Conversation

Notices

1. Embed this notice
   Kris Nóva (nova@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:02:20 JST Kris Nóva

   Spontaneous computer science exercise on the stream today.

   fib(40); in both #C and #Rust

   • Embed this notice
     Kris Nóva (nova@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:04:29 JST Kris Nóva

     in reply to

     C disassembly

   • Embed this notice
     Kris Nóva (nova@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:07:55 JST Kris Nóva

     in reply to

     Rust disassembly (much bigger)

   • Embed this notice
     Kris Nóva (nova@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:10:31 JST Kris Nóva

     in reply to

     fib.rs

   • Embed this notice
     Kris Nóva (nova@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:11:19 JST Kris Nóva

     in reply to

     fib.c

   • Embed this notice
     Kris Nóva (nova@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:11:52 JST Kris Nóva

     in reply to

     I realize this is not the most ideal way to share this - just grabbed some screenshots from the stream - I don't know if I have time to publish it anywhere.

   • Embed this notice
     Dan Ciruli (danciruli@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:32:53 JST Dan Ciruli

     in reply to

     @nova Ideal in that is was convenient for you, the content-producer, which is ideal indeed.

   • Embed this notice
     Dan Lacher (danlacher@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:44:10 JST Dan Lacher

     in reply to

     @nova I was watching the steam for a bit, great times. Two questions; what keyboard are you using and which editor?

   • Embed this notice
     iamericfletcher (iamericfletcher@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:49:37 JST iamericfletcher

     in reply to

     @nova hey - how do you get the execution time details?

   • Embed this notice
     iamericfletcher (iamericfletcher@hachyderm.io)'s status on Sunday, 11-Dec-2022 08:49:59 JST iamericfletcher

     in reply to

     @nova also, how do you print this disassembly info?!?

   • Embed this notice
     sporksmith :unicycle: :rust: (sporksmith@hachyderm.io)'s status on Sunday, 11-Dec-2022 09:20:36 JST sporksmith :unicycle: :rust:

     in reply to

     @nova This looks like it might be a debug build? For a release build I'm seeing:

   • Embed this notice
     Thomas Hurst (freaky@hachyderm.io)'s status on Sunday, 11-Dec-2022 09:35:46 JST Thomas Hurst

     in reply to

     @nova Note your C is using signed int, while your Rust is using unsigned.

     'unsigned/fib-gcc12' ran
     1.19 ± 0.05 times faster than 'signed/fib-gcc12'
     1.50 ± 0.01 times faster than 'unsigned/fib-rust168'
     1.60 ± 0.04 times faster than 'unsigned/fib-clang13'
     1.94 ± 0.06 times faster than 'signed/fib-rust168'
     1.96 ± 0.02 times faster than 'signed/fib-clang13'

   • Embed this notice
     Kris Nóva (nova@hachyderm.io)'s status on Sunday, 11-Dec-2022 10:10:35 JST Kris Nóva

     in reply to

     • sporksmith :unicycle: :rust:

     @sporksmith Good job!

   • Embed this notice
     Florent Vilmart (flovilmart@hachyderm.io)'s status on Sunday, 11-Dec-2022 13:43:51 JST Florent Vilmart

     in reply to

     @nova pretty sure it was asked before but what’s you terminal emulator?

   • Embed this notice
     John Stracke (metageek@techhub.social)'s status on Monday, 12-Dec-2022 03:26:46 JST John Stracke

     in reply to

     @nova How about comparing them using an iterative algorithm instead?

     Kris Nóva repeated this.
   • Embed this notice
     John Stracke (metageek@techhub.social)'s status on Monday, 12-Dec-2022 03:26:46 JST John Stracke

     in reply to

     @nova

     Rust: cargo build --release. C: gcc -O3.

     User space time from time:

     fib(45):

     R, recur: 1.886s, 1.778s, 1.831s
     R, iter: 0.000s, 0.000s, 0.001s
     C, recur: 0.908s, 0.951s, 0.948s
     C, iter: 0.001s, 0.001s, 0.000s

     fib(50):
     R, recur: 20.036s, 20.392s, 20.470s
     R, iter: 0.001s, 0.000s, 0.000s
     C, recur: 12.016s, 12.056s, 12.159s
     C, iter: 0.000s, 0.000s, 0.000s

     fib(60):
     R, recur: >1m
     R, iter: 0.001s, 0.000s, 0.000s
     C, recur: >1m
     C, iter: 0.001s, 0.000s, 0.000s

   • Embed this notice
     Hazel Weakly (hazelweakly@hachyderm.io)'s status on Monday, 12-Dec-2022 04:48:28 JST Hazel Weakly

     in reply to

     @nova if you want the galaxy brained version of the Fibonacci sequence in rust, I cobbled this one together to resemble some of the fun golf versions you can write in Haskell

     https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=e57e6b4d38b0d0bb49a2490c13435bbf