AMD Threadripper 3960X vs. Intel Core i9 10980XE
Intel Core i9-10980XE benchmarks against AMD Ryzen Threadripper 3960X. Benchmarks by Michael Larabel on Ubuntu 19.10 Linux.
HTML result view exported from: https://openbenchmarking.org/result/1911270-HU-THREADRIP09&obr_sgm=y&obr_swl=y&grs.
Parboil
Test: OpenMP MRI Gridding
VP9 libvpx Encoding
vpxenc VP9 1080p Video Encode
Hackbench
Count: 8 - Type: Thread
MKL-DNN DNNL
Harness: IP Batch 1D - Data Type: f32
CP2K Molecular Dynamics
Fayalite-FIST Data
HPC Challenge
Test / Class: Random Ring Bandwidth
dav1d
Video Input: Summer Nature 1080p
MBW
Test: Memory Copy - Array Size: 4096 MiB
Hackbench
Count: 8 - Type: Process
ParaView
Test: Wavelet Contour - Resolution: 2560 x 1440
ParaView
Test: Wavelet Contour - Resolution: 2560 x 1440
MBW
Test: Memory Copy - Array Size: 1024 MiB
NAS Parallel Benchmarks
Test / Class: SP.B
dav1d
Video Input: Chimera 1080p
DaCapo Benchmark
Java Test: Tradebeans
MBW
Test: Memory Copy - Array Size: 512 MiB
m-queens
Time To Solve
Rust Prime Benchmark
Prime Number Test To 200,000,000
MKL-DNN DNNL
Harness: Convolution Batch conv_3d - Data Type: f32
GraphicsMagick
Operation: Swirl
BRL-CAD
VGR Performance Metric
N-Queens
Elapsed Time
Hackbench
Count: 4 - Type: Process
Hackbench
Count: 16 - Type: Thread
Hackbench
Count: 4 - Type: Thread
Hackbench
Count: 16 - Type: Process
dav1d
Video Input: Chimera 1080p 10-bit
DaCapo Benchmark
Java Test: H2
Stress-NG
Test: Crypto
C-Ray
Total Time - 4K, 16 Rays Per Pixel
Blender
Blend File: Classroom - Compute: CPU-Only
libgav1
Video Input: Chimera 1080p
Tungsten Renderer
Scene: Volumetric Caustic
libgav1
Video Input: Summer Nature 1080p
Parboil
Test: OpenMP CUTCP
Blender
Blend File: Pabellon Barcelona - Compute: CPU-Only
dav1d
Video Input: Summer Nature 4K
Blender
Blend File: Fishy Cat - Compute: CPU-Only
FFTW
Build: Float + SSE - Size: 2D FFT Size 512
libgav1
Video Input: Summer Nature 4K
Coremark
CoreMark Size 666 - Iterations Per Second
Timed MrBayes Analysis
Primate Phylogeny Analysis
SVT-AV1
Encoder Mode: Enc Mode 8 - Input: 1080p
GROMACS
Water Benchmark
Pennant
Test: leblancbig
Pennant
Test: sedovbig
FFTW
Build: Float + SSE - Size: 2D FFT Size 1024
LeelaChessZero
Backend: Random
Blender
Blend File: BMW27 - Compute: CPU-Only
GraphicsMagick
Operation: Sharpen
libgav1
Video Input: Chimera 1080p 10-bit
HPC Challenge
Test / Class: Random Ring Latency
Smallpt
Global Illumination Renderer; 128 Samples
Blender
Blend File: Barbershop - Compute: CPU-Only
Embree
Binary: Pathtracer - Model: Crown
Stress-NG
Test: Glibc Qsort Data Sorting
OSPray
Demo: NASA Streamlines - Renderer: Path Tracer
GraphicsMagick
Operation: HWB Color Space
x265
H.265 1080p Video Encoding
Timed GCC Compilation
Time To Compile
7-Zip Compression
Compress Speed Test
GraphicsMagick
Operation: Enhanced
SVT-AV1
Encoder Mode: Enc Mode 4 - Input: 1080p
GraphicsMagick
Operation: Noise-Gaussian
Stockfish
Total Time
OSPray
Demo: XFrog Forest - Renderer: SciVis
HPC Challenge
Test / Class: G-Ffte
HPC Challenge
Test / Class: G-Ffte
NAS Parallel Benchmarks
Test / Class: MG.C
FFTW
Build: Float + SSE - Size: 2D FFT Size 256
Selenium
Benchmark: Maze Solver - Browser: Google Chrome
Timed Linux Kernel Compilation
Time To Compile
RAMspeed SMP
Type: Triad - Benchmark: Integer
GraphicsMagick
Operation: Resizing
Timed LLVM Compilation
Time To Compile
Timed ImageMagick Compilation
Time To Compile
Rodinia
Test: OpenMP LavaMD
Tungsten Renderer
Scene: Hair
Selenium
Benchmark: Maze Solver - Browser: Firefox
Chaos Group V-RAY
Mode: CPU
RAMspeed SMP
Type: Add - Benchmark: Floating Point
FFTW
Build: Float + SSE - Size: 2D FFT Size 2048
Tinymembench
Standard Memset
Facebook RocksDB
Test: Random Read
IndigoBench
Scene: Bedroom
x264
H.264 Video Encoding
OSPray
Demo: XFrog Forest - Renderer: Path Tracer
Stress-NG
Test: Socket Activity
Tachyon
Total Time
Go Benchmarks
Test: json
RAMspeed SMP
Type: Add - Benchmark: Integer
Darktable
Test: Boat - Acceleration: CPU-only
Facebook RocksDB
Test: Sequential Fill
NeatBench
Acceleration: CPU
MKL-DNN DNNL
Harness: Deconvolution Batch deconv_1d - Data Type: f32
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB
SPECjbb 2015
SPECjbb2015-Composite critical-jOPS
MKL-DNN DNNL
Harness: Recurrent Neural Network Training - Data Type: f32
OSPray
Demo: NASA Streamlines - Renderer: SciVis
Apache Cassandra
Test: Writes
FFTW
Build: Float + SSE - Size: 2D FFT Size 128
DaCapo Benchmark
Java Test: Tradesoap
Embree
Binary: Pathtracer - Model: Asian Dragon
Stress-NG
Test: Vector Math
PostgreSQL pgbench
Scaling: Buffer Test - Test: Normal Load - Mode: Read Only
Embree
Binary: Pathtracer - Model: Asian Dragon Obj
HPC Challenge
Test / Class: EP-STREAM Triad
Appleseed
Scene: Emily
asmFish
1024 Hash Memory, 26 Depth
OSBench
Test: Create Threads
RAMspeed SMP
Type: Average - Benchmark: Floating Point
ASKAP
Test: tConvolve MT - Degridding
ASKAP
Test: tConvolve MPI - Degridding
Appleseed
Scene: Disney Material
Renaissance
Test: In-Memory Database Shootout
OSPray
Demo: San Miguel - Renderer: SciVis
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 512 MiB
Bork File Encrypter
File Encryption Time
SPECjbb 2015
SPECjbb2015-Composite max-jOPS
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB
Embree
Binary: Pathtracer ISPC - Model: Crown
glibc bench
Benchmark: sqrt
Java SciMark
Computational Test: Composite
GNU Octave Benchmark
Build2
Time To Compile
OSBench
Test: Launch Programs
NAMD
ATPase Simulation - 327,506 Atoms
GIMP
Test: rotate
HPC Challenge
Test / Class: G-Random Access
Optcarrot
Optimized Benchmark
ParaView
Test: Wavelet Volume - Resolution: 2560 x 1440
ParaView
Test: Wavelet Volume - Resolution: 2560 x 1440
Stress-NG
Test: CPU Stress
CacheBench
Test: Read
RAMspeed SMP
Type: Scale - Benchmark: Integer
FFTW
Build: Stock - Size: 2D FFT Size 2048
MBW
Test: Memory Copy, Fixed Block Size - Array Size: 128 MiB
Darktable
Test: Masskrug - Acceleration: CPU-only
ASKAP
Test: tConvolve OpenMP - Degridding
RAMspeed SMP
Type: Scale - Benchmark: Floating Point
IndigoBench
Scene: Supercar
Go Benchmarks
Test: build
GIMP
Test: resize
RAMspeed SMP
Type: Triad - Benchmark: Floating Point
Stress-NG
Test: Forking
SVT-VP9
Tuning: Visual Quality Optimized - Input: Bosphorus 1080p
LAME MP3 Encoding
WAV To MP3
Selenium
Benchmark: StyleBench - Browser: Firefox
Rodinia
Test: OpenMP CFD Solver
Xsbench
NAS Parallel Benchmarks
Test / Class: BT.C
NGINX Benchmark
Static Web Page Serving
Stream
Type: Copy
RAMspeed SMP
Type: Copy - Benchmark: Integer
Opus Codec Encoding
WAV, Opus Encode, Opus Decode
OpenCV Benchmark
glibc bench
Benchmark: ffs
RAMspeed SMP
Type: Average - Benchmark: Integer
OSBench
Test: Create Processes
glibc bench
Benchmark: ffsll
FFTW
Build: Stock - Size: 1D FFT Size 32
SVT-VP9
Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p
ctx_clock
Context Switch Time
MKL-DNN DNNL
Harness: Deconvolution Batch deconv_all - Data Type: f32
FFTW
Build: Stock - Size: 2D FFT Size 64
CppPerformanceBenchmarks
Test: Ctype
glibc bench
Benchmark: pthread_once
QMCPACK
Selenium
Benchmark: PSPDFKit WASM - Browser: Google Chrome
FFTW
Build: Float + SSE - Size: 2D FFT Size 4096
rav1e
1080p To AV1 Video Encode
SQLite
Threads / Copies: 1
RAMspeed SMP
Type: Copy - Benchmark: Floating Point
Stress-NG
Test: Context Switching
Selenium
Benchmark: PSPDFKit WASM - Browser: Firefox
FFTW
Build: Stock - Size: 2D FFT Size 4096
glibc bench
Benchmark: modf
Sunflow Rendering System
Global Illumination + Image Synthesis
Timed PHP Compilation
Time To Compile
FLAC Audio Encoding
WAV To FLAC
GIMP
Test: unsharp-mask
Selenium
Benchmark: Speedometer - Browser: Firefox
MBW
Test: Memory Copy - Array Size: 128 MiB
ASKAP
Test: tConvolve MT - Gridding
NAS Parallel Benchmarks
Test / Class: LU.C
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon Obj
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon
Tensorflow
Build: Cifar10
FFTW
Build: Stock - Size: 1D FFT Size 256
Rust Mandelbrot
Time To Complete Serial/Parallel Mandelbrot
Renaissance
Test: Apache Spark PageRank
ASKAP
Test: tConvolve MPI - Gridding
Selenium
Benchmark: Speedometer - Browser: Google Chrome
CacheBench
Test: Read / Modify / Write
PHPBench
PHP Benchmark Suite
OSPray
Demo: Magnetic Reconnection - Renderer: SciVis
Darktable
Test: Server Room - Acceleration: CPU-only
ACES DGEMM
Sustained Floating-Point Rate
FFTW
Build: Stock - Size: 1D FFT Size 128
MKL-DNN DNNL
Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32
Facebook RocksDB
Test: Read While Writing
Bullet Physics Engine
Test: 3000 Fall
Radiance Benchmark
Test: SMP Parallel
OSPray
Demo: San Miguel - Renderer: Path Tracer
CppPerformanceBenchmarks
Test: Function Objects
Himeno Benchmark
Poisson Pressure Solver
FFTW
Build: Stock - Size: 2D FFT Size 512
CppPerformanceBenchmarks
Test: Random Numbers
GIMP
Test: auto-levels
FFTW
Build: Float + SSE - Size: 1D FFT Size 1024
FFTW
Build: Stock - Size: 2D FFT Size 1024
Bullet Physics Engine
Test: 136 Ragdolls
MKL-DNN DNNL
Harness: Convolution Batch conv_alexnet - Data Type: f32
Hackbench
Count: 1 - Type: Process
FFTW
Build: Stock - Size: 2D FFT Size 256
Minion
Benchmark: Quasigroup
Selenium
Benchmark: ARES-6 - Browser: Firefox
Selenium
Benchmark: Jetstream - Browser: Firefox
FFTW
Build: Float + SSE - Size: 1D FFT Size 32
dcraw
RAW To PPM Image Conversion
Git
Time To Complete Common Git Commands
Tungsten Renderer
Scene: Water Caustic
Minion
Benchmark: Solitaire
Parboil
Test: OpenMP Stencil
NAS Parallel Benchmarks
Test / Class: EP.D
GraphicsMagick
Operation: Rotate
Zstd Compression
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19
Intel Open Image Denoise
Scene: Memorial
MKL-DNN DNNL
Harness: Deconvolution Batch deconv_3d - Data Type: f32
TTSIOD 3D Renderer
Phong Rendering With Soft-Shadow Mapping
miniFE
Problem Size: Small
Tinymembench
Standard Memcpy
Radiance Benchmark
Test: Serial
Renaissance
Test: Apache Spark ALS
OSBench
Test: Memory Allocations
glibc bench
Benchmark: log2
Darktable
Test: Server Rack - Acceleration: CPU-only
MKL-DNN DNNL
Harness: Convolution Batch conv_all - Data Type: f32
FFTW
Build: Float + SSE - Size: 1D FFT Size 128
Bullet Physics Engine
Test: 1000 Stack
Stream
Type: Triad
FFTW
Build: Float + SSE - Size: 1D FFT Size 2048
Hackbench
Count: 1 - Type: Thread
Stream
Type: Add
FFTW
Build: Float + SSE - Size: 1D FFT Size 512
Selenium
Benchmark: Octane - Browser: Firefox
Bullet Physics Engine
Test: Prim Trimesh
DaCapo Benchmark
Java Test: Jython
Minion
Benchmark: Graceful
HPC Challenge
Test / Class: EP-DGEMM
Hackbench
Count: 2 - Type: Thread
Selenium
Benchmark: ARES-6 - Browser: Google Chrome
HPC Challenge
Test / Class: Max Ping Pong Bandwidth
LibreOffice
Test: 20 Documents To PDF
Stream
Type: Scale
PyBench
Total For Average Test Times
OSBench
Test: Create Files
MKL-DNN DNNL
Harness: IP Batch All - Data Type: f32
Apache Benchmark
Static Web Page Serving
CppPerformanceBenchmarks
Test: Atol
Bullet Physics Engine
Test: Convex Trimesh
glibc bench
Benchmark: sincos
SciMark
Computational Test: Composite
CacheBench
Test: Write
CppPerformanceBenchmarks
Test: Stepanov Abstraction
Bullet Physics Engine
Test: Raytests
Appleseed
Scene: Material Tester
glibc bench
Benchmark: tanh
HPC Challenge
Test / Class: G-Ptrans
Gzip Compression
Linux Source Tree Archiving To .tar.gz
TSCP
AI Chess Performance
Selenium
Benchmark: Jetstream - Browser: Google Chrome
Stress-NG
Test: Semaphores
SVT-AV1
Encoder Mode: Enc Mode 0 - Input: 1080p
glibc bench
Benchmark: cos
Stress-NG
Test: Matrix Math
Renaissance
Test: Apache Spark Bayes
NAS Parallel Benchmarks
Test / Class: FT.C
LuxCoreRender
Scene: DLSC
eSpeak Speech Engine
Text-To-Speech Synthesis
glibc bench
Benchmark: sin
Go Benchmarks
Test: http
Selenium
Benchmark: StyleBench - Browser: Google Chrome
Bullet Physics Engine
Test: 1000 Convex
Renaissance
Test: Akka Unbalanced Cobwebbed Tree
BYTE Unix Benchmark
Computational Test: Dhrystone 2
Hackbench
Count: 2 - Type: Process
FFTW
Build: Float + SSE - Size: 1D FFT Size 4096
Selenium
Benchmark: Octane - Browser: Google Chrome
FFTW
Build: Stock - Size: 2D FFT Size 128
Go Benchmarks
Test: garbage
CppPerformanceBenchmarks
Test: Stepanov Vector
HPC Challenge
Test / Class: G-HPL
Crafty
Elapsed Time
glibc bench
Benchmark: exp
glibc bench
Benchmark: atanh
glibc bench
Benchmark: asinh
LZMA Compression
256MB File Compression
FFTW
Build: Stock - Size: 1D FFT Size 2048
Tesseract OCR
Time To OCR 7 Images
CppPerformanceBenchmarks
Test: Math Library
XZ Compression
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9
glibc bench
Benchmark: sinh
FFTW
Build: Stock - Size: 1D FFT Size 4096
FFTW
Build: Stock - Size: 1D FFT Size 1024
ParaView
Test: Many Spheres - Resolution: 2560 x 1440
ParaView
Test: Many Spheres - Resolution: 2560 x 1440
SQLite Speedtest
Timed Time - Size 1,000
FFTW
Build: Stock - Size: 1D FFT Size 512
Apache Siege
Concurrent Users: 200
ASKAP
Test: tConvolve OpenMP - Gridding
Apache Cassandra
Test: Mixed 1:3
Apache Cassandra
Test: Reads
Stress-NG
Test: Glibc C String Functions
Redis
Test: SET
Redis
Test: GET
Memtier_benchmark
Protocol: Redis
PostgreSQL pgbench
Scaling: Buffer Test - Test: Normal Load - Mode: Read Write
Hackbench
Count: 32 - Type: Process
Tungsten Renderer
Scene: Non-Exponential
POV-Ray
Trace Time
LuxCoreRender
Scene: Rainbow Colors and Prism
SVT-VP9
Tuning: VMAF Optimized - Input: Bosphorus 1080p
OSPray
Demo: Magnetic Reconnection - Renderer: Path Tracer
Renaissance
Test: Savina Reactors.IO
FFTW
Build: Float + SSE - Size: 1D FFT Size 256
FFTW
Build: Float + SSE - Size: 2D FFT Size 64
FFTW
Build: Float + SSE - Size: 2D FFT Size 32
FFTW
Build: Float + SSE - Size: 1D FFT Size 64
FFTW
Build: Stock - Size: 2D FFT Size 32
FFTW
Build: Stock - Size: 1D FFT Size 64
LeelaChessZero
Backend: BLAS
NAS Parallel Benchmarks
Test / Class: EP.C
SQLite
Threads / Copies: 8
Geometric Mean Of All Test Results
Result Composite - AMD Threadripper 3960X vs. Intel Core i9 10980XE
Number Of First Place Finishes
Wins - 334 Tests
Number Of Last Place Finishes
Losses - 334 Tests
Phoronix Test Suite v10.8.4