Crystal Internal Benchmark (Nov 2016)

Update the snapshot of benchmark by Kostya. See the previous (April 2015) snapshot here.

• gcc (Ubuntu 4.8.2-19ubuntu1) 4.8.2
• Nim Compiler Version 0.14.0 (2016-06-06) [Linux: amd64]
• Crystal 0.19.2 [d81c32c] (2016-09-16)
• go version go1.7 linux/amd64
• gccgo (Ubuntu 4.9.1-0ubuntu1) 4.9.1
• DMD64 D Compiler v2.068.0
• gdc (crosstool-NG crosstool-ng-1.20.0-232-gc746732 - 20150830-2.066.1-dadb5a3784) 5.2.0
• LDC - the LLVM D compiler (0.15.2-beta1)
• V8 version 3.29.62 (candidate)
• rustc 1.11.0 (9b21dcd6a 2016-08-15)
• Scala version 2.11.6 (Java HotSpot(TM) 64-Bit Server VM, Java 1.8.0_45)
• Nodejs v7.0.0
• PyPy 4.0.0 with GCC 4.8.4
• topaz (ruby-1.9.3p125) (git rev b95c858) [x86_64-linux]
• ruby 2.1.2p95 (2014-05-08 revision 45877) [x86_64-linux]
• Python 2.7.6 and  3.5.2
• rubinius 2.2.10 (2.1.0 bf61ae2e 2014-06-27 JI) [x86_64-linux-gnu]
• jruby 1.7.20 (1.9.3p551) 2015-05-04 3086e6a on Java HotSpot(TM) 64-Bit Server VM 1.8.0_45-b14 +jit [linux-amd64]
• jruby 9.0.0.0.pre2 (2.2.2) 2015-04-28 2755ae0 Java HotSpot(TM) 64-Bit Server VM 25.45-b02 on 1.8.0_45-b14 +jit [linux-amd64]
• Java version "1.8.0_45" Java HotSpot(TM) 64-Bit Server VM (build 25.45-b02, mixed mode)
• julia version 0.4.3
• clang version 3.5-1ubuntu1 (trunk) (based on LLVM 3.5)
• Mono JIT compiler version 4.0.1 (tarball Tue May 12 15:39:23 UTC 2015)
• rock 0.9.10-head codename sapphire, built on Wed Jul 1 20:09:58 2015
• Felix version 15.04.03
• Q KDB+ 3.3 2015.09.02 Copyright (C) 1993-2015 Kx Systems
• perl 5, version 18, subversion 2 (v5.18.2) built for x86_64-linux-gnu-thread-multi
• The Glorious Glasgow Haskell Compilation System, version 7.10.2
• Tcl 8.6
• jq version 1.3
• Swift version 2.2-dev (LLVM ae2eb212e4, Clang ef4c02f431, Swift 634acb40a1)
• Kotlin version 1.0.3 (JRE 1.8.0_45-b14)
• PHP 7.0.9-1+deb.sury.org~trusty+1 (cli) ( NTS )
• .Net Core 1.0.0-preview2-003121

Havlak loop finder

Language	Time, s	Memory, MiB
Crystal	15.46	443.2
Nim Gcc	16.59	484.3
Nim Clang	16.91	477.9
C++	17.72	174.5
D Ldc	25.15	214.9
D	28.90	418.2
Go	31.26	349.9
D Gdc	31.79	197.6
Scala	32.18	363.0
Go Gcc	32.94	365.7
C# Mono	40.54	270.0
Python Pypy	45.51	625.9
C# .Net Core	61.38	388.8
Python	396.54	724.0

Matrix multiplication

Language	Time, s	Memory, MiB
Julia Native Thr	0.11	148.3
Julia Native	0.31	175.8
D Mir GLAS	0.32	56.7
D Ldc	2.01	68.9
D	2.30	71.3
D Gdc	2.33	73.0
Python Numpy	3.08	65.3
Java	3.50	136.2
Scala	3.62	136.2
Kotlin	3.62	132.2
C	3.64	69.2
Nim Clang	3.73	142.3
Rust	3.74	76.9
Nim Gcc	3.82	146.0
Go	3.83	73.5
Crystal	3.85	73.9
Go Gcc	3.90	84.5
Swift	4.48	110.8
Javascript Node	5.95	88.3
Javascript V8	6.87	81.5
Python Pypy	7.68	122.6
C# .Net Core	10.87	85.7
C# Mono	15.17	83.6
Julia	20.98	187.4
Ruby Topaz	81.41	206.2
Ruby	338.40	82.8
Python	447.39	74.0
Ruby JRuby	416.12	582.4
Ruby JRuby9k	467.59	608.3
Ruby Rbx	591.70	325.0
Perl	666.46	604.1
Tcl	1066.66	279.9

JSON parse

Language	Time, s	Memory, MiB
D Gdc Fast	0.34	226.7
C++ Rapid SAX	0.72	1.0
Rust Pull	0.78	207.9
Rust Struct	0.79	232.5
C++ Gason	0.83	582.2
C++ Rapid	0.94	243.6
Java	1.47	621.2
Crystal Schema	1.93	331.2
Rust Value	2.51	1967.0
Perl XS	2.68	888.4
Crystal	3.00	1115.2
Javascript Node	3.21	863.7
Crystal Pull	3.30	1.6
Nim Clang	4.12	1089.6
Python3 ujson	4.15	1303.2
Nim Gcc	4.46	1090.1
Python Pypy	4.81	1553.0
Python ujson	5.07	1352.9
Q	5.18	684.0
Go	5.21	479.3
C++ LibJson	5.49	2796.3
Clojure	5.81	1148.5
Python3	5.82	1037.8
C# .Net Core	6.31	834.9
Php	6.37	1502.0
Ruby YAJL	8.23	1085.5
Haskell	8.31	70.5
Python	9.85	1409.1
C# Mono	10.57	812.1
Julia	11.89	2622.4
D	12.42	1417.1
Ruby	12.67	2013.9
JQ	14.92	1714.5
Scala	15.47	1415.8
C++ Boost	16.44	2915.2
Ruby JRuby9K	16.53	2050.5
Go Gcc	17.64	473.1
Ruby JRuby	21.98	2761.1
D Gdc	25.86	926.1
D Ldc	27.23	919.6
Perl	46.02	1635.4
Ruby Rbx	67.13	4681.0

Base64 encode and decode

Language	Time, s	Memory, MiB
C aklomp SSSE3	0.93	32.3
C	1.85	32.2
Crystal	2.30	85.0
D Gdc	2.52	33.3
Ruby	2.73	125.3
D Ldc	3.14	53.1
Perl XS	3.63	47.9
Rust	3.64	42.9
Ruby Rbx	4.29	30.7
Nim Gcc	4.62	52.7
Nim Clang	4.70	52.7
Julia	4.41	190.0
Javascript Node	4.76	551.5
C++ Openssl	5.45	65.2
Php	6.34	53.4
C# .Net Core	6.52	121.1
D	7.18	55.3
Tcl	7.20	66.0
Python Pypy	7.32	582.3
Python	7.62	52.6
Go	8.00	70.0
Python3	8.13	54.5
C# Mono	9.01	71.7
Java	9.06	971.2
Kotlin	9.75	932.9
Scala	10.69	292.5
Ruby JRuby9K	12.16	530.6
Ruby JRuby	12.65	514.9
Perl	33.30	99.7
Go Gcc	39.56	185.5

Mandelbrot in Brainfuck

Language	Time, s	Memory, MiB
C++ Gcc	20.49	1.7
Crystal	23.13	1.5
D Ldc	24.90	1.4
Rust	25.19	4.9
D Gdc	29.49	2.4
Nim Gcc	31.04	2.7
Nim Clang	37.39	2.9
Go Gcc	37.59	11.4
Kotlin	40.79	34.9
Scala	58.51	120.12
Java	58.86	423.9
D Dmd	62.69	1.7
Javascript Node	88.34	18.6
Go	102.85	2.2
Python Pypy	126.47	78.9
C# .Net Core	142.30	17.8
C# Mono	147.69	12.6
Ruby Topaz	305.79	38.8

Brainfuck interpreter

Language	Time, s	Memory, MiB
Kotlin	1.78	28.4
C++ Gcc	1.94	1.0
D Ldc	2.02	0.9
Rust	2.49	4.9
Nim Gcc	2.56	0.7
D Gdc	3.05	1.4
Nim Clang	3.09	0.8
Scala	3.43	120.12
Crystal	3.46	1.3
Java	4.03	513.8
Go Gcc	4.20	10.0
D Dmd	5.23	1.0
Go	5.36	0.9
Javascript V8	6.53	7.9
Javascript Node	7.38	16.9
C# .Net Core	16.03	16.9
C# Mono	19.86	14.6
Python Pypy	20.64	77.9
Ruby Topaz	59.53	36.5
Ruby Rbx	120.38	32.3
Ruby Jruby	129.75	267.6
Ruby Jruby9k	134.18	286.8
Ruby	181.44	7.2
Python	314.79	4.9
Python3	412.13	5.5

Javascript Virtual DOM Framework Benchmark

Recently I found a framework benchmark for javascript (round4 2016-09-12), it shows a lot information. The fastest ranker are:

Framework	Version	Runtime Ratio (% slower than fastest)	MB Start	MB 1K rows
vanillajs		1	2.98	4.85
inferno	1.0.0-beta9	3	3.19	6.60
vanillajs-keyed		4	2.98	4.81
dio.js	3.0.5	4	3.19	7.09
domvm	2.0.0-beta	10	3.15	7.43
kivi	1.0.0-rc2	17	3.17	6.69

Also there are another benchmark (this have fewer framework, last result at this time):

Framework	Init Time	First Render Time	Overall Time
kivi[adv] 1.0.0	2,185	25,795	527,703
Inferno 0.7.22	3,340	31,780	553,513
Imba 0.14.3	15,920	26,250	569,895
Bobril 4.38.0	3,430	26,255	682,847
Vidom 0.3.16	3,400	38,220	729,882
Maquette 2.3.3	1,955	27,410	733,165
Snabbdom 0.5.0	2,565	34,710	759,481
React 15.3.1	38,640	56,065	926,403
React-lite 0.15.6	6,345	40,725	1,105,627
Preact 5.4.0	2,550	53,585	1,150,506
Vanilla[innerHTML] 1.0.0	1,790	16,925	1,500,676
Deku 2.0.0-rc16	3,285	45,950	1,598,787
Mercury 14.1.0	2,840	41,325	2,115,253
yo-yo 1.2.2	1,825	21,835	2,295,505

Apparently there are a lot of them around the internet.

Of course you can build your own framework, here's the tips if you plan doing so.

Rescue the Princess using 8 Programming Languages

For the lolz, source.

LXC Web Panel

As you (probably) already know, LXC (Linux Containers) or OpenVZ an operating-system-level virtualization is really faster than hardware virtualization, see the comparison. For those who hate CLI, you can use web interface called LXC Web Panel (for LXC 0.7 to 0.9, or newer fork 1.0+ here) to manage your containers:

wget https://lxc-webpanel.github.io/tools/install.sh -O - | sudo bash

This software only works on Ubuntu 12.04 or later. Despite of its performance, of course there are limitations, such as: you can only use host OS and architecture on guest. You can find more info on their website or this blog post.

So why LXC instead of Docker or Virtualization? because it's simpler :3 yes, both are different kind of animal, don't forget to check LXD and other alternatives too.

Best Video about Bitcoin and Blockchain

This is old (2013), but quite informative, the explanation about BitCoin and BlockChain.

If Programming Language were Musical Instruments

Sauce: http://9gag.com/gag/aM9YOgP
Go would be a drum :3

ZSH is better BASH!

ZSH is better BASH, you should really change to newer and better shell.
Why it's better? it has better completion, support right prompt, see this presentation for more

Here's how to install it on ArchLinux:

pacman -S zsh zsh-completions
chsh -s `which zsh`
sh -c "$(curl -fsSL https://raw.githubusercontent.com/robbyrussell/oh-my-zsh/master/tools/install.sh)"

Oh-My-Zsh is package manager for zsh, you can change theme, add more aliases, for example:

plugins=(git git-extras redis-cli sudo systemd archlinux)

You can also install another theme, such as bullet-train:

mkdir $ZSH_CUSTOM/themes
cd $ZSH_CUSTOM/themes
aria2c "http://raw.github.com/caiogondim/bullet-train-oh-my-zsh-theme/master/bullet-train.zsh-theme"

Then set on ~/.zshrc:

BULLETTRAIN_DIR_EXTENDED=2
ZSH_THEME="bullet-train"

Remap Capslock to Backspace on Linux

Greatest idea IMHO from Colemak keyboard layout is replacing capslock with backspace key.

Here's how to do it in linux:

setxkbmap -option caps:backspace

But if you really need capslock, you can set pressing both shift key:

setxkbmap -option shift:both_capslock

These can be set on the ~/.xprofile file. For more xkeyboard tips, see this blog post.

On Mac OSX you can use Karabiner, for Windows, you look at this blog post.

Btw, have you heard about cheapest mechanical keyboard Nimxo K005?

Keyboard with built-in Soundcard

I was looking for keyboard with built-in soundcard (not just passthrough), but the moderator deletes both my question and answers. Oh well, I post the updated results here now:

Cooler Master Storm Mech
+ Mechanical
+ White backlight
+ 1ms pooling rate
+ 1.5m cable
+ 553 x 267 x 43 mm
+ 1686 gr
+ 5 macro keys
+ USB 3
official review youtube

Thermaltake eSports Challenger Ultimate
+ Membrane
+ Full color backlight
+ 2ms pooling rate
+ 2m braided cable
+ 505 x 195 x 27 mm
+ Fan
+ 14x5 macro keys
+ 6 multimedia keys
official youtube

Logitech G510s
+ Membrane+ Full color backlight
+ 2ms pooling rate
+ 6 keys anti-ghosting
+ 2m cable
+ 557 x 215 x 31 mm
+ 1205 gr
+ Built-in LCD
+ UV Coated keys
+ 18 macro keys
+ 4 media keys
official youtube

Thermaltake eSports Poseidon Z Forged
+ Mechanical
+ Blue backlight
+ 1ms pooling rate
+ 104 keys anti-ghosting+ 1.8 m cable
+ 503 x 156 x 43 mm
+ 1474 gr
+ 10 macro keys
official review youtube

Thermaltake eSports Meka G-Unit
+ Mechanical
+ White backlight
+ 46 keys anti-ghosting
+ 1ms pooling rate
+ 554 x 271 x 70 mm
+ 1.8m braided cable+ 12 macro keys
+ 7 media keys
official youtube

What's the best? dunno guys XD never tried any of them.