/aoc2016

https://adventofcode.com/2016

Primary LanguagePython

advent of code 2016

https://adventofcode.com/2016

stream / youtube

about

for 2016, I'm planning to implement in python

timing

  • comparing to these numbers isn't necessarily useful
  • normalize your timing to day 1 part 1 and compare
  • alternate implementations are listed in parens
  • these timings are very non-scientific (sample size 1)
$ find -maxdepth 1 -type d -name 'day*' -not -name day00 | sort | xargs --replace bash -xc 'python {}/part1.py {}/input.txt; python {}/part2.py {}/input.txt'
+ python day01/part1.py day01/input.txt
246
> 434 μs
+ python day01/part2.py day01/input.txt
124
> 660 μs
+ python day02/part1.py day02/input.txt
76792
> 1852 μs
+ python day02/part2.py day02/input.txt
A7AC3
> 1795 μs
+ python day03/part1.py day03/input.txt
862
> 3581 μs
+ python day03/part2.py day03/input.txt
1577
> 5994 μs
+ python day04/part1.py day04/input.txt
185371
> 16968 μs
+ python day04/part2.py day04/input.txt
984
> 11186 μs
+ python day05/part1.py day05/input.txt
c6697b55
> 8800 ms
+ python day05/part2.py day05/input.txt
8c35d1ab
> 39564 ms
+ python day06/part1.py day06/input.txt
agmwzecr
> 2039 μs
+ python day06/part2.py day06/input.txt
owlaxqvq
> 2138 μs
+ python day07/part1.py day07/input.txt
118
> 25829 μs
+ python day07/part2.py day07/input.txt
260
> 39286 μs