AMD Ryzen Threadripper 3960X 24-Core testing with a MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS) and AMD Navi 10 8GB on Ubuntu 19.10 via the Phoronix Test Suite.
Threadripper 3960X Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-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 -vDisk Notes: NONE / errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Java Notes: OpenJDK Runtime Environment (build 11.0.5-ea+10-post-Ubuntu-0ubuntu1)Python Notes: Python 2.7.17rc1 + Python 3.7.5rc1Security Notes: 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
AMD Ryzen Threadripper 3960X 24-Core Processor: AMD Ryzen Threadripper 3960X 24-Core @ 3.80GHz (24 Cores / 48 Threads), Motherboard: MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS), Chipset: AMD Starship/Matisse, Memory: 32768MB, Disk: 240GB Force MP510, Graphics: AMD Navi 10 8GB (2100/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS PB278, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Device 2723
OS: Ubuntu 19.10, Kernel: 5.3.0-23-generic (x86_64), Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, OpenGL: 4.5 Mesa 19.2.1 (LLVM 9.0.0), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 2560x1440
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-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 -vProcessor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Java Notes: OpenJDK Runtime Environment (build 11.0.5-ea+10-post-Ubuntu-0ubuntu1)Python Notes: Python 2.7.17rc1 + Python 3.7.5rc1Security Notes: 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
POV-Ray This is a test of POV-Ray, the Persistence of Vision Raytracer. POV-Ray is used to create 3D graphics using ray-tracing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray 3.7.0.7 Trace Time AMD Ryzen Threadripper 3960X 24-Core Threadripper 3960X 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 16.36 16.41 1. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSDL -lXpm -lSM -lICE -lX11 -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lpthread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system
LeelaChessZero LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.22.0 Backend: Random Threadripper 3960X AMD Ryzen Threadripper 3960X 24-Core 20K 40K 60K 80K 100K SE +/- 513.36, N = 3 SE +/- 557.11, N = 3 106908 106669 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Random Numbers AMD Ryzen Threadripper 3960X 24-Core Threadripper 3960X 200 400 600 800 1000 SE +/- 1.36, N = 3 SE +/- 0.65, N = 3 1103.85 1105.43 1. (CXX) g++ options: -std=c++11 -O3
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Function Objects Threadripper 3960X AMD Ryzen Threadripper 3960X 24-Core 4 8 12 16 20 SE +/- 0.12, N = 3 SE +/- 0.02, N = 3 14.68 14.69 1. (CXX) g++ options: -std=c++11 -O3
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Stepanov Abstraction Threadripper 3960X AMD Ryzen Threadripper 3960X 24-Core 7 14 21 28 35 SE +/- 0.15, N = 3 SE +/- 0.01, N = 3 28.30 28.31 1. (CXX) g++ options: -std=c++11 -O3
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve OpenMP - Gridding AMD Ryzen Threadripper 3960X 24-Core 1000 2000 3000 4000 5000 SE +/- 26.85, N = 3 4617.47 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve MPI - Degridding AMD Ryzen Threadripper 3960X 24-Core 700 1400 2100 2800 3500 SE +/- 9.46, N = 3 3311.01 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve MPI - Gridding AMD Ryzen Threadripper 3960X 24-Core 400 800 1200 1600 2000 SE +/- 4.29, N = 3 1865.03 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve MT - Degridding AMD Ryzen Threadripper 3960X 24-Core 700 1400 2100 2800 3500 SE +/- 8.26, N = 3 3307.00 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve MT - Gridding AMD Ryzen Threadripper 3960X 24-Core 400 800 1200 1600 2000 SE +/- 9.48, N = 3 1911.41 1. (CXX) g++ options: -lpthread
Rust Mandelbrot This test profile is of the combined time for the serial and parallel Mandelbrot sets written in Rustlang via willi-kappler/mandel-rust. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rust Mandelbrot Time To Complete Serial/Parallel Mandelbrot AMD Ryzen Threadripper 3960X 24-Core 8 16 24 32 40 SE +/- 0.11, N = 3 32.70 1. (CC) gcc options: -m64 -pie -nodefaultlibs -lutil -ldl -lrt -lpthread -lgcc_s -lc -lm
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ns Per Event, Fewer Is Better OSBench Test: Memory Allocations Threadripper 3960X 16 32 48 64 80 SE +/- 0.15, N = 3 70.60 1. (CC) gcc options: -lm
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes Threadripper 3960X 9 18 27 36 45 SE +/- 0.55, N = 3 41.04 1. (CC) gcc options: -lm
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Launch Programs Threadripper 3960X 11 22 33 44 55 SE +/- 0.47, N = 3 48.15 1. (CC) gcc options: -lm
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads Threadripper 3960X 5 10 15 20 25 SE +/- 0.25, N = 3 20.84 1. (CC) gcc options: -lm
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files Threadripper 3960X 3 6 9 12 15 SE +/- 0.01, N = 3 11.54 1. (CC) gcc options: -lm
BRL-CAD BRL-CAD 7.28.0 is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.28.0 VGR Performance Metric Threadripper 3960X 80K 160K 240K 320K 400K 393502 1. (CXX) g++ options: -std=c++98 -pipe -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -rdynamic -lSM -lICE -lXi -lGLU -lGL -lX11 -lXext -lXrender -lpthread -ldl -luuid -lm
Git This test measures the time needed to carry out some sample Git operations on an example, static repository that happens to be a copy of the GNOME GTK tool-kit repository. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Git Time To Complete Common Git Commands Threadripper 3960X 1.0721 2.1442 3.2163 4.2884 5.3605 SE +/- 0.035, N = 3 4.765 1. git version 2.20.1
OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Google Chrome Threadripper 3960X 30 60 90 120 150 SE +/- 0.33, N = 3 133 1. chrome 78.0.3904.108
OpenBenchmarking.org Seconds, Fewer Is Better Selenium Benchmark: Maze Solver - Browser: Google Chrome Threadripper 3960X 0.9 1.8 2.7 3.6 4.5 SE +/- 0.03, N = 3 4.0 1. chrome 78.0.3904.108
OpenBenchmarking.org Runs / Minute, More Is Better Selenium Benchmark: StyleBench - Browser: Google Chrome Threadripper 3960X 8 16 24 32 40 SE +/- 0.09, N = 3 36.6 1. chrome 78.0.3904.108
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream - Browser: Google Chrome Threadripper 3960X 50 100 150 200 250 SE +/- 0.77, N = 3 217.50 1. chrome 78.0.3904.108
OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Firefox Threadripper 3960X 300 600 900 1200 1500 SE +/- 5.46, N = 3 1324 1. firefox 70.0.1
OpenBenchmarking.org Geometric Mean, More Is Better Selenium Benchmark: Octane - Browser: Google Chrome Threadripper 3960X 10K 20K 30K 40K 50K SE +/- 57.21, N = 3 47433 1. chrome 78.0.3904.108
OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: ARES-6 - Browser: Google Chrome Threadripper 3960X 5 10 15 20 25 SE +/- 0.05, N = 3 19.97 1. chrome 78.0.3904.108
OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Firefox Threadripper 3960X 20 40 60 80 100 SE +/- 0.27, N = 3 97.1 1. firefox 70.0.1
OpenBenchmarking.org Seconds, Fewer Is Better Selenium Benchmark: Maze Solver - Browser: Firefox Threadripper 3960X 0.99 1.98 2.97 3.96 4.95 SE +/- 0.07, N = 3 4.4 1. firefox 70.0.1
OpenBenchmarking.org Runs / Minute, More Is Better Selenium Benchmark: StyleBench - Browser: Firefox Threadripper 3960X 20 40 60 80 100 SE +/- 0.32, N = 3 79.0 1. firefox 70.0.1
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream - Browser: Firefox Threadripper 3960X 50 100 150 200 250 SE +/- 0.36, N = 3 217.57 1. firefox 70.0.1
OpenBenchmarking.org Geometric Mean, More Is Better Selenium Benchmark: Octane - Browser: Firefox Threadripper 3960X 8K 16K 24K 32K 40K SE +/- 203.61, N = 3 36739 1. firefox 70.0.1
OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: ARES-6 - Browser: Firefox Threadripper 3960X 11 22 33 44 55 SE +/- 0.41, N = 3 50.67 1. firefox 70.0.1
PHPBench PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. The number of iterations used is 1,000,000. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite Threadripper 3960X 140K 280K 420K 560K 700K SE +/- 5664.15, N = 3 658408
Apache Benchmark This is a test of ab, which is the Apache benchmark program. This test profile measures how many requests per second a given system can sustain when carrying out 1,000,000 requests with 100 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better Apache Benchmark 2.4.29 Static Web Page Serving Threadripper 3960X 7K 14K 21K 28K 35K SE +/- 49.18, N = 3 34193.15 1. (CC) gcc options: -shared -fPIC -O2 -pthread
NGINX Benchmark This is a test of ab, which is the Apache Benchmark program running against nginx. This test profile measures how many requests per second a given system can sustain when carrying out 2,000,000 requests with 500 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better NGINX Benchmark 1.9.9 Static Web Page Serving Threadripper 3960X 8K 16K 24K 32K 40K SE +/- 62.41, N = 3 36127.85 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native
PyBench This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times Threadripper 3960X 200 400 600 800 1000 SE +/- 10.69, N = 3 968
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Sequential Fill Threadripper 3960X 200K 400K 600K 800K 1000K SE +/- 13038.81, N = 4 1030368 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Read Threadripper 3960X 30M 60M 90M 120M 150M SE +/- 1282448.37, N = 3 143688270 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Glibc Qsort Data Sorting Threadripper 3960X 80 160 240 320 400 SE +/- 1.42, N = 3 380.73 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Glibc C String Functions Threadripper 3960X 500K 1000K 1500K 2000K 2500K SE +/- 28050.29, N = 3 2449871.83 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Context Switching Threadripper 3960X 3M 6M 9M 12M 15M SE +/- 214670.64, N = 15 15705353.86 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Socket Activity Threadripper 3960X 4K 8K 12K 16K 20K SE +/- 48.18, N = 3 18734.34 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Memory Copying Threadripper 3960X 3K 6K 9K 12K 15K SE +/- 31.26, N = 3 12234.52 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Vector Math Threadripper 3960X 14K 28K 42K 56K 70K SE +/- 2.23, N = 3 67241.37 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Matrix Math Threadripper 3960X 30K 60K 90K 120K 150K SE +/- 505.47, N = 3 151388.05 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Semaphores Threadripper 3960X 700K 1400K 2100K 2800K 3500K SE +/- 8385.90, N = 3 3239494.25 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: CPU Stress Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 16.14, N = 3 9167.11 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Tsearch Threadripper 3960X 200 400 600 800 1000 SE +/- 1.41, N = 3 964.42 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Lsearch Threadripper 3960X 80 160 240 320 400 SE +/- 0.48, N = 3 357.06 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Hsearch Threadripper 3960X 40K 80K 120K 160K 200K SE +/- 582.97, N = 3 199421.45 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Forking Threadripper 3960X 20K 40K 60K 80K 100K SE +/- 775.10, N = 3 80851.06 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Bsearch Threadripper 3960X 3K 6K 9K 12K 15K SE +/- 134.01, N = 3 16332.26 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Crypto Threadripper 3960X 1500 3000 4500 6000 7500 SE +/- 27.32, N = 3 7072.93 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
Memtier_benchmark Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool. This current test profile currently just stresses the Redis protocol and basic options exposed wotj a 1:1 Set/Get ratio, 30 pipeline, 100 clients per thread, and thread count equal to the number of CPU cores/threads present. Patches to extend the test are welcome as always. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ops/sec, More Is Better Memtier_benchmark 1.2.17 Protocol: Redis Threadripper 3960X 600K 1200K 1800K 2400K 3000K SE +/- 26919.58, N = 3 2912967.74 1. (CXX) g++ options: -O2 -levent -lpthread -lz -lpcre
Darktable Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Darktable 2.6.0 Test: Server Room - Acceleration: CPU-only Threadripper 3960X 0.5045 1.009 1.5135 2.018 2.5225 SE +/- 0.016, N = 3 2.242
glibc bench The GNU C Library project provides the core libraries for the GNU system and GNU/Linux systems, as well as many other systems that use Linux as the kernel. These libraries provide critical APIs including ISO C11, POSIX.1-2008, BSD, OS-specific APIs and more.
Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: pthread_once Threadripper 3960X 0.3924 0.7848 1.1772 1.5696 1.962 SE +/- 0.00112, N = 3 1.74395
Minion Minion is an open-source constraint solver that is designed to be very scalable. This test profile uses Minion's integrated benchmarking problems to solve. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Quasigroup Threadripper 3960X 20 40 60 80 100 SE +/- 0.62, N = 3 107.18 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Solitaire Threadripper 3960X 14 28 42 56 70 SE +/- 0.35, N = 3 60.51 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Graceful Threadripper 3960X 10 20 30 40 50 SE +/- 0.26, N = 3 41.78 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Process Threadripper 3960X 5 10 15 20 25 SE +/- 0.07, N = 3 22.74 1. (CC) gcc options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Process Threadripper 3960X 3 6 9 12 15 SE +/- 0.07, N = 3 11.15 1. (CC) gcc options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Process Threadripper 3960X 1.2303 2.4606 3.6909 4.9212 6.1515 SE +/- 0.008, N = 3 5.468 1. (CC) gcc options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Process Threadripper 3960X 0.794 1.588 2.382 3.176 3.97 SE +/- 0.014, N = 3 3.529 1. (CC) gcc options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Thread Threadripper 3960X 5 10 15 20 25 SE +/- 0.01, N = 3 22.96 1. (CC) gcc options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Process Threadripper 3960X 0.6215 1.243 1.8645 2.486 3.1075 SE +/- 0.027, N = 14 2.762 1. (CC) gcc options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Thread Threadripper 3960X 3 6 9 12 15 SE +/- 0.04, N = 3 11.67 1. (CC) gcc options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Thread Threadripper 3960X 2 4 6 8 10 SE +/- 0.034, N = 3 6.032 1. (CC) gcc options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Thread Threadripper 3960X 0.8406 1.6812 2.5218 3.3624 4.203 SE +/- 0.057, N = 3 3.736 1. (CC) gcc options: -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Thread Threadripper 3960X 0.6431 1.2862 1.9293 2.5724 3.2155 SE +/- 0.030, N = 3 2.858 1. (CC) gcc options: -lpthread
Opus Codec Encoding Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus and then to decode the generated Opus file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.0.1 WAV, Opus Encode, Opus Decode Threadripper 3960X 2 4 6 8 10 SE +/- 0.013, N = 5 6.272 1. (CC) gcc options: -O3 -ffast-math -fPIE -pie -logg -lm -lopus
Tungsten Renderer Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Volumetric Caustic Threadripper 3960X 0.9352 1.8704 2.8056 3.7408 4.676 SE +/- 0.00428, N = 3 4.15646 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Water Caustic Threadripper 3960X 5 10 15 20 25 SE +/- 0.03, N = 3 19.34 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Hair Threadripper 3960X 3 6 9 12 15 SE +/- 0.02317, N = 3 9.79737 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
Smallpt Smallpt is a C++ global illumination renderer written in less than 100 lines of code. Global illumination is done via unbiased Monte Carlo path tracing and there is multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 128 Samples Threadripper 3960X 0.8269 1.6538 2.4807 3.3076 4.1345 SE +/- 0.005, N = 3 3.675 1. (CXX) g++ options: -fopenmp -O3
C-Ray This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Threadripper 3960X 5 10 15 20 25 SE +/- 0.03, N = 3 21.06 1. (CC) gcc options: -lm -lpthread -O3
x264 This is a simple test of the x264 encoder run on the CPU (OpenCL support disabled) with a sample video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better x264 2018-09-25 H.264 Video Encoding Threadripper 3960X 40 80 120 160 200 SE +/- 1.86, N = 3 193.76 1. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p Threadripper 3960X 70 140 210 280 350 SE +/- 2.33, N = 3 326.41 1. (CC) gcc options: -fPIE -fPIC -fvisibility=hidden -O3 -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p Threadripper 3960X 90 180 270 360 450 SE +/- 2.37, N = 3 392.87 1. (CC) gcc options: -fPIE -fPIC -fvisibility=hidden -O3 -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: VMAF Optimized - Input: Bosphorus 1080p Threadripper 3960X 80 160 240 320 400 SE +/- 4.06, N = 13 382.22 1. (CC) gcc options: -fPIE -fPIC -fvisibility=hidden -O3 -pie -rdynamic -lpthread -lrt -lm
SVT-AV1 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.7 Encoder Mode: Enc Mode 8 - Input: 1080p Threadripper 3960X 20 40 60 80 100 SE +/- 0.12, N = 3 78.79 1. (CXX) g++ options: -fPIE -fPIC -pie
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.7 Encoder Mode: Enc Mode 4 - Input: 1080p Threadripper 3960X 2 4 6 8 10 SE +/- 0.027, N = 3 8.043 1. (CXX) g++ options: -fPIE -fPIC -pie
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.7 Encoder Mode: Enc Mode 0 - Input: 1080p Threadripper 3960X 0.0144 0.0288 0.0432 0.0576 0.072 SE +/- 0.000, N = 3 0.064 1. (CXX) g++ options: -fPIE -fPIC -pie
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer ISPC - Model: Asian Dragon Threadripper 3960X 7 14 21 28 35 SE +/- 0.02, N = 3 32.15 MIN: 31.68 / MAX: 32.72
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer - Model: Asian Dragon Obj Threadripper 3960X 7 14 21 28 35 SE +/- 0.01, N = 3 28.79 MIN: 28.45 / MAX: 29.34
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer - Model: Asian Dragon Threadripper 3960X 7 14 21 28 35 SE +/- 0.01, N = 3 32.15 MIN: 31.73 / MAX: 32.8
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer ISPC - Model: Crown Threadripper 3960X 7 14 21 28 35 SE +/- 0.02, N = 3 28.37 MIN: 27.81 / MAX: 29.1
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer - Model: Crown Threadripper 3960X 7 14 21 28 35 SE +/- 0.07, N = 3 29.77 MIN: 29.2 / MAX: 30.55
TTSIOD 3D Renderer A portable GPL 3D software renderer that supports OpenMP and Intel Threading Building Blocks with many different rendering modes. This version does not use OpenGL but is entirely CPU/software based. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping Threadripper 3960X 200 400 600 800 1000 SE +/- 9.82, N = 4 780.04 1. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++
OSPray Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: Magnetic Reconnection - Renderer: Path Tracer Threadripper 3960X 70 140 210 280 350 SE +/- 0.00, N = 12 333.33 MIN: 166.67 / MAX: 500
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: Path Tracer Threadripper 3960X 3 6 9 12 15 SE +/- 0.00, N = 12 10.75 MIN: 10.53 / MAX: 10.99
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: Path Tracer Threadripper 3960X 0.837 1.674 2.511 3.348 4.185 SE +/- 0.01, N = 3 3.72 MIN: 3.66 / MAX: 3.76
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: SciVis Threadripper 3960X 12 24 36 48 60 SE +/- 0.00, N = 12 52.63 MIN: 43.48 / MAX: 55.56
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: Path Tracer Threadripper 3960X 0.648 1.296 1.944 2.592 3.24 SE +/- 0.00, N = 3 2.88 MIN: 2.87 / MAX: 2.92
OpenBenchmarking.org FPS, More Is Better dav1d 0.5.0 Video Input: Summer Nature 1080p Threadripper 3960X 150 300 450 600 750 SE +/- 1.15, N = 3 677.06 MIN: 426.02 / MAX: 738.11 1. (CC) gcc options: -pthread
OpenBenchmarking.org FPS, More Is Better dav1d 0.5.0 Video Input: Summer Nature 4K Threadripper 3960X 60 120 180 240 300 SE +/- 0.85, N = 3 291.09 MIN: 179.18 / MAX: 309.08 1. (CC) gcc options: -pthread
OpenBenchmarking.org FPS, More Is Better dav1d 0.5.0 Video Input: Chimera 1080p Threadripper 3960X 130 260 390 520 650 SE +/- 1.35, N = 3 587.81 MIN: 448.01 / MAX: 721.98 1. (CC) gcc options: -pthread
MKL-DNN DNNL This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32 Threadripper 3960X 13 26 39 52 65 SE +/- 0.28, N = 3 57.28 MIN: 51.08 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Recurrent Neural Network Training - Data Type: f32 Threadripper 3960X 50 100 150 200 250 SE +/- 1.80, N = 3 226.82 MIN: 196.67 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Deconvolution Batch deconv_all - Data Type: f32 Threadripper 3960X 400 800 1200 1600 2000 SE +/- 6.22, N = 3 1772.78 MIN: 1677 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_alexnet - Data Type: f32 Threadripper 3960X 30 60 90 120 150 SE +/- 1.31, N = 3 142.23 MIN: 123.19 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Deconvolution Batch deconv_3d - Data Type: f32 Threadripper 3960X 0.6444 1.2888 1.9332 2.5776 3.222 SE +/- 0.01436, N = 3 2.86390 MIN: 2.45 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Deconvolution Batch deconv_1d - Data Type: f32 Threadripper 3960X 0.5841 1.1682 1.7523 2.3364 2.9205 SE +/- 0.03171, N = 3 2.59611 MIN: 2.27 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_all - Data Type: f32 Threadripper 3960X 200 400 600 800 1000 SE +/- 2.64, N = 3 1070.54 MIN: 997.44 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_3d - Data Type: f32 Threadripper 3960X 2 4 6 8 10 SE +/- 0.08494, N = 3 6.84735 MIN: 5.9 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: IP Batch All - Data Type: f32 Threadripper 3960X 4 8 12 16 20 SE +/- 0.12, N = 3 16.80 MIN: 15.38 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: IP Batch 1D - Data Type: f32 Threadripper 3960X 0.634 1.268 1.902 2.536 3.17 SE +/- 0.03365, N = 6 2.81769 MIN: 2.36 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: HWB Color Space Threadripper 3960X 300 600 900 1200 1500 SE +/- 1.86, N = 3 1462 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Noise-Gaussian Threadripper 3960X 120 240 360 480 600 SE +/- 0.33, N = 3 565 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Resizing Threadripper 3960X 500 1000 1500 2000 2500 SE +/- 12.39, N = 3 2263 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Enhanced Threadripper 3960X 120 240 360 480 600 SE +/- 0.67, N = 3 569 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Sharpen Threadripper 3960X 80 160 240 320 400 SE +/- 2.19, N = 3 379 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Rotate Threadripper 3960X 150 300 450 600 750 SE +/- 3.84, N = 3 710 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Swirl Threadripper 3960X 300 600 900 1200 1500 SE +/- 5.49, N = 3 1351 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite Threadripper 3960X 150 300 450 600 750 SE +/- 1.06, N = 3 714.92 1. (CC) gcc options: -lm
OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Write Threadripper 3960X 7K 14K 21K 28K 35K SE +/- 55.58, N = 3 32059.70 MIN: 28046.66 / MAX: 33459.4 1. (CC) gcc options: -lrt
OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read Threadripper 3960X 700 1400 2100 2800 3500 SE +/- 3.47, N = 3 3063.54 MIN: 3047.77 / MAX: 3069.92 1. (CC) gcc options: -lrt
Java SciMark This test runs the Java version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This benchmark is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Composite Threadripper 3960X 700 1400 2100 2800 3500 SE +/- 3.56, N = 4 3033.26
QMCPACK QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.8 Threadripper 3960X 400 800 1200 1600 2000 1875.8 1. (CXX) g++ options: -fopenmp -fomit-frame-pointer -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -ffast-math -lm
Timed MrBayes Analysis This test performs a bayesian analysis of a set of primate genome sequences in order to estimate their phylogeny. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis Threadripper 3960X 16 32 48 64 80 SE +/- 0.28, N = 3 71.13 1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: sedovbig Threadripper 3960X 1400 2800 4200 5600 7000 SE +/- 2.00, N = 3 6502.56 1. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 Threadripper 3960X 5K 10K 15K 20K 25K SE +/- 60.75, N = 3 22314 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 2048 Threadripper 3960X 7K 14K 21K 28K 35K SE +/- 375.01, N = 3 34104 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 1024 Threadripper 3960X 9K 18K 27K 36K 45K SE +/- 199.23, N = 3 40897 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 4096 Threadripper 3960X 13K 26K 39K 52K 65K SE +/- 410.08, N = 3 58875 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 2048 Threadripper 3960X 13K 26K 39K 52K 65K SE +/- 623.77, N = 3 61112 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 1024 Threadripper 3960X 13K 26K 39K 52K 65K SE +/- 370.51, N = 3 59937 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 512 Threadripper 3960X 9K 18K 27K 36K 45K SE +/- 137.80, N = 3 43049 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 256 Threadripper 3960X 9K 18K 27K 36K 45K SE +/- 310.57, N = 3 42616 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 128 Threadripper 3960X 10K 20K 30K 40K 50K SE +/- 237.97, N = 3 46970 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 512 Threadripper 3960X 11K 22K 33K 44K 55K SE +/- 123.47, N = 3 51240 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 256 Threadripper 3960X 8K 16K 24K 32K 40K SE +/- 412.34, N = 15 38260 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 128 Threadripper 3960X 6K 12K 18K 24K 30K SE +/- 98.24, N = 3 27460 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 64 Threadripper 3960X 10K 20K 30K 40K 50K SE +/- 259.03, N = 3 47733 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 32 Threadripper 3960X 10K 20K 30K 40K 50K SE +/- 88.57, N = 3 46123 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 32 Threadripper 3960X 3K 6K 9K 12K 15K SE +/- 208.40, N = 4 15042 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 Threadripper 3960X 1500 3000 4500 6000 7500 SE +/- 8.75, N = 3 6784.1 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 2048 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 3.11, N = 3 7864.3 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 1024 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 4.47, N = 3 8097.9 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 4096 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 31.06, N = 3 8818.8 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 2048 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 21.61, N = 3 9048.1 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 1024 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 5.17, N = 3 9356.2 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 512 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 17.50, N = 3 8684.6 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 256 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 8.89, N = 3 8674.1 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 128 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 14.00, N = 3 8902.3 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 512 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 39.54, N = 3 9114.0 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 256 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 12.22, N = 3 8883.8 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 128 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 77.36, N = 3 8569.2 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 64 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 21.91, N = 3 9585.4 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 32 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 17.13, N = 3 11113 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 32 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 5.51, N = 3 11237 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTE FFTE is a package by Daisuke Takahashi to compute Discrete Fourier Transforms of 1-, 2- and 3- dimensional sequences of length (2^p)*(3^q)*(5^r). Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MFLOPS, More Is Better FFTE 6.0 Test: N=256, 1D Complex FFT Routine Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 86.87, N = 3 9627.34 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
NAMD NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.13b1 ATPase Simulation - 327,506 Atoms Threadripper 3960X 0.1625 0.325 0.4875 0.65 0.8125 SE +/- 0.00128, N = 3 0.72235
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP CFD Solver Threadripper 3960X 3 6 9 12 15 SE +/- 0.025, N = 3 9.037 1. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD Threadripper 3960X 3 6 9 12 15 SE +/- 0.01, N = 3 10.10 1. (CXX) g++ options: -O2 -lOpenCL
CloverLeaf CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm8192.in input file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf Lagrangian-Eulerian Hydrodynamics Threadripper 3960X 0.1845 0.369 0.5535 0.738 0.9225 SE +/- 0.00, N = 3 0.82 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding Threadripper 3960X 11 22 33 44 55 SE +/- 0.16, N = 3 49.07 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil Threadripper 3960X 2 4 6 8 10 SE +/- 0.018289, N = 3 7.243440 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP Threadripper 3960X 0.2845 0.569 0.8535 1.138 1.4225 SE +/- 0.000632, N = 3 1.264657 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
HPC Challenge HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Max Ping Pong Bandwidth Threadripper 3960X 5K 10K 15K 20K 25K SE +/- 366.52, N = 3 22107.91 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Bandwidth Threadripper 3960X 0.7569 1.5138 2.2707 3.0276 3.7845 SE +/- 0.00987, N = 3 3.36400 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
OpenBenchmarking.org usecs, Fewer Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Latency Threadripper 3960X 0.1028 0.2056 0.3084 0.4112 0.514 SE +/- 0.00045, N = 3 0.45683 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
OpenBenchmarking.org GUP/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Random Access Threadripper 3960X 0.0322 0.0644 0.0966 0.1288 0.161 SE +/- 0.00070, N = 3 0.14292 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: EP-STREAM Triad Threadripper 3960X 0.41 0.82 1.23 1.64 2.05 SE +/- 0.00309, N = 3 1.82228 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ptrans Threadripper 3960X 1.2465 2.493 3.7395 4.986 6.2325 SE +/- 0.01609, N = 3 5.54022 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: EP-DGEMM Threadripper 3960X 7 14 21 28 35 SE +/- 0.45, N = 3 31.64 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
OpenBenchmarking.org GFLOP/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ffte Threadripper 3960X 3 6 9 12 15 SE +/- 0.19, N = 3 10.73 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ffte Threadripper 3960X 3 6 9 12 15 SE +/- 0.19, N = 3 10.73 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-HPL Threadripper 3960X 14 28 42 56 70 SE +/- 0.16, N = 3 60.46 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 3.1.3
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.B Threadripper 3960X 5K 10K 15K 20K 25K SE +/- 269.46, N = 15 23242.08 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C Threadripper 3960X 5K 10K 15K 20K 25K SE +/- 41.78, N = 3 25120.08 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C Threadripper 3960X 11K 22K 33K 44K 55K SE +/- 232.22, N = 3 51711.88 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C Threadripper 3960X 4K 8K 12K 16K 20K SE +/- 33.84, N = 3 17832.39 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D Threadripper 3960X 500 1000 1500 2000 2500 SE +/- 7.75, N = 3 2214.27 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C Threadripper 3960X 500 1000 1500 2000 2500 SE +/- 5.95, N = 3 2208.11 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: BT.C Threadripper 3960X 11K 22K 33K 44K 55K SE +/- 720.74, N = 3 52747.62 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 13.80, N = 3 11380.32 1. (CC) gcc options: -O3 -march=native
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 512 MiB Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 6.27, N = 3 11306.18 1. (CC) gcc options: -O3 -march=native
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 128 MiB Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 15.80, N = 3 10863.42 1. (CC) gcc options: -O3 -march=native
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 4096 MiB Threadripper 3960X 4K 8K 12K 16K 20K SE +/- 33.30, N = 3 19697.48 1. (CC) gcc options: -O3 -march=native
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 1024 MiB Threadripper 3960X 4K 8K 12K 16K 20K SE +/- 19.92, N = 3 19435.38 1. (CC) gcc options: -O3 -march=native
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 512 MiB Threadripper 3960X 4K 8K 12K 16K 20K SE +/- 62.41, N = 3 19400.43 1. (CC) gcc options: -O3 -march=native
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 128 MiB Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 18.64, N = 3 10893.68 1. (CC) gcc options: -O3 -march=native
OpenBenchmarking.org MiVoxels / Sec, More Is Better ParaView 5.4.1 Test: Wavelet Volume - Resolution: 2560 x 1440 Threadripper 3960X 600 1200 1800 2400 3000 SE +/- 15.86, N = 3 2811.52
OpenBenchmarking.org MiPolys / Sec, More Is Better ParaView 5.4.1 Test: Many Spheres - Resolution: 2560 x 1440 Threadripper 3960X 1300 2600 3900 5200 6500 SE +/- 4.91, N = 3 5900.23
OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.30.1 Threads / Copies: 1 Threadripper 3960X 4 8 12 16 20 SE +/- 0.10, N = 3 15.74 1. (CC) gcc options: -O2 -lz -lm -ldl -lpthread
OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 200 AMD Ryzen Threadripper 3960X 24-Core 12K 24K 36K 48K 60K SE +/- 2566.16, N = 12 55723.63 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: GET Threadripper 3960X 600K 1200K 1800K 2400K 3000K SE +/- 58191.94, N = 15 2649622.70 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Tungsten Renderer Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Non-Exponential Threadripper 3960X 0.8019 1.6038 2.4057 3.2076 4.0095 SE +/- 0.10003, N = 15 3.56422 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 64 Threadripper 3960X 4K 8K 12K 16K 20K SE +/- 715.56, N = 15 18900 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 64 Threadripper 3960X 2K 4K 6K 8K 10K SE +/- 231.46, N = 15 9362.2 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
LeelaChessZero LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.22.0 Backend: BLAS AMD Ryzen Threadripper 3960X 24-Core Threadripper 3960X 8 16 24 32 40 SE +/- 2.09, N = 12 SE +/- 0.94, N = 12 33.39 32.73 1. (CXX) g++ options: -lpthread
Threadripper 3960X Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-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 -vDisk Notes: NONE / errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Java Notes: OpenJDK Runtime Environment (build 11.0.5-ea+10-post-Ubuntu-0ubuntu1)Python Notes: Python 2.7.17rc1 + Python 3.7.5rc1Security Notes: 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
Testing initiated at 23 November 2019 21:04 by user phoronix.
AMD Ryzen Threadripper 3960X 24-Core Processor: AMD Ryzen Threadripper 3960X 24-Core @ 3.80GHz (24 Cores / 48 Threads), Motherboard: MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS), Chipset: AMD Starship/Matisse, Memory: 32768MB, Disk: 240GB Force MP510, Graphics: AMD Navi 10 8GB (2100/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS PB278, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Device 2723
OS: Ubuntu 19.10, Kernel: 5.3.0-23-generic (x86_64), Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, OpenGL: 4.5 Mesa 19.2.1 (LLVM 9.0.0), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 2560x1440
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-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 -vProcessor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Java Notes: OpenJDK Runtime Environment (build 11.0.5-ea+10-post-Ubuntu-0ubuntu1)Python Notes: Python 2.7.17rc1 + Python 3.7.5rc1Security Notes: 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
Testing initiated at 25 November 2019 04:43 by user phoronix.