Kata 📎

@movq@tilde.zone

Interests fluctuate. #Retrocomputing, #Retrogaming, #UT99, #FreeSoftware.

Does #AdventOfCode in December.

Posts auto-destruct after 90 days.

#nobot #noai

Passierschein A38
✅ vorhanden
LANG
de_DE.UTF-8, en_US.UTF-8
3Posts
0Following0Followers
Kata 📎@movq@tilde.zone

Replying to @movq@tilde.zone

Okay, this is looking good.

I took my #AdventOfCode solutions from 2024, which I wrote for IBM Java 1 on OS/2 Warp 4, and gave them a try on Sun Java 1.3 on Windows 95. The build times feel a bit slower (I haven’t measured them, though) but the actual runtimes are a lot faster. 😲

That’s at least a factor of 2-3, sometimes more. The difference on day16/1 is huge – I remember that this day was very memory intensive and IBM Java’s garbage collection was struggling.

(And yes, that is LibreOffice Calc, because I, too, sometimes can’t be bothered.)

A LibreOffice Calc plot of this CSV file: Puzzle,Runtime IBM Java 1 (OS/2 Warp 4),Runtime Sun Java 1.3 (Windows 95) day01/1,0.906,0.44 day01/2,0.933,0.28 day02/1,1.542,0.49 day02/2,1.531,0.33 day03/1,0.467,0.44 day03/2,0.38,0.33 day04/1,0.174,0.33 day04/2,0.142,0.17 day05/1,2.362,0.88 day05/2,2.556,0.44 day06/1,0.127,0.22 day06/2,47.859,15.16 day07/1,2.284,0.66 day07/2,53.917,28.61 day08/1,0.196,0.17 day08/2,0.182,0.11 day09/1,0.223,0.39 day09/2,0.371,0.6 day10/1,0.222,0.16 day10/2,0.211,0.11 day11/1,1.597,0.44 day11/2,61.869,6.65 day12/1,14.712,4.06 day12/2,20.263,5.33 day13/1,2.76,0.99 day13/2,3.691,0.49 day14/1,0.607,0.33 day14/2,27.467,13.84 day15/1,0.18,0.33 day15/2,0.187,0.17 day16/1,127.604,3.35 day16/2,331.811,125.07 day17/1,0.05,0.06 day17/2,1.871,0.33 day18/1,7.717,0.82 day18/2,24.513,1.76 day19/1,8.927,2.47 day19/2,25.961,8.57 day20/1,192.173,43.28 day20/2,212.876,44.44 day21/1,94.446,6.87 day21/2,0,0 day22/1,21.961,16.7 day22/2,0,0 day23/1,0,0 day23/2,0,0 day24/1,0,0 day24/2,0,0 day25/1,13.309,0.88 day25/2,0,0
ALT