advent of code 2023

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > advent of code 2023

Dijkstracula: Mar 18, 2003; You can't spell 'vector field' without me, Professor!

I thought I'd try this out again this year since we're coming to the end of the semester and I need my computer-touching dopamine hit one way or another... imagine if you avoided your family and wrote some computer programs with me this month...

here is a private leaderboard that I just made if yall want to join: use 123188-4fade698 here

Dijkstracula fucked around with this message at 03:42 on Dec 1, 2022

# ? Nov 30, 2022 20:55

Adbot: ADBOT LOVES YOU

# ? May 2, 2024 18:10

post hole digger: Mar 21, 2011

im interested as long as its ok to be bad at programming to particpate. having just wrapped up a successful no dev ember i am ready to jump into the fray again.

# ? Nov 30, 2022 20:59

Dijkstracula: Mar 18, 2003; You can't spell 'vector field' without me, Professor!

oh yeah I am 100% not planning on staying up until midnight to be first on the board or anything like

# ? Nov 30, 2022 21:14

LanceHunter: Nov 12, 2016; Beautiful People Club

post hole digger posted:

im interested as long as its ok to be bad at programming to particpate. having just wrapped up a successful no dev ember i am ready to jump into the fray again.

In previous years people were going for really obscure/exotic ways to solve the problems (since the title for being 1st had mostly become over-optimized by the same type of people who camp leetcode competitions with lots of pre-written code). The first years I participated there was someone who was trying to solve every solution using only Excel.

# ? Nov 30, 2022 21:39

Archduke Frantz Fanon: Sep 7, 2004

im a terrible code toucher so I'd be down for some fun excercises rather than the endless repetition of code ninja or fart codez et al

# ? Nov 30, 2022 22:11

Dijkstracula: Mar 18, 2003; You can't spell 'vector field' without me, Professor!

LanceHunter posted:

The first years I participated there was someone who was trying to solve every solution using only Excel.

My original plan was to use this as an excuse to learn Lean but I suspect I should actually just be good and do it in rust, which is something I actually want more practice in

# ? Nov 30, 2022 22:52

lord fifth: Dec 26, 2019; LUCK ???

seems like a good excuse to touch computer to me

# ? Nov 30, 2022 22:58

DrPossum: May 15, 2004; i am not a surgeon

more like advent of chode

# ? Dec 1, 2022 00:52

Captain Foo: May 11, 2004; we vibin'
we slidin'
we breathin'
we dyin'

DrPossum posted:

more like advent of chode

no that's just when you walk in the room

# ? Dec 1, 2022 01:34

post hole digger: Mar 21, 2011

LanceHunter posted:

In previous years people were going for really obscure/exotic ways to solve the problems (since the title for being 1st had mostly become over-optimized by the same type of people who camp leetcode competitions with lots of pre-written code). The first years I participated there was someone who was trying to solve every solution using only Excel.

if i can get off my rear end and submit a solution to the problem at all i will consider it a huge W

# ? Dec 1, 2022 03:01

post hole digger: Mar 21, 2011

is there a leaderboard or how does that work

# ? Dec 1, 2022 03:01

RPATDO_LAMD: Mar 22, 2013; 🐘🪠🍆

there is a leaderboard, here is the one from 2021:
https://adventofcode.com/2021/leaderboard

however the scoring system is pretty stupid, it's purely a realtime race to submit first -- the first 100 people to submit a valid solution after the problem is posted (at 9pm GMT-8) get points, everyone else gets 0
so if that is midnight for you and you are not already sitting at your computer, when the clock strikes :00 you are definitely getting 0.
for reference, the #100 person on the leaderboard for problem 1 in 2021 submitted in 2 minutes 44 seconds.

you can also join private leaderboards. they use the same dumb scoring system but also just show a list of everyone who's joined and which problems they've successfully gotten the gold or silver star from which is maybe more interesting

# ? Dec 1, 2022 03:24

Dijkstracula: Mar 18, 2003; You can't spell 'vector field' without me, Professor!

here is a private leaderboard that I just made if yall want to join: use 123188-4fade698 here

# ? Dec 1, 2022 03:42

post hole digger: Mar 21, 2011

i joined, but maybe the real leaderboard is the friends we make along the way :allears:

# ? Dec 1, 2022 04:59

post hole digger: Mar 21, 2011

oh were all doing anonymous accounts instead of making github accounts for our SA usernames huh. fine

# ? Dec 1, 2022 04:59

lord fifth: Dec 26, 2019; LUCK ???

it already had me logged in and integrated and i didnt want to dox myself more than i already have on this site :3

im 2338265

# ? Dec 1, 2022 05:01

MrQueasy: Nov 15, 2005; Probiot-ICK

Lol, I have the same problem. Been using my realname github for work buddies and such since 2015.

# ? Dec 1, 2022 05:26

MrQueasy: Nov 15, 2005; Probiot-ICK

Created an alt that I'll post to after finishing my main.

anonymous user #2347253

# ? Dec 1, 2022 05:38

Jabor: Jul 16, 2010; #1 Loser at SpaceChem

gonna be trying this in rust again this year

hopefully it ends up being actual rust and not you-can-write-java-in-any-language rust

# ? Dec 1, 2022 09:38

Cybernetic Vermin: Apr 18, 2005

all ngn/k until it gets inconvenient. then probably python/c.

# ? Dec 1, 2022 11:41

Athas: Aug 6, 2007; fuck that joker

let's do this

# ? Dec 1, 2022 15:14

Dijkstracula: Mar 18, 2003; You can't spell 'vector field' without me, Professor!

here's me

Athas posted:

let's do this

" High-performance purely functional data-parallel array programming " :eyepop:

# ? Dec 1, 2022 15:59

LanceHunter: Nov 12, 2016; Beautiful People Club

Joined the leaderboard, and also looked at my GitHub page for the first time in a year (and realize now just out out-of-date it is).

Day 1 was definitely one of those problems that would normally have a really tight time limit if it were on leetcode or something, because there are a lot of really crunchy and inefficient ways to get it done. Doing Advent of Code I always forget that there aren't any such limits, so you can be as sloppy as you want as long as you get the answer. Then again, being sloppy in the first few days can come back and bite you down the line.

# ? Dec 1, 2022 16:11

PIZZA.BAT: Nov 12, 2016

i use this every year to teach myself a new language that i'm interested in. i've been hearing a lot about rust lately so i'm gonna give that a whirl

# ? Dec 1, 2022 16:18

MrQueasy: Nov 15, 2005; Probiot-ICK

PIZZA.BAT posted:

i use this every year to teach myself a new language that i'm interested in. i've been hearing a lot about rust lately so i'm gonna give that a whirl

Lot of good Rustaceans out there publishing hints and solutions. You're in good company!

I'm doing the "learn a new language" as a backfill this year. The joy of solving the problem as it's released just gives me too much dopamine, so I'll just continue in Kotlin (the language I think in the fastest) for now while re-implementing things in Elixir, which seems like a fun thing to learn.

# ? Dec 1, 2022 17:31

Pinterest Mom: Jun 9, 2009

LanceHunter posted:

Day 1 was definitely one of those problems that would normally have a really tight time limit if it were on leetcode or something, because there are a lot of really crunchy and inefficient ways to get it done. Doing Advent of Code I always forget that there aren't any such limits, so you can be as sloppy as you want as long as you get the answer. Then again, being sloppy in the first few days can come back and bite you down the line.

this is my first time doing aoc, does this mean that i can expect future problems to occasionally build on previous days?

# ? Dec 1, 2022 17:52

LanceHunter: Nov 12, 2016; Beautiful People Club

Pinterest Mom posted:

this is my first time doing aoc, does this mean that i can expect future problems to occasionally build on previous days?

It hasn't always been the case, but I remember it happening in the past.

# ? Dec 1, 2022 18:01

PIZZA.BAT: Nov 12, 2016

holy poo poo the compile error messages are actually extremely helpful

this is a first

edit: i'm building an elf object to store the total calories and a vector of their snacks because i know part 2 is going to need that somehow. i'm wandering around building the implementation function for adding a snack to an elf and didn't know how to get the self reference working in the signature

rust goes, 'uh hey it looks like you're trying to mutate the elf's state so maybe try adding mut to self here?'

and it worked! drat what a breath of fresh air

PIZZA.BAT fucked around with this message at 18:18 on Dec 1, 2022

# ? Dec 1, 2022 18:15

PIZZA.BAT: Nov 12, 2016

Pinterest Mom posted:

this is my first time doing aoc, does this mean that i can expect future problems to occasionally build on previous days?

they sometimes have multi-day problems but they're rare. most of the puzzles are stand-alone

# ? Dec 1, 2022 18:16

FalseNegative: Jul 24, 2007; 2>/dev/null

Python code:


import itertools as i,sys;f=open(sys.argv[1]);d=f.read();l=d.split('\n');e=(list(g) for _,g in i.groupby(l, ''.__eq__));ll=[[int(i) if i != '' else 0 for e in pair for i in e] for pair in zip(e,e)];o=[sum(l) for l in ll];o.sort(reverse=True);t=sum(o[0:3]);print(f'\nPart 1: {max(o)}\nPart 2: {t}\n')

Off to a great start.

# ? Dec 1, 2022 19:08

FalseNegative: Jul 24, 2007; 2>/dev/null

Pinterest Mom posted:

this is my first time doing aoc, does this mean that i can expect future problems to occasionally build on previous days?

I think the biggest example of this happening was in 2019. There was a device called the INTCODE machine that processed program input. It essentially involved writing an emulator for this INTCODE machine over the course of many different days. For example, some instructions would be added / indirect memory addressing / jumps as different puzzles.

This all culminated in a puzzle that required mapping the memory to video and handling user input! One puzzle involved playing a breakout game within in it (the puzzle input was the .bin input for the emulator to run). You had to beat the game and use the score as the solution.

Totally wild, but I think it made the whole year very unapproachable and they stepped back from doing puzzles that intertwined or requiring the use of external libraries like SDL.

# ? Dec 1, 2022 19:26

Cybernetic Vermin: Apr 18, 2005

FalseNegative posted:

Python code:

import itertools as i,sys;f=open(sys.argv[1]);d=f.read();l=d.split('\n');e=(list(g) for _,g in i.groupby(l, ''.__eq__));ll=[[int(i) if i != '' else 0 for e in pair for i in e] for pair in zip(e,e)];o=[sum(l) for l in ll];o.sort(reverse=True);t=sum(o[0:3]);print(f'\nPart 1: {max(o)}\nPart 2: {t}\n')

Off to a great start.

if we're doing python crimes i'll play (granted not nearly as many fancy bits, but plenty pointlessly condensed).

Python code:

el=sorted([sum([int(x.strip()) for x in l.split()]) for l in sys.stdin.read().split("\n\n")])[::-1]
print(f'part 1: {el[0]}\npart 2: {sum(el[0:3])}')

i'd post my actual ngn/k solution, but it'd unnecessarily upset everyone.

Cybernetic Vermin fucked around with this message at 19:43 on Dec 1, 2022

# ? Dec 1, 2022 19:32

LanceHunter: Nov 12, 2016; Beautiful People Club

FalseNegative posted:

I think the biggest example of this happening was in 2019. There was a device called the INTCODE machine that processed program input. It essentially involved writing an emulator for this INTCODE machine over the course of many different days. For example, some instructions would be added / indirect memory addressing / jumps as different puzzles.

This all culminated in a puzzle that required mapping the memory to video and handling user input! One puzzle involved playing a breakout game within in it (the puzzle input was the .bin input for the emulator to run). You had to beat the game and use the score as the solution.

Totally wild, but I think it made the whole year very unapproachable and they stepped back from doing puzzles that intertwined or requiring the use of external libraries like SDL.

That�s the year I was thinking of. It was the second year I�d ever done AoC and I washed out extremely quickly.

# ? Dec 1, 2022 20:10

Dijkstracula: Mar 18, 2003; You can't spell 'vector field' without me, Professor!

man that 2019 one sounds fun but keeping up with it every day sounds tough, as opposed to my plan (weave in and out depending on how much egg nog I've consumed on any given day)

# ? Dec 1, 2022 20:11

FalseNegative: Jul 24, 2007; 2>/dev/null

Cybernetic Vermin posted:

i'd post my actual ngn/k solution, but it'd unnecessarily upset everyone.

Please do, j k, etc fascinate me.

# ? Dec 1, 2022 20:14

Share Bear: Apr 27, 2004

I'm in, I did this in Rust last year to try and sharpen that up and I'll suppose I'll do that again.

# ? Dec 1, 2022 20:31

lord fifth: Dec 26, 2019; LUCK ???

i started out in c but i feel like thatll become unwieldy pretty quickly. maybe i'll switch to ocaml or scheme or something

# ? Dec 1, 2022 20:52

PIZZA.BAT: Nov 12, 2016

rust update: day one is done but man i forgot what a pain it is to not have a garbage collector cleaning up behind me

this is a decent system though compared to manually allocating/freeing every drat thing you touch

# ? Dec 1, 2022 21:32

the talent deficit: Dec 20, 2003; self-deprecation is a very british trait, and problems can arise when the british attempt to do so with a foreign culture

the 2019 one wasn't bad. you could write the emulator in like 30 minutes; it was very simple. you also didn't need to use sdl or any kind of graphics. the breakout puzzle was easier to solve analytically than by actually playing the game

# ? Dec 1, 2022 21:40

Adbot: ADBOT LOVES YOU

# ? May 2, 2024 18:10

spankmeister: Jun 15, 2008

did day1 in python in like 5 minutes but gonna try rust now

# ? Dec 1, 2022 21:53

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > advent of code 2023

«‹›15 »