icelake 2023 Tests for a future article. Intel Core i7-1065G7 testing with a Dell 06CDVY (1.0.9 BIOS) and Intel Iris Plus ICL GT2 16GB on Ubuntu 23.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2310252-NE-ICELAKE2057&grs&sro&rro .
icelake 2023 Processor Motherboard Chipset Memory Disk Graphics Audio Network OS Kernel Desktop Display Server OpenGL OpenCL Compiler File-System Screen Resolution a b Intel Core i7-1065G7 @ 3.90GHz (4 Cores / 8 Threads) Dell 06CDVY (1.0.9 BIOS) Intel Ice Lake-LP DRAM 16GB Toshiba KBG40ZPZ512G NVMe 512GB Intel Iris Plus ICL GT2 16GB (1100MHz) Realtek ALC289 Intel Ice Lake-LP PCH CNVi WiFi Ubuntu 23.04 6.2.0-24-generic (x86_64) GNOME Shell 44.0 X Server + Wayland 4.6 Mesa 23.0.4-0ubuntu1~23.04.1 OpenCL 3.0 GCC 12.3.0 ext4 1920x1200 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-DAPbBt/gcc-12-12.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-DAPbBt/gcc-12-12.3.0/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-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xb8 - Thermald 2.5.2 Security Details - itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Enhanced IBRS + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Mitigation of Microcode + tsx_async_abort: Not affected
icelake 2023 onednn: Deconvolution Batch shapes_1d - f32 - CPU ncnn: CPU - shufflenet-v2 ncnn: Vulkan GPU - mnasnet ncnn: CPU - mnasnet onednn: IP Shapes 3D - u8s8f32 - CPU oidn: RTLightmap.hdr.4096x4096 - CPU-Only onednn: IP Shapes 1D - f32 - CPU onednn: IP Shapes 1D - u8s8f32 - CPU ncnn: Vulkan GPU - FastestDet onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU svt-av1: Preset 12 - Bosphorus 4K onednn: IP Shapes 3D - f32 - CPU svt-av1: Preset 8 - Bosphorus 1080p onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU oidn: RT.ldr_alb_nrm.3840x2160 - CPU-Only oidn: RT.hdr_alb_nrm.3840x2160 - CPU-Only onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU onednn: Recurrent Neural Network Inference - u8s8f32 - CPU onednn: Recurrent Neural Network Training - f32 - CPU onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU ncnn: Vulkan GPU - resnet18 onednn: Recurrent Neural Network Inference - f32 - CPU ncnn: CPU - resnet18 svt-av1: Preset 8 - Bosphorus 4K svt-av1: Preset 4 - Bosphorus 1080p stress-ng: Wide Vector Math stress-ng: Cloning ncnn: CPU - blazeface avifenc: 10, Lossless avifenc: 6 ncnn: Vulkan GPU - alexnet ncnn: CPU - resnet50 ncnn: Vulkan GPU - blazeface avifenc: 6, Lossless svt-av1: Preset 13 - Bosphorus 4K avifenc: 2 ncnn: Vulkan GPU - vgg16 avifenc: 0 ncnn: CPU - vgg16 ncnn: CPU - alexnet stress-ng: Vector Shuffle ncnn: CPU - googlenet ncnn: Vulkan GPU - vision_transformer ncnn: Vulkan GPU - googlenet ncnn: Vulkan GPU - efficientnet-b0 ncnn: CPU - regnety_400m ncnn: Vulkan GPU - resnet50 ncnn: Vulkan GPU - squeezenet_ssd onednn: Recurrent Neural Network Training - u8s8f32 - CPU ncnn: Vulkan GPU - mobilenet svt-av1: Preset 4 - Bosphorus 4K ncnn: Vulkan GPU - regnety_400m ncnn: CPU - mobilenet ncnn: CPU - yolov4-tiny ncnn: Vulkan GPU - yolov4-tiny ncnn: CPU - squeezenet_ssd ncnn: CPU - vision_transformer ncnn: CPU - efficientnet-b0 embree: Pathtracer ISPC - Asian Dragon openradioss: Bird Strike on Windshield embree: Pathtracer ISPC - Asian Dragon Obj embree: Pathtracer - Crown ncnn: Vulkan GPU - shufflenet-v2 ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 fluidx3d: FP32-FP32 aom-av1: Speed 10 Realtime - Bosphorus 4K ncnn: CPU-v2-v2 - mobilenet-v2 ncnn: CPU - FastestDet quantlib: Multi-Threaded embree: Pathtracer - Asian Dragon Obj ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 svt-av1: Preset 13 - Bosphorus 1080p openradioss: Chrysler Neon 1M embree: Pathtracer ISPC - Crown openradioss: Bumper Beam fluidx3d: FP32-FP16S aom-av1: Speed 10 Realtime - Bosphorus 1080p onednn: Deconvolution Batch shapes_3d - f32 - CPU fluidx3d: FP32-FP16C svt-av1: Preset 12 - Bosphorus 1080p openradioss: INIVOL and Fluid Structure Interaction Drop Container aom-av1: Speed 11 Realtime - Bosphorus 4K openradioss: Cell Phone Drop Test easywave: e2Asean Grid + BengkuluSept2007 Source - 240 onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU aom-av1: Speed 9 Realtime - Bosphorus 4K openradioss: Rubber O-Ring Seal Installation quantlib: Single-Threaded aom-av1: Speed 11 Realtime - Bosphorus 1080p aom-av1: Speed 9 Realtime - Bosphorus 1080p embree: Pathtracer - Asian Dragon easywave: e2Asean Grid + BengkuluSept2007 Source - 2400 easywave: e2Asean Grid + BengkuluSept2007 Source - 1200 onednn: Convolution Batch Shapes Auto - f32 - CPU ncnn: CPU-v3-v3 - mobilenet-v3 onednn: IP Shapes 1D - bf16bf16bf16 - CPU a b 21.4405 4.33 5.24 5.83 3.81175 0.04 13.3903 6.00177 5.64 17.715 27.939 9.05715 25.293 9.48511 0.09 0.09 14376.4 7378.4 14396.7 7398.63 13.06 7400.64 12.98 9.095 2.726 88229.8 641.54 1.15 13.14 28.757 10.38 34.19 1.17 39.912 30.093 257.08 70.21 581.559 70.08 10.3 2156.51 17.18 268.47 17.19 10.78 11.97 34.28 13.89 14373.9 24.15 0.727 11.95 24.14 32.22 32.2 13.81 267.05 10.72 3.6724 921.19 3.2088 2.5641 2.71 4.35 257 34.32 4.37 5.15 7472.8 2.9828 3.54 255.397 2917.08 2.8093 512.2 500 144.8 15.1772 210 191.279 1818.21 33.21 337.47 11.91 8.16222 34.32 693.04 2898.3 150.75 146.87 3.2657 560.794 225.678 13.1328 3.54 12.9325 2.73 3.6 4.02 3.0439 0.05 10.7302 4.91146 4.84 15.4635 31.483 8.04914 28.357 8.47414 0.10 0.10 12945.5 6648.27 12976.6 6677.03 11.79 6701.8 11.81 9.974 2.989 96693.23 702.93 1.05 12.02 26.309 9.57 31.55 1.08 36.845 32.582 238.246 65.22 540.786 65.36 9.61 2306.11 16.11 251.88 16.13 10.14 11.26 32.25 13.12 13583.8 22.83 0.769 11.33 22.94 30.62 30.66 13.15 254.37 10.23 3.8233 953.64 3.3091 2.6409 2.79 4.47 264 35.13 4.46 5.25 7336.6 3.035 3.48 252.158 2947.38 2.8357 516.66 496 143.67 15.0825 209 190.598 1824.13 33.31 338.36 11.933 8.149 34.37 694.01 2894.4 150.93 147.01 3.2676 560.991 225.755 13.1309 3.54 OpenBenchmarking.org
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU b a 5 10 15 20 25 12.93 21.44 MIN: 11.01 MIN: 17.2 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
NCNN Target: CPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: shufflenet-v2 b a 0.9743 1.9486 2.9229 3.8972 4.8715 2.73 4.33 MIN: 2.61 / MAX: 12.22 MIN: 3.92 / MAX: 15.31 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: mnasnet b a 1.179 2.358 3.537 4.716 5.895 3.60 5.24 MIN: 3.47 / MAX: 8.6 MIN: 3.48 / MAX: 17.52 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: mnasnet b a 1.3118 2.6236 3.9354 5.2472 6.559 4.02 5.83 MIN: 3.47 / MAX: 16.72 MIN: 5.4 / MAX: 19 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
oneDNN Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU b a 0.8576 1.7152 2.5728 3.4304 4.288 3.04390 3.81175 MIN: 2.19 MIN: 2.41 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
Intel Open Image Denoise Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.1 Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only b a 0.0113 0.0226 0.0339 0.0452 0.0565 0.05 0.04
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU b a 3 6 9 12 15 10.73 13.39 MIN: 7.53 MIN: 7.71 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
oneDNN Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU b a 2 4 6 8 10 4.91146 6.00177 MIN: 3.5 MIN: 3.57 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
NCNN Target: Vulkan GPU - Model: FastestDet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: FastestDet b a 1.269 2.538 3.807 5.076 6.345 4.84 5.64 MIN: 4.59 / MAX: 13.78 MIN: 5.29 / MAX: 16.11 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU b a 4 8 12 16 20 15.46 17.72 MIN: 14.77 MIN: 14.78 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 4K b a 7 14 21 28 35 31.48 27.94 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU b a 3 6 9 12 15 8.04914 9.05715 MIN: 7.4 MIN: 7.61 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 1080p b a 7 14 21 28 35 28.36 25.29 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU b a 3 6 9 12 15 8.47414 9.48511 MIN: 7.22 MIN: 8.16 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
Intel Open Image Denoise Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.1 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only b a 0.0225 0.045 0.0675 0.09 0.1125 0.10 0.09
Intel Open Image Denoise Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.1 Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only b a 0.0225 0.045 0.0675 0.09 0.1125 0.10 0.09
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU b a 3K 6K 9K 12K 15K 12945.5 14376.4 MIN: 12762.5 MIN: 14135.6 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU b a 1600 3200 4800 6400 8000 6648.27 7378.40 MIN: 6485.23 MIN: 7171.27 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU b a 3K 6K 9K 12K 15K 12976.6 14396.7 MIN: 12797.4 MIN: 14160.6 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU b a 1600 3200 4800 6400 8000 6677.03 7398.63 MIN: 6521.21 MIN: 7210.77 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
NCNN Target: Vulkan GPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: resnet18 b a 3 6 9 12 15 11.79 13.06 MIN: 11.39 / MAX: 25.81 MIN: 12.51 / MAX: 26 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU b a 1600 3200 4800 6400 8000 6701.80 7400.64 MIN: 6511.57 MIN: 7186.27 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
NCNN Target: CPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: resnet18 b a 3 6 9 12 15 11.81 12.98 MIN: 11.4 / MAX: 27.26 MIN: 12.11 / MAX: 25.94 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K b a 3 6 9 12 15 9.974 9.095 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 1080p b a 0.6725 1.345 2.0175 2.69 3.3625 2.989 2.726 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Stress-NG Test: Wide Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Wide Vector Math b a 20K 40K 60K 80K 100K 96693.23 88229.80 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lmpfr -lpthread -lrt -lz
Stress-NG Test: Cloning OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Cloning b a 150 300 450 600 750 702.93 641.54 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lmpfr -lpthread -lrt -lz
NCNN Target: CPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: blazeface b a 0.2588 0.5176 0.7764 1.0352 1.294 1.05 1.15 MIN: 0.98 / MAX: 5.03 MIN: 1.08 / MAX: 7.24 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
libavif avifenc Encoder Speed: 10, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 10, Lossless b a 3 6 9 12 15 12.02 13.14 1. (CXX) g++ options: -O3 -fPIC -lm
libavif avifenc Encoder Speed: 6 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6 b a 7 14 21 28 35 26.31 28.76 1. (CXX) g++ options: -O3 -fPIC -lm
NCNN Target: Vulkan GPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: alexnet b a 3 6 9 12 15 9.57 10.38 MIN: 8.44 / MAX: 19.55 MIN: 8.85 / MAX: 21.96 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: resnet50 b a 8 16 24 32 40 31.55 34.19 MIN: 30.01 / MAX: 42.43 MIN: 32.53 / MAX: 48.47 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: blazeface b a 0.2633 0.5266 0.7899 1.0532 1.3165 1.08 1.17 MIN: 1.03 / MAX: 5.53 MIN: 1.07 / MAX: 10.59 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6, Lossless b a 9 18 27 36 45 36.85 39.91 1. (CXX) g++ options: -O3 -fPIC -lm
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 4K b a 8 16 24 32 40 32.58 30.09 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 2 b a 60 120 180 240 300 238.25 257.08 1. (CXX) g++ options: -O3 -fPIC -lm
NCNN Target: Vulkan GPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: vgg16 b a 16 32 48 64 80 65.22 70.21 MIN: 63.02 / MAX: 76.03 MIN: 66.55 / MAX: 80.56 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 0 b a 130 260 390 520 650 540.79 581.56 1. (CXX) g++ options: -O3 -fPIC -lm
NCNN Target: CPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: vgg16 b a 16 32 48 64 80 65.36 70.08 MIN: 62.97 / MAX: 77.5 MIN: 66.44 / MAX: 80.57 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: alexnet b a 3 6 9 12 15 9.61 10.30 MIN: 8.48 / MAX: 20.48 MIN: 8.89 / MAX: 20.31 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Stress-NG Test: Vector Shuffle OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Vector Shuffle b a 500 1000 1500 2000 2500 2306.11 2156.51 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lmpfr -lpthread -lrt -lz
NCNN Target: CPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: googlenet b a 4 8 12 16 20 16.11 17.18 MIN: 15.41 / MAX: 27.79 MIN: 16 / MAX: 29.65 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: vision_transformer OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: vision_transformer b a 60 120 180 240 300 251.88 268.47 MIN: 245.57 / MAX: 275.27 MIN: 262.87 / MAX: 279.73 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: googlenet b a 4 8 12 16 20 16.13 17.19 MIN: 15.44 / MAX: 27.07 MIN: 15.97 / MAX: 28.67 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: efficientnet-b0 b a 3 6 9 12 15 10.14 10.78 MIN: 8.56 / MAX: 21.18 MIN: 10.21 / MAX: 21.92 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: regnety_400m b a 3 6 9 12 15 11.26 11.97 MIN: 10.71 / MAX: 22.01 MIN: 11.19 / MAX: 23.34 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: resnet50 b a 8 16 24 32 40 32.25 34.28 MIN: 30.03 / MAX: 45.13 MIN: 32.55 / MAX: 45.96 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: squeezenet_ssd b a 4 8 12 16 20 13.12 13.89 MIN: 12.52 / MAX: 24.8 MIN: 13.1 / MAX: 28.84 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
oneDNN Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU b a 3K 6K 9K 12K 15K 13583.8 14373.9 MIN: 13104.3 MIN: 14132.5 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
NCNN Target: Vulkan GPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: mobilenet b a 6 12 18 24 30 22.83 24.15 MIN: 21.89 / MAX: 33.08 MIN: 22.83 / MAX: 34.07 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 4K b a 0.173 0.346 0.519 0.692 0.865 0.769 0.727 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
NCNN Target: Vulkan GPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: regnety_400m b a 3 6 9 12 15 11.33 11.95 MIN: 10.77 / MAX: 21.78 MIN: 11.17 / MAX: 23.16 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: mobilenet b a 6 12 18 24 30 22.94 24.14 MIN: 21.77 / MAX: 34.37 MIN: 22.95 / MAX: 34.8 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: yolov4-tiny b a 7 14 21 28 35 30.62 32.22 MIN: 29.78 / MAX: 41.83 MIN: 30.8 / MAX: 42.97 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: yolov4-tiny b a 7 14 21 28 35 30.66 32.20 MIN: 29.83 / MAX: 42 MIN: 31.01 / MAX: 43.29 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: squeezenet_ssd b a 4 8 12 16 20 13.15 13.81 MIN: 12.51 / MAX: 24.51 MIN: 13.11 / MAX: 25.08 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: vision_transformer OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: vision_transformer b a 60 120 180 240 300 254.37 267.05 MIN: 245.85 / MAX: 753.75 MIN: 261.81 / MAX: 284.75 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: efficientnet-b0 b a 3 6 9 12 15 10.23 10.72 MIN: 9.49 / MAX: 21.73 MIN: 10.17 / MAX: 21.68 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Embree Binary: Pathtracer ISPC - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon b a 0.8602 1.7204 2.5806 3.4408 4.301 3.8233 3.6724 MIN: 3.66 / MAX: 4.13 MIN: 3.64 / MAX: 3.74
OpenRadioss Model: Bird Strike on Windshield OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Bird Strike on Windshield b a 200 400 600 800 1000 953.64 921.19
Embree Binary: Pathtracer ISPC - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Obj b a 0.7445 1.489 2.2335 2.978 3.7225 3.3091 3.2088 MIN: 3.19 / MAX: 3.62 MIN: 3.18 / MAX: 3.26
Embree Binary: Pathtracer - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer - Model: Crown b a 0.5942 1.1884 1.7826 2.3768 2.971 2.6409 2.5641 MIN: 2.53 / MAX: 2.86 MIN: 2.54 / MAX: 2.62
NCNN Target: Vulkan GPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: shufflenet-v2 b a 0.6278 1.2556 1.8834 2.5112 3.139 2.79 2.71 MIN: 2.62 / MAX: 12.19 MIN: 2.62 / MAX: 12.47 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 b a 1.0058 2.0116 3.0174 4.0232 5.029 4.47 4.35 MIN: 4.2 / MAX: 14.85 MIN: 4.2 / MAX: 14.21 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
FluidX3D Test: FP32-FP32 OpenBenchmarking.org MLUPs/s, More Is Better FluidX3D 2.9 Test: FP32-FP32 b a 60 120 180 240 300 264 257
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K b a 8 16 24 32 40 35.13 34.32 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
NCNN Target: CPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU-v2-v2 - Model: mobilenet-v2 b a 1.0035 2.007 3.0105 4.014 5.0175 4.46 4.37 MIN: 4.22 / MAX: 14.73 MIN: 4.21 / MAX: 13.61 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: FastestDet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: FastestDet b a 1.1813 2.3626 3.5439 4.7252 5.9065 5.25 5.15 MIN: 5.09 / MAX: 12.35 MIN: 4.89 / MAX: 15.82 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
QuantLib Configuration: Multi-Threaded OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Multi-Threaded b a 1600 3200 4800 6400 8000 7336.6 7472.8 1. (CXX) g++ options: -O3 -march=native -fPIE -pie
Embree Binary: Pathtracer - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer - Model: Asian Dragon Obj b a 0.6829 1.3658 2.0487 2.7316 3.4145 3.0350 2.9828 MIN: 2.95 / MAX: 3.29 MIN: 2.96 / MAX: 3.03
NCNN Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 b a 0.7965 1.593 2.3895 3.186 3.9825 3.48 3.54 MIN: 3.4 / MAX: 7.46 MIN: 3.4 / MAX: 13.64 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 1080p b a 60 120 180 240 300 252.16 255.40 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenRadioss Model: Chrysler Neon 1M OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Chrysler Neon 1M b a 600 1200 1800 2400 3000 2947.38 2917.08
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Crown b a 0.638 1.276 1.914 2.552 3.19 2.8357 2.8093 MIN: 2.76 / MAX: 3.16 MIN: 2.78 / MAX: 2.89
OpenRadioss Model: Bumper Beam OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Bumper Beam b a 110 220 330 440 550 516.66 512.20
FluidX3D Test: FP32-FP16S OpenBenchmarking.org MLUPs/s, More Is Better FluidX3D 2.9 Test: FP32-FP16S b a 110 220 330 440 550 496 500
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p b a 30 60 90 120 150 143.67 144.80 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU b a 4 8 12 16 20 15.08 15.18 MIN: 14.35 MIN: 14.54 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
FluidX3D Test: FP32-FP16C OpenBenchmarking.org MLUPs/s, More Is Better FluidX3D 2.9 Test: FP32-FP16C b a 50 100 150 200 250 209 210
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 1080p b a 40 80 120 160 200 190.60 191.28 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenRadioss Model: INIVOL and Fluid Structure Interaction Drop Container OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: INIVOL and Fluid Structure Interaction Drop Container b a 400 800 1200 1600 2000 1824.13 1818.21
AOM AV1 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 4K b a 8 16 24 32 40 33.31 33.21 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenRadioss Model: Cell Phone Drop Test OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Cell Phone Drop Test b a 70 140 210 280 350 338.36 337.47
easyWave Input: e2Asean Grid + BengkuluSept2007 Source - Time: 240 OpenBenchmarking.org Seconds, Fewer Is Better easyWave r34 Input: e2Asean Grid + BengkuluSept2007 Source - Time: 240 b a 3 6 9 12 15 11.93 11.91 1. (CXX) g++ options: -O3 -fopenmp
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU b a 2 4 6 8 10 8.14900 8.16222 MIN: 7.2 MIN: 7.27 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K b a 8 16 24 32 40 34.37 34.32 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenRadioss Model: Rubber O-Ring Seal Installation OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Rubber O-Ring Seal Installation b a 150 300 450 600 750 694.01 693.04
QuantLib Configuration: Single-Threaded OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Single-Threaded b a 600 1200 1800 2400 3000 2894.4 2898.3 1. (CXX) g++ options: -O3 -march=native -fPIE -pie
AOM AV1 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 1080p b a 30 60 90 120 150 150.93 150.75 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p b a 30 60 90 120 150 147.01 146.87 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Embree Binary: Pathtracer - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer - Model: Asian Dragon b a 0.7352 1.4704 2.2056 2.9408 3.676 3.2676 3.2657 MIN: 3.2 / MAX: 3.53 MIN: 3.23 / MAX: 3.33
easyWave Input: e2Asean Grid + BengkuluSept2007 Source - Time: 2400 OpenBenchmarking.org Seconds, Fewer Is Better easyWave r34 Input: e2Asean Grid + BengkuluSept2007 Source - Time: 2400 b a 120 240 360 480 600 560.99 560.79 1. (CXX) g++ options: -O3 -fopenmp
easyWave Input: e2Asean Grid + BengkuluSept2007 Source - Time: 1200 OpenBenchmarking.org Seconds, Fewer Is Better easyWave r34 Input: e2Asean Grid + BengkuluSept2007 Source - Time: 1200 b a 50 100 150 200 250 225.76 225.68 1. (CXX) g++ options: -O3 -fopenmp
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU b a 3 6 9 12 15 13.13 13.13 MIN: 12.68 MIN: 12.76 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
NCNN Target: CPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU-v3-v3 - Model: mobilenet-v3 b a 0.7965 1.593 2.3895 3.186 3.9825 3.54 3.54 MIN: 3.39 / MAX: 13.39 MIN: 3.4 / MAX: 13.43 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Phoronix Test Suite v10.8.5