Core i9 10900K - Ubuntu 20.04 LTS vs. Ubuntu 20.10
Benchmarks by Michael Larabel.
HTML result view exported from: https://openbenchmarking.org/result/2010155-FI-2010126FI35&grs.
oneDNN
Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPU
oneDNN
Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU
oneDNN
Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU
Hackbench
Count: 8 - Type: Thread
oneDNN
Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPU
Hackbench
Count: 8 - Type: Process
oneDNN
Harness: IP Batch All - Data Type: f32 - Engine: CPU
oneDNN
Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU
oneDNN
Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU
GLmark2
Resolution: 3840 x 2160
GraphicsMagick
Operation: Resizing
GLmark2
Resolution: 2560 x 1440
RealSR-NCNN
Scale: 4x - TAA: No
GLmark2
Resolution: 1920 x 1080
Hackbench
Count: 2 - Type: Process
Hackbench
Count: 2 - Type: Thread
Intel Open Image Denoise
Scene: Memorial
Mlpack Benchmark
Benchmark: scikit_ica
Apache CouchDB
Bulk Size: 100 - Inserts: 1000 - Rounds: 24
GraphicsMagick
Operation: HWB Color Space
GraphicsMagick
Operation: Swirl
Xonotic
Resolution: 3840 x 2160 - Effects Quality: High
Xonotic
Resolution: 3840 x 2160 - Effects Quality: Ultra
Algebraic Multi-Grid Benchmark
GraphicsMagick
Operation: Enhanced
ASKAP
Test: tConvolve OpenMP - Gridding
ET: Legacy
Renderer: Renderer2 - Resolution: 3840 x 2160
MPV
Video Input: Big Buck Bunny Sunflower 4K - Decode: Software Only
librsvg
Operation: SVG Files To PNG
Coremark
CoreMark Size 666 - Iterations Per Second
GraphicsMagick
Operation: Sharpen
CacheBench
Test: Read
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon
ASKAP
Test: tConvolve OpenMP - Degridding
Hackbench
Count: 1 - Type: Thread
Go Benchmarks
Test: http
Xonotic
Resolution: 3840 x 2160 - Effects Quality: Ultimate
ebizzy
Hackbench
Count: 1 - Type: Process
MPV
Video Input: Big Buck Bunny Sunflower 1080p - Decode: Software Only
Polyhedron Fortran Benchmarks
Benchmark: air
SciMark
Computational Test: Fast Fourier Transform
Nebular Empirical Analysis Tool
POV-Ray
Trace Time
OpenVKL
Benchmark: vklBenchmark
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon Obj
PyPerformance
Benchmark: pathlib
Botan
Test: CAST-256
FLAC Audio Encoding
WAV To FLAC
GraphicsMagick
Operation: Noise-Gaussian
Gzip Compression
Linux Source Tree Archiving To .tar.gz
Embree
Binary: Pathtracer - Model: Asian Dragon
Crafty
Elapsed Time
SVT-HEVC
1080p 8-bit YUV To HEVC Video Encode
Perl Benchmarks
Test: Pod2html
Botan
Test: Blowfish
glibc bench
Benchmark: cos
Timed Linux Kernel Compilation
Time To Compile
Himeno Benchmark
Poisson Pressure Solver
glibc bench
Benchmark: sin
Polyhedron Fortran Benchmarks
Benchmark: gas_dyn2
dav1d
Video Input: Chimera 1080p
PyPerformance
Benchmark: crypto_pyaes
OSPray
Demo: NASA Streamlines - Renderer: SciVis
LibreOffice
Test: 20 Documents To PDF
CloverLeaf
Lagrangian-Eulerian Hydrodynamics
Tesseract
Resolution: 3840 x 2160
OSPray
Demo: XFrog Forest - Renderer: Path Tracer
asmFish
1024 Hash Memory, 26 Depth
glibc bench
Benchmark: exp
Timed ImageMagick Compilation
Time To Compile
Polyhedron Fortran Benchmarks
Benchmark: fatigue2
LAME MP3 Encoding
WAV To MP3
OSPray
Demo: NASA Streamlines - Renderer: Path Tracer
Botan
Test: AES-256
OSPray
Demo: XFrog Forest - Renderer: SciVis
Polyhedron Fortran Benchmarks
Benchmark: channel2
PolyBench-C
Test: Covariance Computation
Java 2D Microbenchmark
Rendering Test: Vector Graphics Rendering
SVT-AV1
Encoder Mode: Enc Mode 8 - Input: 1080p
Smallpt
Global Illumination Renderer; 128 Samples
Embree
Binary: Pathtracer ISPC - Model: Crown
Tesseract
Resolution: 1920 x 1080
PlaidML
FP16: No - Mode: Inference - Network: VGG19 - Device: CPU
OSPray
Demo: San Miguel - Renderer: Path Tracer
m-queens
Time To Solve
OSPray
Demo: San Miguel - Renderer: SciVis
Selenium
Benchmark: StyleBench - Browser: Firefox
PolyBench-C
Test: Correlation Computation
Numpy Benchmark
N-Queens
Elapsed Time
LuxCoreRender
Scene: DLSC
LuxCoreRender
Scene: Rainbow Colors and Prism
Caffe
Model: AlexNet - Acceleration: CPU - Iterations: 100
Caffe
Model: AlexNet - Acceleration: CPU - Iterations: 200
libavif avifenc
Encoder Speed: 8
OSPray
Demo: Magnetic Reconnection - Renderer: SciVis
SVT-AV1
Encoder Mode: Enc Mode 4 - Input: 1080p
7-Zip Compression
Compress Speed Test
Caffe
Model: GoogleNet - Acceleration: CPU - Iterations: 200
Polyhedron Fortran Benchmarks
Benchmark: linpk
CP2K Molecular Dynamics
Fayalite-FIST Data
SVT-VP9
Tuning: Visual Quality Optimized - Input: Bosphorus 1080p
Perl Benchmarks
Test: Interpreter
Botan
Test: KASUMI
Minion
Benchmark: Graceful
Ogg Encoding
WAV To Ogg
libavif avifenc
Encoder Speed: 10
Botan
Test: Twofish
GraphicsMagick
Operation: Rotate
Java 2D Microbenchmark
Rendering Test: All Rendering Tests
Bullet Physics Engine
Test: 3000 Fall
AOM AV1
Encoder Mode: Speed 4 Two-Pass
Basemark GPU
Renderer: Vulkan - Resolution: 3840 x 2160 - Graphics Preset: Medium
PlaidML
FP16: No - Mode: Inference - Network: VGG16 - Device: CPU
Basemark GPU
Renderer: Vulkan - Resolution: 2560 x 1440 - Graphics Preset: Medium
libavif avifenc
Encoder Speed: 2
Java 2D Microbenchmark
Rendering Test: Image Rendering
DaCapo Benchmark
Java Test: Tradesoap
PolyBench-C
Test: 3 Matrix Multiplications
Embree
Binary: Pathtracer - Model: Asian Dragon Obj
AOM AV1
Encoder Mode: Speed 6 Two-Pass
Timed Apache Compilation
Time To Compile
LAMMPS Molecular Dynamics Simulator
Model: Rhodopsin Protein
libavif avifenc
Encoder Speed: 0
Minion
Benchmark: Solitaire
SciMark
Computational Test: Monte Carlo
Basemark GPU
Renderer: Vulkan - Resolution: 1920 x 1080 - Graphics Preset: High
AOM AV1
Encoder Mode: Speed 0 Two-Pass
Unigine Superposition
Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: Low - Renderer: OpenGL
Embree
Binary: Pathtracer - Model: Crown
Mobile Neural Network
Model: MobileNetV2_224
FFTE
N=256, 3D Complex FFT Routine
dav1d
Video Input: Summer Nature 1080p
LuaJIT
Test: Sparse Matrix Multiply
ctx_clock
Context Switch Time
PyPerformance
Benchmark: float
SVT-VP9
Tuning: VMAF Optimized - Input: Bosphorus 1080p
SVT-VP9
Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p
Caffe
Model: GoogleNet - Acceleration: CPU - Iterations: 100
YafaRay
Total Time For Sample Scene
Basemark GPU
Renderer: Vulkan - Resolution: 3840 x 2160 - Graphics Preset: High
Basemark GPU
Renderer: Vulkan - Resolution: 2560 x 1440 - Graphics Preset: High
Unigine Superposition
Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: Medium - Renderer: OpenGL
SciMark
Computational Test: Composite
Polyhedron Fortran Benchmarks
Benchmark: doduc
Bullet Physics Engine
Test: 1000 Convex
Selenium
Benchmark: Basemark - Browser: Firefox
Xonotic
Resolution: 3840 x 2160 - Effects Quality: Low
x265
H.265 1080p Video Encoding
SciMark
Computational Test: Sparse Matrix Multiply
Primesieve
1e12 Prime Number Generation
LuaJIT
Test: Dense LU Matrix Factorization
LevelDB
Benchmark: Hot Read
Tesseract OCR
Time To OCR 7 Images
Hierarchical INTegration
Test: FLOAT
Mobile Neural Network
Model: SqueezeNetV1.0
Polyhedron Fortran Benchmarks
Benchmark: capacita
LevelDB
Benchmark: Seek Random
Basemark GPU
Renderer: Vulkan - Resolution: 1920 x 1080 - Graphics Preset: Medium
dav1d
Video Input: Chimera 1080p 10-bit
VkFFT
Tesseract
Resolution: 2560 x 1440
TensorFlow Lite
Model: NASNet Mobile
Go Benchmarks
Test: garbage
LuaJIT
Test: Composite
Minion
Benchmark: Quasigroup
TNN
Target: CPU - Model: MobileNet v2
Bullet Physics Engine
Test: Convex Trimesh
Bullet Physics Engine
Test: 1000 Stack
GIMP
Test: resize
Zstd Compression
Compression Level: 19
SciMark
Computational Test: Dense LU Matrix Factorization
SVT-AV1
Encoder Mode: Enc Mode 0 - Input: 1080p
Appleseed
Scene: Material Tester
LuaJIT
Test: Monte Carlo
DaCapo Benchmark
Java Test: Tradebeans
KeyDB
Polyhedron Fortran Benchmarks
Benchmark: aermod
Appleseed
Scene: Disney Material
PyPerformance
Benchmark: chaos
Polyhedron Fortran Benchmarks
Benchmark: ac
SQLite Speedtest
Timed Time - Size 1,000
CLOMP
Static OMP Speedup
SciMark
Computational Test: Jacobi Successive Over-Relaxation
Selenium
Benchmark: Octane - Browser: Firefox
PlaidML
FP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPU
Stockfish
Total Time
PyBench
Total For Average Test Times
ParaView
Test: Wavelet Contour - Resolution: 1920 x 1080
ParaView
Test: Wavelet Contour - Resolution: 1920 x 1080
glibc bench
Benchmark: tanh
AOM AV1
Encoder Mode: Speed 6 Realtime
Bullet Physics Engine
Test: Raytests
Polyhedron Fortran Benchmarks
Benchmark: tfft2
Go Benchmarks
Test: build
XZ Compression
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9
PlaidML
FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU
LevelDB
Benchmark: Sequential Fill
GIMP
Test: unsharp-mask
Scikit-Learn
Java SciMark
Computational Test: Dense LU Matrix Factorization
NeatBench
Acceleration: CPU
LevelDB
Benchmark: Sequential Fill
LuaJIT
Test: Jacobi Successive Over-Relaxation
Apache Siege
Concurrent Users: 250
Polyhedron Fortran Benchmarks
Benchmark: mdbx
ASTC Encoder
Preset: Fast
PlaidML
FP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPU
Radiance Benchmark
Test: Serial
RNNoise
miniFE
Problem Size: Small
Go Benchmarks
Test: json
Selenium
Benchmark: WASM collisionDetection - Browser: Firefox
BRL-CAD
VGR Performance Metric
Timed HMMer Search
Pfam Database Search
GIMP
Test: auto-levels
PyPerformance
Benchmark: django_template
Selenium
Benchmark: Jetstream 2 - Browser: Firefox
Java 2D Microbenchmark
Rendering Test: Text Rendering
FS-Mark
Test: 1000 Files, 1MB Size
LevelDB
Benchmark: Random Fill
Selenium
Benchmark: CanvasMark - Browser: Firefox
LevelDB
Benchmark: Fill Sync
Timed MrBayes Analysis
Primate Phylogeny Analysis
LevelDB
Benchmark: Random Fill
C-Ray
Total Time - 4K, 16 Rays Per Pixel
Bullet Physics Engine
Test: Prim Trimesh
Polyhedron Fortran Benchmarks
Benchmark: protein
ASKAP
Test: tConvolve MT - Degridding
Timed MAFFT Alignment
Multiple Sequence Alignment - LSU RNA
dav1d
Video Input: Summer Nature 4K
Polyhedron Fortran Benchmarks
Benchmark: rnflow
Incompact3D
Input: Cylinder
LevelDB
Benchmark: Overwrite
glibc bench
Benchmark: modf
Mobile Neural Network
Model: resnet-v2-50
Hackbench
Count: 16 - Type: Process
G'MIC
Test: 3D Elevated Function In Random Colors, 100 Times
Mlpack Benchmark
Benchmark: scikit_svm
NAMD
ATPase Simulation - 327,506 Atoms
FFTW
Build: Float + SSE - Size: 2D FFT Size 4096
GROMACS
Water Benchmark
BYTE Unix Benchmark
Computational Test: Dhrystone 2
Hackbench
Count: 32 - Type: Process
Selenium
Benchmark: WASM imageConvolute - Browser: Firefox
LevelDB
Benchmark: Random Read
AOBench
Size: 2048 x 2048 - Total Time
Git
Time To Complete Common Git Commands
LevelDB
Benchmark: Overwrite
Bullet Physics Engine
Test: 136 Ragdolls
Polyhedron Fortran Benchmarks
Benchmark: induct2
PyPerformance
Benchmark: json_loads
Mobile Neural Network
Model: mobilenet-v1-1.0
Zstd Compression
Compression Level: 3
Selenium
Benchmark: ARES-6 - Browser: Firefox
DeepSpeech
Acceleration: CPU
ASKAP
Test: tConvolve MPI - Degridding
NGINX Benchmark
Static Web Page Serving
Basis Universal
Settings: UASTC Level 3
GIMP
Test: rotate
Apache Benchmark
Static Web Page Serving
Java SciMark
Computational Test: Composite
Apache Siege
Concurrent Users: 200
TNN
Target: CPU - Model: SqueezeNet v1.1
glibc bench
Benchmark: log2
PyPerformance
Benchmark: 2to3
GPAW
Input: Carbon Nanotube
Basis Universal
Settings: ETC1S
Mobile Neural Network
Model: inception-v3
glibc bench
Benchmark: pthread_once
Appleseed
Scene: Emily
PyPerformance
Benchmark: python_startup
PlaidML
FP16: No - Mode: Inference - Network: Inception V3 - Device: CPU
Mlpack Benchmark
Benchmark: scikit_linearridgeregression
glibc bench
Benchmark: sinh
glibc bench
Benchmark: ffsll
glibc bench
Benchmark: sqrt
Basis Universal
Settings: UASTC Level 0
glibc bench
Benchmark: asinh
TensorFlow Lite
Model: Mobilenet Quant
Selenium
Benchmark: PSPDFKit WASM - Browser: Firefox
TSCP
AI Chess Performance
LevelDB
Benchmark: Random Delete
Polyhedron Fortran Benchmarks
Benchmark: mp_prop_design
PyPerformance
Benchmark: raytrace
RawTherapee
Total Benchmark Time
Selenium
Benchmark: Kraken - Browser: Firefox
Basis Universal
Settings: UASTC Level 2
InfluxDB
Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000
PyPerformance
Benchmark: nbody
PyPerformance
Benchmark: go
Radiance Benchmark
Test: SMP Parallel
OpenSSL
RSA 4096-bit Performance
InfluxDB
Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000
Dolfyn
Computational Fluid Dynamics
ASKAP
Test: tConvolve MT - Gridding
PHPBench
PHP Benchmark Suite
CacheBench
Test: Write
G'MIC
Test: Plotting Isosurface Of A 3D Volume, 1000 Times
Hugin
Panorama Photo Assistant + Stitching Time
Selenium
Benchmark: WebXPRT - Browser: Firefox
ASKAP
Test: tConvolve MPI - Gridding
C-Blosc
Compressor: blosclz
Java SciMark
Computational Test: Monte Carlo
CacheBench
Test: Read / Modify / Write
ASTC Encoder
Preset: Exhaustive
TensorFlow Lite
Model: Mobilenet Float
PyPerformance
Benchmark: pickle_pure_python
FFTW
Build: Stock - Size: 2D FFT Size 4096
Sudokut
Total Time
ASTC Encoder
Preset: Medium
Java SciMark
Computational Test: Jacobi Successive Over-Relaxation
TensorFlow Lite
Model: SqueezeNet
TensorFlow Lite
Model: Inception V4
glibc bench
Benchmark: sincos
GNU Octave Benchmark
High Performance Conjugate Gradient
Selenium
Benchmark: Jetstream - Browser: Firefox
Bork File Encrypter
File Encryption Time
ASTC Encoder
Preset: Thorough
libjpeg-turbo tjbench
Test: Decompression Throughput
Mlpack Benchmark
Benchmark: scikit_qda
InfluxDB
Concurrent Streams: 1024 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000
TensorFlow Lite
Model: Inception ResNet V2
DaCapo Benchmark
Java Test: Jython
Basis Universal
Settings: UASTC Level 2 + RDO Post-Processing
glibc bench
Benchmark: atanh
Timed FFmpeg Compilation
Time To Compile
LuaJIT
Test: Fast Fourier Transform
Java SciMark
Computational Test: Fast Fourier Transform
WireGuard + Linux Networking Stack Stress Test
Chaos Group V-RAY
Mode: CPU
Polyhedron Fortran Benchmarks
Benchmark: test_fpu2
SQLite
Threads / Copies: 1
Java SciMark
Computational Test: Sparse Matrix Multiply
PlaidML
FP16: No - Mode: Inference - Network: Mobilenet - Device: CPU
glibc bench
Benchmark: ffs
AOM AV1
Encoder Mode: Speed 8 Realtime
OCRMyPDF
Processing 60 Page PDF Document
Darktable
Test: Server Room - Acceleration: CPU-only
Darktable
Test: Server Rack - Acceleration: CPU-only
Darktable
Test: Masskrug - Acceleration: CPU-only
Darktable
Test: Boat - Acceleration: CPU-only
Systemd Total Boot Time
Test: Userspace
Systemd Total Boot Time
Test: Firmware
Systemd Total Boot Time
Test: Loader
Systemd Total Boot Time
Test: Kernel
Systemd Total Boot Time
Test: Total
Selenium
Benchmark: Speedometer - Browser: Firefox
Selenium
Benchmark: Maze Solver - Browser: Firefox
Apache Siege
Concurrent Users: 10
PyPerformance
Benchmark: regex_compile
Redis
Test: GET
Redis
Test: LPUSH
Redis
Test: LPOP
Memtier_benchmark
Protocol: Redis
OCRMyPDF
Processing 60 Page PDF Document
Inkscape
Operation: SVG Files To PNG
Darktable
Test: Server Room - Acceleration: CPU-only
Darktable
Test: Server Rack - Acceleration: CPU-only
Darktable
Test: Masskrug - Acceleration: CPU-only
Darktable
Test: Boat - Acceleration: CPU-only
LevelDB
Benchmark: Fill Sync
Montage Astronomical Image Mosaic Engine
Mosaic of M17, K band, 1.5 deg x 1.5 deg
GnuPG
2GB File Encryption
Cython benchmark
Build2
Time To Compile
OSPray
Demo: Magnetic Reconnection - Renderer: Path Tracer
Nettle
Test: poly1305-aes
Nettle
Test: sha512
Nettle
Test: chacha
Nettle
Test: aes256
Java Gradle Build
Gradle Build: Reactor
Monte Carlo Simulations of Ionised Nebulae
Input: Dust 2D tau100.0
ParaView
Test: Many Spheres - Resolution: 1920 x 1080
ParaView
Test: Many Spheres - Resolution: 1920 x 1080
Compile Bench
Test: Read Compiled Tree
Compile Bench
Test: Initial Create
Selenium
Benchmark: MotionMark - Browser: Firefox
Apache Siege
Concurrent Users: 100
Apache Siege
Concurrent Users: 50
Redis
Test: SET
Redis
Test: SADD
Hackbench
Count: 4 - Type: Process
Hackbench
Count: 16 - Type: Thread
Hackbench
Count: 4 - Type: Thread
TTSIOD 3D Renderer
Phong Rendering With Soft-Shadow Mapping
oneDNN
Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU
oneDNN
Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU
oneDNN
Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU
oneDNN
Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPU
oneDNN
Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU
oneDNN
Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPU
oneDNN
Harness: IP Batch 1D - Data Type: f32 - Engine: CPU
DaCapo Benchmark
Java Test: H2
ParaView
Test: Wavelet Volume - Resolution: 1920 x 1080
ParaView
Test: Wavelet Volume - Resolution: 1920 x 1080
Compile Bench
Test: Compile
IOR
Read Test
IOR
Write Test
FS-Mark
Test: 4000 Files, 32 Sub Dirs, 1MB Size
FS-Mark
Test: 5000 Files, 1MB Size, 4 Threads
Phoronix Test Suite v10.8.4