3950X svt AMD Ryzen 9 3950X 16-Core testing with a ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS) and Sapphire AMD Radeon RX 470/480/570/570X/580/580X/590 4GB on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2103153-HA-3950XSVT011&grr&sro .
3950X svt Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Vulkan Compiler File-System Screen Resolution 1 2 3 AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads) ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS) AMD Starship/Matisse 16GB Samsung SSD 970 EVO 250GB Sapphire AMD Radeon RX 470/480/570/570X/580/580X/590 4GB (1260/1750MHz) AMD Ellesmere HDMI Audio VA2431 Intel I211 + Realtek RTL8822BE 802.11a/b/g/n/ac Ubuntu 20.04 5.9.0-050900rc6daily20200922-generic (x86_64) 20200921 GNOME Shell 3.36.4 X Server 1.20.9 4.6 Mesa 20.0.8 (LLVM 10.0.0) 1.2.128 GCC 9.3.0 ext4 1920x1080 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 ondemand (Boost: Enabled) - CPU Microcode: 0x8701021 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 STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
3950X svt onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU sysbench: CPU onednn: Recurrent Neural Network Training - f32 - CPU onednn: Recurrent Neural Network Training - u8s8f32 - CPU onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU onednn: Recurrent Neural Network Inference - f32 - CPU onednn: Recurrent Neural Network Inference - u8s8f32 - CPU svt-hevc: 1 - Bosphorus 1080p onednn: Deconvolution Batch shapes_1d - f32 - CPU onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU onednn: IP Shapes 1D - f32 - CPU onednn: IP Shapes 1D - u8s8f32 - CPU onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU sysbench: RAM / Memory onednn: IP Shapes 3D - f32 - CPU onednn: IP Shapes 3D - u8s8f32 - CPU onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU onednn: Convolution Batch Shapes Auto - f32 - CPU svt-hevc: 7 - Bosphorus 1080p svt-vp9: Visual Quality Optimized - Bosphorus 1080p svt-vp9: VMAF Optimized - Bosphorus 1080p svt-vp9: PSNR/SSIM Optimized - Bosphorus 1080p onednn: Deconvolution Batch shapes_3d - f32 - CPU onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU svt-hevc: 10 - Bosphorus 1080p 1 2 3 2207.39 35178.03 4564.79 4546.42 4601.11 2169.85 2166.90 14.83 6.84325 1.85431 4.59347 1.34993 0.712223 1.57400 8802.72 8.20256 0.632691 21.8822 20.1843 185.78 198.94 213.01 217.00 4.36341 2.56538 313.54 2172.40 35172.42 4609.00 4603.70 4619.77 2202.79 2156.84 14.83 6.86241 1.83752 4.67291 1.35442 0.717815 1.57255 8816.82 8.47307 0.666738 22.0844 20.2969 184.77 198.84 212.78 216.41 4.37641 2.55256 313.10 2196.62 35182.76 4642.11 4696.75 4610.74 2173.76 2171.52 14.87 6.86142 1.84177 4.67373 1.35123 0.720222 1.57042 8822.47 8.57986 0.689412 22.1441 20.3484 185.40 199.40 212.46 216.88 4.36025 2.55169 313.32 OpenBenchmarking.org
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU 1 2 3 500 1000 1500 2000 2500 SE +/- 22.26, N = 8 SE +/- 13.75, N = 3 SE +/- 18.66, N = 13 2207.39 2172.40 2196.62 MIN: 2144.07 MIN: 2144.29 MIN: 2144.76 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Sysbench Test: CPU OpenBenchmarking.org Events Per Second, More Is Better Sysbench 1.0.20 Test: CPU 1 2 3 8K 16K 24K 32K 40K SE +/- 2.48, N = 3 SE +/- 8.57, N = 3 SE +/- 2.97, N = 3 35178.03 35172.42 35182.76 1. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU 1 2 3 1000 2000 3000 4000 5000 SE +/- 22.44, N = 3 SE +/- 13.05, N = 3 SE +/- 16.33, N = 3 4564.79 4609.00 4642.11 MIN: 4505 MIN: 4555.25 MIN: 4601.29 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU 1 2 3 1000 2000 3000 4000 5000 SE +/- 19.10, N = 3 SE +/- 38.20, N = 3 SE +/- 6.15, N = 3 4546.42 4603.70 4696.75 MIN: 4491.77 MIN: 4517.77 MIN: 4659.16 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU 1 2 3 1000 2000 3000 4000 5000 SE +/- 27.16, N = 3 SE +/- 24.40, N = 3 SE +/- 41.24, N = 3 4601.11 4619.77 4610.74 MIN: 4523.45 MIN: 4564.55 MIN: 4494.78 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU 1 2 3 500 1000 1500 2000 2500 SE +/- 8.25, N = 3 SE +/- 18.47, N = 3 SE +/- 8.37, N = 3 2169.85 2202.79 2173.76 MIN: 2147.84 MIN: 2154.97 MIN: 2144.26 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU 1 2 3 500 1000 1500 2000 2500 SE +/- 8.14, N = 3 SE +/- 2.21, N = 3 SE +/- 10.21, N = 3 2166.90 2156.84 2171.52 MIN: 2143.78 MIN: 2144.96 MIN: 2143.49 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
SVT-HEVC Tuning: 1 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 1 - Input: Bosphorus 1080p 1 2 3 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 14.83 14.83 14.87 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU 1 2 3 2 4 6 8 10 SE +/- 0.00880, N = 3 SE +/- 0.00168, N = 3 SE +/- 0.00853, N = 3 6.84325 6.86241 6.86142 MIN: 3.54 MIN: 3.85 MIN: 3.53 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.4172 0.8344 1.2516 1.6688 2.086 SE +/- 0.01267, N = 3 SE +/- 0.00518, N = 3 SE +/- 0.00383, N = 3 1.85431 1.83752 1.84177 MIN: 1.79 MIN: 1.79 MIN: 1.79 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU 1 2 3 1.0516 2.1032 3.1548 4.2064 5.258 SE +/- 0.00553, N = 3 SE +/- 0.00547, N = 3 SE +/- 0.00892, N = 3 4.59347 4.67291 4.67373 MIN: 4.33 MIN: 4.46 MIN: 4.46 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.3047 0.6094 0.9141 1.2188 1.5235 SE +/- 0.00281, N = 3 SE +/- 0.00659, N = 3 SE +/- 0.00222, N = 3 1.34993 1.35442 1.35123 MIN: 1.32 MIN: 1.32 MIN: 1.32 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU 1 2 3 0.162 0.324 0.486 0.648 0.81 SE +/- 0.001931, N = 3 SE +/- 0.001394, N = 3 SE +/- 0.002170, N = 3 0.712223 0.717815 0.720222 MIN: 0.69 MIN: 0.69 MIN: 0.7 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.3542 0.7084 1.0626 1.4168 1.771 SE +/- 0.00339, N = 3 SE +/- 0.00439, N = 3 SE +/- 0.00081, N = 3 1.57400 1.57255 1.57042 MIN: 1.52 MIN: 1.52 MIN: 1.52 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Sysbench Test: RAM / Memory OpenBenchmarking.org MiB/sec, More Is Better Sysbench 1.0.20 Test: RAM / Memory 1 2 3 2K 4K 6K 8K 10K SE +/- 15.29, N = 3 SE +/- 9.00, N = 3 SE +/- 15.46, N = 3 8802.72 8816.82 8822.47 1. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU 1 2 3 2 4 6 8 10 SE +/- 0.03344, N = 3 SE +/- 0.00869, N = 3 SE +/- 0.01948, N = 3 8.20256 8.47307 8.57986 MIN: 7.77 MIN: 8.11 MIN: 8.21 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.1551 0.3102 0.4653 0.6204 0.7755 SE +/- 0.007730, N = 3 SE +/- 0.003939, N = 3 SE +/- 0.008883, N = 3 0.632691 0.666738 0.689412 MIN: 0.57 MIN: 0.61 MIN: 0.64 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU 1 2 3 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.13, N = 3 21.88 22.08 22.14 MIN: 21.68 MIN: 21.83 MIN: 21.88 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU 1 2 3 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 20.18 20.30 20.35 MIN: 19.8 MIN: 19.84 MIN: 19.91 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
SVT-HEVC Tuning: 7 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 7 - Input: Bosphorus 1080p 1 2 3 40 80 120 160 200 SE +/- 0.15, N = 3 SE +/- 0.59, N = 3 SE +/- 0.49, N = 3 185.78 184.77 185.40 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
SVT-VP9 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p 1 2 3 40 80 120 160 200 SE +/- 0.10, N = 3 SE +/- 0.14, N = 3 SE +/- 0.11, N = 3 198.94 198.84 199.40 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
SVT-VP9 Tuning: VMAF Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 1080p 1 2 3 50 100 150 200 250 SE +/- 0.25, N = 3 SE +/- 0.35, N = 3 SE +/- 0.15, N = 3 213.01 212.78 212.46 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
SVT-VP9 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p 1 2 3 50 100 150 200 250 SE +/- 0.59, N = 3 SE +/- 0.27, N = 3 SE +/- 0.54, N = 3 217.00 216.41 216.88 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU 1 2 3 0.9847 1.9694 2.9541 3.9388 4.9235 SE +/- 0.01253, N = 3 SE +/- 0.00857, N = 3 SE +/- 0.01339, N = 3 4.36341 4.37641 4.36025 MIN: 4.25 MIN: 4.28 MIN: 4.24 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.5772 1.1544 1.7316 2.3088 2.886 SE +/- 0.00851, N = 3 SE +/- 0.00351, N = 3 SE +/- 0.00214, N = 3 2.56538 2.55256 2.55169 MIN: 2.46 MIN: 2.45 MIN: 2.44 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
SVT-HEVC Tuning: 10 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 10 - Input: Bosphorus 1080p 1 2 3 70 140 210 280 350 SE +/- 0.88, N = 3 SE +/- 0.11, N = 3 SE +/- 0.59, N = 3 313.54 313.10 313.32 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
Phoronix Test Suite v10.8.4