AMD Ryzen 9 5950X vs. Core i9 12900K Linux 6.0

Linux 6.0 CPU AMD Ryzen 9 5950X vs. Intel Core i9 12900K Linux benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2209120-NE-5950X129022&sor&grs.

ASKAP

Test: tConvolve MT - Gridding

miniFE

Problem Size: Small

ASKAP

Test: tConvolve MT - Degridding

Xcompact3d Incompact3d

Input: input.i3d 193 Cells Per Direction

NAS Parallel Benchmarks

Test / Class: MG.C

LeelaChessZero

Backend: Eigen

ASKAP

Test: Hogbom Clean OpenMP

CloverLeaf

Lagrangian-Eulerian Hydrodynamics

LeelaChessZero

Backend: BLAS

NAS Parallel Benchmarks

Test / Class: SP.B

High Performance Conjugate Gradient

Darktable

Test: Server Room - Acceleration: CPU-only

Xcompact3d Incompact3d

Input: input.i3d 129 Cells Per Direction

ASKAP

Test: tConvolve OpenMP - Degridding

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 4K

AI Benchmark Alpha

Device Training Score

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K

Pennant

Test: leblancbig

GNU Radio

Test: Hilbert Transform

Algebraic Multi-Grid Benchmark

NAS Parallel Benchmarks

Test / Class: BT.C

NCNN

Target: CPU - Model: resnet18

Stress-NG

Test: Memory Copying

NCNN

Target: CPU - Model: vgg16

Darktable

Test: Masskrug - Acceleration: CPU-only

NAS Parallel Benchmarks

Test / Class: IS.D

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 1080p

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p

NAS Parallel Benchmarks

Test / Class: LU.C

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 4K

NCNN

Target: CPU - Model: resnet50

Node.js Express HTTP Load Test

NCNN

Target: CPU - Model: alexnet

GNU Radio

Test: Five Back to Back FIR Filters

NAS Parallel Benchmarks

Test / Class: FT.C

LuaRadio

Test: Hilbert Transform

SVT-HEVC

Tuning: 10 - Input: Bosphorus 4K

Nettle

Test: aes256

ASKAP

Test: tConvolve OpenMP - Gridding

Darktable

Test: Boat - Acceleration: CPU-only

AI Benchmark Alpha

Device AI Score

Intel Open Image Denoise

Run: RT.ldr_alb_nrm.3840x2160

Intel Open Image Denoise

Run: RT.hdr_alb_nrm.3840x2160

DaCapo Benchmark

Java Test: Jython

Mobile Neural Network

Model: mobilenetV3

GPAW

Input: Carbon Nanotube

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 4K

NCNN

Target: CPU - Model: googlenet

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 1080p

PHPBench

PHP Benchmark Suite

NCNN

Target: CPU - Model: shufflenet-v2

PyPerformance

Benchmark: chaos

ONNX Runtime

Model: GPT-2 - Device: CPU - Executor: Standard

ONNX Runtime

Model: yolov4 - Device: CPU - Executor: Standard

G'MIC

Test: 2D Function Plotting, 1000 Times

Redis

Test: GET - Parallel Connections: 50

m-queens

Time To Solve

PyPerformance

Benchmark: pickle_pure_python

Nettle

Test: poly1305-aes

WireGuard + Linux Networking Stack Stress Test

Redis

Test: GET - Parallel Connections: 500

PyBench

Total For Average Test Times

PyPerformance

Benchmark: crypto_pyaes

PyPerformance

Benchmark: raytrace

Sysbench

Test: CPU

AOM AV1

Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K

AOM AV1

Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K

Redis

Test: SET - Parallel Connections: 50

SVT-AV1

Encoder Mode: Preset 10 - Input: Bosphorus 4K

Cryptsetup

PBKDF2-whirlpool

Mobile Neural Network

Model: MobileNetV2_224

Redis

Test: SET - Parallel Connections: 500

PyPerformance

Benchmark: nbody

TensorFlow Lite

Model: Mobilenet Float

PyPerformance

Benchmark: regex_compile

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

Chia Blockchain VDF

Test: Square Assembly Optimized

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

GraphicsMagick

Operation: Rotate

Sysbench

Test: RAM / Memory

Etcpak

Benchmark: Multi-Threaded - Configuration: ETC2

PyPerformance

Benchmark: 2to3

Node.js V8 Web Tooling Benchmark

PyPerformance

Benchmark: json_loads

Hierarchical INTegration

Test: FLOAT

TensorFlow Lite

Model: SqueezeNet

simdjson

Throughput Test: LargeRandom

PyPerformance

Benchmark: pathlib

ClickHouse

100M Rows Web Analytics Dataset, First Run / Cold Cache

QuantLib

JPEG XL libjxl

Input: PNG - Encode Speed: 5

Aircrack-ng

Nettle

Test: chacha

ClickHouse

100M Rows Web Analytics Dataset, Third Run

Liquid-DSP

Threads: 1 - Buffer Length: 256 - Filter Length: 57

simdjson

Throughput Test: DistinctUserID

GNU Radio

Test: FM Deemphasis Filter

Liquid-DSP

Threads: 4 - Buffer Length: 256 - Filter Length: 57

TensorFlow Lite

Model: Inception V4

Liquid-DSP

Threads: 2 - Buffer Length: 256 - Filter Length: 57

ClickHouse

100M Rows Web Analytics Dataset, Second Run

LAMMPS Molecular Dynamics Simulator

Model: 20k Atoms

N-Queens

Elapsed Time

simdjson

Throughput Test: PartialTweets

GEGL

Operation: Cartoon

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 1080p

Java Gradle Build

Gradle Build: Reactor

OSPRay

Benchmark: gravity_spheres_volume/dim_512/scivis/real_time

LAMMPS Molecular Dynamics Simulator

Model: Rhodopsin Protein

GraphicsMagick

Operation: Resizing

simdjson

Throughput Test: TopTweet

IndigoBench

Acceleration: CPU - Scene: Bedroom

AOM AV1

Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K

Stress-NG

Test: Vector Math

RawTherapee

Total Benchmark Time

OSPRay

Benchmark: gravity_spheres_volume/dim_512/ao/real_time

Selenium

Benchmark: Speedometer - Browser: Google Chrome

Stress-NG

Test: CPU Stress

Cryptsetup

AES-XTS 512b Decryption

Stargate Digital Audio Workstation

Sample Rate: 44100 - Buffer Size: 512

Stargate Digital Audio Workstation

Sample Rate: 44100 - Buffer Size: 1024

Selenium

Benchmark: ARES-6 - Browser: Firefox

Selenium

Benchmark: WASM imageConvolute - Browser: Firefox

DaCapo Benchmark

Java Test: Tradesoap

OSPRay

Benchmark: particle_volume/scivis/real_time

Zstd Compression

Compression Level: 19, Long Mode - Compression Speed

JPEG XL Decoding libjxl

CPU Threads: All

G'MIC

Test: Plotting Isosurface Of A 3D Volume, 1000 Times

FinanceBench

Benchmark: Repo OpenMP

Selenium

Benchmark: Speedometer - Browser: Firefox

Stress-NG

Test: Matrix Math

NAMD

ATPase Simulation - 327,506 Atoms