TR 3990X June AMD Ryzen Threadripper 3990X 64-Core testing with a System76 Thelio Major (F3 Z5 BIOS) and AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2006198-PTS-TR3990XJ55&grs .
TR 3990X June Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution Threadripper 3990X AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads) System76 Thelio Major (F3 Z5 BIOS) AMD Starship/Matisse 126GB Samsung SSD 970 EVO Plus 500GB AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (1750/875MHz) AMD Navi 10 HDMI Audio G237HL Intel I211 + Intel Wi-Fi 6 AX200 Ubuntu 20.04 5.4.0-29-generic (x86_64) GNOME Shell 3.36.1 X Server 1.20.8 amdgpu 19.1.0 4.6 Mesa 20.0.4 (LLVM 9.0.1) GCC 9.3.0 ext4 1920x1080 OpenBenchmarking.org - --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,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 - Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025 - Python 3.8.2 - 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 + tsx_async_abort: Not affected
TR 3990X June neatbench: CPU pyperformance: pickle_pure_python pyperformance: django_template pyperformance: python_startup pyperformance: regex_compile pyperformance: crypto_pyaes pyperformance: json_loads pyperformance: raytrace pyperformance: pathlib pyperformance: nbody pyperformance: float pyperformance: chaos pyperformance: 2to3 pyperformance: go build-clash: Time To Compile onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU onednn: Recurrent Neural Network Inference - f32 - CPU onednn: Recurrent Neural Network Training - f32 - CPU onednn: Deconvolution Batch deconv_3d - u8s8f32 - CPU onednn: Deconvolution Batch deconv_1d - u8s8f32 - CPU onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU onednn: Deconvolution Batch deconv_3d - f32 - CPU onednn: Deconvolution Batch deconv_1d - f32 - CPU onednn: Convolution Batch Shapes Auto - f32 - CPU onednn: IP Batch All - u8s8f32 - CPU onednn: IP Batch 1D - u8s8f32 - CPU onednn: IP Batch All - f32 - CPU onednn: IP Batch 1D - f32 - CPU compress-zstd: 19 compress-zstd: 3 wireguard: Threadripper 3990X 33.6 431 45.2 12 159 100.3 22.6 439 16.0 103 108 103 294 232 413.588 0.587907 0.393824 111.528 323.224 1.20727 1.43473 6.54516 1.99309 1.47260 1.08630 7.71645 2.26385 22.8213 1.16298 82.1 7197.7 239.420 OpenBenchmarking.org
NeatBench Acceleration: CPU OpenBenchmarking.org FPS, More Is Better NeatBench 5 Acceleration: CPU Threadripper 3990X 8 16 24 32 40 SE +/- 0.15, N = 3 33.6
PyPerformance Benchmark: pickle_pure_python OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: pickle_pure_python Threadripper 3990X 90 180 270 360 450 SE +/- 0.58, N = 3 431
PyPerformance Benchmark: django_template OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: django_template Threadripper 3990X 10 20 30 40 50 SE +/- 0.19, N = 3 45.2
PyPerformance Benchmark: python_startup OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: python_startup Threadripper 3990X 3 6 9 12 15 12
PyPerformance Benchmark: regex_compile OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: regex_compile Threadripper 3990X 40 80 120 160 200 159
PyPerformance Benchmark: crypto_pyaes OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: crypto_pyaes Threadripper 3990X 20 40 60 80 100 SE +/- 0.35, N = 3 100.3
PyPerformance Benchmark: json_loads OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: json_loads Threadripper 3990X 5 10 15 20 25 SE +/- 0.03, N = 3 22.6
PyPerformance Benchmark: raytrace OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: raytrace Threadripper 3990X 100 200 300 400 500 SE +/- 1.20, N = 3 439
PyPerformance Benchmark: pathlib OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: pathlib Threadripper 3990X 4 8 12 16 20 SE +/- 0.03, N = 3 16.0
PyPerformance Benchmark: nbody OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: nbody Threadripper 3990X 20 40 60 80 100 103
PyPerformance Benchmark: float OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: float Threadripper 3990X 20 40 60 80 100 108
PyPerformance Benchmark: chaos OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: chaos Threadripper 3990X 20 40 60 80 100 103
PyPerformance Benchmark: 2to3 OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: 2to3 Threadripper 3990X 60 120 180 240 300 SE +/- 0.33, N = 3 294
PyPerformance Benchmark: go OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: go Threadripper 3990X 50 100 150 200 250 232
Timed Clash Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Clash Compilation Time To Compile Threadripper 3990X 90 180 270 360 450 SE +/- 1.25, N = 3 413.59
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU Threadripper 3990X 0.1323 0.2646 0.3969 0.5292 0.6615 SE +/- 0.000969, N = 3 0.587907 MIN: 0.56 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 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU Threadripper 3990X 0.0886 0.1772 0.2658 0.3544 0.443 SE +/- 0.001977, N = 3 0.393824 MIN: 0.38 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 1.5 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU Threadripper 3990X 20 40 60 80 100 SE +/- 1.68, N = 3 111.53 MIN: 108.33 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: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU Threadripper 3990X 70 140 210 280 350 SE +/- 0.61, N = 3 323.22 MIN: 320.82 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPU Threadripper 3990X 0.2716 0.5432 0.8148 1.0864 1.358 SE +/- 0.01236, N = 3 1.20727 MIN: 1.15 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPU Threadripper 3990X 0.3228 0.6456 0.9684 1.2912 1.614 SE +/- 0.00351, N = 3 1.43473 MIN: 1.39 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 1.5 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU Threadripper 3990X 2 4 6 8 10 SE +/- 0.02018, N = 3 6.54516 MIN: 6.43 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU Threadripper 3990X 0.4484 0.8968 1.3452 1.7936 2.242 SE +/- 0.00203, N = 3 1.99309 MIN: 1.92 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU Threadripper 3990X 0.3313 0.6626 0.9939 1.3252 1.6565 SE +/- 0.00724, N = 3 1.47260 MIN: 1.42 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 1.5 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU Threadripper 3990X 0.2444 0.4888 0.7332 0.9776 1.222 SE +/- 0.01024, N = 10 1.08630 MIN: 0.95 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPU Threadripper 3990X 2 4 6 8 10 SE +/- 0.02072, N = 3 7.71645 MIN: 7.44 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPU Threadripper 3990X 0.5094 1.0188 1.5282 2.0376 2.547 SE +/- 0.01750, N = 3 2.26385 MIN: 2.04 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Batch All - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch All - Data Type: f32 - Engine: CPU Threadripper 3990X 5 10 15 20 25 SE +/- 0.09, N = 3 22.82 MIN: 22.2 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Batch 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: f32 - Engine: CPU Threadripper 3990X 0.2617 0.5234 0.7851 1.0468 1.3085 SE +/- 0.00525, N = 3 1.16298 MIN: 1.12 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Zstd Compression Compression Level: 19 OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 19 Threadripper 3990X 20 40 60 80 100 SE +/- 0.20, N = 3 82.1 1. (CC) gcc options: -O3 -pthread -lz -llzma
Zstd Compression Compression Level: 3 OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 3 Threadripper 3990X 1500 3000 4500 6000 7500 SE +/- 54.98, N = 3 7197.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
WireGuard + Linux Networking Stack Stress Test OpenBenchmarking.org Seconds, Fewer Is Better WireGuard + Linux Networking Stack Stress Test Threadripper 3990X 50 100 150 200 250 SE +/- 1.31, N = 3 239.42
Phoronix Test Suite v10.8.5