Core i9 10900K vs. Ryzen 9 3900X/3950X Linux Benchmarks
Tests fot a future article on Phoronix by Michael Larabel.
HTML result view exported from: https://openbenchmarking.org/result/2005283-PTS-BIGCOMPA49&obr_sgm=y&obr_swl=y&sro&grs.
ebizzy
CP2K Molecular Dynamics
Fayalite-FIST Data
ParaView
Test: Wavelet Volume - Resolution: 3840 x 2160
ParaView
Test: Wavelet Volume - Resolution: 3840 x 2160
Mlpack Benchmark
Benchmark: scikit_linearridgeregression
BLAKE2
Cpuminer-Opt
Algorithm: m7m
Parboil
Test: OpenMP CUTCP
GraphicsMagick
Operation: Swirl
Facebook RocksDB
Test: Random Fill Sync
Stress-NG
Test: Crypto
m-queens
Time To Solve
libgav1
Video Input: Summer Nature 1080p
toyBrot Fractal Generator
Implementation: TBB
toyBrot Fractal Generator
Implementation: C++ Threads
toyBrot Fractal Generator
Implementation: C++ Tasks
OSBench
Test: Create Processes
toyBrot Fractal Generator
Implementation: OpenMP
Rodinia
Test: OpenMP CFD Solver
Bork File Encrypter
File Encryption Time
C-Ray
Total Time - 4K, 16 Rays Per Pixel
DaCapo Benchmark
Java Test: Tradebeans
NAS Parallel Benchmarks
Test / Class: SP.B
oneDNN MKL-DNN
Harness: IP Batch 1D - Data Type: u8s8f32
Multichase Pointer Chaser
Test: 256MB Array, 256 Byte Stride
oneDNN MKL-DNN
Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32
FFmpeg
H.264 HD To NTSC DV
Stress-NG
Test: Vector Math
Zstd Compression
Compression Level: 3
Core-Latency
Average Latency Between CPU Cores
IPC_benchmark
Type: Unnamed Pipe - Message Bytes: 4096
IPC_benchmark
Type: FIFO Named Pipe - Message Bytes: 4096
N-Queens
Elapsed Time
LeelaChessZero
Backend: Eigen
Multichase Pointer Chaser
Test: 1GB Array, 256 Byte Stride
LuxCoreRender
Scene: DLSC
Smallpt
Global Illumination Renderer; 128 Samples
Blender
Blend File: Classroom - Compute: CPU-Only
Blender
Blend File: Pabellon Barcelona - Compute: CPU-Only
GraphicsMagick
Operation: Enhanced
Tungsten Renderer
Scene: Hair
Polyhedron Fortran Benchmarks
Benchmark: induct2
Hackbench
Count: 1 - Type: Process
libgav1
Video Input: Chimera 1080p 10-bit
Rodinia
Test: OpenMP Streamcluster
Multichase Pointer Chaser
Test: 1GB Array, 256 Byte Stride, 2 Threads
Redis
Test: GET
Blender
Blend File: Fishy Cat - Compute: CPU-Only
Primesieve
1e12 Prime Number Generation
Redis
Test: SET
LuxCoreRender
Scene: Rainbow Colors and Prism
libgav1
Video Input: Summer Nature 4K
GraphicsMagick
Operation: Noise-Gaussian
Appleseed
Scene: Disney Material
NeatBench
Acceleration: CPU
Tachyon
Total Time
Stockfish
Total Time
Basis Universal
Settings: UASTC Level 3
GraphicsMagick
Operation: Sharpen
CacheBench
Test: Read
Zstd Compression
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19
IndigoBench
Scene: Bedroom
ACES DGEMM
Sustained Floating-Point Rate
GnuPG
2GB File Encryption
Rodinia
Test: OpenMP LavaMD
oneDNN MKL-DNN
Harness: IP Batch 1D - Data Type: f32
Appleseed
Scene: Emily
Cpuminer-Opt
Algorithm: sha256t
GraphicsMagick
Operation: Resizing
Facebook RocksDB
Test: Random Read
Multichase Pointer Chaser
Test: 1GB Array, 256 Byte Stride, 4 Threads
Stress-NG
Test: CPU Stress
dav1d
Video Input: Chimera 1080p
Timed MPlayer Compilation
Time To Compile
Java SciMark
Computational Test: Monte Carlo
BRL-CAD
VGR Performance Metric
Polyhedron Fortran Benchmarks
Benchmark: linpk
Cpuminer-Opt
Algorithm: skein
7-Zip Compression
Compress Speed Test
Blender
Blend File: Barbershop - Compute: CPU-Only
Timed LLVM Compilation
Time To Compile
Hackbench
Count: 1 - Type: Thread
Renaissance
Test: Genetic Algorithm Using Jenetics + Futures
OSBench
Test: Launch Programs
Numenta Anomaly Benchmark
Detector: Bayesian Changepoint
NAS Parallel Benchmarks
Test / Class: FT.C
Coremark
CoreMark Size 666 - Iterations Per Second
Nettle
Test: poly1305-aes
Nettle
Test: chacha
Aircrack-ng
Cpuminer-Opt
Algorithm: lbry
Timed Linux Kernel Compilation
Time To Compile
Stress-NG
Test: Context Switching
John The Ripper
Test: Blowfish
Blender
Blend File: BMW27 - Compute: CPU-Only
Stress-NG
Test: Socket Activity
dav1d
Video Input: Summer Nature 1080p
John The Ripper
Test: MD5
Polyhedron Fortran Benchmarks
Benchmark: ac
oneDNN MKL-DNN
Harness: IP Batch All - Data Type: u8s8f32
Sysbench
Test: CPU
Cpuminer-Opt
Algorithm: deep
LeelaChessZero
Backend: BLAS
OpenSSL
RSA 4096-bit Performance
Embree
Binary: Pathtracer - Model: Crown
Embree
Binary: Pathtracer ISPC - Model: Crown
Stress-NG
Test: Matrix Math
Hierarchical INTegration
Test: FLOAT
Go Benchmarks
Test: json
Pennant
Test: sedovbig
Java Gradle Build
Gradle Build: Reactor
dav1d
Video Input: Chimera 1080p
YafaRay
Total Time For Sample Scene
Pennant
Test: leblancbig
Tungsten Renderer
Scene: Volumetric Caustic
dav1d
Video Input: Summer Nature 1080p
OpenVKL
Benchmark: vklBenchmark
NAMD
ATPase Simulation - 327,506 Atoms
Nettle
Test: aes256
rays1bench
Large Scene
Tungsten Renderer
Scene: Non-Exponential
Basis Universal
Settings: UASTC Level 2
Selenium
Benchmark: Maze Solver - Browser: Google Chrome
Facebook RocksDB
Test: Read While Writing
AOM AV1
Encoder Mode: Speed 6 Realtime
IndigoBench
Scene: Supercar
Zstd Compression
Compression Level: 19
ASKAP
Test: tConvolve MT - Gridding
asmFish
1024 Hash Memory, 26 Depth
Unigine Superposition
Resolution: 1920 x 1080 - Mode: Windowed - Quality: Low - Renderer: OpenGL
ASKAP
Test: tConvolve MPI - Gridding
Timed FFmpeg Compilation
Time To Compile
FFTW
Build: Float + SSE - Size: 2D FFT Size 4096
SQLite Speedtest
Timed Time - Size 1,000
AOM AV1
Encoder Mode: Speed 8 Realtime
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon
NWChem
Input: C240 Buckyball
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon Obj
Embree
Binary: Pathtracer - Model: Asian Dragon
PyPerformance
Benchmark: pickle_pure_python
Polyhedron Fortran Benchmarks
Benchmark: fatigue2
Appleseed
Scene: Material Tester
ASKAP
Test: tConvolve MPI - Degridding
Optcarrot
Optimized Benchmark
Embree
Binary: Pathtracer - Model: Asian Dragon Obj
ASKAP
Test: tConvolve MT - Degridding
Renaissance
Test: Random Forest
LeelaChessZero
Backend: Random
x264
H.264 Video Encoding
GROMACS
Water Benchmark
Timed HMMer Search
Pfam Database Search
Parallel BZIP2 Compression
256MB File Compression
TTSIOD 3D Renderer
Phong Rendering With Soft-Shadow Mapping
PlaidML
FP16: No - Mode: Inference - Network: VGG16 - Device: CPU
GIMP
Test: unsharp-mask
PyPerformance
Benchmark: float
GraphicsMagick
Operation: HWB Color Space
Basemark GPU
Renderer: OpenGL - Resolution: 3840 x 2160 - Graphics Preset: High
PyPerformance
Benchmark: go
PlaidML
FP16: No - Mode: Inference - Network: VGG19 - Device: CPU
Selenium
Benchmark: WASM collisionDetection - Browser: Google Chrome
Botan
Test: AES-256
OSBench
Test: Memory Allocations
Facebook RocksDB
Test: Sequential Fill
Selenium
Benchmark: StyleBench - Browser: Firefox
TSCP
AI Chess Performance
DeepSpeech
Acceleration: CPU
Selenium
Benchmark: Maze Solver - Browser: Firefox
NAS Parallel Benchmarks
Test / Class: MG.C
ASKAP
Test: tConvolve OpenMP - Degridding
PyPerformance
Benchmark: raytrace
LULESH
DaCapo Benchmark
Java Test: Tradesoap
GraphicsMagick
Operation: Rotate
PlaidML
FP16: No - Mode: Inference - Network: Inception V3 - Device: CPU
PyBench
Total For Average Test Times
Timed GDB GNU Debugger Compilation
Time To Compile
dav1d
Video Input: Chimera 1080p 10-bit
oneDNN MKL-DNN
Harness: Recurrent Neural Network Training - Data Type: f32
DaCapo Benchmark
Java Test: Jython
Renaissance
Test: Akka Unbalanced Cobwebbed Tree
GIMP
Test: auto-levels
Perl Benchmarks
Test: Pod2html
Polyhedron Fortran Benchmarks
Benchmark: air
PyPerformance
Benchmark: json_loads
Build2
Time To Compile
dav1d
Video Input: Chimera 1080p 10-bit
Polyhedron Fortran Benchmarks
Benchmark: aermod
PyPerformance
Benchmark: crypto_pyaes
Timed ImageMagick Compilation
Time To Compile
oneDNN MKL-DNN
Harness: Deconvolution Batch deconv_1d - Data Type: f32
Go Benchmarks
Test: garbage
GEGL
Operation: Rotate 90 Degrees
Polyhedron Fortran Benchmarks
Benchmark: rnflow
VP9 libvpx Encoding
Speed: Speed 5
Renaissance
Test: Apache Spark ALS
Sockperf
Test: Latency Ping Pong
Swet
Average
PyPerformance
Benchmark: django_template
x265
H.265 1080p Video Encoding
lzbench
Test: Brotli 0 - Process: Decompression
lzbench
Test: XZ 0 - Process: Compression
Selenium
Benchmark: WASM collisionDetection - Browser: Firefox
GEGL
Operation: Reflect
GEGL
Operation: Wavelet Blur
Tesseract
Resolution: 3840 x 2160
GEGL
Operation: Tile Glass
PyPerformance
Benchmark: 2to3
High Performance Conjugate Gradient
PyPerformance
Benchmark: regex_compile
G'MIC
Test: Plotting Isosurface Of A 3D Volume, 1000 Times
lzbench
Test: Brotli 2 - Process: Decompression
Timed Apache Compilation
Time To Compile
IPC_benchmark
Type: Unnamed Unix Domain Socket - Message Bytes: 4096
Selenium
Benchmark: WebXPRT - Browser: Firefox
Selenium
Benchmark: ARES-6 - Browser: Firefox
GEGL
Operation: Antialias
Botan
Test: Blowfish
Crypto++
Test: Integer + Elliptic Curve Public Key Algorithms
Gzip Compression
Linux Source Tree Archiving To .tar.gz
lzbench
Test: Crush 0 - Process: Decompression
NAS Parallel Benchmarks
Test / Class: BT.C
Numenta Anomaly Benchmark
Detector: Earthgecko Skyline
Selenium
Benchmark: ARES-6 - Browser: Google Chrome
PlaidML
FP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPU
Intel Open Image Denoise
Scene: Memorial
Fhourstones
Complex Connect-4 Solving
Numenta Anomaly Benchmark
Detector: Windowed Gaussian
Timed MrBayes Analysis
Primate Phylogeny Analysis
Numenta Anomaly Benchmark
Detector: Relative Entropy
Selenium
Benchmark: Jetstream - Browser: Google Chrome
lzbench
Test: Crush 0 - Process: Compression
AOBench
Size: 2048 x 2048 - Total Time
CacheBench
Test: Read / Modify / Write
lzbench
Test: Libdeflate 1 - Process: Decompression
CacheBench
Test: Write
Botan
Test: KASUMI
Sockperf
Test: Throughput
GEGL
Operation: Cartoon
oneDNN MKL-DNN
Harness: Deconvolution Batch deconv_3d - Data Type: f32
lzbench
Test: Zstd 1 - Process: Decompression
Selenium
Benchmark: PSPDFKit WASM - Browser: Firefox
Botan
Test: Twofish
oneDNN MKL-DNN
Harness: IP Batch All - Data Type: f32
Algebraic Multi-Grid Benchmark
Minion
Benchmark: Quasigroup
Crypto++
Test: Keyed Algorithms
Tungsten Renderer
Scene: Water Caustic
Himeno Benchmark
Poisson Pressure Solver
AOM AV1
Encoder Mode: Speed 6 Two-Pass
GEGL
Operation: Color Enhance
Botan
Test: CAST-256
Selenium
Benchmark: Octane - Browser: Firefox
PyPerformance
Benchmark: nbody
ASKAP
Test: tConvolve OpenMP - Gridding
XZ Compression
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9
Timed PHP Compilation
Time To Compile
IPC_benchmark
Type: TCP Socket - Message Bytes: 4096
GEGL
Operation: Crop
Polyhedron Fortran Benchmarks
Benchmark: mdbx
OSBench
Test: Create Files
Mlpack Benchmark
Benchmark: scikit_svm
GIMP
Test: resize
Selenium
Benchmark: Jetstream 2 - Browser: Firefox
GIMP
Test: rotate
Multichase Pointer Chaser
Test: 4MB Array, 64 Byte Stride
Java SciMark
Computational Test: Fast Fourier Transform
Timed GCC Compilation
Time To Compile
Basis Universal
Settings: UASTC Level 0
Mlpack Benchmark
Benchmark: scikit_qda
Cython benchmark
dav1d
Video Input: Summer Nature 4K
Selenium
Benchmark: Octane - Browser: Google Chrome
Crypto++
Test: All Algorithms
Numpy Benchmark
lzbench
Test: Zstd 8 - Process: Decompression
Unigine Valley
Resolution: 1920 x 1080 - Mode: Windowed - Renderer: OpenGL
AOM AV1
Encoder Mode: Speed 4 Two-Pass
lzbench
Test: Zstd 1 - Process: Compression
Git
Time To Complete Common Git Commands
Selenium
Benchmark: PSPDFKit WASM - Browser: Google Chrome
Selenium
Benchmark: Jetstream - Browser: Firefox
Parboil
Test: OpenMP Stencil
ctx_clock
Context Switch Time
Selenium
Benchmark: Jetstream 2 - Browser: Google Chrome
Polyhedron Fortran Benchmarks
Benchmark: test_fpu2
Polyhedron Fortran Benchmarks
Benchmark: mp_prop_design
Polyhedron Fortran Benchmarks
Benchmark: gas_dyn2
Nebular Empirical Analysis Tool
FFTW
Build: Stock - Size: 2D FFT Size 4096
Bullet Physics Engine
Test: 136 Ragdolls
Radiance Benchmark
Test: Serial
oneDNN MKL-DNN
Harness: Recurrent Neural Network Inference - Data Type: f32
libjpeg-turbo tjbench
Test: Decompression Throughput
FLAC Audio Encoding
WAV To FLAC
Radiance Benchmark
Test: SMP Parallel
Polyhedron Fortran Benchmarks
Benchmark: doduc
lzbench
Test: Libdeflate 1 - Process: Compression
Java SciMark
Computational Test: Dense LU Matrix Factorization
Basis Universal
Settings: ETC1S
VP9 libvpx Encoding
Speed: Speed 0
lzbench
Test: Zstd 8 - Process: Compression
Polyhedron Fortran Benchmarks
Benchmark: protein
Minion
Benchmark: Graceful
Bullet Physics Engine
Test: 1000 Stack
Scikit-Learn
AOM AV1
Encoder Mode: Speed 0 Two-Pass
lzbench
Test: Brotli 2 - Process: Compression
Unigine Superposition
Resolution: 1920 x 1080 - Mode: Windowed - Quality: High - Renderer: OpenGL
Polyhedron Fortran Benchmarks
Benchmark: channel2
Basis Universal
Settings: UASTC Level 2 + RDO Post-Processing
dav1d
Video Input: Summer Nature 4K
C-Blosc
Compressor: blosclz
Selenium
Benchmark: Kraken - Browser: Google Chrome
Crypto++
Test: Unkeyed Algorithms
Selenium
Benchmark: Speedometer - Browser: Firefox
GEGL
Operation: Scale
Perl Benchmarks
Test: Interpreter
Renaissance
Test: Twitter HTTP Requests
Bullet Physics Engine
Test: 1000 Convex
Minion
Benchmark: Solitaire
PlaidML
FP16: No - Mode: Inference - Network: Mobilenet - Device: CPU
Mlpack Benchmark
Benchmark: scikit_ica
Numenta Anomaly Benchmark
Detector: EXPoSE
Unigine Superposition
Resolution: 1920 x 1080 - Mode: Windowed - Quality: Medium - Renderer: OpenGL
Java SciMark
Computational Test: Jacobi Successive Over-Relaxation
PlaidML
FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU
ParaView
Test: Many Spheres - Resolution: 3840 x 2160
ParaView
Test: Many Spheres - Resolution: 3840 x 2160
Bullet Physics Engine
Test: 3000 Fall
Parboil
Test: OpenMP LBM
Selenium
Benchmark: WASM imageConvolute - Browser: Firefox
Bullet Physics Engine
Test: Prim Trimesh
Xonotic
Resolution: 3840 x 2160 - Effects Quality: Ultimate
Bullet Physics Engine
Test: Raytests
Bullet Physics Engine
Test: Convex Trimesh
LibreOffice
Test: 20 Documents To PDF
ParaView
Test: Wavelet Contour - Resolution: 3840 x 2160
ParaView
Test: Wavelet Contour - Resolution: 3840 x 2160
RawTherapee
Total Benchmark Time
Renaissance
Test: Scala Dotty
Selenium
Benchmark: CanvasMark - Browser: Firefox
Basemark GPU
Renderer: Vulkan - Resolution: 3840 x 2160 - Graphics Preset: Medium
lzbench
Test: Brotli 0 - Process: Compression
Selenium
Benchmark: WebXPRT - Browser: Google Chrome
Selenium
Benchmark: Speedometer - Browser: Google Chrome
Xsbench
Xonotic
Resolution: 3840 x 2160 - Effects Quality: Low
Java SciMark
Computational Test: Composite
Selenium
Benchmark: Kraken - Browser: Firefox
Nettle
Test: sha512
Selenium
Benchmark: WASM imageConvolute - Browser: Google Chrome
lzbench
Test: XZ 0 - Process: Decompression
Basemark GPU
Renderer: Vulkan - Resolution: 3840 x 2160 - Graphics Preset: High
Java SciMark
Computational Test: Sparse Matrix Multiply
Xonotic
Resolution: 3840 x 2160 - Effects Quality: Ultra
Unigine Heaven
Resolution: 1920 x 1080 - Mode: Windowed - Renderer: OpenGL
Xonotic
Resolution: 3840 x 2160 - Effects Quality: High
Selenium
Benchmark: CanvasMark - Browser: Google Chrome
LAME MP3 Encoding
WAV To MP3
Selenium
Benchmark: StyleBench - Browser: Google Chrome
Unigine Superposition
Resolution: 1920 x 1080 - Mode: Windowed - Quality: Ultra - Renderer: OpenGL
OSBench
Test: Create Threads
Selenium
Benchmark: Basemark - Browser: Google Chrome
Selenium
Benchmark: MotionMark - Browser: Firefox
Selenium
Benchmark: Basemark - Browser: Firefox
Facebook RocksDB
Test: Random Fill
Stress-NG
Test: CPU Cache
Stress-NG
Test: Atomic
Hackbench
Count: 32 - Type: Process
oneDNN MKL-DNN
Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32
Renaissance
Test: In-Memory Database Shootout
Renaissance
Test: Apache Spark PageRank
Renaissance
Test: Savina Reactors.IO
Renaissance
Test: Apache Spark Bayes
DaCapo Benchmark
Java Test: H2
libgav1
Video Input: Chimera 1080p
Parboil
Test: OpenMP MRI Gridding
ET: Legacy
Renderer: Renderer2 - Resolution: 3840 x 2160
IOR
Read Test
IOR
Write Test
Geometric Mean Of All Test Results
Result Composite - Core i9 10900K vs. Ryzen 9 3900X/3950X Linux Benchmarks
Number Of First Place Finishes
Wins - 381 Tests
Number Of Last Place Finishes
Losses - 381 Tests
Phoronix Test Suite v10.8.5