Big Threadripper AMD Ryzen Threadripper 3970X 32-Core testing with a ASUS ROG ZENITH II EXTREME (0702 BIOS) and llvmpipe 63GB on Ubuntu 19.10 via the Phoronix Test Suite. Threadripper 3970X: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (0702 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: 1000GB Force MP600, Graphics: llvmpipe 63GB, Audio: AMD Navi 10 HDMI Audio, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 19.10, Kernel: 5.3.0-18-generic (x86_64), Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, OpenGL: 3.3 Mesa 19.2.8 (LLVM 9.0 128 bits), Compiler: GCC 9.2.1 20191008 + Clang 9.0.0-2, File-System: ext4, Screen Resolution: 3840x2160 16 Cores: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (16 Cores), Motherboard: ASUS ROG ZENITH II EXTREME (0702 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: 1000GB Force MP600, Graphics: llvmpipe 63GB, Audio: AMD Navi 10 HDMI Audio, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 19.10, Kernel: 5.3.0-40-generic (x86_64), Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, OpenGL: 3.3 Mesa 19.2.8 (LLVM 9.0 128 bits), Compiler: GCC 9.2.1 20191008 + Clang 9.0.0-2, File-System: ext4, Screen Resolution: 3840x2160 Stress-NG 0.07.26 Test: Forking Bogo Ops/s > Higher Is Better Threadripper 3970X . 97017.28 |================================================ 16 Cores ........... 10970.25 |===== LevelDB 1.22 Benchmark: Random Delete Microseconds Per Op < Lower Is Better Threadripper 3970X . 256.44 |================================================== 16 Cores ........... 39.90 |======== LevelDB 1.22 Benchmark: Sequential Fill Microseconds Per Op < Lower Is Better Threadripper 3970X . 256.33 |================================================== 16 Cores ........... 42.30 |======== LevelDB 1.22 Benchmark: Overwrite Microseconds Per Op < Lower Is Better Threadripper 3970X . 263.94 |================================================== 16 Cores ........... 43.85 |======== LevelDB 1.22 Benchmark: Random Fill Microseconds Per Op < Lower Is Better Threadripper 3970X . 259.84 |================================================== 16 Cores ........... 43.87 |======== MariaDB 10.3.8 Clients: 64 Queries Per Second > Higher Is Better Threadripper 3970X . 207 |========== 16 Cores ........... 1080 |==================================================== LevelDB 1.22 Benchmark: Hot Read Microseconds Per Op < Lower Is Better Threadripper 3970X . 44.286 |================================================== 16 Cores ........... 8.974 |========== LevelDB 1.22 Benchmark: Random Read Microseconds Per Op < Lower Is Better Threadripper 3970X . 44.353 |================================================== 16 Cores ........... 9.313 |========== OSBench Test: Launch Programs us Per Event < Lower Is Better Threadripper 3970X . 49.30 |=========== 16 Cores ........... 228.58 |================================================== LevelDB 1.22 Benchmark: Seek Random Microseconds Per Op < Lower Is Better Threadripper 3970X . 68.59 |=================================================== 16 Cores ........... 14.89 |=========== BlogBench 1.1 Test: Write Final Score > Higher Is Better Threadripper 3970X . 49874 |=================================================== 16 Cores ........... 11700 |============ OSBench Test: Create Threads us Per Event < Lower Is Better Threadripper 3970X . 11.33 |============= 16 Cores ........... 46.12 |=================================================== Apache Cassandra 3.11.4 Test: Writes Op/s > Higher Is Better Threadripper 3970X . 293003 |================================================== 16 Cores ........... 7744 |= LevelDB 1.22 Benchmark: Fill Sync MB/s > Higher Is Better Threadripper 3970X . 3.6 |===================================================== 16 Cores ........... 1.0 |=============== HPC Challenge 1.5.0 Test / Class: G-Ptrans GB/s > Higher Is Better Threadripper 3970X . 7.44119 |================================================= 16 Cores ........... 2.12629 |============== Tungsten Renderer 0.2.2 Scene: Non-Exponential Seconds < Lower Is Better Threadripper 3970X . 2.64289 |=============== 16 Cores ........... 8.92215 |================================================= N-Queens 1.0 Elapsed Time Seconds < Lower Is Better Threadripper 3970X . 3.496 |=============== 16 Cores ........... 11.321 |================================================== m-queens 1.2 Time To Solve Seconds < Lower Is Better Threadripper 3970X . 17.95 |================= 16 Cores ........... 55.08 |=================================================== Stress-NG 0.07.26 Test: Matrix Math Bogo Ops/s > Higher Is Better Threadripper 3970X . 234235.76 |=============================================== 16 Cores ........... 76991.04 |=============== PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only TPS > Higher Is Better Threadripper 3970X . 792360.46 |=============================================== 16 Cores ........... 272392.97 |================ Stress-NG 0.07.26 Test: Tsearch Bogo Ops/s > Higher Is Better Threadripper 3970X . 1111.04 |================================================= 16 Cores ........... 383.18 |================= Embree 3.6.1 Binary: Pathtracer - Model: Crown Frames Per Second > Higher Is Better Threadripper 3970X . 37.66 |=================================================== 16 Cores ........... 13.23 |================== Systemd Total Boot Time Test: Loader ms < Lower Is Better Threadripper 3970X . 9976 |==================================================== 16 Cores ........... 3542 |================== Stress-NG 0.07.26 Test: Bsearch Bogo Ops/s > Higher Is Better Threadripper 3970X . 20222.03 |================================================ 16 Cores ........... 7239.75 |================= Embree 3.6.1 Binary: Pathtracer ISPC - Model: Crown Frames Per Second > Higher Is Better Threadripper 3970X . 35.55 |=================================================== 16 Cores ........... 12.83 |================== OSPray 1.8.5 Demo: NASA Streamlines - Renderer: Path Tracer FPS > Higher Is Better Threadripper 3970X . 13.58 |=================================================== 16 Cores ........... 4.97 |=================== Stress-NG 0.07.26 Test: Glibc C String Functions Bogo Ops/s > Higher Is Better Threadripper 3970X . 3406391.46 |============================================== 16 Cores ........... 1252570.26 |================= OSPray 1.8.5 Demo: XFrog Forest - Renderer: SciVis FPS > Higher Is Better Threadripper 3970X . 8.99 |==================================================== 16 Cores ........... 3.32 |=================== OSPray 1.8.5 Demo: San Miguel - Renderer: SciVis FPS > Higher Is Better Threadripper 3970X . 50.00 |=================================================== 16 Cores ........... 18.52 |=================== Stress-NG 0.07.26 Test: Glibc Qsort Data Sorting Bogo Ops/s > Higher Is Better Threadripper 3970X . 489.81 |================================================== 16 Cores ........... 181.55 |=================== Blender 2.80 Blend File: Barbershop - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 159.93 |=================== 16 Cores ........... 430.81 |================================================== Blender 2.82 Blend File: Pabellon Barcelona - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 153.88 |=================== 16 Cores ........... 414.42 |================================================== OSPray 1.8.5 Demo: XFrog Forest - Renderer: Path Tracer FPS > Higher Is Better Threadripper 3970X . 4.69 |==================================================== 16 Cores ........... 1.75 |=================== OSPray 1.8.5 Demo: San Miguel - Renderer: Path Tracer FPS > Higher Is Better Threadripper 3970X . 3.67 |==================================================== 16 Cores ........... 1.37 |=================== OSPray 1.8.5 Demo: NASA Streamlines - Renderer: SciVis FPS > Higher Is Better Threadripper 3970X . 66.67 |=================================================== 16 Cores ........... 25.00 |=================== IndigoBench 4.0.64 Scene: Bedroom M samples/s > Higher Is Better Threadripper 3970X . 4.906 |=================================================== 16 Cores ........... 1.852 |=================== Blender 2.82 Blend File: Classroom - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 119.47 |=================== 16 Cores ........... 314.82 |================================================== Coremark 1.0 CoreMark Size 666 - Iterations Per Second Iterations/Sec > Higher Is Better Threadripper 3970X . 1331439.75 |============================================== 16 Cores ........... 508354.40 |================== Blender 2.80 Blend File: Pabellon Barcelona - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 146.02 |=================== 16 Cores ........... 382.22 |================================================== Facebook RocksDB 6.3.6 Test: Read While Writing Op/s > Higher Is Better Threadripper 3970X . 6204076 |================================================= 16 Cores ........... 2381360 |=================== Blender 2.80 Blend File: Classroom - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 109.39 |=================== 16 Cores ........... 283.82 |================================================== Stockfish 9 Total Time Nodes Per Second > Higher Is Better Threadripper 3970X . 95728787 |================================================ 16 Cores ........... 36932742 |=================== Blender 2.82 Blend File: Barbershop - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 194.17 |=================== 16 Cores ........... 501.07 |================================================== asmFish 2018-07-23 1024 Hash Memory, 26 Depth Nodes/second > Higher Is Better Threadripper 3970X . 95916197 |================================================ 16 Cores ........... 37415509 |=================== Smallpt 1.0 Global Illumination Renderer; 128 Samples Seconds < Lower Is Better Threadripper 3970X . 2.894 |==================== 16 Cores ........... 7.407 |=================================================== Blender 2.82 Blend File: BMW27 - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 47.57 |==================== 16 Cores ........... 119.58 |================================================== Stress-NG 0.07.26 Test: Vector Math Bogo Ops/s > Higher Is Better Threadripper 3970X . 88021.72 |================================================ 16 Cores ........... 35109.17 |=================== Blender 2.82 Blend File: Fishy Cat - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 71.28 |==================== 16 Cores ........... 178.46 |================================================== 7-Zip Compression 16.02 Compress Speed Test MIPS > Higher Is Better Threadripper 3970X . 196186 |================================================== 16 Cores ........... 78488 |==================== QMCPACK 3.8 Total Execution Time - Seconds < Lower Is Better Threadripper 3970X . 2581.7 |================================================== 16 Cores ........... 1039.4 |==================== Facebook RocksDB 6.3.6 Test: Random Read Op/s > Higher Is Better Threadripper 3970X . 180626444 |=============================================== 16 Cores ........... 72774397 |=================== Cpuminer-Opt 3.8.8.1 Algorithm: m7m kH/s - Hash Speed > Higher Is Better Threadripper 3970X . 1624.30 |================================================= 16 Cores ........... 654.62 |==================== OSPray 1.8.5 Demo: Magnetic Reconnection - Renderer: SciVis FPS > Higher Is Better Threadripper 3970X . 32.62 |=================================================== 16 Cores ........... 13.16 |===================== POV-Ray 3.7.0.7 Trace Time Seconds < Lower Is Better Threadripper 3970X . 13.24 |===================== 16 Cores ........... 32.25 |=================================================== Blender 2.80 Blend File: Fishy Cat - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 67.04 |===================== 16 Cores ........... 163.13 |================================================== Stress-NG 0.07.26 Test: CPU Stress Bogo Ops/s > Higher Is Better Threadripper 3970X . 12151.31 |================================================ 16 Cores ........... 5019.09 |==================== Blender 2.80 Blend File: BMW27 - Compute: CPU-Only Seconds < Lower Is Better Threadripper 3970X . 40.40 |===================== 16 Cores ........... 97.69 |=================================================== BRL-CAD 7.28.0 VGR Performance Metric VGR Performance Metric > Higher Is Better Threadripper 3970X . 507345 |================================================== 16 Cores ........... 210066 |===================== Tungsten Renderer 0.2.2 Scene: Volumetric Caustic Seconds < Lower Is Better Threadripper 3970X . 3.56382 |==================== 16 Cores ........... 8.56420 |================================================= Stress-NG 0.07.26 Test: Hsearch Bogo Ops/s > Higher Is Better Threadripper 3970X . 253747.42 |=============================================== 16 Cores ........... 106584.80 |==================== Hackbench Count: 32 - Type: Process Seconds < Lower Is Better Threadripper 3970X . 31.39 |====================== 16 Cores ........... 74.35 |=================================================== Tungsten Renderer 0.2.2 Scene: Hair Seconds < Lower Is Better Threadripper 3970X . 8.18217 |==================== 16 Cores ........... 19.25580 |================================================ rays1bench 2020-01-09 Large Scene mrays/s > Higher Is Better Threadripper 3970X . 225.76 |================================================== 16 Cores ........... 95.94 |===================== Tachyon 0.99b6 Total Time Seconds < Lower Is Better Threadripper 3970X . 25.04 |====================== 16 Cores ........... 58.32 |=================================================== ACES DGEMM 1.0 Sustained Floating-Point Rate GFLOP/s > Higher Is Better Threadripper 3970X . 10.868534 |=============================================== 16 Cores ........... 4.666959 |==================== IndigoBench 4.0.64 Scene: Supercar M samples/s > Higher Is Better Threadripper 3970X . 9.317 |=================================================== 16 Cores ........... 4.106 |====================== Go Benchmarks Test: json Nanoseconds/Operation < Lower Is Better Threadripper 3970X . 1564168 |====================== 16 Cores ........... 3523585 |================================================= Chaos Group V-RAY 4.10.07 Mode: CPU Ksamples > Higher Is Better Threadripper 3970X . 45841 |=================================================== 16 Cores ........... 20433 |======================= Cpuminer-Opt 3.8.8.1 Algorithm: lbry kH/s - Hash Speed > Higher Is Better Threadripper 3970X . 89853 |=================================================== 16 Cores ........... 41063 |======================= Timed MPlayer Compilation 1.4 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 11.58 |======================== 16 Cores ........... 25.09 |=================================================== SVT-VP9 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p Frames Per Second > Higher Is Better Threadripper 3970X . 368.41 |================================================== 16 Cores ........... 172.13 |======================= Parboil 2.5 Test: OpenMP CUTCP Seconds < Lower Is Better Threadripper 3970X . 0.993342 |====================== 16 Cores ........... 2.122203 |================================================ SVT-HEVC 1.4.1 1080p 8-bit YUV To HEVC Video Encode Frames Per Second > Higher Is Better Threadripper 3970X . 174.75 |================================================== 16 Cores ........... 82.13 |======================= Stress-NG 0.07.26 Test: Socket Activity Bogo Ops/s > Higher Is Better Threadripper 3970X . 22217.76 |================================================ 16 Cores ........... 10510.35 |======================= Pennant 1.0.1 Test: leblancbig Hydro Cycle Time - Seconds < Lower Is Better Threadripper 3970X . 8623.65 |================================================= 16 Cores ........... 4088.41 |======================= MKL-DNN DNNL 1.1 Harness: Recurrent Neural Network Training - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 299.86 |================================================== 16 Cores ........... 142.35 |======================== Stress-NG 0.07.26 Test: Memory Copying Bogo Ops/s > Higher Is Better Threadripper 3970X . 8324.57 |================================================= 16 Cores ........... 3963.81 |======================= R Benchmark Seconds < Lower Is Better Threadripper 3970X . 0.2858 |================================================== 16 Cores ........... 0.1361 |======================== Stress-NG 0.07.26 Test: Crypto Bogo Ops/s > Higher Is Better Threadripper 3970X . 9207.85 |================================================= 16 Cores ........... 4392.36 |======================= Cpuminer-Opt 3.8.8.1 Algorithm: skein kH/s - Hash Speed > Higher Is Better Threadripper 3970X . 102703 |================================================== 16 Cores ........... 49073 |======================== Cpuminer-Opt 3.8.8.1 Algorithm: sha256t kH/s - Hash Speed > Higher Is Better Threadripper 3970X . 223623 |================================================== 16 Cores ........... 106950 |======================== Appleseed 2.0 Beta Scene: Emily Seconds < Lower Is Better Threadripper 3970X . 138.48 |======================== 16 Cores ........... 289.31 |================================================== Appleseed 2.0 Beta Scene: Disney Material Seconds < Lower Is Better Threadripper 3970X . 70.30 |======================== 16 Cores ........... 146.68 |================================================== C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Seconds < Lower Is Better Threadripper 3970X . 16.38 |======================== 16 Cores ........... 34.16 |=================================================== Sysbench 2018-07-28 Test: CPU Events Per Second > Higher Is Better Threadripper 3970X . 68765.38 |================================================ 16 Cores ........... 33077.17 |======================= Aircrack-ng 1.5.2 k/s > Higher Is Better Threadripper 3970X . 112289.30 |=============================================== 16 Cores ........... 54172.20 |======================= Primesieve 7.4 1e12 Prime Number Generation Seconds < Lower Is Better Threadripper 3970X . 6.123 |======================== 16 Cores ........... 12.538 |================================================== Pennant 1.0.1 Test: sedovbig Hydro Cycle Time - Seconds < Lower Is Better Threadripper 3970X . 8885.04 |================================================= 16 Cores ........... 4341.34 |======================== OpenSSL 1.1.1 RSA 4096-bit Performance Signs Per Second > Higher Is Better Threadripper 3970X . 9307.4 |================================================== 16 Cores ........... 4550.9 |======================== GraphicsMagick 1.3.33 Operation: Swirl Iterations Per Minute > Higher Is Better Threadripper 3970X . 1590 |==================================================== 16 Cores ........... 784 |========================== Timed Linux Kernel Compilation 5.4 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 24.54 |========================= 16 Cores ........... 49.36 |=================================================== NAMD 2.13 ATPase Simulation - 327,506 Atoms days/ns < Lower Is Better Threadripper 3970X . 0.62357 |========================= 16 Cores ........... 1.23901 |================================================= Rodinia 2.4 Test: OpenMP LavaMD Seconds < Lower Is Better Threadripper 3970X . 8.337 |========================= 16 Cores ........... 16.534 |================================================== ebizzy 0.3 Records/s > Higher Is Better Threadripper 3970X . 2542372 |================================================= 16 Cores ........... 1284378 |========================= NAS Parallel Benchmarks 3.4 Test / Class: EP.D Total Mop/s > Higher Is Better Threadripper 3970X . 2907.58 |================================================= 16 Cores ........... 1480.10 |========================= SVT-AV1 0.8 Encoder Mode: Enc Mode 8 - Input: 1080p Frames Per Second > Higher Is Better Threadripper 3970X . 69.74 |=================================================== 16 Cores ........... 35.89 |========================== Rust Prime Benchmark Prime Number Test To 200,000,000 Seconds < Lower Is Better Threadripper 3970X . 10.27 |========================== 16 Cores ........... 19.96 |=================================================== Go Benchmarks Test: http Nanoseconds/Operation < Lower Is Better Threadripper 3970X . 7825 |==================================================== 16 Cores ........... 4059 |=========================== NAS Parallel Benchmarks 3.4 Test / Class: EP.C Total Mop/s > Higher Is Better Threadripper 3970X . 2831.86 |================================================= 16 Cores ........... 1486.44 |========================== Intel MPI Benchmarks 2019.3 Test: IMB-P2P PingPong Average Msg/sec > Higher Is Better Threadripper 3970X . 17789641.38 |============================================= 16 Cores ........... 9362495.74 |======================== Timed LLVM Compilation 6.0.1 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 122.77 |========================== 16 Cores ........... 232.33 |================================================== GraphicsMagick 1.3.33 Operation: Enhanced Iterations Per Minute > Higher Is Better Threadripper 3970X . 669 |===================================================== 16 Cores ........... 357 |============================ GraphicsMagick 1.3.33 Operation: Sharpen Iterations Per Minute > Higher Is Better Threadripper 3970X . 468 |===================================================== 16 Cores ........... 252 |============================= DaCapo Benchmark 9.12-MR1 Java Test: Tradebeans msec < Lower Is Better Threadripper 3970X . 4496 |==================================================== 16 Cores ........... 2452 |============================ HPC Challenge 1.5.0 Test / Class: EP-STREAM Triad GB/s > Higher Is Better Threadripper 3970X . 1.90348 |=========================== 16 Cores ........... 3.47614 |================================================= Rodinia 2.4 Test: OpenMP Streamcluster Seconds < Lower Is Better Threadripper 3970X . 20.40 |=================================================== 16 Cores ........... 11.27 |============================ HPC Challenge 1.5.0 Test / Class: Random Ring Bandwidth GB/s > Higher Is Better Threadripper 3970X . 2.20814 |=========================== 16 Cores ........... 3.97686 |================================================= GraphicsMagick 1.3.33 Operation: Noise-Gaussian Iterations Per Minute > Higher Is Better Threadripper 3970X . 631 |===================================================== 16 Cores ........... 351 |============================= Ethr 2019-01-02 Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 2 Mbits/sec > Higher Is Better Threadripper 3970X . 1844.58 |=========================== 16 Cores ........... 3314.79 |================================================= Facebook RocksDB 6.3.6 Test: Random Fill Op/s > Higher Is Better Threadripper 3970X . 1083225 |================================================= 16 Cores ........... 606976 |=========================== Timed FFmpeg Compilation 4.2.2 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 21.97 |============================= 16 Cores ........... 38.02 |=================================================== LeelaChessZero 0.23.2 Backend: BLAS Nodes Per Second > Higher Is Better Threadripper 3970X . 2105.11 |============================ 16 Cores ........... 3630.24 |================================================= ArrayFire 3.7 Test: BLAS CPU GFLOPS > Higher Is Better Threadripper 3970X . 802.01 |================================================== 16 Cores ........... 481.37 |============================== HPC Challenge 1.5.0 Test / Class: EP-DGEMM GFLOPS > Higher Is Better Threadripper 3970X . 29.80 |=============================== 16 Cores ........... 49.58 |=================================================== Timed GCC Compilation 8.2 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 589.11 |============================== 16 Cores ........... 979.62 |================================================== Timed ImageMagick Compilation 6.9.0 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 13.43 |=============================== 16 Cores ........... 22.30 |=================================================== John The Ripper 1.9.0-jumbo-1 Test: Blowfish Real C/S > Higher Is Better Threadripper 3970X . 43255 |=================================================== 16 Cores ........... 26111 |=============================== Facebook RocksDB 6.3.6 Test: Sequential Fill Op/s > Higher Is Better Threadripper 3970X . 1147975 |================================================= 16 Cores ........... 702809 |============================== Parboil 2.5 Test: OpenMP MRI Gridding Seconds < Lower Is Better Threadripper 3970X . 60.18 |=================================================== 16 Cores ........... 37.09 |=============================== GROMACS 2020 Water Benchmark Ns Per Day > Higher Is Better Threadripper 3970X . 3.288 |=================================================== 16 Cores ........... 2.048 |================================ Stress-NG 0.07.26 Test: System V Message Passing Bogo Ops/s > Higher Is Better Threadripper 3970X . 26168429.69 |============================================= 16 Cores ........... 16301159.73 |============================ t-test1 2017-01-13 Threads: 2 Seconds < Lower Is Better Threadripper 3970X . 8.743 |=================================================== 16 Cores ........... 5.449 |================================ t-test1 2017-01-13 Threads: 1 Seconds < Lower Is Better Threadripper 3970X . 25.60 |=================================================== 16 Cores ........... 16.06 |================================ Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Seconds < Lower Is Better Threadripper 3970X . 8.544 |=============================== 16 Cores ........... 13.565 |================================================== Xsbench 2017-07-06 Lookups/s > Higher Is Better Threadripper 3970X . 6079554 |================================================= 16 Cores ........... 3840404 |=============================== dav1d 0.5.0 Video Input: Summer Nature 4K FPS > Higher Is Better Threadripper 3970X . 324.66 |================================================== 16 Cores ........... 205.50 |================================ x264 2019-12-17 H.264 Video Encoding Frames Per Second > Higher Is Better Threadripper 3970X . 205.31 |================================================== 16 Cores ........... 131.88 |================================ Mlpack Benchmark Benchmark: scikit_qda Seconds < Lower Is Better Threadripper 3970X . 53.02 |=================================================== 16 Cores ........... 34.16 |================================= John The Ripper 1.9.0-jumbo-1 Test: MD5 Real C/S > Higher Is Better Threadripper 3970X . 2520750 |================================================= 16 Cores ........... 1626000 |================================ AOM AV1 2020-01-10 Encoder Mode: Speed 8 Realtime Frames Per Second > Higher Is Better Threadripper 3970X . 40.88 |================================= 16 Cores ........... 62.37 |=================================================== Go Benchmarks Test: garbage Nanoseconds/Operation < Lower Is Better Threadripper 3970X . 589712 |================================= 16 Cores ........... 897936 |================================================== LevelDB 1.22 Benchmark: Sequential Fill MB/s > Higher Is Better Threadripper 3970X . 27.6 |================================== 16 Cores ........... 41.8 |==================================================== HPC Challenge 1.5.0 Test / Class: G-Random Access GUP/s > Higher Is Better Threadripper 3970X . 0.17138 |================================================= 16 Cores ........... 0.11362 |================================ Memcached mcperf 1.5.22 Method: Set - Connections: 64 Operations Per Second > Higher Is Better Threadripper 3970X . 48126.0 |================================= 16 Cores ........... 72538.5 |================================================= Timed PHP Compilation 7.4.2 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 32.04 |================================== 16 Cores ........... 48.21 |=================================================== LevelDB 1.22 Benchmark: Overwrite MB/s > Higher Is Better Threadripper 3970X . 26.8 |=================================== 16 Cores ........... 40.3 |==================================================== Perl Benchmarks Test: Interpreter Seconds < Lower Is Better Threadripper 3970X . 0.00136623 |============================================== 16 Cores ........... 0.00091186 |=============================== MKL-DNN DNNL 1.1 Harness: IP Batch All - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 17.33 |=================================================== 16 Cores ........... 11.57 |================================== DaCapo Benchmark 9.12-MR1 Java Test: H2 msec < Lower Is Better Threadripper 3970X . 3886 |==================================================== 16 Cores ........... 2606 |=================================== Timed Apache Compilation 2.4.41 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 18.58 |================================== 16 Cores ........... 27.54 |=================================================== LevelDB 1.22 Benchmark: Random Fill MB/s > Higher Is Better Threadripper 3970X . 27.3 |=================================== 16 Cores ........... 40.3 |==================================================== Memcached mcperf 1.5.22 Method: Add - Connections: 64 Operations Per Second > Higher Is Better Threadripper 3970X . 48669.6 |================================= 16 Cores ........... 71799.0 |================================================= Parboil 2.5 Test: OpenMP Stencil Seconds < Lower Is Better Threadripper 3970X . 5.282141 |================================= 16 Cores ........... 7.782015 |================================================ Renaissance 0.10.0 Test: In-Memory Database Shootout ms < Lower Is Better Threadripper 3970X . 5898.14 |================================================= 16 Cores ........... 4006.60 |================================= Intel Open Image Denoise 1.0.0 Scene: Memorial Images / Sec > Higher Is Better Threadripper 3970X . 21.96 |=================================================== 16 Cores ........... 14.97 |=================================== Renaissance 0.10.0 Test: Akka Unbalanced Cobwebbed Tree ms < Lower Is Better Threadripper 3970X . 12746.61 |================================================ 16 Cores ........... 8732.84 |================================= Parboil 2.5 Test: OpenMP LBM Seconds < Lower Is Better Threadripper 3970X . 52.97 |=================================================== 16 Cores ........... 36.35 |=================================== FFmpeg 4.0.2 H.264 HD To NTSC DV Seconds < Lower Is Better Threadripper 3970X . 7.331 |=================================================== 16 Cores ........... 5.041 |=================================== Build2 0.12 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 49.58 |=================================== 16 Cores ........... 71.88 |=================================================== MKL-DNN DNNL 1.1 Harness: Deconvolution Batch deconv_all - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 1713.90 |================================================= 16 Cores ........... 1182.61 |================================== HPC Challenge 1.5.0 Test / Class: G-Ffte GFLOPS > Higher Is Better Threadripper 3970X . 16.12 |=================================================== 16 Cores ........... 11.17 |=================================== ASKAP 2018-11-10 Test: tConvolve OpenMP - Gridding Million Grid Points Per Second > Higher Is Better Threadripper 3970X . 4771.16 |================================================= 16 Cores ........... 3330.88 |================================== Intel MPI Benchmarks 2019.3 Test: IMB-MPI1 Exchange Average usec > Higher Is Better Threadripper 3970X . 149.28 |================================================== 16 Cores ........... 104.95 |=================================== MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_all - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 968.72 |=================================== 16 Cores ........... 1369.92 |================================================= MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_3d - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 6.82846 |=================================== 16 Cores ........... 9.50372 |================================================= NAS Parallel Benchmarks 3.4 Test / Class: SP.B Total Mop/s > Higher Is Better Threadripper 3970X . 33710.70 |================================================ 16 Cores ........... 24225.64 |================================== libgav1 2019-10-05 Video Input: Summer Nature 1080p FPS > Higher Is Better Threadripper 3970X . 76.46 |==================================== 16 Cores ........... 105.76 |================================================== Mlpack Benchmark Benchmark: scikit_ica Seconds < Lower Is Better Threadripper 3970X . 49.20 |=================================================== 16 Cores ........... 36.05 |===================================== NAS Parallel Benchmarks 3.4 Test / Class: BT.C Total Mop/s > Higher Is Better Threadripper 3970X . 74815.99 |================================================ 16 Cores ........... 54825.13 |=================================== LAMMPS Molecular Dynamics Simulator 9Jan2020 Model: Rhodopsin Protein ns/day > Higher Is Better Threadripper 3970X . 17.27 |=================================================== 16 Cores ........... 12.75 |====================================== FS-Mark 3.3 Test: 1000 Files, 1MB Size, No Sync/FSync Files/s > Higher Is Better Threadripper 3970X . 2724.8 |================================================== 16 Cores ........... 2012.9 |===================================== MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 56.58 |====================================== 16 Cores ........... 76.35 |=================================================== Appleseed 2.0 Beta Scene: Material Tester Seconds < Lower Is Better Threadripper 3970X . 131.52 |===================================== 16 Cores ........... 176.69 |================================================== GIMP 2.10.8 Test: unsharp-mask Seconds < Lower Is Better Threadripper 3970X . 20.17 |=================================================== 16 Cores ........... 15.12 |====================================== GIMP 2.10.8 Test: auto-levels Seconds < Lower Is Better Threadripper 3970X . 16.99 |=================================================== 16 Cores ........... 12.77 |====================================== Sockperf 3.4 Test: Latency Ping Pong usec < Lower Is Better Threadripper 3970X . 2.931 |=================================================== 16 Cores ........... 2.210 |====================================== Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better Threadripper 3970X . 4.842 |=================================================== 16 Cores ........... 3.678 |======================================= Compile Bench 0.6 Test: Initial Create MB/s > Higher Is Better Threadripper 3970X . 716.22 |================================================== 16 Cores ........... 548.75 |====================================== dav1d 0.5.0 Video Input: Summer Nature 1080p FPS > Higher Is Better Threadripper 3970X . 726.09 |================================================== 16 Cores ........... 556.99 |====================================== Intel MPI Benchmarks 2019.3 Test: IMB-MPI1 Sendrecv Average usec > Higher Is Better Threadripper 3970X . 82.61 |=================================================== 16 Cores ........... 63.49 |======================================= MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_alexnet - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 137.79 |======================================= 16 Cores ........... 178.67 |================================================== Tungsten Renderer 0.2.2 Scene: Water Caustic Seconds < Lower Is Better Threadripper 3970X . 18.68 |======================================= 16 Cores ........... 24.13 |=================================================== librsvg Operation: SVG Files To PNG Seconds < Lower Is Better Threadripper 3970X . 23.87 |=================================================== 16 Cores ........... 18.68 |======================================== XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 Seconds < Lower Is Better Threadripper 3970X . 18.65 |======================================== 16 Cores ........... 23.63 |=================================================== Flexible IO Tester Type: Sequential Read - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better Threadripper 3970X . 459333 |================================================== 16 Cores ........... 364333 |======================================== Flexible IO Tester 3.12 Type: Sequential Read - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory MB/s > Higher Is Better Threadripper 3970X . 1794 |==================================================== 16 Cores ........... 1442 |========================================== 16 Cores ........... 1424 |========================================= Rust Mandelbrot Time To Complete Serial/Parallel Mandelbrot Seconds < Lower Is Better Threadripper 3970X . 32.09 |========================================= 16 Cores ........... 40.17 |=================================================== ctx_clock Context Switch Time Clocks < Lower Is Better Threadripper 3970X . 148 |========================================== 16 Cores ........... 185 |===================================================== Flexible IO Tester 3.12 Type: Random Read - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory MB/s > Higher Is Better Threadripper 3970X . 56.2 |==================================================== 16 Cores ........... 45.2 |========================================== IOzone 3.465 Record Size: 4Kb - File Size: 512MB - Disk Test: Write Performance MB/s > Higher Is Better Threadripper 3970X . 1594.31 |================================================= 16 Cores ........... 1283.87 |======================================= Flexible IO Tester Type: Random Read - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better Threadripper 3970X . 14400 |=================================================== 16 Cores ........... 11600 |========================================= GNU Octave Benchmark 4.4.1 Seconds < Lower Is Better Threadripper 3970X . 13.07 |=================================================== 16 Cores ........... 10.58 |========================================= Rodinia 2.4 Test: OpenMP CFD Solver Seconds < Lower Is Better Threadripper 3970X . 9.784 |========================================= 16 Cores ........... 12.077 |================================================== IOzone 3.465 Record Size: 64Kb - File Size: 512MB - Disk Test: Write Performance MB/s > Higher Is Better Threadripper 3970X . 1677.55 |================================================= 16 Cores ........... 1361.61 |======================================== HPC Challenge 1.5.0 Test / Class: G-HPL GFLOPS > Higher Is Better Threadripper 3970X . 73.20 |========================================= 16 Cores ........... 89.99 |=================================================== PostMark 1.51 Disk Transaction Performance TPS > Higher Is Better Threadripper 3970X . 7812 |==================================================== 16 Cores ........... 6357 |========================================== NeatBench 5 Acceleration: CPU FPS > Higher Is Better Threadripper 3970X . 37.6 |==================================================== 16 Cores ........... 30.7 |========================================== GIMP 2.10.8 Test: resize Seconds < Lower Is Better Threadripper 3970X . 8.470 |=================================================== 16 Cores ........... 6.923 |========================================== GraphicsMagick 1.3.33 Operation: Resizing Iterations Per Minute > Higher Is Better Threadripper 3970X . 2243 |==================================================== 16 Cores ........... 1847 |=========================================== Darktable 2.6.0 Test: Masskrug - Acceleration: CPU-only Seconds < Lower Is Better Threadripper 3970X . 3.088 |========================================== 16 Cores ........... 3.746 |=================================================== OSBench Test: Create Files us Per Event < Lower Is Better Threadripper 3970X . 11.45 |========================================== 16 Cores ........... 13.88 |=================================================== NAS Parallel Benchmarks 3.4 Test / Class: LU.C Total Mop/s > Higher Is Better Threadripper 3970X . 65854.97 |================================================ 16 Cores ........... 54335.46 |======================================== Memcached mcperf 1.5.10 Method: Delete Operations Per Second > Higher Is Better Threadripper 3970X . 74147.4 |========================================= 16 Cores ........... 88516.6 |================================================= Stress-NG 0.07.26 Test: Semaphores Bogo Ops/s > Higher Is Better Threadripper 3970X . 3379188.34 |======================================= 16 Cores ........... 3998333.22 |============================================== miniFE 2.2 Problem Size: Small CG Mflops > Higher Is Better Threadripper 3970X . 9322.75 |========================================= 16 Cores ........... 10962.70 |================================================ RAMspeed SMP 3.5.0 Type: Scale - Benchmark: Integer MB/s > Higher Is Better Threadripper 3970X . 42704.74 |================================================ 16 Cores ........... 36372.44 |========================================= VP9 libvpx Encoding 1.8.2 Speed: Speed 5 Frames Per Second > Higher Is Better Threadripper 3970X . 24.48 |============================================ 16 Cores ........... 28.62 |=================================================== CP2K Molecular Dynamics 6.1 Fayalite-FIST Data Seconds < Lower Is Better Threadripper 3970X . 494.65 |================================================== 16 Cores ........... 423.75 |=========================================== Go Benchmarks Test: build Nanoseconds/Operation < Lower Is Better Threadripper 3970X . 20067677101 |============================================= 16 Cores ........... 17244462905 |======================================= LULESH 2.0.3 z/s > Higher Is Better Threadripper 3970X . 9.3908987 |======================================== 16 Cores ........... 10.9021150 |============================================== MKL-DNN DNNL 1.1 Harness: Deconvolution Batch deconv_1d - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 2.78688 |========================================== 16 Cores ........... 3.22435 |================================================= LevelDB 1.22 Benchmark: Fill Sync Microseconds Per Op < Lower Is Better Threadripper 3970X . 1938.17 |================================================= 16 Cores ........... 1677.71 |========================================== DeepSpeech 0.6 Acceleration: CPU Seconds < Lower Is Better Threadripper 3970X . 56.68 |=================================================== 16 Cores ........... 49.10 |============================================ Intel MPI Benchmarks 2019.3 Test: IMB-MPI1 Exchange Average Mbytes/sec > Higher Is Better Threadripper 3970X . 6258.07 |========================================== 16 Cores ........... 7217.49 |================================================= HPC Challenge 1.5.0 Test / Class: Random Ring Latency usecs < Lower Is Better Threadripper 3970X . 0.46962 |================================================= 16 Cores ........... 0.40788 |=========================================== Memcached mcperf 1.5.10 Method: Get Operations Per Second > Higher Is Better Threadripper 3970X . 77680.2 |=========================================== 16 Cores ........... 89370.8 |================================================= ASKAP 2018-11-10 Test: tConvolve MT - Gridding Million Grid Points Per Second > Higher Is Better Threadripper 3970X . 2651.59 |================================================= 16 Cores ........... 2306.50 |=========================================== Numenta Anomaly Benchmark 1.1 Detector: Windowed Gaussian Seconds < Lower Is Better Threadripper 3970X . 6.111 |============================================ 16 Cores ........... 7.013 |=================================================== GEGL Operation: Scale Seconds < Lower Is Better Threadripper 3970X . 5.661 |=================================================== 16 Cores ........... 4.940 |============================================= RAR Compression 5.6.1 Linux Source Tree Archiving To RAR Seconds < Lower Is Better Threadripper 3970X . 68.09 |=================================================== 16 Cores ........... 59.84 |============================================= ASKAP 2018-11-10 Test: tConvolve MPI - Gridding Million Grid Points Per Second > Higher Is Better Threadripper 3970X . 2682.40 |================================================= 16 Cores ........... 2368.48 |=========================================== Apache Benchmark 2.4.29 Static Web Page Serving Requests Per Second > Higher Is Better Threadripper 3970X . 36286.78 |========================================== 16 Cores ........... 41019.20 |================================================ FS-Mark 3.3 Test: 4000 Files, 32 Sub Dirs, 1MB Size Files/s > Higher Is Better Threadripper 3970X . 560.7 |============================================= 16 Cores ........... 633.2 |=================================================== x265 3.1.2 H.265 1080p Video Encoding Frames Per Second > Higher Is Better Threadripper 3970X . 64.90 |============================================= 16 Cores ........... 73.21 |=================================================== Memcached mcperf 1.5.22 Method: Get - Connections: 64 Operations Per Second > Higher Is Better Threadripper 3970X . 78526.0 |============================================ 16 Cores ........... 88109.7 |================================================= GIMP 2.10.8 Test: rotate Seconds < Lower Is Better Threadripper 3970X . 12.44 |=================================================== 16 Cores ........... 11.19 |============================================== NAS Parallel Benchmarks 3.4 Test / Class: FT.C Total Mop/s > Higher Is Better Threadripper 3970X . 34599.49 |================================================ 16 Cores ........... 31324.20 |=========================================== Sysbench 2018-07-28 Test: Memory Events Per Second > Higher Is Better Threadripper 3970X . 9119242.23 |============================================== 16 Cores ........... 8262426.98 |========================================== Renaissance 0.10.0 Test: Apache Spark ALS ms < Lower Is Better Threadripper 3970X . 4708.54 |================================================= 16 Cores ........... 4278.41 |============================================= Flexible IO Tester 3.18 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory MB/s > Higher Is Better Threadripper 3970X . 1052 |==================================================== 16 Cores ........... 961 |================================================ Numenta Anomaly Benchmark 1.1 Detector: Earthgecko Skyline Seconds < Lower Is Better Threadripper 3970X . 69.66 |=============================================== 16 Cores ........... 75.99 |=================================================== Nebular Empirical Analysis Tool 2020-02-29 Seconds < Lower Is Better Threadripper 3970X . 12.82 |=============================================== 16 Cores ........... 13.98 |=================================================== Node.js Express HTTP Load Test Requests Per Second > Higher Is Better Threadripper 3970X . 10908 |=============================================== 16 Cores ........... 11858 |=================================================== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better Threadripper 3970X . 356.71 |================================================== 16 Cores ........... 328.58 |============================================== Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis Seconds < Lower Is Better Threadripper 3970X . 75.41 |=================================================== 16 Cores ........... 69.52 |=============================================== GEGL Operation: Tile Glass Seconds < Lower Is Better Threadripper 3970X . 31.03 |=================================================== 16 Cores ........... 28.69 |=============================================== Tensorflow Build: Cifar10 Seconds < Lower Is Better Threadripper 3970X . 42.02 |=============================================== 16 Cores ........... 45.44 |=================================================== ASKAP 2018-11-10 Test: tConvolve OpenMP - Degridding Million Grid Points Per Second > Higher Is Better Threadripper 3970X . 3764.47 |================================================= 16 Cores ........... 3482.69 |============================================= Systemd Total Boot Time Test: Kernel ms < Lower Is Better Threadripper 3970X . 5710 |================================================ 16 Cores ........... 6170 |==================================================== TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping FPS > Higher Is Better Threadripper 3970X . 667.34 |================================================== 16 Cores ........... 617.85 |============================================== ASKAP 2018-11-10 Test: tConvolve MT - Degridding Million Grid Points Per Second > Higher Is Better Threadripper 3970X . 3738.11 |================================================= 16 Cores ........... 3463.50 |============================================= OpenCV Benchmark 3.3.0 Seconds < Lower Is Better Threadripper 3970X . 66.36 |=================================================== 16 Cores ........... 61.54 |=============================================== Qmlbench 2 Test: Canvas Text Simple Frames > Higher Is Better Threadripper 3970X . 358.8 |=============================================== 16 Cores ........... 386.4 |=================================================== Algebraic Multi-Grid Benchmark Figure Of Merit > Higher Is Better Threadripper 3970X . 24786.28 |============================================= 16 Cores ........... 26671.34 |================================================ Perl Benchmarks Test: Pod2html Seconds < Lower Is Better Threadripper 3970X . 0.11410246 |=========================================== 16 Cores ........... 0.12256629 |============================================== Flexible IO Tester 3.18 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better Threadripper 3970X . 294333 |=============================================== 16 Cores ........... 315733 |================================================== GEGL Operation: Antialias Seconds < Lower Is Better Threadripper 3970X . 39.00 |=================================================== 16 Cores ........... 36.36 |================================================ Flexible IO Tester 3.18 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory MB/s > Higher Is Better Threadripper 3970X . 1151 |================================================= 16 Cores ........... 1234 |==================================================== Numenta Anomaly Benchmark 1.1 Detector: Relative Entropy Seconds < Lower Is Better Threadripper 3970X . 12.42 |=================================================== 16 Cores ........... 11.59 |================================================ ASKAP 2018-11-10 Test: tConvolve MPI - Degridding Million Grid Points Per Second > Higher Is Better Threadripper 3970X . 3795.21 |================================================= 16 Cores ........... 3544.17 |============================================== ArrayFire 3.7 Test: Conjugate Gradient CPU ms < Lower Is Better Threadripper 3970X . 25.18 |=================================================== 16 Cores ........... 23.52 |================================================ High Performance Conjugate Gradient 3.1 GFLOP/s > Higher Is Better Threadripper 3970X . 10.67070 |================================================ 16 Cores ........... 9.97276 |============================================= GEGL Operation: Crop Seconds < Lower Is Better Threadripper 3970X . 8.003 |=================================================== 16 Cores ........... 7.491 |================================================ IOR 3.2.1 Read Test MB/s > Higher Is Better Threadripper 3970X . 1047.58 |============================================== 16 Cores ........... 1117.52 |================================================= FS-Mark 3.3 Test: 1000 Files, 1MB Size Files/s > Higher Is Better Threadripper 3970X . 576.3 |================================================ 16 Cores ........... 611.7 |=================================================== Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better Threadripper 3970X . 4607.31 |============================================== 16 Cores ........... 4880.02 |================================================= Stream 2013-01-17 Type: Copy MB/s > Higher Is Better Threadripper 3970X . 52075.8 |================================================= 16 Cores ........... 49216.8 |============================================== GraphicsMagick 1.3.33 Operation: HWB Color Space Iterations Per Minute > Higher Is Better Threadripper 3970X . 1499 |==================================================== 16 Cores ........... 1417 |================================================= IPC_benchmark Type: Unnamed Unix Domain Socket - Message Bytes: 128 Messages Per Second > Higher Is Better Threadripper 3970X . 2416213 |============================================== 16 Cores ........... 2555431 |================================================= Redis 5.0.5 Test: GET Requests Per Second > Higher Is Better Threadripper 3970X . 2229021.42 |============================================== 16 Cores ........... 2108260.53 |============================================ GraphicsMagick 1.3.33 Operation: Rotate Iterations Per Minute > Higher Is Better Threadripper 3970X . 693 |================================================== 16 Cores ........... 732 |===================================================== Sudokut 0.4 Total Time Seconds < Lower Is Better Threadripper 3970X . 12.90 |=================================================== 16 Cores ........... 12.23 |================================================ Systemd Total Boot Time Test: Total ms < Lower Is Better Threadripper 3970X . 37055 |================================================ 16 Cores ........... 39048 |=================================================== Bork File Encrypter 1.4 File Encryption Time Seconds < Lower Is Better Threadripper 3970X . 8.121 |================================================ 16 Cores ........... 8.556 |=================================================== Redis 5.0.5 Test: LPUSH Requests Per Second > Higher Is Better Threadripper 3970X . 1417209.04 |============================================ 16 Cores ........... 1492623.12 |============================================== BYTE Unix Benchmark 3.6 Computational Test: Dhrystone 2 LPS > Higher Is Better Threadripper 3970X . 44416335.2 |============================================== 16 Cores ........... 42231623.5 |============================================ NAS Parallel Benchmarks 3.4 Test / Class: MG.C Total Mop/s > Higher Is Better Threadripper 3970X . 30218.91 |================================================ 16 Cores ........... 28788.06 |============================================== Systemd Total Boot Time Test: Userspace ms < Lower Is Better Threadripper 3970X . 31345 |================================================= 16 Cores ........... 32878 |=================================================== Zstd Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Seconds < Lower Is Better Threadripper 3970X . 170.78 |================================================== 16 Cores ........... 163.05 |================================================ OSBench Test: Memory Allocations Ns Per Event < Lower Is Better Threadripper 3970X . 70.32 |=================================================== 16 Cores ........... 67.16 |================================================= Stream 2013-01-17 Type: Triad MB/s > Higher Is Better Threadripper 3970X . 57321.8 |================================================= 16 Cores ........... 54847.9 |=============================================== Stream 2013-01-17 Type: Scale MB/s > Higher Is Better Threadripper 3970X . 51667.1 |================================================= 16 Cores ........... 49498.8 |=============================================== Flexible IO Tester Type: Sequential Write - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better Threadripper 3970X . 307000 |================================================ 16 Cores ........... 320333 |================================================== Flexible IO Tester 3.12 Type: Sequential Write - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory MB/s > Higher Is Better Threadripper 3970X . 1200 |================================================== 16 Cores ........... 1246 |==================================================== 16 Cores ........... 1251 |==================================================== IPC_benchmark Type: Unnamed Pipe - Message Bytes: 128 Messages Per Second > Higher Is Better Threadripper 3970X . 3558893 |=============================================== 16 Cores ........... 3701949 |================================================= Inkscape Operation: SVG Files To PNG Seconds < Lower Is Better Threadripper 3970X . 26.27 |=================================================== 16 Cores ........... 25.26 |================================================= SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better Threadripper 3970X . 983.93 |================================================== 16 Cores ........... 946.17 |================================================ RawTherapee Total Benchmark Time Seconds < Lower Is Better Threadripper 3970X . 40.81 |================================================= 16 Cores ........... 42.41 |=================================================== PlaidML FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU FPS > Higher Is Better Threadripper 3970X . 7.40 |================================================== 16 Cores ........... 7.69 |==================================================== MKL-DNN DNNL 1.1 Harness: IP Batch 1D - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 3.41655 |=============================================== 16 Cores ........... 3.55042 |================================================= NAS Parallel Benchmarks 3.4 Test / Class: IS.D Total Mop/s > Higher Is Better Threadripper 3970X . 1464.45 |=============================================== 16 Cores ........... 1521.34 |================================================= Loopback TCP Network Performance Time To Transfer 10GB Via Loopback Seconds < Lower Is Better Threadripper 3970X . 10.40 |================================================= 16 Cores ........... 10.81 |=================================================== Stream 2013-01-17 Type: Add MB/s > Higher Is Better Threadripper 3970X . 56819.2 |================================================= 16 Cores ........... 54720.5 |=============================================== GEGL Operation: Wavelet Blur Seconds < Lower Is Better Threadripper 3970X . 62.40 |=================================================== 16 Cores ........... 60.11 |================================================= SQLite 3.30.1 Threads / Copies: 1 Seconds < Lower Is Better Threadripper 3970X . 14.41 |=================================================== 16 Cores ........... 13.92 |================================================= IPC_benchmark Type: TCP Socket - Message Bytes: 128 Messages Per Second > Higher Is Better Threadripper 3970X . 3578760 |=============================================== 16 Cores ........... 3705204 |================================================= Mlpack Benchmark Benchmark: scikit_linearridgeregression Seconds < Lower Is Better Threadripper 3970X . 1.47 |==================================================== 16 Cores ........... 1.42 |================================================== LuaJIT 2.1-git Test: Fast Fourier Transform Mflops > Higher Is Better Threadripper 3970X . 287.97 |================================================== 16 Cores ........... 278.18 |================================================ Memtier_benchmark 1.2.17 Protocol: Redis Ops/sec > Higher Is Better Threadripper 3970X . 2351667.58 |============================================== 16 Cores ........... 2272891.42 |============================================ Darktable 2.6.0 Test: Server Room - Acceleration: CPU-only Seconds < Lower Is Better Threadripper 3970X . 1.991 |================================================= 16 Cores ........... 2.057 |=================================================== PyPerformance 1.0.0 Benchmark: float Milliseconds < Lower Is Better Threadripper 3970X . 103.0 |=================================================== 16 Cores ........... 99.7 |================================================= Redis 5.0.5 Test: SET Requests Per Second > Higher Is Better Threadripper 3970X . 1674234.13 |============================================= 16 Cores ........... 1723424.59 |============================================== Node.js Octane Benchmark Score > Higher Is Better Threadripper 3970X . 50125 |================================================== 16 Cores ........... 51582 |=================================================== SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better Threadripper 3970X . 718.03 |================================================== 16 Cores ........... 697.80 |================================================= RAMspeed SMP 3.5.0 Type: Copy - Benchmark: Integer MB/s > Higher Is Better Threadripper 3970X . 41870.30 |=============================================== 16 Cores ........... 43057.69 |================================================ DaCapo Benchmark 9.12-MR1 Java Test: Jython msec < Lower Is Better Threadripper 3970X . 4023 |==================================================== 16 Cores ........... 3921 |=================================================== GEGL Operation: Cartoon Seconds < Lower Is Better Threadripper 3970X . 90.95 |=================================================== 16 Cores ........... 88.68 |================================================== LuaJIT 2.1-git Test: Sparse Matrix Multiply Mflops > Higher Is Better Threadripper 3970X . 1352.77 |================================================= 16 Cores ........... 1319.00 |================================================ DaCapo Benchmark 9.12-MR1 Java Test: Tradesoap msec < Lower Is Better Threadripper 3970X . 3301 |==================================================== 16 Cores ........... 3219 |=================================================== HPC Challenge 1.5.0 Test / Class: Max Ping Pong Bandwidth MB/s > Higher Is Better Threadripper 3970X . 21719.65 |=============================================== 16 Cores ........... 22271.54 |================================================ Fhourstones 3.1 Complex Connect-4 Solving Kpos / sec > Higher Is Better Threadripper 3970X . 14702.3 |================================================ 16 Cores ........... 15067.4 |================================================= Cython benchmark 0.27 Seconds < Lower Is Better Threadripper 3970X . 40.86 |=================================================== 16 Cores ........... 39.94 |================================================== BLAKE2 20170307 Cycles Per Byte < Lower Is Better Threadripper 3970X . 8.35 |=================================================== 16 Cores ........... 8.54 |==================================================== LuaJIT 2.1-git Test: Monte Carlo Mflops > Higher Is Better Threadripper 3970X . 526.75 |================================================== 16 Cores ........... 515.37 |================================================= LuaJIT 2.1-git Test: Dense LU Matrix Factorization Mflops > Higher Is Better Threadripper 3970X . 3996.69 |================================================= 16 Cores ........... 3910.40 |================================================ CppPerformanceBenchmarks 9 Test: Stepanov Vector Seconds < Lower Is Better Threadripper 3970X . 77.19 |================================================== 16 Cores ........... 78.89 |=================================================== LuaJIT 2.1-git Test: Composite Mflops > Higher Is Better Threadripper 3970X . 1659.51 |================================================= 16 Cores ........... 1624.14 |================================================ Git Time To Complete Common Git Commands Seconds < Lower Is Better Threadripper 3970X . 4.914 |================================================== 16 Cores ........... 5.018 |=================================================== eSpeak Speech Engine 1.48.04 Text-To-Speech Synthesis Seconds < Lower Is Better Threadripper 3970X . 27.35 |================================================== 16 Cores ........... 27.91 |=================================================== GEGL Operation: Reflect Seconds < Lower Is Better Threadripper 3970X . 29.98 |=================================================== 16 Cores ........... 29.41 |================================================== lzbench 1.8 Test: Zstd 8 - Process: Compression MB/s > Higher Is Better Threadripper 3970X . 107 |==================================================== 16 Cores ........... 109 |===================================================== LibreOffice Test: 20 Documents To PDF Seconds < Lower Is Better Threadripper 3970X . 6.663 |=================================================== 16 Cores ........... 6.541 |================================================== BlogBench 1.1 Test: Read Final Score > Higher Is Better Threadripper 3970X . 1260198 |================================================ 16 Cores ........... 1283483 |================================================= Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride, 4 Threads ns < Lower Is Better Threadripper 3970X . 74.56 |=================================================== 16 Cores ........... 73.27 |================================================== SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better Threadripper 3970X . 1298.02 |================================================= 16 Cores ........... 1275.97 |================================================ LuaJIT 2.1-git Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better Threadripper 3970X . 2133.37 |================================================= 16 Cores ........... 2097.74 |================================================ glibc bench 1.0 Benchmark: ffs nanoseconds < Lower Is Better Threadripper 3970X . 1.75870 |================================================ 16 Cores ........... 1.78754 |================================================= CppPerformanceBenchmarks 9 Test: Atol Seconds < Lower Is Better Threadripper 3970X . 58.06 |================================================== 16 Cores ........... 58.99 |=================================================== Botan 2.13.0 Test: Twofish MiB/s > Higher Is Better Threadripper 3970X . 391.43 |================================================== 16 Cores ........... 385.68 |================================================= lzbench 1.8 Test: Brotli 0 - Process: Compression MB/s > Higher Is Better Threadripper 3970X . 547 |===================================================== 16 Cores ........... 539 |==================================================== SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better Threadripper 3970X . 157.71 |================================================== 16 Cores ........... 155.44 |================================================= glibc bench 1.0 Benchmark: asinh nanoseconds < Lower Is Better Threadripper 3970X . 8.41835 |================================================ 16 Cores ........... 8.53711 |================================================= SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better Threadripper 3970X . 793.80 |================================================== 16 Cores ........... 782.85 |================================================= glibc bench 1.0 Benchmark: cos nanoseconds < Lower Is Better Threadripper 3970X . 42.25 |================================================== 16 Cores ........... 42.81 |=================================================== TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better Threadripper 3970X . 1336877 |================================================= 16 Cores ........... 1319372 |================================================ Hierarchical INTegration 1.0 Test: FLOAT QUIPs > Higher Is Better Threadripper 3970X . 386120425.42 |============================================ 16 Cores ........... 381206340.37 |=========================================== Redis 5.0.5 Test: SADD Requests Per Second > Higher Is Better Threadripper 3970X . 1851271.35 |============================================= 16 Cores ........... 1874999.40 |============================================== MBW 2018-09-08 Test: Memory Copy - Array Size: 4096 MiB MiB/s > Higher Is Better Threadripper 3970X . 21587.38 |=============================================== 16 Cores ........... 21862.43 |================================================ GEGL Operation: Rotate 90 Degrees Seconds < Lower Is Better Threadripper 3970X . 38.17 |=================================================== 16 Cores ........... 37.69 |================================================== Java SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better Threadripper 3970X . 2683.18 |================================================ 16 Cores ........... 2716.60 |================================================= Scikit-Learn 0.22.1 Seconds < Lower Is Better Threadripper 3970X . 7.973 |=================================================== 16 Cores ........... 7.875 |================================================== Optcarrot Optimized Benchmark FPS > Higher Is Better Threadripper 3970X . 103.98 |================================================== 16 Cores ........... 102.75 |================================================= glibc bench 1.0 Benchmark: sincos nanoseconds < Lower Is Better Threadripper 3970X . 12.26 |================================================== 16 Cores ........... 12.41 |=================================================== GEGL Operation: Color Enhance Seconds < Lower Is Better Threadripper 3970X . 56.07 |=================================================== 16 Cores ........... 55.43 |================================================== Timed GDB GNU Debugger Compilation 9.1 Time To Compile Seconds < Lower Is Better Threadripper 3970X . 83.12 |=================================================== 16 Cores ........... 82.19 |================================================== Tesseract OCR 4.1.0 Time To OCR 7 Images Seconds < Lower Is Better Threadripper 3970X . 24.08 |=================================================== 16 Cores ........... 23.82 |================================================== glibc bench 1.0 Benchmark: atanh nanoseconds < Lower Is Better Threadripper 3970X . 10.09 |================================================== 16 Cores ........... 10.20 |=================================================== glibc bench 1.0 Benchmark: ffsll nanoseconds < Lower Is Better Threadripper 3970X . 1.75897 |================================================ 16 Cores ........... 1.77800 |================================================= Radiance Benchmark 5.0 Test: SMP Parallel Seconds < Lower Is Better Threadripper 3970X . 173.32 |================================================== 16 Cores ........... 171.49 |================================================= IPC_benchmark Type: FIFO Named Pipe - Message Bytes: 128 Messages Per Second > Higher Is Better Threadripper 3970X . 3388324 |================================================= 16 Cores ........... 3353946 |================================================= Bullet Physics Engine 2.81 Test: Prim Trimesh Seconds < Lower Is Better Threadripper 3970X . 0.796787 |================================================ 16 Cores ........... 0.804927 |================================================ Bullet Physics Engine 2.81 Test: 3000 Fall Seconds < Lower Is Better Threadripper 3970X . 3.394250 |================================================ 16 Cores ........... 3.428156 |================================================ Bullet Physics Engine 2.81 Test: Convex Trimesh Seconds < Lower Is Better Threadripper 3970X . 0.952042 |================================================ 16 Cores ........... 0.961552 |================================================ Bullet Physics Engine 2.81 Test: 136 Ragdolls Seconds < Lower Is Better Threadripper 3970X . 2.173580 |================================================ 16 Cores ........... 2.195278 |================================================ PolyBench-C 4.2 Test: Covariance Computation Seconds < Lower Is Better Threadripper 3970X . 2.141 |=================================================== 16 Cores ........... 2.120 |================================================== Bullet Physics Engine 2.81 Test: Raytests Seconds < Lower Is Better Threadripper 3970X . 2.106475 |================================================ 16 Cores ........... 2.127226 |================================================ Bullet Physics Engine 2.81 Test: 1000 Convex Seconds < Lower Is Better Threadripper 3970X . 3.787201 |================================================ 16 Cores ........... 3.822767 |================================================ Minion 1.8 Benchmark: Quasigroup Seconds < Lower Is Better Threadripper 3970X . 107.53 |================================================== 16 Cores ........... 108.52 |================================================== CppPerformanceBenchmarks 9 Test: Math Library Seconds < Lower Is Better Threadripper 3970X . 273.23 |================================================== 16 Cores ........... 275.73 |================================================== Bullet Physics Engine 2.81 Test: 1000 Stack Seconds < Lower Is Better Threadripper 3970X . 4.094888 |================================================ 16 Cores ........... 4.132133 |================================================ Numenta Anomaly Benchmark 1.1 Detector: Bayesian Changepoint Seconds < Lower Is Better Threadripper 3970X . 24.90 |=================================================== 16 Cores ........... 24.67 |=================================================== glibc bench 1.0 Benchmark: modf nanoseconds < Lower Is Better Threadripper 3970X . 2.23442 |================================================= 16 Cores ........... 2.25415 |================================================= glibc bench 1.0 Benchmark: sqrt nanoseconds < Lower Is Better Threadripper 3970X . 2.22027 |================================================= 16 Cores ........... 2.23979 |================================================= CppPerformanceBenchmarks 9 Test: Stepanov Abstraction Seconds < Lower Is Better Threadripper 3970X . 28.62 |=================================================== 16 Cores ........... 28.87 |=================================================== Botan 2.13.0 Test: Blowfish MiB/s > Higher Is Better Threadripper 3970X . 475.10 |================================================== 16 Cores ........... 471.10 |================================================== glibc bench 1.0 Benchmark: exp nanoseconds < Lower Is Better Threadripper 3970X . 4.99317 |================================================= 16 Cores ........... 4.95189 |================================================= lzbench 1.8 Test: Crush 0 - Process: Compression MB/s > Higher Is Better Threadripper 3970X . 121 |===================================================== 16 Cores ........... 120 |===================================================== glibc bench 1.0 Benchmark: pthread_once nanoseconds < Lower Is Better Threadripper 3970X . 1.76133 |================================================= 16 Cores ........... 1.77594 |================================================= Minion 1.8 Benchmark: Graceful Seconds < Lower Is Better Threadripper 3970X . 41.34 |=================================================== 16 Cores ........... 41.67 |=================================================== lzbench 1.8 Test: Brotli 0 - Process: Decompression MB/s > Higher Is Better Threadripper 3970X . 633 |===================================================== 16 Cores ........... 628 |===================================================== MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB MiB/s > Higher Is Better Threadripper 3970X . 12649.31 |================================================ 16 Cores ........... 12748.61 |================================================ Renaissance 0.10.0 Test: Apache Spark PageRank ms < Lower Is Better Threadripper 3970X . 16363.84 |================================================ 16 Cores ........... 16238.53 |================================================ lzbench 1.8 Test: XZ 0 - Process: Decompression MB/s > Higher Is Better Threadripper 3970X . 131 |===================================================== 16 Cores ........... 130 |===================================================== Tinymembench 2018-05-28 Standard Memcpy MB/s > Higher Is Better Threadripper 3970X . 12225.4 |================================================= 16 Cores ........... 12317.7 |================================================= Systemd Total Boot Time Test: Firmware ms < Lower Is Better Threadripper 3970X . 20002 |=================================================== 16 Cores ........... 20153 |=================================================== Botan 2.13.0 Test: AES-256 MiB/s > Higher Is Better Threadripper 3970X . 5902.14 |================================================= 16 Cores ........... 5858.03 |================================================= CppPerformanceBenchmarks 9 Test: Function Objects Seconds < Lower Is Better Threadripper 3970X . 14.87 |=================================================== 16 Cores ........... 14.98 |=================================================== Minion 1.8 Benchmark: Solitaire Seconds < Lower Is Better Threadripper 3970X . 60.65 |=================================================== 16 Cores ........... 61.09 |=================================================== SQLite Speedtest 3.30 Timed Time - Size 1,000 Seconds < Lower Is Better Threadripper 3970X . 58.59 |=================================================== 16 Cores ........... 58.17 |=================================================== NAS Parallel Benchmarks 3.4 Test / Class: CG.C Total Mop/s > Higher Is Better Threadripper 3970X . 10815.59 |================================================ 16 Cores ........... 10888.19 |================================================ lzbench 1.8 Test: Zstd 1 - Process: Decompression MB/s > Higher Is Better Threadripper 3970X . 1503 |==================================================== 16 Cores ........... 1493 |==================================================== CppPerformanceBenchmarks 9 Test: Ctype Seconds < Lower Is Better Threadripper 3970X . 32.17 |=================================================== 16 Cores ........... 32.38 |=================================================== GnuPG Linux 4.3 Package File Encryption Seconds < Lower Is Better Threadripper 3970X . 12.38 |=================================================== 16 Cores ........... 12.47 |=================================================== CacheBench Test: Write MB/s > Higher Is Better Threadripper 3970X . 31832.50 |================================================ 16 Cores ........... 31631.79 |================================================ Mlpack Benchmark Benchmark: scikit_svm Seconds < Lower Is Better Threadripper 3970X . 11.50 |=================================================== 16 Cores ........... 11.43 |=================================================== glibc bench 1.0 Benchmark: sin nanoseconds < Lower Is Better Threadripper 3970X . 42.04 |=================================================== 16 Cores ........... 42.28 |=================================================== dcraw RAW To PPM Image Conversion Seconds < Lower Is Better Threadripper 3970X . 39.66 |=================================================== 16 Cores ........... 39.43 |=================================================== Renaissance 0.10.0 Test: Scala Dotty ms < Lower Is Better Threadripper 3970X . 5168.07 |================================================= 16 Cores ........... 5197.60 |================================================= Java SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better Threadripper 3970X . 6674.41 |================================================= 16 Cores ........... 6637.13 |================================================= Botan 2.13.0 Test: CAST-256 MiB/s > Higher Is Better Threadripper 3970X . 154.15 |================================================== 16 Cores ........... 154.99 |================================================== Ogg Encoding 1.3.3 WAV To Ogg Seconds < Lower Is Better Threadripper 3970X . 5.225 |=================================================== 16 Cores ........... 5.253 |=================================================== Redis 5.0.5 Test: SET Requests Per Second > Higher Is Better Threadripper 3970X . 1687462.13 |============================================== 16 Cores ........... 1696335.08 |============================================== Swet 1.5.16 Average Operations Per Second > Higher Is Better Threadripper 3970X . 811442548 |=============================================== 16 Cores ........... 815597466 |=============================================== FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 Mflops > Higher Is Better Threadripper 3970X . 6875.5 |================================================== 16 Cores ........... 6841.8 |================================================== GNU GMP GMPbench 6.1.2 Total Time GMPbench Score > Higher Is Better Threadripper 3970X . 5794.6 |================================================== 16 Cores ........... 5767.1 |================================================== lzbench 1.8 Test: Brotli 2 - Process: Compression MB/s > Higher Is Better Threadripper 3970X . 216 |===================================================== 16 Cores ........... 215 |===================================================== LZMA Compression 256MB File Compression Seconds < Lower Is Better Threadripper 3970X . 221.25 |================================================== 16 Cores ........... 220.23 |================================================== Gzip Compression Linux Source Tree Archiving To .tar.gz Seconds < Lower Is Better Threadripper 3970X . 33.91 |=================================================== 16 Cores ........... 34.07 |=================================================== Crypto++ 8.2 Test: Keyed Algorithms MiB/second > Higher Is Better Threadripper 3970X . 696.16 |================================================== 16 Cores ........... 693.03 |================================================== Java SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better Threadripper 3970X . 1686.79 |================================================= 16 Cores ........... 1679.21 |================================================= CppPerformanceBenchmarks 9 Test: Random Numbers Seconds < Lower Is Better Threadripper 3970X . 1110.91 |================================================= 16 Cores ........... 1115.82 |================================================= glibc bench 1.0 Benchmark: sinh nanoseconds < Lower Is Better Threadripper 3970X . 7.79696 |================================================= 16 Cores ........... 7.82957 |================================================= GNU MPC 1.1.0 Multi-Precision Benchmark Global Score > Higher Is Better Threadripper 3970X . 9050 |==================================================== 16 Cores ........... 9013 |==================================================== lzbench 1.8 Test: Brotli 2 - Process: Decompression MB/s > Higher Is Better Threadripper 3970X . 739 |===================================================== 16 Cores ........... 736 |===================================================== Numpy Benchmark Score > Higher Is Better Threadripper 3970X . 432.11 |================================================== 16 Cores ........... 430.36 |================================================== Sockperf 3.4 Test: Throughput Messages Per Second > Higher Is Better Threadripper 3970X . 534240 |================================================== 16 Cores ........... 532089 |================================================== Unpacking The Linux Kernel linux-4.15.tar.xz Seconds < Lower Is Better Threadripper 3970X . 4.984 |=================================================== 16 Cores ........... 4.964 |=================================================== lzbench 1.8 Test: Crush 0 - Process: Decompression MB/s > Higher Is Better Threadripper 3970X . 500 |===================================================== 16 Cores ........... 498 |===================================================== Tinymembench 2018-05-28 Standard Memset MB/s > Higher Is Better Threadripper 3970X . 20042.4 |================================================= 16 Cores ........... 20120.6 |================================================= GnuPG 1.4.22 2GB File Encryption Seconds < Lower Is Better Threadripper 3970X . 11.43 |=================================================== 16 Cores ........... 11.47 |=================================================== lzbench 1.8 Test: Libdeflate 1 - Process: Compression MB/s > Higher Is Better Threadripper 3970X . 266 |===================================================== 16 Cores ........... 267 |===================================================== lzbench 1.8 Test: Zstd 8 - Process: Decompression MB/s > Higher Is Better Threadripper 3970X . 1656 |==================================================== 16 Cores ........... 1650 |==================================================== Renaissance 0.10.0 Test: Random Forest ms < Lower Is Better Threadripper 3970X . 3840.46 |================================================= 16 Cores ........... 3826.80 |================================================= FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 Mflops > Higher Is Better Threadripper 3970X . 21610 |=================================================== 16 Cores ........... 21534 |=================================================== CacheBench Test: Read MB/s > Higher Is Better Threadripper 3970X . 3035.03 |================================================= 16 Cores ........... 3024.68 |================================================= Polyhedron Fortran Benchmarks Benchmark: ac Seconds < Lower Is Better Threadripper 3970X . 5.91 |==================================================== 16 Cores ........... 5.89 |==================================================== Java SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better Threadripper 3970X . 1957.92 |================================================= 16 Cores ........... 1951.48 |================================================= PyBench 2018-02-16 Total For Average Test Times Milliseconds < Lower Is Better Threadripper 3970X . 958 |===================================================== 16 Cores ........... 955 |===================================================== glibc bench 1.0 Benchmark: tanh nanoseconds < Lower Is Better Threadripper 3970X . 10.68 |=================================================== 16 Cores ........... 10.72 |=================================================== PolyBench-C 4.2 Test: Correlation Computation Seconds < Lower Is Better Threadripper 3970X . 2.120 |=================================================== 16 Cores ........... 2.114 |=================================================== glibc bench 1.0 Benchmark: log2 nanoseconds < Lower Is Better Threadripper 3970X . 5.88218 |================================================= 16 Cores ........... 5.89799 |================================================= Crafty 25.2 Elapsed Time Nodes Per Second > Higher Is Better Threadripper 3970X . 8892788 |================================================= 16 Cores ........... 8869846 |================================================= libjpeg-turbo tjbench 2.0.2 Test: Decompression Throughput Megapixels/sec > Higher Is Better Threadripper 3970X . 218.90 |================================================== 16 Cores ........... 218.34 |================================================== Botan 2.13.0 Test: KASUMI MiB/s > Higher Is Better Threadripper 3970X . 100.10 |================================================== 16 Cores ........... 99.85 |================================================== CacheBench Test: Read / Modify / Write MB/s > Higher Is Better Threadripper 3970X . 61162.94 |================================================ 16 Cores ........... 61009.39 |================================================ FLAC Audio Encoding 1.3.2 WAV To FLAC Seconds < Lower Is Better Threadripper 3970X . 7.761 |=================================================== 16 Cores ........... 7.776 |=================================================== PolyBench-C 4.2 Test: 3 Matrix Multiplications Seconds < Lower Is Better Threadripper 3970X . 2.139 |=================================================== 16 Cores ........... 2.135 |=================================================== NGINX Benchmark 1.9.9 Static Web Page Serving Requests Per Second > Higher Is Better Threadripper 3970X . 36707.66 |================================================ 16 Cores ........... 36762.59 |================================================ Java SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better Threadripper 3970X . 3009.12 |================================================= 16 Cores ........... 3005.07 |================================================= Java SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better Threadripper 3970X . 2043.32 |================================================= 16 Cores ........... 2040.93 |================================================= Radiance Benchmark 5.0 Test: Serial Seconds < Lower Is Better Threadripper 3970X . 564.24 |================================================== 16 Cores ........... 564.78 |================================================== LAME MP3 Encoding 3.100 WAV To MP3 Seconds < Lower Is Better Threadripper 3970X . 7.188 |=================================================== 16 Cores ........... 7.182 |=================================================== lzbench 1.8 Test: Libdeflate 1 - Process: Decompression MB/s > Higher Is Better Threadripper 3970X . 1245 |==================================================== 16 Cores ........... 1244 |==================================================== AOBench Size: 2048 x 2048 - Total Time Seconds < Lower Is Better Threadripper 3970X . 31.71 |=================================================== 16 Cores ........... 31.73 |=================================================== PHPBench 0.8.1 PHP Benchmark Suite Score > Higher Is Better Threadripper 3970X . 641756 |================================================== 16 Cores ........... 641380 |================================================== Flexible IO Tester 3.12 Type: Sequential Write - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better 16 Cores . 319333 |============================================================ Flexible IO Tester 3.12 Type: Sequential Read - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better 16 Cores . 369000 |============================================================ Flexible IO Tester 3.12 Type: Random Write - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better 16 Cores . 273333 |============================================================ Flexible IO Tester 3.12 Type: Random Read - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better 16 Cores . 11600 |============================================================= BenchmarkMutex Benchmark: Semaphore Release And Acquire ns < Lower Is Better Threadripper 3970X . 11 |====================================================== 16 Cores ........... 11 |====================================================== BenchmarkMutex Benchmark: Shared Mutex Lock Shared ns < Lower Is Better Threadripper 3970X . 19 |====================================================== 16 Cores ........... 19 |====================================================== System Libxml2 Parsing Filesize: 1 MB ms < Lower Is Better Threadripper 3970X . 424 |===================================================== 16 Cores ........... 424 |===================================================== FFTE 6.0 Test: N=256, 1D Complex FFT Routine MFLOPS > Higher Is Better Threadripper 3970X . 9483.92 |================================================= lzbench 1.8 Test: Zstd 1 - Process: Compression MB/s > Higher Is Better Threadripper 3970X . 543 |===================================================== 16 Cores ........... 543 |===================================================== lzbench 1.8 Test: XZ 0 - Process: Compression MB/s > Higher Is Better Threadripper 3970X . 41 |====================================================== 16 Cores ........... 41 |====================================================== iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 64 Mbits/sec > Higher Is Better Threadripper 3970X . 67.1 |==================================================== 16 Cores ........... 67.1 |==================================================== Schbench Message Threads: 8 - Workers Per Message Thread: 8 usec, 99.9th Latency Percentile < Lower Is Better Threadripper 3970X . 3587 |==== 16 Cores ........... 46315 |=================================================== OSBench Test: Create Processes us Per Event < Lower Is Better Threadripper 3970X . 27.96 |========================== 16 Cores ........... 54.98 |=================================================== Sunflow Rendering System 0.07.2 Global Illumination + Image Synthesis Seconds < Lower Is Better Threadripper 3970X . 0.517 |============================ 16 Cores ........... 0.954 |=================================================== Apache Siege 2.4.29 Concurrent Users: 250 Transactions Per Second > Higher Is Better Threadripper 3970X . 55626.42 |================================================ 16 Cores ........... 54186.82 |=============================================== Apache HBase 2.2.3 Test: Random Write - Clients: 16 Microseconds - Average Latency < Lower Is Better Threadripper 3970X . 72 |================================================ 16 Cores ........... 81 |====================================================== Apache HBase 2.2.3 Test: Random Write - Clients: 16 Rows Per Second > Higher Is Better Threadripper 3970X . 418033 |============================================ 16 Cores ........... 475673 |================================================== Memcached mcperf 1.5.10 Method: Replace Operations Per Second > Higher Is Better Threadripper 3970X . 47267.1 |=========================== 16 Cores ........... 84273.8 |================================================= Memcached mcperf 1.5.10 Method: Prepend Operations Per Second > Higher Is Better Threadripper 3970X . 46495.9 |========================== 16 Cores ........... 88961.4 |================================================= Memcached mcperf 1.5.10 Method: Append Operations Per Second > Higher Is Better Threadripper 3970X . 49940.5 |============================ 16 Cores ........... 85868.9 |================================================= Memcached mcperf 1.5.10 Method: Set Operations Per Second > Higher Is Better Threadripper 3970X . 45117.5 |============================= 16 Cores ........... 75424.0 |================================================= Memcached mcperf 1.5.10 Method: Add Operations Per Second > Higher Is Better Threadripper 3970X . 45047.0 |============================== 16 Cores ........... 72837.7 |================================================= Facebook RocksDB 6.3.6 Test: Random Fill Sync Op/s > Higher Is Better Threadripper 3970X . 29167 |=================================================== 16 Cores ........... 9106 |================ Apache Cassandra 3.11.4 Test: Reads Op/s > Higher Is Better Threadripper 3970X . 20360 |=================================================== 16 Cores ........... 3166 |======== Redis 5.0.5 Test: GET Requests Per Second > Higher Is Better Threadripper 3970X . 2005988.50 |========================================== 16 Cores ........... 2184449.11 |============================================== Redis 5.0.5 Test: LPOP Requests Per Second > Higher Is Better Threadripper 3970X . 2175519.08 |============================================== 16 Cores ........... 2097030.00 |============================================ Stress-NG 0.07.26 Test: Context Switching Bogo Ops/s > Higher Is Better Threadripper 3970X . 21267958.31 |============================================= 16 Cores ........... 2258922.93 |===== Stress-NG 0.07.26 Test: Lsearch Bogo Ops/s > Higher Is Better Threadripper 3970X . 434.81 |================================================== 16 Cores ........... 138.27 |================ Darktable 2.6.0 Test: Server Rack - Acceleration: CPU-only Seconds < Lower Is Better Threadripper 3970X . 0.120 |=================================================== 16 Cores ........... 0.097 |========================================= Darktable 2.6.0 Test: Boat - Acceleration: CPU-only Seconds < Lower Is Better Threadripper 3970X . 6.010 |=================================================== 16 Cores ........... 5.060 |=========================================== PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write TPS > Higher Is Better Threadripper 3970X . 11196.80 |========================================== 16 Cores ........... 12881.18 |================================================ Intel MPI Benchmarks 2019.3 Test: IMB-MPI1 Sendrecv Average Mbytes/sec > Higher Is Better Threadripper 3970X . 4955.00 |======================================= 16 Cores ........... 6263.73 |================================================= Intel MPI Benchmarks 2019.3 Test: IMB-MPI1 PingPong Average Mbytes/sec > Higher Is Better Threadripper 3970X . 5106.63 |=========================================== 16 Cores ........... 5828.12 |================================================= KeyDB 5.3.1 Ops/sec > Higher Is Better Threadripper 3970X . 119630.18 |================================== 16 Cores ........... 167550.10 |=============================================== Cpuminer-Opt 3.8.8.1 Algorithm: myr-gr kH/s - Hash Speed > Higher Is Better Threadripper 3970X . 13923.00 |=========================================== 16 Cores ........... 15369.09 |================================================ Cpuminer-Opt 3.8.8.1 Algorithm: deep kH/s - Hash Speed > Higher Is Better Threadripper 3970X . 29323.00 |================================================ 16 Cores ........... 6879.91 |=========== LuxCoreRender 2.2 Scene: Rainbow Colors and Prism M samples/sec > Higher Is Better Threadripper 3970X . 2.07 |============================================= 16 Cores ........... 2.40 |==================================================== LuxCoreRender 2.2 Scene: DLSC M samples/sec > Higher Is Better Threadripper 3970X . 2.51 |==================================================== 16 Cores ........... 2.30 |================================================ OSPray 1.8.5 Demo: Magnetic Reconnection - Renderer: Path Tracer FPS > Higher Is Better Threadripper 3970X . 444.44 |================================================== 16 Cores ........... 250.00 |============================ MKL-DNN DNNL 1.1 Harness: Deconvolution Batch deconv_3d - Data Type: f32 ms < Lower Is Better Threadripper 3970X . 4.91239 |================================================= 16 Cores ........... 3.69166 |===================================== Renaissance 0.10.0 Test: Savina Reactors.IO ms < Lower Is Better Threadripper 3970X . 13551.88 |========================================= 16 Cores ........... 15896.12 |================================================ Renaissance 0.10.0 Test: Apache Spark Bayes ms < Lower Is Better Threadripper 3970X . 2677.51 |========================================== 16 Cores ........... 3157.26 |================================================= CLOMP 3.3 Static OMP Speedup Speedup > Higher Is Better Threadripper 3970X . 0.71 |============ 16 Cores ........... 3.04 |==================================================== CloverLeaf Lagrangian-Eulerian Hydrodynamics Seconds < Lower Is Better Threadripper 3970X . 1.05 |=========================================== 16 Cores ........... 1.26 |==================================================== iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 64 Mbits/sec > Higher Is Better Threadripper 3970X . 54840 |=================================================== 16 Cores ........... 47216 |============================================ Sockperf 3.4 Test: Latency Under Load usec < Lower Is Better Threadripper 3970X . 13.23 |=================================================== 16 Cores ........... 10.78 |========================================== IOzone 3.465 Record Size: 4Kb - File Size: 512MB - Disk Test: Write Performance MB/s > Higher Is Better Threadripper 3970X . 1529.57 |================================================= 16 Cores ........... 1280.55 |========================================= Dbench 4.0 Client Count: 12 MB/s > Higher Is Better Threadripper 3970X . 665.37 |======================== 16 Cores ........... 1382.09 |================================================= Threaded I/O Tester 20170503 Test: Write - Size Per Thread: 32MB - Thread Count: 4 MB/s > Higher Is Better Threadripper 3970X . 10.55 |=================================================== 16 Cores ........... 10.41 |================================================== IOzone 3.465 Record Size: 1MB - File Size: 512MB - Disk Test: Write Performance MB/s > Higher Is Better Threadripper 3970X . 1554.84 |================================================= 16 Cores ........... 1335.37 |========================================== IOR 3.2.1 Write Test MB/s > Higher Is Better Threadripper 3970X . 182.76 |=============== 16 Cores ........... 622.53 |================================================== Dbench 4.0 Client Count: 12 MB/s > Higher Is Better Threadripper 3970X . 595.26 |==================== 16 Cores ........... 1466.07 |================================================= FS-Mark 3.3 Test: 5000 Files, 1MB Size, 4 Threads Files/s > Higher Is Better Threadripper 3970X . 485.9 |========================================= 16 Cores ........... 602.3 |=================================================== Flexible IO Tester Type: Random Write - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better Threadripper 3970X . 263667 |================================================ 16 Cores ........... 277000 |================================================== Flexible IO Tester 3.12 Type: Random Write - IO Engine: POSIX AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory MB/s > Higher Is Better Threadripper 3970X . 1030 |================================================== 16 Cores ........... 1068 |=================================================== 16 Cores ........... 1082 |==================================================== Flexible IO Tester 3.18 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better Threadripper 3970X . 269200 |================================================== 16 Cores ........... 245750 |============================================== Flexible IO Tester 3.18 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better Threadripper 3970X . 344533 |=========================================== 16 Cores ........... 397600 |================================================== Flexible IO Tester 3.18 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory MB/s > Higher Is Better Threadripper 3970X . 1346 |============================================= 16 Cores ........... 1553 |==================================================== Flexible IO Tester 3.18 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory IOPS > Higher Is Better Threadripper 3970X . 180633 |=========================== 16 Cores ........... 335000 |================================================== Flexible IO Tester 3.18 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory MB/s > Higher Is Better Threadripper 3970X . 706 |============================ 16 Cores ........... 1309 |====================================================