Ryzen 9 5900X / Ryzen 9 5950X Linux Performance
AMD Ryzen 9 5900X initial Linux benchmarks by Michael Larabel for a future article preparation for 5 November launch day.
HTML result view exported from: https://openbenchmarking.org/result/2011040-FI-AMDRYZEN976&grs&rdt.
ACES DGEMM
Sustained Floating-Point Rate
Aircrack-ng
oneDNN
Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU
John The Ripper
Test: MD5
OpenVINO
Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU
OpenVINO
Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU
Intel Open Image Denoise
Scene: Memorial
oneDNN
Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU
oneDNN
Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU
OpenVINO
Model: Person Detection 0106 FP16 - Device: CPU
OpenVINO
Model: Person Detection 0106 FP32 - Device: CPU
OpenVINO
Model: Face Detection 0106 FP16 - Device: CPU
OpenVINO
Model: Face Detection 0106 FP32 - Device: CPU
OpenVKL
Benchmark: vklBenchmark
OpenSSL
RSA 4096-bit Performance
John The Ripper
Test: Blowfish
TSCP
AI Chess Performance
Embree
Binary: Pathtracer ISPC - Model: Crown
BRL-CAD
VGR Performance Metric
N-Queens
Elapsed Time
Embree
Binary: Pathtracer - Model: Crown
Chaos Group V-RAY
Mode: CPU
Kvazaar
Video Input: Bosphorus 4K - Video Preset: Slow
SciMark
Computational Test: Dense LU Matrix Factorization
Kvazaar
Video Input: Bosphorus 4K - Video Preset: Medium
m-queens
Time To Solve
SVT-AV1
Encoder Mode: Enc Mode 8 - Input: 1080p
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon
TTSIOD 3D Renderer
Phong Rendering With Soft-Shadow Mapping
Blender
Blend File: Fishy Cat - Compute: CPU-Only
GraphicsMagick
Operation: Enhanced
asmFish
1024 Hash Memory, 26 Depth
Tachyon
Total Time
YafaRay
Total Time For Sample Scene
Rodinia
Test: OpenMP LavaMD
Stockfish
Total Time
oneDNN
Harness: IP Batch 1D - Data Type: f32 - Engine: CPU
SVT-AV1
Encoder Mode: Enc Mode 4 - Input: 1080p
Smallpt
Global Illumination Renderer; 128 Samples
Tungsten Renderer
Scene: Volumetric Caustic
Embree
Binary: Pathtracer - Model: Asian Dragon
Blender
Blend File: Classroom - Compute: CPU-Only
LuxCoreRender
Scene: Rainbow Colors and Prism
ASTC Encoder
Preset: Exhaustive
Kvazaar
Video Input: Bosphorus 4K - Video Preset: Ultra Fast
Basis Universal
Settings: UASTC Level 3
Timed MPlayer Compilation
Time To Compile
ASTC Encoder
Preset: Thorough
Tungsten Renderer
Scene: Non-Exponential
Kvazaar
Video Input: Bosphorus 4K - Video Preset: Very Fast
LuxCoreRender
Scene: DLSC
Kvazaar
Video Input: Bosphorus 1080p - Video Preset: Slow
Blender
Blend File: BMW27 - Compute: CPU-Only
TensorFlow Lite
Model: Inception ResNet V2
Kvazaar
Video Input: Bosphorus 1080p - Video Preset: Medium
LAMMPS Molecular Dynamics Simulator
Model: Rhodopsin Protein
TensorFlow Lite
Model: Mobilenet Float
TensorFlow Lite
Model: Inception V4
NAMD
ATPase Simulation - 327,506 Atoms
POV-Ray
Trace Time
x265
Video Input: Bosphorus 4K
oneDNN
Harness: IP Batch All - Data Type: f32 - Engine: CPU
C-Ray
Total Time - 4K, 16 Rays Per Pixel
GraphicsMagick
Operation: Resizing
Timed Linux Kernel Compilation
Time To Compile
TensorFlow Lite
Model: SqueezeNet
Rodinia
Test: OpenMP CFD Solver
Tungsten Renderer
Scene: Hair
Timed LLVM Compilation
Time To Compile
Numenta Anomaly Benchmark
Detector: Earthgecko Skyline
TensorFlow Lite
Model: Mobilenet Quant
Timed FFmpeg Compilation
Time To Compile
Appleseed
Scene: Emily
Basis Universal
Settings: UASTC Level 2
7-Zip Compression
Compress Speed Test
x264
H.264 Video Encoding
SVT-VP9
Tuning: Visual Quality Optimized - Input: Bosphorus 1080p
Appleseed
Scene: Disney Material
Sunflow Rendering System
Global Illumination + Image Synthesis
Polyhedron Fortran Benchmarks
Benchmark: linpk
Appleseed
Scene: Material Tester
NeatBench
Acceleration: CPU
Kvazaar
Video Input: Bosphorus 1080p - Video Preset: Ultra Fast
Kvazaar
Video Input: Bosphorus 1080p - Video Preset: Very Fast
GROMACS
Water Benchmark
Numenta Anomaly Benchmark
Detector: Relative Entropy
Timed ImageMagick Compilation
Time To Compile
Mobile Neural Network
Model: inception-v3
Mobile Neural Network
Model: SqueezeNetV1.0
Numenta Anomaly Benchmark
Detector: Windowed Gaussian
SciMark
Computational Test: Fast Fourier Transform
libavif avifenc
Encoder Speed: 0
OCRMyPDF
Processing 60 Page PDF Document
NCNN
Target: CPU - Model: mobilenet
libavif avifenc
Encoder Speed: 2
SVT-VP9
Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p
dav1d
Video Input: Chimera 1080p
Timed PHP Compilation
Time To Compile
Mobile Neural Network
Model: MobileNetV2_224
TensorFlow Lite
Model: NASNet Mobile
dav1d
Video Input: Summer Nature 4K
Mobile Neural Network
Model: resnet-v2-50
NCNN
Target: CPU - Model: efficientnet-b0
DaCapo Benchmark
Java Test: Tradesoap
Numenta Anomaly Benchmark
Detector: Bayesian Changepoint
Rodinia
Test: OpenMP Leukocyte
Nebular Empirical Analysis Tool
NCNN
Target: CPU-v2-v2 - Model: mobilenet-v2
LibRaw
Post-Processing Benchmark
NCNN
Target: CPU - Model: squeezenet
Zstd Compression
Compression Level: 19
AOM AV1
Encoder Mode: Speed 8 Realtime
Zstd Compression
Compression Level: 3
AOM AV1
Encoder Mode: Speed 4 Two-Pass
PlaidML
FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU
PlaidML
FP16: No - Mode: Inference - Network: Inception V3 - Device: CPU
NCNN
Target: CPU - Model: yolov4-tiny
NCNN
Target: CPU - Model: resnet50
Tungsten Renderer
Scene: Water Caustic
AOM AV1
Encoder Mode: Speed 6 Two-Pass
NCNN
Target: CPU-v3-v3 - Model: mobilenet-v3
Timed GDB GNU Debugger Compilation
Time To Compile
AOM AV1
Encoder Mode: Speed 6 Realtime
WireGuard + Linux Networking Stack Stress Test
PlaidML
FP16: No - Mode: Inference - Network: Mobilenet - Device: CPU
Perl Benchmarks
Test: Pod2html
Radiance Benchmark
Test: Serial
Basis Universal
Settings: UASTC Level 0
Numpy Benchmark
DaCapo Benchmark
Java Test: Jython
Timed Apache Compilation
Time To Compile
PyPerformance
Benchmark: raytrace
NCNN
Target: CPU - Model: googlenet
Radiance Benchmark
Test: SMP Parallel
Timed MAFFT Alignment
Multiple Sequence Alignment - LSU RNA
VP9 libvpx Encoding
Speed: Speed 5
SciMark
Computational Test: Composite
SQLite Speedtest
Timed Time - Size 1,000
PyPerformance
Benchmark: nbody
VP9 libvpx Encoding
Speed: Speed 0
Inkscape
Operation: SVG Files To PNG
GIMP
Test: resize
libjpeg-turbo tjbench
Test: Decompression Throughput
Crafty
Elapsed Time
ASTC Encoder
Preset: Medium
Selenium
Benchmark: ARES-6 - Browser: Google Chrome
Basis Universal
Settings: ETC1S
Selenium
Benchmark: Speedometer - Browser: Google Chrome
Selenium
Benchmark: StyleBench - Browser: Google Chrome
Timed MrBayes Analysis
Primate Phylogeny Analysis
Fhourstones
Complex Connect-4 Solving
GraphicsMagick
Operation: Rotate
Polyhedron Fortran Benchmarks
Benchmark: protein
Monte Carlo Simulations of Ionised Nebulae
Input: Dust 2D tau100.0
PyPerformance
Benchmark: pathlib
LibreOffice
Test: 20 Documents To PDF
dav1d
Video Input: Chimera 1080p 10-bit
Hierarchical INTegration
Test: FLOAT
PyPerformance
Benchmark: regex_compile
Selenium
Benchmark: Jetstream 2 - Browser: Google Chrome
Montage Astronomical Image Mosaic Engine
Mosaic of M17, K band, 1.5 deg x 1.5 deg
Rodinia
Test: OpenMP HotSpot3D
RawTherapee
Total Benchmark Time
Selenium
Benchmark: Kraken - Browser: Google Chrome
PyPerformance
Benchmark: python_startup
PyBench
Total For Average Test Times
Selenium
Benchmark: WASM collisionDetection - Browser: Firefox
WebP Image Encode
Encode Settings: Quality 100, Lossless
LAME MP3 Encoding
WAV To MP3
eSpeak-NG Speech Engine
Text-To-Speech Synthesis
libavif avifenc
Encoder Speed: 8
Caffe
Model: AlexNet - Acceleration: CPU - Iterations: 200
WebP Image Encode
Encode Settings: Quality 100, Highest Compression
Perl Benchmarks
Test: Interpreter
GNU Octave Benchmark
C-Blosc
Compressor: blosclz
FLAC Audio Encoding
WAV To FLAC
DaCapo Benchmark
Java Test: H2
GIMP
Test: rotate
Tesseract OCR
Time To OCR 7 Images
WebP Image Encode
Encode Settings: Quality 100
libavif avifenc
Encoder Speed: 10
CloverLeaf
Lagrangian-Eulerian Hydrodynamics
WebP Image Encode
Encode Settings: Default
FFTE
N=256, 3D Complex FFT Routine
Dolfyn
Computational Fluid Dynamics
ctx_clock
Context Switch Time
Caffe
Model: GoogleNet - Acceleration: CPU - Iterations: 200
BYTE Unix Benchmark
Computational Test: Dhrystone 2
GIMP
Test: auto-levels
G'MIC
Test: Plotting Isosurface Of A 3D Volume, 1000 Times
WebP Image Encode
Encode Settings: Quality 100, Lossless, Highest Compression
NCNN
Target: CPU - Model: resnet18
GEGL
Operation: Scale
Polyhedron Fortran Benchmarks
Benchmark: air
SciMark
Computational Test: Monte Carlo
RNNoise
Selenium
Benchmark: WASM collisionDetection - Browser: Google Chrome
GEGL
Operation: Color Enhance
Selenium
Benchmark: Jetstream 2 - Browser: Firefox
GEGL
Operation: Rotate 90 Degrees
GEGL
Operation: Crop
Selenium
Benchmark: Kraken - Browser: Firefox
GEGL
Operation: Antialias
Selenium
Benchmark: WASM imageConvolute - Browser: Google Chrome
GEGL
Operation: Reflect
Mlpack Benchmark
Benchmark: scikit_svm
Selenium
Benchmark: WASM imageConvolute - Browser: Firefox
Botan
Test: Twofish
DeepSpeech
Acceleration: CPU
Botan
Test: Blowfish
TNN
Target: CPU - Model: MobileNet v2
TNN
Target: CPU - Model: SqueezeNet v1.1
Botan
Test: CAST-256
Selenium
Benchmark: Maze Solver - Browser: Firefox
Selenium
Benchmark: Maze Solver - Browser: Google Chrome
Mlpack Benchmark
Benchmark: scikit_qda
Phoronix Test Suite v10.8.5