GCC 9 Skylake-AVX512 Tests
Intel Core i9-7980XE testing with a ASUS PRIME X299-A (1602 BIOS) and NVIDIA NV120 12GB on Ubuntu 18.10 via the Phoronix Test Suite.


-O0: 

	Processor: Intel Core i9-7980XE @ 4.20GHz (18 Cores / 36 Threads), Motherboard: ASUS PRIME X299-A (1602 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 16384MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: NVIDIA NV120 12GB, Audio: Realtek ALC1220, Monitor: ASUS PB278, Network: Intel I219-V

	OS: Ubuntu 18.10, Kernel: 4.18.0-15-generic (x86_64), Desktop: GNOME Shell 3.30.1, Display Server: X Server 1.20.1, Display Driver: modesetting 1.20.1, OpenGL: 4.3 Mesa 18.2.2, Compiler: GCC 9.0.1 20190217, File-System: ext4, Screen Resolution: 2560x1440

-O3 -march=x86-64: 

	Processor: Intel Core i9-7980XE @ 4.20GHz (18 Cores / 36 Threads), Motherboard: ASUS PRIME X299-A (1602 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 16384MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: NVIDIA NV120 12GB, Audio: Realtek ALC1220, Monitor: ASUS PB278, Network: Intel I219-V

	OS: Ubuntu 18.10, Kernel: 4.18.0-15-generic (x86_64), Desktop: GNOME Shell 3.30.1, Display Server: X Server 1.20.1, Display Driver: modesetting 1.20.1, OpenGL: 4.3 Mesa 18.2.2, Compiler: GCC 9.0.1 20190217, File-System: ext4, Screen Resolution: 2560x1440

-O3 -march=skylake: 

	Processor: Intel Core i9-7980XE @ 4.20GHz (18 Cores / 36 Threads), Motherboard: ASUS PRIME X299-A (1602 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 16384MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: NVIDIA NV120 12GB, Audio: Realtek ALC1220, Monitor: ASUS PB278, Network: Intel I219-V

	OS: Ubuntu 18.10, Kernel: 4.18.0-15-generic (x86_64), Desktop: GNOME Shell 3.30.1, Display Server: X Server 1.20.1, Display Driver: modesetting 1.20.1, OpenGL: 4.3 Mesa 18.2.2, Compiler: GCC 9.0.1 20190217, File-System: ext4, Screen Resolution: 2560x1440

-O3 -march=skylake-avx512: 

	Processor: Intel Core i9-7980XE @ 4.20GHz (18 Cores / 36 Threads), Motherboard: ASUS PRIME X299-A (1602 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 16384MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: NVIDIA NV120 12GB, Audio: Realtek ALC1220, Monitor: ASUS PB278, Network: Intel I219-V

	OS: Ubuntu 18.10, Kernel: 4.18.0-15-generic (x86_64), Desktop: GNOME Shell 3.30.1, Display Server: X Server 1.20.1, Display Driver: modesetting 1.20.1, OpenGL: 4.3 Mesa 18.2.2, Compiler: GCC 9.0.1 20190217, File-System: ext4, Screen Resolution: 2560x1440


t-test1 2017-01-13
Threads: 1
Seconds < Lower Is Better
-O0 ....................... 52.48 |============================================
-O3 -march=x86-64 ......... 50.09 |==========================================
-O3 -march=skylake ........ 49.83 |==========================================
-O3 -march=skylake-avx512 . 49.96 |==========================================


t-test1 2017-01-13
Threads: 2
Seconds < Lower Is Better
-O0 ....................... 18.11 |============================================
-O3 -march=x86-64 ......... 17.17 |==========================================
-O3 -march=skylake ........ 17.09 |==========================================
-O3 -march=skylake-avx512 . 17.07 |=========================================


NAS Parallel Benchmarks 3.3.1
Test / Class: BT.A
Total Mop/s > Higher Is Better
-O0 ....................... 4654 |=============================================
-O3 -march=x86-64 ......... 4636 |=============================================
-O3 -march=skylake ........ 4641 |=============================================
-O3 -march=skylake-avx512 . 4647 |=============================================


NAS Parallel Benchmarks 3.3.1
Test / Class: EP.C
Total Mop/s > Higher Is Better
-O0 ....................... 1178 |============================================
-O3 -march=x86-64 ......... 1153 |===========================================
-O3 -march=skylake ........ 1192 |=============================================
-O3 -march=skylake-avx512 . 1198 |=============================================


NAS Parallel Benchmarks 3.3.1
Test / Class: FT.A
Total Mop/s > Higher Is Better
-O0 ....................... 6333 |============================================
-O3 -march=x86-64 ......... 6458 |=============================================
-O3 -march=skylake ........ 6388 |============================================
-O3 -march=skylake-avx512 . 6509 |=============================================


NAS Parallel Benchmarks 3.3.1
Test / Class: FT.B
Total Mop/s > Higher Is Better
-O0 ....................... 7042 |=============================================
-O3 -march=x86-64 ......... 7051 |=============================================
-O3 -march=skylake ........ 7049 |=============================================
-O3 -march=skylake-avx512 . 7037 |=============================================


NAS Parallel Benchmarks 3.3.1
Test / Class: LU.A
Total Mop/s > Higher Is Better
-O0 ....................... 36068 |============================================
-O3 -march=x86-64 ......... 35984 |============================================
-O3 -march=skylake ........ 36077 |============================================
-O3 -march=skylake-avx512 . 36190 |============================================


NAS Parallel Benchmarks 3.3.1
Test / Class: LU.C
Total Mop/s > Higher Is Better
-O0 ....................... 33583 |============================================
-O3 -march=x86-64 ......... 33535 |============================================
-O3 -march=skylake ........ 33551 |============================================
-O3 -march=skylake-avx512 . 33175 |===========================================


NAS Parallel Benchmarks 3.3.1
Test / Class: SP.A
Total Mop/s > Higher Is Better
-O0 ....................... 4089 |============================================
-O3 -march=x86-64 ......... 4130 |=============================================
-O3 -march=skylake ........ 4110 |=============================================
-O3 -march=skylake-avx512 . 4147 |=============================================


Parboil 2.5
Test: OpenMP LBM
Seconds < Lower Is Better
-O0 ....................... 73.01 |============================================
-O3 -march=x86-64 ......... 72.98 |============================================
-O3 -march=skylake ........ 73.27 |============================================
-O3 -march=skylake-avx512 . 73.00 |============================================


Parboil 2.5
Test: OpenMP CUTCP
Seconds < Lower Is Better
-O0 ....................... 2.40 |=============================================
-O3 -march=x86-64 ......... 2.39 |=============================================
-O3 -march=skylake ........ 2.38 |============================================
-O3 -march=skylake-avx512 . 2.41 |=============================================


Parboil 2.5
Test: OpenMP Stencil
Seconds < Lower Is Better
-O0 ....................... 6.93 |=============================================
-O3 -march=x86-64 ......... 6.91 |=============================================
-O3 -march=skylake ........ 6.97 |=============================================
-O3 -march=skylake-avx512 . 6.94 |=============================================


Parboil 2.5
Test: OpenMP MRI Gridding
Seconds < Lower Is Better
-O0 ....................... 156 |==============================================
-O3 -march=x86-64 ......... 155 |=============================================
-O3 -march=skylake ........ 157 |==============================================
-O3 -march=skylake-avx512 . 154 |=============================================


CloverLeaf
Lagrangian-Eulerian Hydrodynamics
Seconds < Lower Is Better
-O0 ....................... 2.41 |=============================================
-O3 -march=x86-64 ......... 2.42 |=============================================
-O3 -march=skylake ........ 2.41 |=============================================
-O3 -march=skylake-avx512 . 2.42 |=============================================


Rodinia 2.4
Test: OpenMP LavaMD
Seconds < Lower Is Better
-O0 ....................... 26.19 |============================================
-O3 -march=x86-64 ......... 26.31 |============================================
-O3 -march=skylake ........ 25.74 |===========================================
-O3 -march=skylake-avx512 . 25.94 |===========================================


Rodinia 2.4
Test: OpenMP CFD Solver
Seconds < Lower Is Better
-O0 ....................... 12.32 |============================================
-O3 -march=x86-64 ......... 12.38 |============================================
-O3 -march=skylake ........ 12.33 |============================================
-O3 -march=skylake-avx512 . 12.31 |============================================


Rodinia 2.4
Test: OpenMP Streamcluster
Seconds < Lower Is Better
-O0 ....................... 14.09 |============================================
-O3 -march=x86-64 ......... 14.16 |============================================
-O3 -march=skylake ........ 14.07 |============================================
-O3 -march=skylake-avx512 . 14.11 |============================================


FFTW 3.3.6
Build: Stock - Size: 1D FFT Size 32
Mflops > Higher Is Better
-O0 ....................... 3014  |=========
-O3 -march=x86-64 ......... 10701 |===============================
-O3 -march=skylake ........ 13574 |=======================================
-O3 -march=skylake-avx512 . 15228 |============================================


FFTW 3.3.6
Build: Stock - Size: 1D FFT Size 64
Mflops > Higher Is Better
-O0 ....................... 3424  |===========
-O3 -march=x86-64 ......... 9034  |============================
-O3 -march=skylake ........ 12318 |=======================================
-O3 -march=skylake-avx512 . 14057 |============================================


FFTW 3.3.6
Build: Stock - Size: 2D FFT Size 32
Mflops > Higher Is Better
-O0 ....................... 3438  |==========
-O3 -march=x86-64 ......... 11277 |=================================
-O3 -march=skylake ........ 13854 |========================================
-O3 -march=skylake-avx512 . 15119 |============================================


FFTW 3.3.6
Build: Stock - Size: 2D FFT Size 64
Mflops > Higher Is Better
-O0 ....................... 3335  |=============
-O3 -march=x86-64 ......... 7431  |==============================
-O3 -march=skylake ........ 10917 |============================================
-O3 -march=skylake-avx512 . 10786 |===========================================


FFTW 3.3.6
Build: Stock - Size: 2D FFT Size 4096
Mflops > Higher Is Better
-O0 ....................... 2171 |==============
-O3 -march=x86-64 ......... 5833 |=======================================
-O3 -march=skylake ........ 6475 |===========================================
-O3 -march=skylake-avx512 . 6744 |=============================================


FFTW 3.3.6
Build: Float + SSE - Size: 1D FFT Size 32
Mflops > Higher Is Better
-O0 ....................... 3421  |========
-O3 -march=skylake ........ 16773 |========================================
-O3 -march=skylake-avx512 . 18242 |============================================


FFTW 3.3.6
Build: Float + SSE - Size: 1D FFT Size 64
Mflops > Higher Is Better
-O0 ....................... 3250  |=======
-O3 -march=skylake ........ 20081 |============================================
-O3 -march=skylake-avx512 . 19856 |============================================


FFTW 3.3.6
Build: Float + SSE - Size: 2D FFT Size 32
Mflops > Higher Is Better
-O0 ....................... 3146  |===
-O3 -march=skylake ........ 38683 |=====================================
-O3 -march=skylake-avx512 . 46170 |============================================


FFTW 3.3.6
Build: Float + SSE - Size: 2D FFT Size 64
Mflops > Higher Is Better
-O0 ....................... 3640  |====
-O3 -march=skylake ........ 37821 |=======================================
-O3 -march=skylake-avx512 . 42467 |============================================


FFTW 3.3.6
Build: Float + SSE - Size: 2D FFT Size 4096
Mflops > Higher Is Better
-O0 ....................... 3226  |=======
-O3 -march=skylake ........ 19132 |===========================================
-O3 -march=skylake-avx512 . 19440 |============================================


Timed HMMer Search 2.3.2
Pfam Database Search
Seconds < Lower Is Better
-O0 ....................... 9.30  |============================
-O3 -march=x86-64 ......... 14.48 |============================================
-O3 -march=skylake ........ 14.16 |===========================================
-O3 -march=skylake-avx512 . 14.60 |============================================


Timed MAFFT Alignment 7.392
Multiple Sequence Alignment
Seconds < Lower Is Better
-O0 ....................... 2.47 |=============================================
-O3 -march=x86-64 ......... 2.48 |=============================================
-O3 -march=skylake ........ 2.38 |===========================================
-O3 -march=skylake-avx512 . 2.42 |============================================


BYTE Unix Benchmark 3.6
Computational Test: Dhrystone 2
LPS > Higher Is Better
-O0 ....................... 39431843 |=========================================
-O3 -march=x86-64 ......... 39354548 |=========================================
-O3 -march=skylake ........ 38166612 |========================================
-O3 -march=skylake-avx512 . 38455237 |========================================


BYTE Unix Benchmark 3.6
Computational Test: Integer Arithmetic
LPS > Higher Is Better
-O0 ....................... 1 |================================================
-O3 -march=x86-64 ......... 1 |================================================
-O3 -march=skylake ........ 1 |================================================
-O3 -march=skylake-avx512 . 1 |================================================


BYTE Unix Benchmark 3.6
Computational Test: Register Arithmetic
LPS > Higher Is Better
-O0 ....................... 1 |================================================
-O3 -march=x86-64 ......... 1 |================================================
-O3 -march=skylake ........ 1 |================================================
-O3 -march=skylake-avx512 . 1 |================================================


BYTE Unix Benchmark 3.6
Computational Test: Floating-Point Arithmetic
LPS > Higher Is Better
-O0 ....................... 1 |================================================
-O3 -march=x86-64 ......... 1 |================================================
-O3 -march=skylake ........ 1 |================================================
-O3 -march=skylake-avx512 . 1 |================================================


SciMark 2.0
Computational Test: Composite
Mflops > Higher Is Better
-O0 ....................... 703  |===========
-O3 -march=x86-64 ......... 2516 |=========================================
-O3 -march=skylake ........ 2752 |=============================================
-O3 -march=skylake-avx512 . 2751 |=============================================


SciMark 2.0
Computational Test: Monte Carlo
Mflops > Higher Is Better
-O0 ....................... 153 |=======
-O3 -march=x86-64 ......... 947 |=============================================
-O3 -march=skylake ........ 970 |==============================================
-O3 -march=skylake-avx512 . 971 |==============================================


SciMark 2.0
Computational Test: Fast Fourier Transform
Mflops > Higher Is Better
-O0 ....................... 337 |====================
-O3 -march=x86-64 ......... 773 |=============================================
-O3 -march=skylake ........ 791 |==============================================
-O3 -march=skylake-avx512 . 783 |==============================================


SciMark 2.0
Computational Test: Sparse Matrix Multiply
Mflops > Higher Is Better
-O0 ....................... 797  |==========
-O3 -march=x86-64 ......... 3416 |============================================
-O3 -march=skylake ........ 3495 |=============================================
-O3 -march=skylake-avx512 . 3478 |=============================================


SciMark 2.0
Computational Test: Dense LU Matrix Factorization
Mflops > Higher Is Better
-O0 ....................... 984  |=======
-O3 -march=x86-64 ......... 5779 |=========================================
-O3 -march=skylake ........ 6284 |=============================================
-O3 -march=skylake-avx512 . 6301 |=============================================


SciMark 2.0
Computational Test: Jacobi Successive Over-Relaxation
Mflops > Higher Is Better
-O0 ....................... 1243 |=========================
-O3 -march=x86-64 ......... 1662 |==================================
-O3 -march=skylake ........ 2219 |=============================================
-O3 -march=skylake-avx512 . 2220 |=============================================


John The Ripper 1.8.0-jumbo-1
Test: Blowfish
Real C/S > Higher Is Better
-O0 ....................... 5639  |=========
-O3 -march=x86-64 ......... 28966 |============================================
-O3 -march=skylake ........ 28950 |============================================
-O3 -march=skylake-avx512 . 29031 |============================================


John The Ripper 1.8.0-jumbo-1
Test: Traditional DES
Real C/S > Higher Is Better
-O0 ....................... 99444000  |=======================================
-O3 -march=x86-64 ......... 100829667 |========================================
-O3 -march=skylake ........ 100709667 |========================================
-O3 -march=skylake-avx512 . 100746333 |========================================


John The Ripper 1.8.0-jumbo-1
Test: MD5
Real C/S > Higher Is Better
-O0 ....................... 893520 |===========================================
-O3 -march=x86-64 ......... 902217 |===========================================
-O3 -march=skylake ........ 901612 |===========================================
-O3 -march=skylake-avx512 . 885449 |==========================================


AOM AV1 2019-02-11
AV1 Video Encoding
Frames Per Second > Higher Is Better
-O0 ....................... 0.22 |========================================
-O3 -march=x86-64 ......... 0.22 |========================================
-O3 -march=skylake ........ 0.24 |===========================================
-O3 -march=skylake-avx512 . 0.25 |=============================================


SVT-AV1 2019-02-15
1080p 8-bit YUV To AV1 Video Encode
Frames Per Second > Higher Is Better
-O0 ....................... 8.53 |=============================================
-O3 -march=x86-64 ......... 8.52 |=============================================
-O3 -march=skylake ........ 8.38 |============================================
-O3 -march=skylake-avx512 . 8.48 |=============================================


SVT-HEVC 2019-02-03
1080p 8-bit YUV To HEVC Video Encode
Frames Per Second > Higher Is Better
-O0 ....................... 307 |=============================================
-O3 -march=x86-64 ......... 311 |==============================================
-O3 -march=skylake ........ 309 |==============================================
-O3 -march=skylake-avx512 . 309 |==============================================


SVT-VP9 2019-02-17
1080p 8-bit YUV To VP9 Video Encode
Frames Per Second > Higher Is Better
-O0 ....................... 306 |=============================================
-O3 -march=x86-64 ......... 306 |=============================================
-O3 -march=skylake ........ 315 |==============================================
-O3 -march=skylake-avx512 . 313 |==============================================


VP9 libvpx Encoding 1.8.0
vpxenc VP9 1080p Video Encode
Frames Per Second > Higher Is Better
-O0 ....................... 22.97 |=======================================
-O3 -march=x86-64 ......... 22.99 |========================================
-O3 -march=skylake ........ 23.14 |========================================
-O3 -march=skylake-avx512 . 25.60 |============================================


x264 2018-09-25
H.264 Video Encoding
Frames Per Second > Higher Is Better
-O0 ....................... 87.34  |===============================
-O3 -march=x86-64 ......... 122.00 |===========================================
-O3 -march=skylake ........ 123.00 |===========================================
-O3 -march=skylake-avx512 . 123.00 |===========================================


x265 3.0
H.265 1080p Video Encoding
Frames Per Second > Higher Is Better
-O0 ....................... 38.81 |===========================================
-O3 -march=x86-64 ......... 39.10 |============================================
-O3 -march=skylake ........ 39.30 |============================================
-O3 -march=skylake-avx512 . 39.29 |============================================


GraphicsMagick 1.3.30
Operation: Swirl
Iterations Per Minute > Higher Is Better
-O0 ....................... 130 |==========================
-O3 -march=x86-64 ......... 220 |============================================
-O3 -march=skylake ........ 229 |==============================================
-O3 -march=skylake-avx512 . 226 |=============================================


GraphicsMagick 1.3.30
Operation: Rotate
Iterations Per Minute > Higher Is Better
-O0 ....................... 151 |===========================
-O3 -march=x86-64 ......... 252 |=============================================
-O3 -march=skylake ........ 256 |==============================================
-O3 -march=skylake-avx512 . 257 |==============================================


GraphicsMagick 1.3.30
Operation: Sharpen
Iterations Per Minute > Higher Is Better
-O0 ....................... 80  |===================
-O3 -march=x86-64 ......... 176 |===========================================
-O3 -march=skylake ........ 189 |==============================================
-O3 -march=skylake-avx512 . 189 |==============================================


GraphicsMagick 1.3.30
Operation: Enhanced
Iterations Per Minute > Higher Is Better
-O0 ....................... 101 |=====================
-O3 -march=x86-64 ......... 201 |===========================================
-O3 -march=skylake ........ 217 |==============================================
-O3 -march=skylake-avx512 . 211 |=============================================


GraphicsMagick 1.3.30
Operation: Resizing
Iterations Per Minute > Higher Is Better
-O0 ....................... 141 |==========================
-O3 -march=x86-64 ......... 245 |============================================
-O3 -march=skylake ........ 254 |==============================================
-O3 -march=skylake-avx512 . 252 |==============================================


GraphicsMagick 1.3.30
Operation: Noise-Gaussian
Iterations Per Minute > Higher Is Better
-O0 ....................... 103 |============================
-O3 -march=x86-64 ......... 164 |============================================
-O3 -march=skylake ........ 171 |==============================================
-O3 -march=skylake-avx512 . 170 |==============================================


GraphicsMagick 1.3.30
Operation: HWB Color Space
Iterations Per Minute > Higher Is Better
-O0 ....................... 152 |==========================
-O3 -march=x86-64 ......... 260 |=============================================
-O3 -march=skylake ........ 266 |==============================================
-O3 -march=skylake-avx512 . 265 |==============================================


Himeno Benchmark 3.0
Poisson Pressure Solver
MFLOPS > Higher Is Better
-O0 ....................... 468  |======
-O3 -march=x86-64 ......... 3084 |==========================================
-O3 -march=skylake ........ 3315 |=============================================
-O3 -march=skylake-avx512 . 3316 |=============================================


7-Zip Compression 16.02
Compress Speed Test
MIPS > Higher Is Better
-O0 ....................... 93770 |============================================
-O3 -march=x86-64 ......... 93592 |============================================
-O3 -march=skylake ........ 93408 |============================================
-O3 -march=skylake-avx512 . 93504 |============================================


Stockfish 9
Total Time
Nodes Per Second > Higher Is Better
-O0 ....................... 45894946 |=========================================
-O3 -march=x86-64 ......... 46220162 |=========================================
-O3 -march=skylake ........ 46299014 |=========================================
-O3 -march=skylake-avx512 . 46318478 |=========================================


ebizzy 0.3

Records/s > Higher Is Better
-O0 ....................... 569980 |==========================================
-O3 -march=x86-64 ......... 565019 |=========================================
-O3 -march=skylake ........ 589046 |===========================================
-O3 -march=skylake-avx512 . 567481 |=========================================


Timed Apache Compilation 2.4.7
Time To Compile
Seconds < Lower Is Better
-O0 ....................... 13.24 |========================
-O3 -march=x86-64 ......... 24.05 |============================================
-O3 -march=skylake ........ 23.74 |===========================================
-O3 -march=skylake-avx512 . 23.97 |============================================


Timed GCC Compilation 8.2
Time To Compile
Seconds < Lower Is Better
-O0 ....................... 886 |==========================================
-O3 -march=x86-64 ......... 960 |==============================================
-O3 -march=skylake ........ 952 |==============================================
-O3 -march=skylake-avx512 . 961 |==============================================


Timed ImageMagick Compilation 6.9.0
Time To Compile
Seconds < Lower Is Better
-O0 ....................... 6.67  |===========
-O3 -march=x86-64 ......... 27.11 |============================================
-O3 -march=skylake ........ 27.22 |============================================
-O3 -march=skylake-avx512 . 27.27 |============================================


Timed PHP Compilation 7.1.9
Time To Compile
Seconds < Lower Is Better
-O0 ....................... 16.65 |============
-O3 -march=x86-64 ......... 63.29 |============================================
-O3 -march=skylake ........ 63.26 |============================================
-O3 -march=skylake-avx512 . 63.57 |============================================


C-Ray 1.1
Total Time - 4K, 16 Rays Per Pixel
Seconds < Lower Is Better
-O0 ....................... 127.42 |===========================================
-O3 -march=x86-64 ......... 42.43  |==============
-O3 -march=skylake ........ 31.23  |===========
-O3 -march=skylake-avx512 . 33.94  |===========


Parallel BZIP2 Compression 1.1.12
256MB File Compression
Seconds < Lower Is Better
-O0 ....................... 2.40 |=============================================
-O3 -march=x86-64 ......... 2.39 |=============================================
-O3 -march=skylake ........ 2.36 |============================================
-O3 -march=skylake-avx512 . 2.41 |=============================================


AOBench
Size: 2048 x 2048 - Total Time
Seconds < Lower Is Better
-O0 ....................... 72.37 |============================================
-O3 -march=x86-64 ......... 37.68 |=======================
-O3 -march=skylake ........ 34.92 |=====================
-O3 -march=skylake-avx512 . 34.92 |=====================


Bullet Physics Engine 2.81
Test: Raytests
Seconds < Lower Is Better
-O0 ....................... 2.33 |========================================
-O3 -march=x86-64 ......... 2.34 |========================================
-O3 -march=skylake ........ 2.27 |=======================================
-O3 -march=skylake-avx512 . 2.61 |=============================================


Bullet Physics Engine 2.81
Test: 3000 Fall
Seconds < Lower Is Better
-O0 ....................... 4.12 |=============================================
-O3 -march=x86-64 ......... 4.10 |=============================================
-O3 -march=skylake ........ 3.84 |==========================================
-O3 -march=skylake-avx512 . 3.82 |==========================================


Bullet Physics Engine 2.81
Test: 1000 Stack
Seconds < Lower Is Better
-O0 ....................... 4.54 |=============================================
-O3 -march=x86-64 ......... 4.53 |=============================================
-O3 -march=skylake ........ 4.07 |========================================
-O3 -march=skylake-avx512 . 4.06 |========================================


Bullet Physics Engine 2.81
Test: 1000 Convex
Seconds < Lower Is Better
-O0 ....................... 4.02 |=============================================
-O3 -march=x86-64 ......... 4.01 |=============================================
-O3 -march=skylake ........ 3.82 |===========================================
-O3 -march=skylake-avx512 . 3.85 |===========================================


Bullet Physics Engine 2.81
Test: 136 Ragdolls
Seconds < Lower Is Better
-O0 ....................... 2.57 |=============================================
-O3 -march=x86-64 ......... 2.56 |=============================================
-O3 -march=skylake ........ 2.32 |=========================================
-O3 -march=skylake-avx512 . 2.31 |========================================


Bullet Physics Engine 2.81
Test: Prim Trimesh
Seconds < Lower Is Better
-O0 ....................... 0.89 |=============================================
-O3 -march=x86-64 ......... 0.89 |=============================================
-O3 -march=skylake ........ 0.84 |==========================================
-O3 -march=skylake-avx512 . 0.84 |==========================================


Bullet Physics Engine 2.81
Test: Convex Trimesh
Seconds < Lower Is Better
-O0 ....................... 1.05 |=============================================
-O3 -march=x86-64 ......... 1.05 |=============================================
-O3 -march=skylake ........ 1.00 |===========================================
-O3 -march=skylake-avx512 . 1.00 |===========================================


XZ Compression 5.2.4
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9
Seconds < Lower Is Better
-O0 ....................... 119.12 |===========================================
-O3 -march=x86-64 ......... 85.42  |===============================
-O3 -march=skylake ........ 84.95  |===============================
-O3 -march=skylake-avx512 . 83.95  |==============================


Zstd Compression 1.3.4
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19
Seconds < Lower Is Better
-O0 ....................... 19.58 |============================================
-O3 -march=x86-64 ......... 10.39 |=======================
-O3 -march=skylake ........ 10.66 |========================
-O3 -march=skylake-avx512 . 10.49 |========================


dav1d 0.1
Video Input: Summer Nature 4K
Seconds < Lower Is Better
-O0 ....................... 49.47 |============================================
-O3 -march=x86-64 ......... 26.95 |========================
-O3 -march=skylake ........ 26.39 |=======================
-O3 -march=skylake-avx512 . 26.75 |========================


dav1d 0.1
Video Input: Summer Nature 1080p
Seconds < Lower Is Better
-O0 ....................... 23.72 |============================================
-O3 -march=x86-64 ......... 13.89 |==========================
-O3 -march=skylake ........ 13.65 |=========================
-O3 -march=skylake-avx512 . 13.82 |==========================


dcraw
RAW To PPM Image Conversion
Seconds < Lower Is Better
-O0 ....................... 39.04 |============================================
-O3 -march=x86-64 ......... 11.56 |=============
-O3 -march=skylake ........ 10.83 |============
-O3 -march=skylake-avx512 . 10.79 |============


FLAC Audio Encoding 1.3.2
WAV To FLAC
Seconds < Lower Is Better
-O0 ....................... 58.09 |============================================
-O3 -march=x86-64 ......... 10.06 |========
-O3 -march=skylake ........ 9.10  |=======
-O3 -march=skylake-avx512 . 9.02  |=======


LAME MP3 Encoding 3.100
WAV To MP3
Seconds < Lower Is Better
-O0 ....................... 27.45 |============================================
-O3 -march=x86-64 ......... 8.82  |==============
-O3 -march=skylake ........ 9.36  |===============
-O3 -march=skylake-avx512 . 9.41  |===============


m-queens 1.2
Time To Solve
Seconds < Lower Is Better
-O0 ....................... 48.38 |============================================
-O3 -march=x86-64 ......... 48.40 |============================================
-O3 -march=skylake ........ 48.38 |============================================
-O3 -march=skylake-avx512 . 48.37 |============================================


Mencoder 1.3.0
AVI To LAVC
Seconds < Lower Is Better
-O0 ....................... 32.24 |============================================
-O3 -march=x86-64 ......... 16.29 |======================
-O3 -march=skylake ........ 16.02 |======================
-O3 -march=skylake-avx512 . 16.23 |======================


N-Queens 1.0
Elapsed Time
Seconds < Lower Is Better
-O0 ....................... 8.41 |=============================================
-O3 -march=x86-64 ......... 8.40 |=============================================
-O3 -march=skylake ........ 8.40 |=============================================
-O3 -march=skylake-avx512 . 8.40 |=============================================


OpenSSL 1.1.1
RSA 4096-bit Performance
Signs Per Second > Higher Is Better
-O0 ....................... 4578 |============================================
-O3 -march=x86-64 ......... 4693 |=============================================
-O3 -march=skylake ........ 4695 |=============================================
-O3 -march=skylake-avx512 . 4693 |=============================================


Multichase Pointer Chaser
Test: 4MB Array, 64 Byte Stride
ns < Lower Is Better
-O0 ....................... 8.27 |=============================================
-O3 -march=x86-64 ......... 8.34 |=============================================
-O3 -march=skylake ........ 8.32 |=============================================
-O3 -march=skylake-avx512 . 8.33 |=============================================


Multichase Pointer Chaser
Test: 1GB Array, 256 Byte Stride
ns < Lower Is Better
-O0 ....................... 63.90 |============================================
-O3 -march=x86-64 ......... 64.45 |============================================
-O3 -march=skylake ........ 64.16 |============================================
-O3 -march=skylake-avx512 . 64.11 |============================================


Multichase Pointer Chaser
Test: 256MB Array, 256 Byte Stride
ns < Lower Is Better
-O0 ....................... 59.97 |============================================
-O3 -march=x86-64 ......... 59.79 |============================================
-O3 -march=skylake ........ 60.00 |============================================
-O3 -march=skylake-avx512 . 59.57 |============================================


Multichase Pointer Chaser
Test: 1GB Array, 256 Byte Stride, 2 Threads
ns < Lower Is Better
-O0 ....................... 66.76 |============================================
-O3 -march=x86-64 ......... 67.19 |============================================
-O3 -march=skylake ........ 66.31 |===========================================
-O3 -march=skylake-avx512 . 67.06 |============================================


Multichase Pointer Chaser
Test: 1GB Array, 256 Byte Stride, 4 Threads
ns < Lower Is Better
-O0 ....................... 70.54 |============================================
-O3 -march=x86-64 ......... 70.55 |============================================
-O3 -march=skylake ........ 70.36 |============================================
-O3 -march=skylake-avx512 . 70.47 |============================================


libjpeg-turbo tjbench 1.5.3
Test: Decompression Throughput
Megapixels/sec > Higher Is Better
-O0 ....................... 151 |===================================
-O3 -march=x86-64 ......... 185 |===========================================
-O3 -march=skylake ........ 194 |=============================================
-O3 -march=skylake-avx512 . 197 |==============================================


PostgreSQL pgbench 10.3
Scaling: Buffer Test - Test: Normal Load - Mode: Read Only
TPS > Higher Is Better
-O0 ....................... 301052 |============================
-O3 -march=x86-64 ......... 457955 |===========================================
-O3 -march=skylake ........ 460061 |===========================================
-O3 -march=skylake-avx512 . 459370 |===========================================


PostgreSQL pgbench 10.3
Scaling: Buffer Test - Test: Normal Load - Mode: Read Write
TPS > Higher Is Better
-O0 ....................... 12257 |============================================
-O3 -march=x86-64 ......... 12198 |============================================
-O3 -march=skylake ........ 12143 |============================================
-O3 -march=skylake-avx512 . 12200 |============================================


Redis 4.0.8
Test: LPOP
Requests Per Second > Higher Is Better
-O0 ....................... 2196687 |========================================
-O3 -march=x86-64 ......... 2313644 |==========================================
-O3 -march=skylake ........ 2218415 |========================================
-O3 -march=skylake-avx512 . 2232979 |=========================================


Redis 4.0.8
Test: SADD
Requests Per Second > Higher Is Better
-O0 ....................... 1629567 |==========================================
-O3 -march=x86-64 ......... 1626054 |==========================================
-O3 -march=skylake ........ 1645082 |==========================================
-O3 -march=skylake-avx512 . 1640318 |==========================================


Redis 4.0.8
Test: LPUSH
Requests Per Second > Higher Is Better
-O0 ....................... 1417088 |========================================
-O3 -march=x86-64 ......... 1470671 |==========================================
-O3 -march=skylake ........ 1427923 |=========================================
-O3 -march=skylake-avx512 . 1399734 |========================================


Redis 4.0.8
Test: GET
Requests Per Second > Higher Is Better
-O0 ....................... 2070792 |==========================================
-O3 -march=x86-64 ......... 2029011 |=========================================
-O3 -march=skylake ........ 2092803 |==========================================
-O3 -march=skylake-avx512 . 2068424 |==========================================


Redis 4.0.8
Test: SET
Requests Per Second > Higher Is Better
-O0 ....................... 1504661 |=========================================
-O3 -march=x86-64 ......... 1534324 |==========================================
-O3 -march=skylake ........ 1514425 |=========================================
-O3 -march=skylake-avx512 . 1511340 |=========================================


ctx_clock
Context Switch Time
Clocks < Lower Is Better
-O0 ....................... 400 |==============================================
-O3 -march=x86-64 ......... 403 |==============================================
-O3 -march=skylake ........ 399 |==============================================
-O3 -march=skylake-avx512 . 397 |=============================================


Sysbench 2018-07-28
Test: Memory
Events Per Second > Higher Is Better
-O0 ....................... 7152403  |===============
-O3 -march=x86-64 ......... 19443592 |=========================================
-O3 -march=skylake ........ 19327199 |=========================================
-O3 -march=skylake-avx512 . 19353234 |=========================================


Sysbench 2018-07-28
Test: CPU
Events Per Second > Higher Is Better
-O0 ....................... 25334 |================================
-O3 -march=x86-64 ......... 35011 |============================================
-O3 -march=skylake ........ 32596 |=========================================
-O3 -march=skylake-avx512 . 33869 |===========================================


Xsbench 2017-07-06

Lookups/s > Higher Is Better
-O0 ....................... 4509047 |==========================================
-O3 -march=x86-64 ......... 4508790 |==========================================
-O3 -march=skylake ........ 4503111 |==========================================
-O3 -march=skylake-avx512 . 4500275 |==========================================


Hierarchical INTegration 1.0
Test: FLOAT
QUIPs > Higher Is Better
-O0 ....................... 436365279 |========================================
-O3 -march=x86-64 ......... 436477351 |========================================
-O3 -march=skylake ........ 436446118 |========================================
-O3 -march=skylake-avx512 . 437202778 |========================================


Hierarchical INTegration 1.0
Test: DOUBLE
QUIPs > Higher Is Better
-O0 ....................... 981051255 |========================================
-O3 -march=x86-64 ......... 981861431 |========================================
-O3 -march=skylake ........ 984996712 |========================================
-O3 -march=skylake-avx512 . 981700599 |========================================


NGINX Benchmark 1.9.9
Static Web Page Serving
Requests Per Second > Higher Is Better
-O0 ....................... 33311 |===========================================
-O3 -march=x86-64 ......... 34381 |============================================
-O3 -march=skylake ........ 34444 |============================================
-O3 -march=skylake-avx512 . 33972 |===========================================


Apache Benchmark 2.4.29
Static Web Page Serving
Requests Per Second > Higher Is Better
-O0 ....................... 22853 |==========================================
-O3 -march=x86-64 ......... 23270 |==========================================
-O3 -march=skylake ........ 23505 |===========================================
-O3 -march=skylake-avx512 . 24212 |============================================


Apache Siege 2.4.29
Concurrent Users: 250
Transactions Per Second > Higher Is Better
-O0 ....................... 68741 |============================================
-O3 -march=x86-64 ......... 42582 |===========================
-O3 -march=skylake ........ 37845 |========================
-O3 -march=skylake-avx512 . 41973 |===========================