CPU BLAS - dCOPY rodinia: OpenCL Particle Filter vkpeak: fp64-scalar vkpeak: fp16-vec4 vkpeak: fp64-vec4 vkpeak: int32-scalar ncnn: Vulkan GPU - vgg16 blender: Classroom - NVIDIA OptiX realsr-ncnn: 4x - No ncnn: Vulkan GPU - mobilenet shoc: OpenCL - Max SP Flops vkpeak: fp32-vec4 vkpeak: fp32-scalar viennacl: OpenCL BLAS - sCOPY ncnn: Vulkan GPU - blazeface ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 vkpeak: fp16-scalar blender: Barbershop - NVIDIA OptiX realsr-ncnn: 4x - Yes waifu2x-ncnn: 2x - 3 - Yes viennacl: OpenCL BLAS - dGEMM-TN viennacl: OpenCL BLAS - dGEMM-NN viennacl: OpenCL BLAS - dGEMM-NT luxcorerender: Rainbow Colors and Prism - GPU shoc: OpenCL - MD5 Hash viennacl: CPU BLAS - dGEMV-T vkpeak: int16-vec4 vkpeak: int32-vec4 arrayfire: Conjugate Gradient OpenCL vkpeak: int16-scalar luxcorerender: LuxCore Benchmark - GPU redshift: hashcat: TrueCrypt RIPEMD160 + XTS v-ray: NVIDIA RTX GPU lczero: OpenCL ncnn: Vulkan GPU - resnet18 ncnn: Vulkan GPU - efficientnet-b0 viennacl: CPU BLAS - dDOT viennacl: CPU BLAS - dGEMV-N hashcat: 7-Zip ncnn: Vulkan GPU - shufflenet-v2 ncnn: Vulkan GPU - resnet50 luxcorerender: Orange Juice - GPU viennacl: OpenCL BLAS - dAXPY viennacl: OpenCL BLAS - sDOT mandelgpu: GPU viennacl: OpenCL BLAS - dGEMV-T fahbench: ncnn: Vulkan GPU - mnasnet namd-cuda: ATPase Simulation - 327,506 Atoms blender: Classroom - CUDA indigobench: OpenCL GPU - Supercar viennacl: OpenCL BLAS - dDOT vkresample: 2x - Double viennacl: OpenCL BLAS - dGEMV-N ncnn: Vulkan GPU - regnety_400m clpeak: Global Memory Bandwidth viennacl: CPU BLAS - dGEMM-TN cl-mem: Write viennacl: OpenCL BLAS - dGEMM-TT shoc: OpenCL - Reduction hashcat: MD5 blender: Barbershop - CUDA clpeak: Double-Precision Double blender: BMW27 - CUDA blender: Fishy Cat - CUDA octanebench: Total Score indigobench: OpenCL GPU - Bedroom blender: Pabellon Barcelona - NVIDIA OptiX shoc: OpenCL - Texture Read Bandwidth shoc: OpenCL - S3D viennacl: OpenCL BLAS - sAXPY hashcat: SHA-512 viennacl: CPU BLAS - dGEMM-NT viennacl: CPU BLAS - dGEMM-TT blender: Pabellon Barcelona - CUDA hashcat: SHA1 shoc: OpenCL - FFT SP vkresample: 2x - Single shoc: OpenCL - Bus Speed Download shoc: OpenCL - Bus Speed Readback financebench: Black-Scholes OpenCL clpeak: Single-Precision Float cl-mem: Copy cl-mem: Read v-ray: NVIDIA CUDA GPU shoc: OpenCL - Triad ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 luxcorerender: DLSC - GPU blender: Fishy Cat - NVIDIA OptiX blender: BMW27 - NVIDIA OptiX RTX 3090 2 3 4.21 6.75 4.16 102 67.9 8.93 593 53.2 155 43351 18162.91 8329.86 1.78 36.3 24.3 3.730 653.53 40572.58 653.50 20763.65 4.06 28.60 5.592 4.18 40181.1 27572.34 20910.29 363 0.91 1.86 20776.35 315.25 28.508 3.306 600 602 605 32.61 44.5445 89.1 16931.70 20652.47 1.472 13670.44 11.01 138 854867 2941 53955 1.56 3.11 47.5 82.1 1153133 1.69 3.44 10.38 721 372 483619873.6 377 345.3526 1.96 0.12056 51.31 53.168 658 119.648 240 2.41 810.35 58.4 740.1 603 391.546 66726633333 373.96 657.71 18.31 34.53 681.17956 20.935 43.45 2236.38 430.301 499 2879033333 53.7 56.6 120.33 22839700000 2349.20 9.272 26.3155 27.0998 6.271 35211.37 363.8 827.3 2115 25.4664 2.10 11.50 21.66 12.20 3.9 6.77 4 105 69.4 8.66 592 53.3 156 43413 18313.8 8432.96 1.81 36.9 24.7 3.672 658.57 40867.15 658.55 20924.81 4.04 28.58 5.523 4.16 40013.3 27739.81 21142.21 363 0.92 1.84 20950.52 314.61 28.245 3.277 597 600 602 32.81 44.2953 89.8 17004.32 20673.05 1.483 13709.9 10.93 139 860700 2939 53950 1.55 3.13 47.8 81.7 1160100 1.68 3.44 10.36 720 370 483164709 376 343.5447 1.95 0.12076 51.07 53.39 658 119.394 239 2.41 813.51 58.2 739 601 390.914 66914300000 372.81 656.4 18.33 34.44 682.731009 20.987 43.51 2237.75 430.156 499 2884800000 53.8 56.7 120.54 22860600000 2352 9.271 26.338 27.1248 6.271 35226 363.7 827.3 2116 25.4781 2.1 11.5 18.94 9.5 3.85 6.68 3.99 101 67.2 8.86 607 54.6 152 42399 18533.09 8278.67 1.8 36.8 24.6 3.675 648.73 40257.04 648.72 20614.05 4.1 28.97 5.573 4.13 39723.3 27428.37 20946.96 367 0.91 1.85 20725.45 317.94 28.299 3.287 602 605 607 32.87 44.1942 89.8 16875.09 20517.61 1.473 13608.26 10.97 138 855400 2959 53598 1.55 3.12 47.5 82.2 1157500 1.69 3.43 10.42 724 371 485737287.8 377 344.8072 1.96 0.12115 51.13 53.414 660 119.107 240 2.41 813.54 58.4 737.6 390.293 66706700000 373.1 658.23 18.36 34.5 682.901259 20.976 43.55 2241.48 429.423 498 2882500000 53.8 56.6 120.48 22821000000 2348.03 9.286 26.2982 27.0876 6.263 35183.92 363.6 827.7 2116 25.4707 2.1 11.5 18.99 9.51 OpenBenchmarking.org
NCNN Target: Vulkan GPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: googlenet RTX 3090 2 3 0.9473 1.8946 2.8419 3.7892 4.7365 SE +/- 0.06, N = 3 4.21 3.87 3.85 MIN: 3.58 / MAX: 15.09 MIN: 3.62 / MAX: 14.75 MIN: 3.62 / MAX: 14.78 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: Vulkan GPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: yolov4-tiny RTX 3090 2 3 2 4 6 8 10 SE +/- 0.03, N = 3 6.75 7.00 6.68 MIN: 6.39 / MAX: 7.17 MIN: 5.95 / MAX: 18.59 MIN: 6.33 / MAX: 7.23 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: Vulkan GPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: squeezenet_ssd RTX 3090 2 3 0.936 1.872 2.808 3.744 4.68 SE +/- 0.12, N = 2 4.16 4.00 3.99 MIN: 3.94 / MAX: 5.71 MIN: 3.92 / MAX: 4.17 MIN: 3.91 / MAX: 4.27 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
ViennaCL Test: CPU BLAS - sAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sAXPY RTX 3090 2 3 20 40 60 80 100 SE +/- 0.33, N = 3 102 105 101 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - sCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sCOPY RTX 3090 2 3 15 30 45 60 75 SE +/- 0.06, N = 3 67.9 69.4 67.2 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
LuxCoreRender Scene: Danish Mood - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.5 Scene: Danish Mood - Acceleration: GPU RTX 3090 2 3 2 4 6 8 10 SE +/- 0.13, N = 3 8.93 8.66 8.86 MIN: 2.97 / MAX: 10.79 MIN: 2.96 / MAX: 10.52 MIN: 3.16 / MAX: 10.65
ViennaCL Test: OpenCL BLAS - dCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dCOPY RTX 3090 2 3 130 260 390 520 650 SE +/- 0.33, N = 3 593 594 608 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 RTX 3090 2 3 12 24 36 48 60 SE +/- 0.52, N = 3 53.2 53.3 54.6 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 RTX 3090 2 3 30 60 90 120 150 SE +/- 0.67, N = 3 155 156 152 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
VkFFT OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.1.1 RTX 3090 2 3 9K 18K 27K 36K 45K SE +/- 375.64, N = 3 43351 43413 42399 1. (CXX) g++ options: -O3 -pthread
clpeak OpenCL Test: Integer Compute INT OpenBenchmarking.org GIOPS, More Is Better clpeak OpenCL Test: Integer Compute INT RTX 3090 2 3 4K 8K 12K 16K 20K SE +/- 148.97, N = 9 18162.91 18313.80 18533.09 1. (CXX) g++ options: -O3 -rdynamic -lOpenCL
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 RTX 3090 2 3 2K 4K 6K 8K 10K SE +/- 56.22, N = 3 8329.86 8432.96 8278.67 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
NCNN Target: Vulkan GPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: alexnet RTX 3090 2 3 0.4073 0.8146 1.2219 1.6292 2.0365 SE +/- 0.01, N = 3 1.78 1.78 1.80 MIN: 1.75 / MAX: 2.11 MIN: 1.76 / MAX: 1.96 MIN: 1.77 / MAX: 1.97 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
ViennaCL Test: CPU BLAS - dAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dAXPY RTX 3090 2 3 8 16 24 32 40 SE +/- 0.25, N = 3 36.3 36.9 36.8 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 RTX 3090 2 3 6 12 18 24 30 SE +/- 0.09, N = 3 24.3 24.7 24.6 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Rodinia Test: OpenCL Particle Filter OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenCL Particle Filter RTX 3090 2 3 0.8393 1.6786 2.5179 3.3572 4.1965 SE +/- 0.044, N = 4 3.730 3.672 3.675 1. (CXX) g++ options: -O2 -lOpenCL
vkpeak fp64-scalar OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20210424 fp64-scalar RTX 3090 2 3 140 280 420 560 700 SE +/- 2.50, N = 3 653.53 658.57 648.73
vkpeak fp16-vec4 OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20210424 fp16-vec4 RTX 3090 2 3 9K 18K 27K 36K 45K SE +/- 147.11, N = 3 40572.58 40867.15 40257.04
vkpeak fp64-vec4 OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20210424 fp64-vec4 RTX 3090 2 3 140 280 420 560 700 SE +/- 2.55, N = 3 653.50 658.55 648.72
vkpeak int32-scalar OpenBenchmarking.org GIOPS, More Is Better vkpeak 20210424 int32-scalar RTX 3090 2 3 4K 8K 12K 16K 20K SE +/- 81.20, N = 3 20763.65 20924.81 20614.05
NCNN Target: Vulkan GPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: vgg16 RTX 3090 2 3 0.9225 1.845 2.7675 3.69 4.6125 SE +/- 0.01, N = 3 4.06 4.06 4.10 MIN: 4.01 / MAX: 10.61 MIN: 4.02 / MAX: 7.24 MIN: 4.01 / MAX: 14.21 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Blender Blend File: Classroom - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Classroom - Compute: NVIDIA OptiX RTX 3090 2 3 7 14 21 28 35 SE +/- 0.04, N = 3 28.60 28.58 28.97
RealSR-NCNN Scale: 4x - TAA: No OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No RTX 3090 2 3 1.2582 2.5164 3.7746 5.0328 6.291 SE +/- 0.008, N = 3 5.592 5.523 5.573
NCNN Target: Vulkan GPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: mobilenet RTX 3090 2 3 0.9405 1.881 2.8215 3.762 4.7025 SE +/- 0.00, N = 3 4.18 4.16 4.13 MIN: 4.11 / MAX: 4.51 MIN: 4.11 / MAX: 4.52 MIN: 4.09 / MAX: 4.58 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
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 RTX 3090 2 3 9K 18K 27K 36K 45K SE +/- 475.87, N = 3 40181.1 40013.3 39723.3 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
vkpeak fp32-vec4 OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20210424 fp32-vec4 RTX 3090 2 3 6K 12K 18K 24K 30K SE +/- 183.95, N = 3 27572.34 27739.81 27428.37
vkpeak fp32-scalar OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20210424 fp32-scalar RTX 3090 2 3 5K 10K 15K 20K 25K SE +/- 144.67, N = 3 20910.29 21142.21 20946.96
ViennaCL Test: OpenCL BLAS - sCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - sCOPY RTX 3090 2 3 80 160 240 320 400 SE +/- 0.58, N = 3 363 363 364 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
NCNN Target: Vulkan GPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: blazeface RTX 3090 2 3 0.207 0.414 0.621 0.828 1.035 SE +/- 0.01, N = 3 0.91 0.91 0.91 MIN: 0.89 / MAX: 1.96 MIN: 0.89 / MAX: 1.76 MIN: 0.9 / MAX: 2.07 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 RTX 3090 2 3 0.4185 0.837 1.2555 1.674 2.0925 SE +/- 0.00, N = 3 1.86 1.85 1.85 MIN: 1.82 / MAX: 5.71 MIN: 1.83 / MAX: 2.91 MIN: 1.83 / MAX: 3.41 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
vkpeak fp16-scalar OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20210424 fp16-scalar RTX 3090 2 3 4K 8K 12K 16K 20K SE +/- 67.18, N = 3 20776.35 20950.52 20725.45
Blender Blend File: Barbershop - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Barbershop - Compute: NVIDIA OptiX RTX 3090 2 3 70 140 210 280 350 SE +/- 0.18, N = 3 315.25 314.61 317.94
RealSR-NCNN Scale: 4x - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: Yes RTX 3090 2 3 7 14 21 28 35 SE +/- 0.04, N = 3 28.51 28.25 28.30
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 RTX 3090 2 3 0.7439 1.4878 2.2317 2.9756 3.7195 SE +/- 0.016, N = 3 3.306 3.277 3.287
ViennaCL Test: OpenCL BLAS - dGEMM-TN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMM-TN RTX 3090 2 3 130 260 390 520 650 SE +/- 1.53, N = 3 600 602 602 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 RTX 3090 2 3 130 260 390 520 650 SE +/- 1.67, N = 3 602 605 605 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 RTX 3090 2 3 130 260 390 520 650 SE +/- 1.33, N = 3 605 607 607 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
LuxCoreRender Scene: Rainbow Colors and Prism - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.5 Scene: Rainbow Colors and Prism - Acceleration: GPU RTX 3090 2 3 8 16 24 32 40 SE +/- 0.18, N = 3 32.61 32.81 32.87 MIN: 29.7 / MAX: 35.64 MIN: 30.72 / MAX: 35.61 MIN: 30.74 / MAX: 35.64
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 RTX 3090 2 3 10 20 30 40 50 SE +/- 0.00, N = 3 44.54 44.30 44.19 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
ViennaCL Test: CPU BLAS - dGEMV-T OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-T RTX 3090 2 3 20 40 60 80 100 SE +/- 0.03, N = 3 89.1 89.8 89.8 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
vkpeak int16-vec4 OpenBenchmarking.org GIOPS, More Is Better vkpeak 20210424 int16-vec4 RTX 3090 2 3 4K 8K 12K 16K 20K SE +/- 41.21, N = 3 16931.70 17004.32 16875.09
vkpeak int32-vec4 OpenBenchmarking.org GIOPS, More Is Better vkpeak 20210424 int32-vec4 RTX 3090 2 3 4K 8K 12K 16K 20K SE +/- 55.18, N = 3 20652.47 20673.05 20517.61
ArrayFire Test: Conjugate Gradient OpenCL OpenBenchmarking.org ms, Fewer Is Better ArrayFire 3.7 Test: Conjugate Gradient OpenCL RTX 3090 2 3 0.3337 0.6674 1.0011 1.3348 1.6685 SE +/- 0.001, N = 3 1.472 1.483 1.473 1. (CXX) g++ options: -rdynamic
vkpeak int16-scalar OpenBenchmarking.org GIOPS, More Is Better vkpeak 20210424 int16-scalar RTX 3090 2 3 3K 6K 9K 12K 15K SE +/- 20.28, N = 3 13670.44 13709.90 13608.26
LuxCoreRender Scene: LuxCore Benchmark - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.5 Scene: LuxCore Benchmark - Acceleration: GPU RTX 3090 2 3 3 6 9 12 15 SE +/- 0.03, N = 3 11.01 10.93 10.97 MIN: 3.31 / MAX: 13.09 MIN: 3.29 / MAX: 12.95 MIN: 3.33 / MAX: 13.02
RedShift Demo OpenBenchmarking.org Seconds, Fewer Is Better RedShift Demo 3.0 RTX 3090 2 3 30 60 90 120 150 SE +/- 0.33, N = 3 138 139 138
Hashcat Benchmark: TrueCrypt RIPEMD160 + XTS OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: TrueCrypt RIPEMD160 + XTS RTX 3090 2 3 200K 400K 600K 800K 1000K SE +/- 1741.97, N = 3 854867 860700 855400
Chaos Group V-RAY Mode: NVIDIA RTX GPU OpenBenchmarking.org vrays, More Is Better Chaos Group V-RAY 5 Mode: NVIDIA RTX GPU RTX 3090 2 3 600 1200 1800 2400 3000 SE +/- 11.15, N = 3 2941 2939 2959
LeelaChessZero Backend: OpenCL OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.28 Backend: OpenCL RTX 3090 2 3 12K 24K 36K 48K 60K SE +/- 24.25, N = 3 53955 53950 53598 1. (CXX) g++ options: -flto -pthread
NCNN Target: Vulkan GPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: resnet18 RTX 3090 2 3 0.351 0.702 1.053 1.404 1.755 SE +/- 0.02, N = 3 1.56 1.55 1.55 MIN: 1.52 / MAX: 10.44 MIN: 1.52 / MAX: 1.82 MIN: 1.54 / MAX: 1.85 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: Vulkan GPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: efficientnet-b0 RTX 3090 2 3 0.7043 1.4086 2.1129 2.8172 3.5215 SE +/- 0.01, N = 3 3.11 3.11 3.12 MIN: 3.09 / MAX: 4.93 MIN: 3.1 / MAX: 3.23 MIN: 3.1 / MAX: 4.74 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
ViennaCL Test: CPU BLAS - dDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dDOT RTX 3090 2 3 11 22 33 44 55 SE +/- 0.23, N = 3 47.5 47.8 47.5 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 RTX 3090 2 3 20 40 60 80 100 SE +/- 0.10, N = 3 82.1 81.7 82.2 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Hashcat Benchmark: 7-Zip OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: 7-Zip RTX 3090 2 3 200K 400K 600K 800K 1000K SE +/- 1017.08, N = 3 1153133 1160100 1157500
NCNN Target: Vulkan GPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: shufflenet-v2 RTX 3090 2 3 0.3803 0.7606 1.1409 1.5212 1.9015 SE +/- 0.01, N = 3 1.69 1.69 1.69 MIN: 1.67 / MAX: 3.01 MIN: 1.67 / MAX: 2.34 MIN: 1.67 / MAX: 2.42 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: Vulkan GPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: resnet50 RTX 3090 2 3 0.7763 1.5526 2.3289 3.1052 3.8815 SE +/- 0.01, N = 3 3.44 3.45 3.43 MIN: 3.42 / MAX: 3.57 MIN: 3.44 / MAX: 3.58 MIN: 3.42 / MAX: 3.7 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
LuxCoreRender Scene: Orange Juice - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.5 Scene: Orange Juice - Acceleration: GPU RTX 3090 2 3 3 6 9 12 15 SE +/- 0.03, N = 3 10.38 10.36 10.42 MIN: 8.42 / MAX: 13.74 MIN: 8.48 / MAX: 13.54 MIN: 8.51 / MAX: 13.74
ViennaCL Test: OpenCL BLAS - dAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dAXPY RTX 3090 2 3 160 320 480 640 800 SE +/- 0.33, N = 3 721 721 724 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 RTX 3090 2 3 80 160 240 320 400 SE +/- 0.33, N = 3 372 371 371 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
MandelGPU OpenCL Device: GPU OpenBenchmarking.org Samples/sec, More Is Better MandelGPU 1.3pts1 OpenCL Device: GPU RTX 3090 2 3 100M 200M 300M 400M 500M SE +/- 222116.26, N = 3 483619873.6 483164709.0 485737287.8 1. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
ViennaCL Test: OpenCL BLAS - dGEMV-T OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMV-T RTX 3090 2 3 80 160 240 320 400 SE +/- 0.33, N = 3 377 377 378 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
FAHBench OpenBenchmarking.org Ns Per Day, More Is Better FAHBench 2.3.2 RTX 3090 2 3 80 160 240 320 400 SE +/- 0.17, N = 3 345.35 343.54 344.81
NCNN Target: Vulkan GPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: mnasnet RTX 3090 2 3 0.441 0.882 1.323 1.764 2.205 SE +/- 0.01, N = 3 1.96 1.96 1.96 MIN: 1.94 / MAX: 3.05 MIN: 1.94 / MAX: 2.15 MIN: 1.94 / MAX: 2.95 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NAMD CUDA ATPase Simulation - 327,506 Atoms OpenBenchmarking.org days/ns, Fewer Is Better NAMD CUDA 2.14 ATPase Simulation - 327,506 Atoms RTX 3090 2 3 0.0273 0.0546 0.0819 0.1092 0.1365 SE +/- 0.00027, N = 3 0.12056 0.12076 0.12115
Blender Blend File: Classroom - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Classroom - Compute: CUDA RTX 3090 2 3 12 24 36 48 60 SE +/- 0.05, N = 3 51.31 51.07 51.13
IndigoBench Acceleration: OpenCL GPU - Scene: Supercar OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: OpenCL GPU - Scene: Supercar RTX 3090 2 3 12 24 36 48 60 SE +/- 0.06, N = 3 53.17 53.39 53.41
ViennaCL Test: OpenCL BLAS - dDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dDOT RTX 3090 2 3 140 280 420 560 700 SE +/- 0.67, N = 3 658 660 661 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
VkResample Upscale: 2x - Precision: Double OpenBenchmarking.org ms, Fewer Is Better VkResample 1.0 Upscale: 2x - Precision: Double RTX 3090 2 3 30 60 90 120 150 SE +/- 0.14, N = 3 119.65 119.39 119.11 1. (CXX) g++ options: -O3 -pthread
ViennaCL Test: OpenCL BLAS - dGEMV-N OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMV-N RTX 3090 2 3 50 100 150 200 250 SE +/- 0.33, N = 3 240 240 240 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
NCNN Target: Vulkan GPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU - Model: regnety_400m RTX 3090 2 3 0.5423 1.0846 1.6269 2.1692 2.7115 SE +/- 0.01, N = 3 2.41 2.40 2.41 MIN: 2.37 / MAX: 3.46 MIN: 2.39 / MAX: 2.53 MIN: 2.39 / MAX: 3.92 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
clpeak OpenCL Test: Global Memory Bandwidth OpenBenchmarking.org GBPS, More Is Better clpeak OpenCL Test: Global Memory Bandwidth RTX 3090 2 3 200 400 600 800 1000 SE +/- 3.03, N = 3 810.35 813.51 813.54 1. (CXX) g++ options: -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 RTX 3090 2 3 13 26 39 52 65 SE +/- 0.09, N = 3 58.4 58.2 58.4 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
cl-mem Benchmark: Write OpenBenchmarking.org GB/s, More Is Better cl-mem 2017-01-13 Benchmark: Write RTX 3090 2 3 160 320 480 640 800 SE +/- 0.23, N = 3 740.1 739.0 737.6 1. (CC) gcc options: -O2 -flto -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMM-TT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMM-TT RTX 3090 2 130 260 390 520 650 SE +/- 1.67, N = 3 603 601 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
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 RTX 3090 2 3 80 160 240 320 400 SE +/- 0.20, N = 3 391.55 390.91 390.29 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
Hashcat Benchmark: MD5 OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: MD5 RTX 3090 2 3 14000M 28000M 42000M 56000M 70000M SE +/- 4511220.58, N = 3 66726633333 66914300000 66706700000
Blender Blend File: Barbershop - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Barbershop - Compute: CUDA RTX 3090 2 3 80 160 240 320 400 SE +/- 0.27, N = 3 373.96 372.81 373.10
clpeak OpenCL Test: Double-Precision Double OpenBenchmarking.org GFLOPS, More Is Better clpeak OpenCL Test: Double-Precision Double RTX 3090 2 3 140 280 420 560 700 SE +/- 0.02, N = 3 657.71 656.40 658.23 1. (CXX) g++ options: -O3 -rdynamic -lOpenCL
Blender Blend File: BMW27 - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: BMW27 - Compute: CUDA RTX 3090 2 3 5 10 15 20 25 SE +/- 0.01, N = 3 18.31 18.33 18.36
Blender Blend File: Fishy Cat - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Fishy Cat - Compute: CUDA RTX 3090 2 3 8 16 24 32 40 SE +/- 0.01, N = 3 34.53 34.44 34.50
OctaneBench Total Score OpenBenchmarking.org Score, More Is Better OctaneBench 2020.1 Total Score RTX 3090 2 3 150 300 450 600 750 681.18 682.73 682.90
IndigoBench Acceleration: OpenCL GPU - Scene: Bedroom OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: OpenCL GPU - Scene: Bedroom RTX 3090 2 3 5 10 15 20 25 SE +/- 0.05, N = 3 20.94 20.99 20.98
Blender Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX RTX 3090 2 3 10 20 30 40 50 SE +/- 0.09, N = 3 43.45 43.51 43.55
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 RTX 3090 2 3 500 1000 1500 2000 2500 SE +/- 5.62, N = 3 2236.38 2237.75 2241.48 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
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 RTX 3090 2 3 90 180 270 360 450 SE +/- 0.25, N = 3 430.30 430.16 429.42 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
ViennaCL Test: OpenCL BLAS - sAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - sAXPY RTX 3090 2 3 110 220 330 440 550 SE +/- 0.33, N = 3 499 499 498 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Hashcat Benchmark: SHA-512 OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: SHA-512 RTX 3090 2 3 600M 1200M 1800M 2400M 3000M SE +/- 3299158.14, N = 3 2879033333 2884800000 2882500000
ViennaCL Test: CPU BLAS - dGEMM-NT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NT RTX 3090 2 3 12 24 36 48 60 SE +/- 0.07, N = 3 53.7 53.8 53.8 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 RTX 3090 2 3 13 26 39 52 65 SE +/- 0.09, N = 3 56.6 56.7 56.6 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Blender Blend File: Pabellon Barcelona - Compute: CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Pabellon Barcelona - Compute: CUDA RTX 3090 2 3 30 60 90 120 150 SE +/- 0.05, N = 3 120.33 120.54 120.48
Hashcat Benchmark: SHA1 OpenBenchmarking.org H/s, More Is Better Hashcat 6.1.1 Benchmark: SHA1 RTX 3090 2 3 5000M 10000M 15000M 20000M 25000M SE +/- 24017979.38, N = 3 22839700000 22860600000 22821000000
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 RTX 3090 2 3 500 1000 1500 2000 2500 SE +/- 1.26, N = 3 2349.20 2352.00 2348.03 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
VkResample Upscale: 2x - Precision: Single OpenBenchmarking.org ms, Fewer Is Better VkResample 1.0 Upscale: 2x - Precision: Single RTX 3090 2 3 3 6 9 12 15 SE +/- 0.015, N = 3 9.272 9.271 9.286 1. (CXX) g++ options: -O3 -pthread
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 RTX 3090 2 3 6 12 18 24 30 SE +/- 0.03, N = 3 26.32 26.34 26.30 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 RTX 3090 2 3 6 12 18 24 30 SE +/- 0.01, N = 3 27.10 27.12 27.09 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
FinanceBench Benchmark: Black-Scholes OpenCL OpenBenchmarking.org ms, Fewer Is Better FinanceBench 2016-07-25 Benchmark: Black-Scholes OpenCL RTX 3090 2 3 2 4 6 8 10 SE +/- 0.001, N = 3 6.271 6.271 6.263 1. (CXX) g++ options: -O3 -march=native -fopenmp
clpeak OpenCL Test: Single-Precision Float OpenBenchmarking.org GFLOPS, More Is Better clpeak OpenCL Test: Single-Precision Float RTX 3090 2 3 8K 16K 24K 32K 40K SE +/- 13.13, N = 3 35211.37 35226.00 35183.92 1. (CXX) g++ options: -O3 -rdynamic -lOpenCL
cl-mem Benchmark: Copy OpenBenchmarking.org GB/s, More Is Better cl-mem 2017-01-13 Benchmark: Copy RTX 3090 2 3 80 160 240 320 400 SE +/- 0.15, N = 3 363.8 363.7 363.6 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 RTX 3090 2 3 200 400 600 800 1000 SE +/- 0.42, N = 3 827.3 827.3 827.7 1. (CC) gcc options: -O2 -flto -lOpenCL
Chaos Group V-RAY Mode: NVIDIA CUDA GPU OpenBenchmarking.org vpaths, More Is Better Chaos Group V-RAY 5 Mode: NVIDIA CUDA GPU RTX 3090 2 3 500 1000 1500 2000 2500 SE +/- 0.88, N = 3 2115 2116 2116
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 RTX 3090 2 3 6 12 18 24 30 SE +/- 0.02, N = 3 25.47 25.48 25.47 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
NCNN Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 RTX 3090 2 3 0.4725 0.945 1.4175 1.89 2.3625 SE +/- 0.01, N = 3 2.10 2.10 2.10 MIN: 2.07 / MAX: 3.1 MIN: 2.08 / MAX: 3.15 MIN: 2.08 / MAX: 2.65 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
LuxCoreRender Scene: DLSC - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.5 Scene: DLSC - Acceleration: GPU RTX 3090 2 3 3 6 9 12 15 SE +/- 0.01, N = 3 11.50 11.50 11.50 MIN: 11.31 / MAX: 11.72 MIN: 11.34 / MAX: 11.68 MIN: 11.35 / MAX: 11.71
Blender Blend File: Fishy Cat - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Fishy Cat - Compute: NVIDIA OptiX RTX 3090 2 3 5 10 15 20 25 SE +/- 2.66, N = 15 21.66 18.94 18.99
Blender Blend File: BMW27 - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: BMW27 - Compute: NVIDIA OptiX RTX 3090 2 3 3 6 9 12 15 SE +/- 2.70, N = 15 12.20 9.50 9.51
Phoronix Test Suite v10.8.4