rt up AMD Ryzen 9 9950X 16-Core testing with a ASUS ROG STRIX X670E-E GAMING WIFI (2401 BIOS) and AMD Radeon PRO W7900 45GB on Ubuntu 24.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2410151-PTS-RTUP561730&grs .
rt up Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL OpenCL Compiler File-System Screen Resolution a b c d AMD Ryzen 9 9950X 16-Core @ 5.75GHz (16 Cores / 32 Threads) ASUS ROG STRIX X670E-E GAMING WIFI (2401 BIOS) AMD Device 14d8 2 x 32GB DDR5-6400MT/s Corsair CMK64GX5M2B6400C32 Western Digital WD_BLACK SN850X 2000GB + 257GB Flash Drive AMD Radeon PRO W7900 45GB (2200/3200MHz) AMD Navi 31 HDMI/DP DELL U2723QE Intel I225-V + Intel Wi-Fi 6E Ubuntu 24.04 6.10.1-061001-generic (x86_64) GNOME Shell 46.0 X Server 1.21.1.11 + Wayland 4.6 Mesa 24.2.0-devel (LLVM 18.1.7 DRM 3.57) OpenCL 2.1 AMD-APP (3625.0) GCC 13.2.0 ext4 3840x2160 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-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-backtrace --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-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/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-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: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xb404022 Security Details - gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + 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 / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
rt up litert: Mobilenet Quant litert: NASNet Mobile litert: DeepLab V3 litert: Quantized COCO SSD MobileNet v1 litert: Inception V4 xnnpack: FP32MobileNetV1 litert: Inception ResNet V2 onednn: Recurrent Neural Network Inference - CPU onednn: IP Shapes 3D - CPU xnnpack: QS8MobileNetV2 xnnpack: FP32MobileNetV3Large litert: Mobilenet Float onednn: Recurrent Neural Network Training - CPU xnnpack: FP32MobileNetV3Small onednn: IP Shapes 1D - CPU xnnpack: FP16MobileNetV2 onednn: Convolution Batch Shapes Auto - CPU xnnpack: FP32MobileNetV2 xnnpack: FP16MobileNetV3Small xnnpack: FP16MobileNetV1 xnnpack: FP16MobileNetV3Large onednn: Deconvolution Batch shapes_1d - CPU onednn: Deconvolution Batch shapes_3d - CPU litert: SqueezeNet a b c d 593.870 10524.1 1993.85 1237.78 15724.4 1024 14335.0 401.080 3.17092 747 1601 1002.97 752.845 902 0.636587 1070 5.24008 1385 851 1008 1395 1.82100 1.40825 1399.04 599.288 10303.4 1916.71 1265.69 15880.3 995 14386.0 401.052 3.18695 759 1622 1003.998 753.326 913 0.637647 1066 5.23454 1355 858 1008 1397 1.81784 1.41365 1377.42 585.758 10342.5 1971.80 1273.91 15836.1 1001 14366.4 399.152 3.17892 752 1618 1008.93 752.460 907 0.638661 1075 5.24225 1355 861 1000 1398 1.81460 1.40659 1390.86 738.067 12903.4 2250.63 1448.45 17160.6 1072 15403.4 422.910 3.35833 791 1693 1060.57 795.016 950 0.669839 1112 5.45077 1409 884 1038 1440 1.86755 1.44600 1408.46 OpenBenchmarking.org
LiteRT Model: Mobilenet Quant OpenBenchmarking.org Microseconds, Fewer Is Better LiteRT 2024-10-15 Model: Mobilenet Quant a b c d 160 320 480 640 800 SE +/- 2.40, N = 3 SE +/- 0.53, N = 3 SE +/- 6.33, N = 3 SE +/- 10.36, N = 3 593.87 599.29 585.76 738.07
LiteRT Model: NASNet Mobile OpenBenchmarking.org Microseconds, Fewer Is Better LiteRT 2024-10-15 Model: NASNet Mobile a b c d 3K 6K 9K 12K 15K SE +/- 42.72, N = 3 SE +/- 51.10, N = 3 SE +/- 86.24, N = 3 SE +/- 102.57, N = 3 10524.1 10303.4 10342.5 12903.4
LiteRT Model: DeepLab V3 OpenBenchmarking.org Microseconds, Fewer Is Better LiteRT 2024-10-15 Model: DeepLab V3 a b c d 500 1000 1500 2000 2500 SE +/- 10.75, N = 3 SE +/- 19.56, N = 3 SE +/- 10.88, N = 3 SE +/- 30.70, N = 3 1993.85 1916.71 1971.80 2250.63
LiteRT Model: Quantized COCO SSD MobileNet v1 OpenBenchmarking.org Microseconds, Fewer Is Better LiteRT 2024-10-15 Model: Quantized COCO SSD MobileNet v1 a b c d 300 600 900 1200 1500 SE +/- 1.69, N = 3 SE +/- 7.74, N = 3 SE +/- 12.41, N = 6 SE +/- 10.93, N = 3 1237.78 1265.69 1273.91 1448.45
LiteRT Model: Inception V4 OpenBenchmarking.org Microseconds, Fewer Is Better LiteRT 2024-10-15 Model: Inception V4 a b c d 4K 8K 12K 16K 20K SE +/- 91.27, N = 3 SE +/- 106.26, N = 3 SE +/- 31.83, N = 3 SE +/- 90.13, N = 3 15724.4 15880.3 15836.1 17160.6
XNNPACK Model: FP32MobileNetV1 OpenBenchmarking.org us, Fewer Is Better XNNPACK b7b048 Model: FP32MobileNetV1 a b c d 200 400 600 800 1000 SE +/- 2.33, N = 3 SE +/- 9.53, N = 3 SE +/- 5.51, N = 3 SE +/- 14.74, N = 3 1024 995 1001 1072 1. (CXX) g++ options: -O3 -lrt -lm
LiteRT Model: Inception ResNet V2 OpenBenchmarking.org Microseconds, Fewer Is Better LiteRT 2024-10-15 Model: Inception ResNet V2 a b c d 3K 6K 9K 12K 15K SE +/- 163.02, N = 3 SE +/- 120.32, N = 3 SE +/- 120.76, N = 3 SE +/- 38.80, N = 3 14335.0 14386.0 14366.4 15403.4
oneDNN Harness: Recurrent Neural Network Inference - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.6 Harness: Recurrent Neural Network Inference - Engine: CPU a b c d 90 180 270 360 450 SE +/- 0.27, N = 3 SE +/- 0.66, N = 3 SE +/- 0.53, N = 3 SE +/- 0.44, N = 3 401.08 401.05 399.15 422.91 MIN: 386.16 MIN: 385.35 MIN: 384.47 MIN: 402.96 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl
oneDNN Harness: IP Shapes 3D - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.6 Harness: IP Shapes 3D - Engine: CPU a b c d 0.7556 1.5112 2.2668 3.0224 3.778 SE +/- 0.01148, N = 3 SE +/- 0.01628, N = 3 SE +/- 0.01639, N = 3 SE +/- 0.01850, N = 3 3.17092 3.18695 3.17892 3.35833 MIN: 3.01 MIN: 3.01 MIN: 3.01 MIN: 3.02 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl
XNNPACK Model: QS8MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK b7b048 Model: QS8MobileNetV2 a b c d 200 400 600 800 1000 SE +/- 6.66, N = 3 SE +/- 2.19, N = 3 SE +/- 4.84, N = 3 SE +/- 4.00, N = 3 747 759 752 791 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP32MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK b7b048 Model: FP32MobileNetV3Large a b c d 400 800 1200 1600 2000 SE +/- 14.53, N = 3 SE +/- 2.65, N = 3 SE +/- 5.13, N = 3 SE +/- 6.69, N = 3 1601 1622 1618 1693 1. (CXX) g++ options: -O3 -lrt -lm
LiteRT Model: Mobilenet Float OpenBenchmarking.org Microseconds, Fewer Is Better LiteRT 2024-10-15 Model: Mobilenet Float a b c d 200 400 600 800 1000 SE +/- 1.06, N = 3 SE +/- 3.97, N = 3 SE +/- 3.98, N = 3 SE +/- 3.94, N = 3 1002.97 1004.00 1008.93 1060.57
oneDNN Harness: Recurrent Neural Network Training - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.6 Harness: Recurrent Neural Network Training - Engine: CPU a b c d 200 400 600 800 1000 SE +/- 1.48, N = 3 SE +/- 0.30, N = 3 SE +/- 0.72, N = 3 SE +/- 1.11, N = 3 752.85 753.33 752.46 795.02 MIN: 726.85 MIN: 729.83 MIN: 726.66 MIN: 763.71 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl
XNNPACK Model: FP32MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK b7b048 Model: FP32MobileNetV3Small a b c d 200 400 600 800 1000 SE +/- 3.61, N = 3 SE +/- 1.45, N = 3 SE +/- 3.28, N = 3 SE +/- 3.18, N = 3 902 913 907 950 1. (CXX) g++ options: -O3 -lrt -lm
oneDNN Harness: IP Shapes 1D - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.6 Harness: IP Shapes 1D - Engine: CPU a b c d 0.1507 0.3014 0.4521 0.6028 0.7535 SE +/- 0.002186, N = 3 SE +/- 0.003981, N = 3 SE +/- 0.003543, N = 3 SE +/- 0.004097, N = 3 0.636587 0.637647 0.638661 0.669839 MIN: 0.6 MIN: 0.6 MIN: 0.59 MIN: 0.59 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl
XNNPACK Model: FP16MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK b7b048 Model: FP16MobileNetV2 a b c d 200 400 600 800 1000 SE +/- 5.49, N = 3 SE +/- 6.11, N = 3 SE +/- 0.88, N = 3 SE +/- 2.19, N = 3 1070 1066 1075 1112 1. (CXX) g++ options: -O3 -lrt -lm
oneDNN Harness: Convolution Batch Shapes Auto - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.6 Harness: Convolution Batch Shapes Auto - Engine: CPU a b c d 1.2264 2.4528 3.6792 4.9056 6.132 SE +/- 0.01241, N = 3 SE +/- 0.01667, N = 3 SE +/- 0.01347, N = 3 SE +/- 0.00366, N = 3 5.24008 5.23454 5.24225 5.45077 MIN: 4.93 MIN: 4.93 MIN: 4.94 MIN: 4.96 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl
XNNPACK Model: FP32MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK b7b048 Model: FP32MobileNetV2 a b c d 300 600 900 1200 1500 SE +/- 8.45, N = 3 SE +/- 8.33, N = 3 SE +/- 6.39, N = 3 SE +/- 30.28, N = 3 1385 1355 1355 1409 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK b7b048 Model: FP16MobileNetV3Small a b c d 200 400 600 800 1000 SE +/- 1.76, N = 3 SE +/- 4.70, N = 3 SE +/- 1.67, N = 3 SE +/- 4.41, N = 3 851 858 861 884 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV1 OpenBenchmarking.org us, Fewer Is Better XNNPACK b7b048 Model: FP16MobileNetV1 a b c d 200 400 600 800 1000 SE +/- 6.17, N = 3 SE +/- 5.04, N = 3 SE +/- 7.22, N = 3 SE +/- 6.08, N = 3 1008 1008 1000 1038 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK b7b048 Model: FP16MobileNetV3Large a b c d 300 600 900 1200 1500 SE +/- 6.84, N = 3 SE +/- 2.65, N = 3 SE +/- 5.21, N = 3 SE +/- 5.51, N = 3 1395 1397 1398 1440 1. (CXX) g++ options: -O3 -lrt -lm
oneDNN Harness: Deconvolution Batch shapes_1d - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.6 Harness: Deconvolution Batch shapes_1d - Engine: CPU a b c d 0.4202 0.8404 1.2606 1.6808 2.101 SE +/- 0.01078, N = 3 SE +/- 0.00686, N = 3 SE +/- 0.00929, N = 3 SE +/- 0.01009, N = 3 1.82100 1.81784 1.81460 1.86755 MIN: 1.4 MIN: 1.4 MIN: 1.38 MIN: 1.39 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl
oneDNN Harness: Deconvolution Batch shapes_3d - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.6 Harness: Deconvolution Batch shapes_3d - Engine: CPU a b c d 0.3254 0.6508 0.9762 1.3016 1.627 SE +/- 0.00511, N = 3 SE +/- 0.00648, N = 3 SE +/- 0.00405, N = 3 SE +/- 0.00140, N = 3 1.40825 1.41365 1.40659 1.44600 MIN: 1.33 MIN: 1.33 MIN: 1.33 MIN: 1.33 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl
LiteRT Model: SqueezeNet OpenBenchmarking.org Microseconds, Fewer Is Better LiteRT 2024-10-15 Model: SqueezeNet a b c d 300 600 900 1200 1500 SE +/- 2.23, N = 3 SE +/- 12.91, N = 6 SE +/- 5.30, N = 3 SE +/- 3.75, N = 3 1399.04 1377.42 1390.86 1408.46
Phoronix Test Suite v10.8.5