File: calendar.png (3.25 MB, 9777x3472)

3.25 MB PNG

official /aocg/ calendar

>>107523907
eric has lost the magic

we could have ended /aocg/ on a high note in 2024. Instead we got this. What a joke.

File: Screenshot 2025-12-12 080730.png (34 KB, 530x565)

34 KB PNG

what a letdown
naive solution that avoided the problem was sufficient and there was no part 2

Man I'm so mad that this puzzle was easy. I'm crashing out. I'm experiencing gamer rage.

File: 1746936943156824.png (372 KB, 1272x1562)

372 KB PNG

I can't believe this worked. I though eliminating obvious cases before actually trying to fit shapes - which I have no idea how to do - and just entered the value as a meme. And I got the star lmao. This would've filtered me for sure if inputs were actually challenging. Spent more time in parsing then writing the solution.

File: whoville2.gif (407 KB, 500x369)

407 KB GIF

>https://files.catbox.moe/3z365u.mp4
God bless us, everyone!

>>107523907
To the anon making obonofcode. Try your best to make it AI resistant so it at least can't one shot it, but you'd need to work through step by step with AI to solve.

I think AoC is dead after this year anyway.

File: aoc_2025_day12_pythiomatic.png (163 KB, 1668x846)

163 KB PNG

Verbose sol'n.

>>107523907
kek I'm enjoying the ai slop. artists btfo. also the last question was based.

I might've liked this problem more if there wasn't a genuinely NP-hard problem that was actually real and didn't have any trick to it just two days ago
I'm coming into today's problem thinking that I really do need to solve it as described but then it ends up being a quirk-of-the-input day (a quirk that the example doesn't include btw)
there's no moment where I think "but eric wouldn't REALLY expect me to solve this, right? I'd better take a look at my input very closely to be sure" if that exact thought led nowhere just two days prior
instead, I randomly stumble upon it when halfway through writing a pruning DP recursive who-cares algorithm, and I have to trash the whole thing for the quick and easy five line solution
my work feels unfinished, like I cheated or gave up and imported solution

>>107524021
this, AI OC is better than no OC

>>107524026
my thoughts as well. If I wasn't pseudo-filtered from the "import z3" day I wouldn't even have considered trying to solve this for real, especially next to the rest of the puzzles this year. I guess I thought he would give us 2 hard puzzles rather than 1, silly me.

>>107524026
>anti-import cels seething
based eric

>>107523955
I'll do a proper solver later today, just out of spite

>>107524059
part 2: sum the minimum area you can fit each set of blocks into.

>>107524059
wdym proper? I knew tiling was a NP hard problem but I was still trying to make something work. fuck you eric, again

> only 22 hours to go

HOLD THE LINE

pythiomatic python solution

print(
    sum(
        int.__mul__(*map(int, region[:-1].split("x")))
        >= 9 * sum(map(int, present_counts))
        for region, *present_counts in map(
            str.split, open(0).read().strip().split("\n\n")[-1].splitlines()
        )
    )
)

>>107523914
you do good anon <3

missing c23 anon :<

>>107524070
it might be NP-hard but pretty small inputs.
oth those that can pass have so much spare space that I think anything will find one solution quickly.
so I dont really know what to solve.

>>107524095(Me)
pythiomatic python solution: hacker's edition

print(
    eval(
        "("
        + open(0)
        .read()
        .strip()
        .split("\n\n")[-1]
        .replace("\n", ")+(")
        .replace("x", "*")
        .replace(": ", "//9>=")
        .replace(" ", "+")
        + ")"
    )
)

are you shitting me? there's really no case where sizes fit total but can't be stacked properly?

i guess it's supposed to be a split challenge so you can play with constraint solving if you want more but still it's weird

>>107524152
lol

>>107524152
holy based

File: sadcowboy.jpg (7 KB, 220x229)

7 KB JPG

>over already
Shit. I'm gonna miss you guys. I look forward to these threads all year long. It's the only time the actual, genuine programmers on /g/ congregate.

>>107524152
Rustroons real quiet after this dropped

>>107524157
>are you shitting me? there's really no case where sizes fit total but can't be stacked properly?
don't even have to do that, can just go full retard and count whether the sum of all of the points needed between all of the shapes is greater than or equal to the area. fails the test case but clears the input

File: drink.jpg (15 KB, 480x360)

15 KB JPG

So, now what do I do? Leetcode daily problems? Old AoC events? An actual, honest-to-God programming project, not just toy puzzle programs?

File: I_KNEEL.png (403 KB, 600x906)

403 KB PNG

>>107524152

File: weredone.png (583 KB, 1080x607)

583 KB PNG

>>107524212
day 13 tomorrow, of course.

>>107524212
look forward to this anon's project ig https://obonofcode.com
other than that, https://atcoder.jp/contests/ beginner contests have a fixed schedule on saturday.
programming project would be nice too. I can just never get to old aoc problems. they feel out of season

going to enjoy getting a full night's sleep for once

>>107524212
dbpro.app/advent-of-sql

learn sql

>no maze
>no recursive parsing
>no image creation/finding
>no reverse engineering
>no language interpreting
>no cellular automata

>>107524013
>make it AI resistant
hide the niggerword throughout your text and ai will refuse to process it

>>107524195
Yeah same, these threads are the only evidence I've seen in a long time of /g/ users having any level of technical knowledge or ability at all.

>>107524259
image creation puzzles are inaccessible for blind programmers, please understand

File: seethe level, maximum.jpg (107 KB, 636x800)

107 KB JPG

shant do day 10
shant do day 12
shall send hatemail to eric

File: fuck you eric.png (94 KB, 800x540)

94 KB PNG

BORING

>>107523907
what are you making these with sloppa-anon

AND WE END WITH A FUCKING HOW MANY BUCKETS DO YOU HAVE PUZZLE!
THANK YOU ERIC FOR RUINING CHRISTMAS YET AGAIN!
This was the buckiest year of all years.

File: carbon.png (154 KB, 598x1052)

154 KB PNG

ERIC YOU FUCK holy FUCKING SHIT

Day   -Part 1-   -Part 2-
 12   02:17:54   02:18:08

>realize it's an Algorithm X/dancing links thing
>spend like an hour trying to see if there's some library to import solution
>yes but only in the depths of python dependency hell
>okay fuck it I'll write my own
>spend another hour studying the algorithm
>start to write it
>wait, before I do, let's see how many lines I can just skip
>just sum the raw tile count and see if any are too big for the box
>pretty sure every line would pass
>it would be stupid to leave any in the input that were obviously too big, right?
>run code
>476? wtf?
>try submitting it
>That's the right answer!
>WHAT
>Okay so what's part 2?
>there is no part 2
ERIC
Eric you nigger
I'm so fucking pissed

>>107524293
I'm so sorry, Anon

>>107524152(Me)
pythiomatic python solution: trans edition

print(
    eval(
        "("
        + open(0)
        .read()
        .strip()
        .split("\n\n")[-1]
        .replace("\n", ")+(")
        .replace(":", "//9>=")
        .translate(str.maketrans("x ", "*+"))
        + ")"
    )
)

>>107524196

Eric, you're a madlad. I love you man. What you managed to pull on us on the last day is incredible. I don't remember the last time I laughed so hard and was so happy at making something work in a long time. You've actually made me belly laugh. Thank you. Truly thank you, what a fantastic way celebrate 10 years of AoC, what a way to go. I can't stop saying thank you, because the joy your work brings, and the absolute troll today, it's phenomenal. Marry Christmas, and a Happy New Year! Stay safe, everyone.

>>107524322
reddit post

>>107524322
holy shit lynch all redditors for real

>>107524326
4chan post: >>107524333

>>107524333
holy shit is this a recursive dp

File: 1719450042673550.jpg (51 KB, 1032x774)

51 KB JPG

>>107524333

>>107524333
hahaha

>>107524333
checked

How long would it take to check every possible arrangement in every possible location?

>>107524333
>and trips on top of it
holy based it's a christmas miracle

>>107524365
a long time. Hope this helps!

>>107524377
Hmmm, sounds painful.

>>107524365
largest grid size is 50x50
there's around 200 boxes per line on average

worst case scenario (if all the boxes were 1x1 tiles) each box would have (50x50) = 2500 valid positions
the second box would have 2499 remaining positions for all the first box's positions, so two boxes make (2500*2499) = 6,247,500 positions
with 200 boxes, there's a total of (comparatively a little less than) 2500^200 positions, which is about ~10^679 possibilities
obviously too many to check each one individually

there's a lot of pruning possible, considering the boxes are closer to a size of 3x3 and won't tile perfectly, and identical box types don't need to be tiled in locations that the previous box type has already tried (because that would be an identical state)

and here I thought we might have an interesting problem... FUCK YOU ERIC YOU FUCKING HACK

>>107524292
kek bucket-broken

>>107524417
What about rotations?

File: 20151218_200346-BlendColl(...).jpg (78 KB, 1200x1200)

78 KB JPG

Eric here. You tards spent more hours trying to rotate and place shapes?! Get rekted chuds! HAHAHAHA. Merry Christmas you filthy animals!

i think tomorrow will be the great filter day, things usually pick up after the first half of calendar

>>107523428
>solve eric's puzzles and give us the solutions
>do it in RTL so we can use your work for publications or other bullshit
no fucking thank you, jane street. my day time job is RTL, I'm not doing AoC in RTL as well. That wouldn't help me do anything but waste time implementing solutions that should never be done in RTL

>>107524481
trvke

Did anyone actually write a full backtracking broot, but with the dumb checks to skip it when you get lucky (if enough 3x3 tiles, say yes, if not enough 1x1 units, say no)?
Or did everyone just assert (or hope) that those conditions are enough?

File: 55123.png (32 KB, 678x288)

32 KB PNG

thoughts on this shape?

>>107524505
completely normal if you look at previous years

>>107524505
>more day 11 golds than day 10 golds
uh oh filty

>>107524505
what size grid does this fit into?

Would anyone happened to have a big zip of every bigboy? I wanna go back and do them optiomally.

File: 1762449550268313.png (30 KB, 375x97)

30 KB PNG

>>107524531
my sophisticated move elements around in ms paint algorithm found this 40x5 solution

>>107524533
https://github.com/Error916/Aoc2025

>>107524569
>day1.7z
>day2.7z
>day3.7z
>day4.7z
>day5.7z
>day6.7z
>day7.7z
>day8.7z
>day9.7z

>>107524598
Cluster virus.

1
>>107394076
>>107394087

2
>>107408306
>>107408317

3
>>107425550
>>107425561

4
>>107439676
>>107439688

5
>>107450786
>>107450806

6
>>107464449
>>107464456

7
>>107474326
>>107474337

8
>>107487388
>>107487394

9
>>107499472
>>107499480

10
>>107511949
>>107511958

>>107524212
>Old AoC events?
Literally every AoC year is significantly higher-quality than this one.
Also Everybody Codes. I hope the theme next year is something a little less gay than dragonducks.

AOC flavor of C++23:

#include "aoc.h"
MAIN
    p1 = +((in / "\n\n").back() / '\n' * λx(
        *((x | ": ").h / 'x' * λx(+x)) >= +((x | ": ").t / ' ' * λx(+x * 9))
    ));
END

>>107524689
Gay spiral dick dragonduck rape could have been a fun puzzle backdrop, though.

File: day2025real.png (3.31 MB, 1494x5070)

3.31 MB PNG

I made an actual solution to the problem that uses a SAT solver to see if a placement is feasible or not because the solution felt unsatisfying.

The idea is every possible placement (top left corner) per rotation per piece is a boolean (~60k). Then for each spot add a constraint that the sum of booleans (placements) that cover that position is at most 1. And another constraint that for the sum of booleans (placements) per piece type is equal to exactly the required amount

``` python d12.py 2235.57s user 55.69s system 159% cpu 23:58.30 total ```

>>107524322
Go Fuck Yourself

>>107524753
>import solution
you did not solve it

>>107524663
>>107524668
christmas saved, come back with last two days draw anon

File: 1536195029090.jpg (44 KB, 542x537)

44 KB JPG

>>107524753
>```

>>107524753
pretty cool. I also thought of posing it as a BSAT problem but I did not have much experience with it. I also thought of flattening the grid as a bitmask (int) and then turning presents into bitmasks too. then shift the presents and bitwise or with grid if no common bits for that shift.

so eric gave an example where just checking if the number of tiles is less than the dimensions doesn't work, but then that's what he expects us to do

>>107524533
https://0x0.st/KC22.tar.zstd

File: out.png (177 KB, 2969x1805)

177 KB PNG

I did not enjoy this year

File: file.png (1.43 MB, 4005x3298)

1.43 MB PNG

>>107524790
problem?

>>107524790
bravo eric

>>107524794
>.ztsd
nice virus

>>107524790
This isn't the first time where solving the examples is considerably harder than the real input, but it's an uncommon pattern that I wasn't expecting on the last day at all.

>>107524785
I considered the bitmask solution too, but back of the envelope even the last example is on the order of 10^12 permutations, checking each one (like backtracking), 10^6 ops checks per second still takes like 10 days

It is kind of import solution but modeling it as a SAT problem in a solver is actually pretty intuitive and the docs are good, I used CP SAT which does make it pretty easy

>>107524761
Sorry for not solving it in my own custom language

>>107524850
usually when he does it, you can look at the inputs and realize it's a specific case like a dag or whatever, in this case there is no way you can "prove" that the heuristic works because nothing is obvious

>if a grid doesn't have enough squares for the contents of all presents, reject it
>if a grid is large enough to fit all presents without overlap, accept it
>there is nothing left over that is not one of these in the real input
Is that all?

>>107524070
You can actually solve it for real with constraint programming. I finished writing it then came here only to find out that you can do a dumb retarded solution. Very cool. It takes a few minutes per region.
I don't know why he didn't just make us do the actual problem but make the inputs such that they could all be solved in around an hour instead of 1.5 days.

File: 3e5f301a91c448c89ea336913(...).png (2.71 MB, 1024x1536)

2.71 MB PNG

>>107524800
tried to make it edit trollface onto the eric picture, this shit is demonic

>>107524984
As much as I hate low effort AI shit, I miss the edits of Eric's face, especially AI videos.

Last year there was this edit of Eric jumping into a bunch of big titty black gfs and laughing at the viewer that I enjoyed.

File: file.png (38 KB, 648x597)

38 KB PNG

this is so coal

>>107525091
import solution niggers BTFO

>>107525091
He should've made more interesting puzzles worth solving

>>107525091
kek

>>107525091
does he know that the whole point of csp solvers is to speed up broots?

File: file.png (61 KB, 333x340)

61 KB PNG

>>107525091
um its different because uhh uhhhh it just is

>>107525091
If you don't do these on pencil+paper you personally made from a tree you cut down using tools you fashioned by banging rocks together, you didn't solve the puzzle

>elves are rushing to put presents under the tree
>almost two weeks before Christmas
hmmmmm

>>107525150
there is literally no way Eric knows what integer programming is if he doesn't know the chinese remainder theorem

>day 12
>Elapsed: 93.460545191s
>You got a gold star!

fear my bruting powers, eric

>>107525151
he's not even trying anymore

>>107525091
> solve Ax = b
> positive integer solutions only
> u cant use an ILP
I hope the gasline to your house blows up

>>107525174
wait I could have just summed the tiles up? FUCK YOU eric I actually went through my input and modified it to get an easier solution to just brute place into each grid

I'm glad I spoiled it by reading the thread.
I like to make sure I have the example working before I even touch the input file and in this case the example did require more than just adding up the size of the presents.
So basically if I hadn't read the thread I'd be working on this for hours feeling like a retard for not being able to figure it out while everyone else gets their stars.
Not cool eric.

>>107525196
filtered

> read day 12
> wtf, how do I even approach it
> check if all trees can potentially have all listed presents, like if all # can fit without accounting for shape
> they don't
> check how many can to start with this number
> try to enter it as an answer just in case
> mfw it is the right answer
Ok, this was funny. But how do I solve it in a legit way?

File: asmlets BTFO.png (125 KB, 1515x322)

125 KB PNG

>>107525091
Why did you edit the original?

>>107525357
>advent of *code*

>>107525369
Assembly *code*. Original *coders* used punch cards, and quite honesty if you need more you're effectively cheating.

>>107525379
Original programmers were women.

>>107524013
Why bother? If you don't want AI just don't use it?

>>107525397
Original COMPUTERS were women.
Programmers were men.

>>107525091
nice discord filename, faggot

>>107525451
Takes one to know one. I'm so straight I don't even know what you mean

File: 1762756283357137.jpg (385 KB, 1500x1384)

385 KB JPG

>>107525348
I don't know what the "legit" answer is but I didn't do it that way and got a pseudo algorithm. If you want a "real" algorithm look up NFP and Irregular 2D Packing. After a bunch of research I decided it was too much work so tried a naive solution and got it.

I ended up editing my input data after noticing some patterns. My starting shapes were:

0:  1:  2:  3:  4:  5:
### ##. ### ### ### ..#
### .## ..# .#. ##. .##
..# ### ### ### #.. ##.

(these blocks might be fucked, kind of guessed the spacing, but you get the idea)
Noticing that some of the shapes fit together nicely I created some new shapes:

0,1: 2,2: 3,3:  4,5:
###  .###  .#.# ####
###  ####  .### ####
###  ####  #### .###
.##  ###.  ###.
###        #.#.

where the new shapes are combinations of the base shapes. These ones tend to be more efficient for packing, so then it's just a matter of counting each shape as incrementing the base shapes one each (or twice in the case of 2,2 and 3,3). My packing algorithm was dead simple, just pick the top left spot in the grid that it can be placed, weighted top over left. In terms of what shapes, it prioritized by absolute bounding box first (something I picked up from reading a paper about 2d packing) and places as many as it can/needs to then just moves down. It takes about ~2 seconds to run but I'm proud of my spaghetti nonetheless. NFP would have involved figuring out the optimal spot for each shape in each iteration so I just opted to brute force the location by iterating location per location until an open spot was found. As it turns out, rotations are completely irrelevant when you do it this way, I made some code to calculate rotations and never even used it - rotating the shapes to create the new more efficient ones was all that was necessary. Any stragglers not picked up by a multi shape ended up just getting placed without rotation or thought and it still worked.

>>107525300
i gave up within 10 minutes of reading it
good thing im so lazy or i might have wasted time on this nonsense

>>107525348
>how do I solve it in a legit way?
Write a backtracing brute that just tries every combination
But before calling it, check some simple cases if it is certainly possible or certainly impossible.
The example will call the brute, the actual inputs won't.

>>107525091
import solution cucks eternally BTFO

File: file.png (180 KB, 1225x1230)

180 KB PNG

I don't want to admit it but I think d10 has filtered me

>>107525640
>When I was a child, I used to speak like a child, think like a child, reason like a child; when I became a man, I did away with childish things.

>>107525678
what

File: you.png (394 KB, 720x720)

394 KB PNG

>>107525694

File: aoc2025-12.png (153 KB, 554x987)

153 KB PNG

Sneaky sneaky
I spent an hour and a half thinking about ways to approach this. Then I noticed two lines in my input which had basically the same region size, but one had a way higher present count
Uh, Merry Christmas everyone

>very slow on sample input
>very fast on real input
why have you done this, eric?

>>107525743
you're almost there. think dumber not harder.

>>107525743
my code doesnt even work for the sample input but it works for the real one

>>107524455
your butthole is unprotected

What language should I learn? I used to use Crystal, but development on that language is dead.

File: IMG_7175.jpg (127 KB, 800x600)

127 KB JPG

We should all do a previous year together now. As long as we agree on a year, it won’t feel stale and lonely.
Who’s with me?

>>107525875
Rust of course

>>107525893
I am

>>107525707
Im using all the colors

>>107525893
but we already did those

File: Screenshot_20251212_145047.png (12 KB, 562x87)

12 KB PNG

Intended solution

well now that this shit is over I will now check the new warframe update

the year of np complete problems

>>107525893
I am. in for anything pre 2020.

>>107525875
try zig

>>107525640
anon see that smart redditor's solution and impl it. 12th day is still not over yet

>>107526208
>try zig
Can Zig look elegant?

I had the idea last night to try something like simulated annealing. Randomly place pieces, consider overlaps as high energy, do Monte Carlo simulation to lower total energy, see if you get into a totally valid state. Dunno if it would be efficient enough. In the cold light of day, I don’t know if I have the motivation to implement it. I just feel hollow and empty now that it is all over in mid-December.

I should be making this kind of post on Christmas morning with you fuckers, not a fucking work day.

>>107526215
depends on how you define elegance and whether you give it an honest try. it might not be for everyone and I avoided it for years because how jank its syntax felt. but reading a bit about it started making a lot of sense. and it's definitely not as vast as rust, in which I still haven't gained confidence after years of programming on and off. I tried zig for aoc this year and it was very painless.

>>107526245
tragic indeed, anon. about today's problem, if an arrangement is trivially possible then it will be possible using whatever tiling method you use. because it will never be worse than tiling all presents separately in 3x3 subgrids, unless you deliberately make it. and if it is not trivially possible for a tree/case then it will not be possible using whatever algorithm you use because the grid will not have enough squares that are required for all the presents (you can verify this by just doing h*w >= sum of no. of # in all presents).
so any solution randomized or greedy will not yield any benefit for the given sets of inputs. for more random cases it might lead to nice visualizations but you cannot guarantee optimal packing. check the other anon's bsat solution >>107524753

>>107526108
>problem(s)
>plural
name 12

>>107525893
2019

>>107525893
I did all years except 2020

>>107526334
10,12,10,10,12,10,12,12,12,12,10,10

>>107526245
I've implemented simulated annealing before.
It's powerful at quickly finding a very good arrangement, but is very unreliable at finding the true optimum unless you spend ages fine-tuning it.
It could work, but I feel like it would be littered with incorrect results that just barely don't fit because of bad luck.

This reminds me of Jewish problems https://arxiv.org/abs/1110.1556
Is Eric antisemitic?

>>107526108
>NPC problems
No
>Problems for NPCs
Yes

>>107526351
only 10/p2

>>107526378
both parts are np complete chuddie and today is np complete too

I gave up on day 2

File: __hiiragi_kagami_izumi_ko(...).jpg (224 KB, 1838x1034)

224 KB JPG

Thanks for another amazing Advent of Code, guys!

If you've added your email to obonofcode.com I'll be sending you updates mid-January!

Can't wait to see you all again soon!

>>107526464
>tranime
no thanks

>>107526500
and yet you slopchuds love making stuff look anime in a last-ditch effort to make it look passable

>>107526500
anime website

>>107526520
I'm not a slopchud though

>>107526528
you are in my headcanon

>>107526372
not with those advertisers

>>107526527
he said 'tranime'. he didn't post Cowbop Bebop or even The Dirty Pair or something, he posted Lucky Star.

File: anime.jpg (116 KB, 550x340)

116 KB JPG

>>107526527

>>107524663
>>107524668
will miss you drawanon <3

wish slopanon would make a compilation too. I ignored them at first but they started growing on me. very quirky. would like to how the anon generated them

>>107526527
nope, >>>/a/

>>107525348
My thoughts are to use this solution as part of the broot.

(total area - total #) is your degrees of freedom

then as you broot place the blocks, they will cut off squares and make them inaccessible, reducing the degrees of freedom, and you can prune the whole branch once it goes below 0, which it will very early on in the broot for failed cases.

>>107526387
just shut up, amerifat. you are wrong, accept it

>>107526387
today was indeed no problem complete in 5 minutes.

File: redditwew.png (54 KB, 873x210)

54 KB PNG

the most absolute state absolute state post i think i ever seen
fuck this problem, the state space was insane

>>107524333
impressive, very nice

>>107525893
i'm in

>>107526387
day 10 part 1 is bounded by 2 to the power of n (max light states) and the problem is solved by finding the shortest path in an unweighted graph in linear time, not np complete, retard.

it's over
what now
I'm just startic my holidays now btw

>>107526867
part 13 is coming out tomorrow. i have eric tied up in my basement with electrodes on his nips, and if he WILL give me his password to reveal the rest of the puzzles

>>107526867
now you write a real program

>>107527200
A program to do what?

I don’t know how to make things, or even what things to make. I just like puzzles.

>>107527200
but I just came back from work

eric should've quit AOC on 10th anniversary instead of this half-assed attempt at continuity
even one insane puzzle per year would be better
most fun was in the second half, it felt like a chore this year

File: d12.png (90 KB, 838x517)

90 KB PNG

k.

>>107527249
>A program to do what?
whatever the hell you want
>>107527262
and now you can program something useful for yourself

>>107527478
post your git

>>107527478
> whatever the hell you want

Yeah, but, like, what do I want? I have no idea. I’m not creative. I will just end up doing nothing because I have no ideas.

>>107527546
rewrite a c/c++ codebase in Rust

File: Screenshot_251212_103549.png (33 KB, 500x194)

33 KB PNG

>12 hours to day 13

did you ever doubt

File: GE43AKC.jpg (127 KB, 858x950)

127 KB JPG

Solving day 12

>>107525640
Day 10 was stupid. It was literally all "meh too easy" and then they throw a "fuck you, linear programming, nigger" filter on day 10.

does anyone else have a day 13?
my puzzle won't be ready on the hour, so if someone could post their day 13 and i will post on day 14?
otherwise i will post mine an hour late or so

File: Untitled.png (28 KB, 453x467)

28 KB PNG

Lame year, merry Christmas, I guess...

>>107528512
>10 minutes for day 1 p2
Anon, i..

>>107528138
i just won't do day 10, i'm glad p2 got spoiled for me so I can just refuse to do it

>>107528412
I might have something, but I usually only wake up a few hours after the normal time.

>>107528412
i have one but i don't think /g/ is up for it, it's hard

File: instathotzmodeljeetlora.png (474 KB, 2005x893)

474 KB PNG

>download instathot jeet z-model lora from civitiai
>download example images for WF
>author has FluxGuidance node, it doesn't make any changes if you use or not
>author made up a schizo negative prompt, he connects it to ConditioningZeroOut node


Why are they are like this?

See you guys next year if this bitter, fun-hating, mindlessly abrasive wasteland of screencapped ragebait from other social media sites is still limping along by then. Merry Christmas!

>>107527546
just make a game and render using your favorite terminal emulator and a nice font.

>>107528581
where is yours?

see you next year

im not leaving

File: 45345243423.png (6 KB, 60x60)

6 KB PNG

Anyone is interested in/tried out everybody(.)codes? Someone shilled it in the AoC threads this year. Since AoC is only 12 days now, I think I might try it next year, it begins in November.

can someone generate me an elf picture to post along with my bonus puzzle?
if possible an elf holding cubes numbered 1-4

>>107528904
do the ones from this year

>>107528695
Man, I’ve been trying and failing to make gaem for years. I have no ideas and can’t into art.

>>107528725
in yer mum

File: aoc-day-12-sloppa-2.jpg (1.08 MB, 1408x768)

1.08 MB JPG

Daily sloppa

>>107524279
>>107526645
I used Nano banana pro. https://gemini.google/overview/image-generation/ It really didn't take much from me, it's pretty smart.

If thread is still up when I'm done with work I'll post compilation links

>>107529102
can you make me slop? >>107528920

>>107528904
I checked out day 1 a while ago but I'm severely allergic to AI prose
I don't even care if they use AI for anything else but I don't want to read LLM fiction

>>107529088
>he can't btfo my casual times
Pajeet, I....

File: 1764524929691031.jpg (663 KB, 1408x768)

663 KB JPG

>>107529122

>>107529253
many thanks, anon

>she doesn't have all stars from all years yet
Get to work
https://adventofcode.com/2015/day/1

>>107528904
i didn't do everybody codes; i'll try to do all of their available puzzles before Christmas

File: 1745110854719708.png (1.53 MB, 903x906)

1.53 MB PNG

>>107529122

File: file.png (42 KB, 1200x1200)

42 KB PNG

>>107523907
Missed last night because I was kinda tuckered out after Geoffslop. Working on day 12 now

>>107526736
that's literally what np complete means retard, p means polynomial time, nothing about 2^n is polynomial. also finding the shortest path is not always a polynomial problem, check https://en.wikipedia.org/wiki/Shortest_path_problem#Paths_with_constraints

>>107529307
cheers

i'll have like 30 minutes when i wake up to check the input and answer
see you tomorrow, unless the yookay gov blocks 4chan overnight

>>107529102
These are great

I'm gonna make my own AoC! With Blackjack and hookers!

File: image_2025-12-13_011937109.png (19 KB, 376x384)

19 KB PNG

Started late on day 2 and day 8. Other than that, this year was probably the easiest AoC of all time. Only problem I found challenging was day 10 part 2, but that too I was able to solve using scipy pretty easily.

>>107528904
I did it, it was way better than whatever eric spat out this year

File: IMG_8830.jpg (67 KB, 585x198)

67 KB JPG

>>107529360

File: 9212b6c8efb57b38209d1b167(...).gif (1.01 MB, 470x395)

1.01 MB GIF

Congrats to everyone who finished.

Sadly on day 10 my 98 years old grandfather had an episode and it felt to me to accompany him during nights at the hospital. So that was the end of this year's AOC for me. At least he was already discharged without any serious health issue besides his age.

>>107528130
This is great.

File: __madobe_nanami_claudia_m(...).jpg (1.78 MB, 1920x1080)

1.78 MB JPG

>>107529330
>>107529360
>>107529102
Best part of this thread.

Merry Christmas, everyone! Hope to catch you all this summer, and will def catch you next year!

>>107529292
I'm a 524 starer, I can't cope with a bad aoc year by solving a different one instead unfortunately

>>107529330
>>107529360
SOUL

>>107529459
https://everybody.codes/

File: tjsj0iiwm1971.jpg (52 KB, 640x624)

52 KB JPG

>>107529438
You had hospital waiting room downtime. Filtered

i wasted over 6 hours in total solving day 9 part 2.
i got the correct answer 3 separate times.

my area function was wrong
my area function was wrong
my area function was wrong
my area function was wrong
my area function was wrong
FUCK

>>107529505
>abs(x - y + 1)

>>107529335
FINDING
THE
SHORTEST
PATH
IN
AN
UNWEIGHTED
GRAPH
IS
NOT
NP
COMPLETE

File: 13dd4f5fc9e5b035225cf5594(...).jpg (853 KB, 2448x3264)

853 KB JPG

>>107529451
>>107529360
>>107529330
>>107529102
>>107529253
yay merry christmas art anons! thanks for the memes see you next year!

>>107529525
it isn't, it's polynomial in the number of edges and vertices, which guess what, it's exponential

>>107529512
just call me a faggot
it hurts less

>>107529555
ur a faget

>>107529545
you should make a thread on /sci/ with your new findings, anon.

>>107529505
mine was too, anon. mine was too...

>>107529525
It is when the size of the graph is exponential in the size of the input

>>107529598
i don't even know if this is bait or mental retardation
https://cstheory.stackexchange.com/a/10402
finding if a solution exists to the problem is O(N^3), finding the smallest solution is O(2^N), which can be reduced to a SAT problem which puts it in NP-complete.
by your logic, every single problem is P because you can solve it with a BFS/Dijkstra on all the possible states

>>107529627
which only applies to part 2

File: file.png (42 KB, 1200x1200)

42 KB PNG

One more star

>>107529651
https://math.stackexchange.com/questions/3228355/finding-short-vectors-in-mboxgf2
ok, found it for you, you can go read it later

>>107529662
no, i don't think i will pollute my head with whatever jew maths you study

Can anyone reupload all of drawanon's drawings this year? I want to save them.

File: cat.jpg (127 KB, 1024x999)

127 KB JPG

>day10 part 2
make it stop

>>107529330
>>107529360
>>107529659
Love you anon! Hope to see you again next year!! <3

>>107529737
see >107524663 and >107524668

File: irs.jpg (186 KB, 1920x1080)

186 KB JPG

>day 10 part 2
>spend the better part of a full day fucking around and trying different solutions
>do some research, find and implement Gaussian Reduction with backtracking to solve the system of equations
>turns out a lot of them have free variables which, surprise surprise, is an lp problem
>Eric: if u use a lp lib you didn't beat the game
fuck you I got my star

Was going to try redrawing the calendar, but I'm a little too gassed for that, so I'm happy with calling it here. I'll whip up a catbox in a sec

File: file.png (48 KB, 1200x1200)

48 KB PNG

>>107529933

>>107529659
>>107529938
Blessed!
Thank you for 24 elves!

>>107529102
thank you, slop comic anon, looked forward to these everyday

>>107529938
https://catbox.moe/c/rldl38

>>107529938
i feel nothing when I look at your tranime sketches

File: swaprows.png (18 KB, 604x560)

18 KB PNG

chat I am getting a 2 in my pivot column during reduction of one of the real inputs and idk what I should do about it since the rest of the row is indivisible
I suppose it is legal to swap columns as it just means swapping the position of two buttons
or maybe I just have to leave it like that

the rest of the code looks like it works on the test cases it just struggles with reducing the real ones
to be fair the more I think about it the more it feels like I dont have to care about it and I can leave it as a 2

File: 1749245523628378.png (50 KB, 863x774)

50 KB PNG

After staring at the problem for 40 minutes and getting distracted 5 times, I realized I've officially burned out on AoC now.

It was fun, I wish the contrast between the "easy" and "hard" problems wasn't as stark, or that he could've interleaved them more, put day 9 and 10 on the weekend, make the mondays easier, that sort of thing.
Overall I'm happy that there's less problems now, and that I might actually get through them by the end of the month. That said, what puts me off from doing these in other times of the year is that I really need to ask questions to solve some of these...

File: 1751353432287776.png (12 KB, 239x270)

12 KB PNG

>>107530155
It's crazy to think I've been doing these for almost a decade already, but never managed to finish a single year...

>>107530000
>day 10
>holy sovl and bigdigits

>>107529938
I feel nothing but joy when I look at your chranime sketches.

>>107530155
how have you not done 8?

>>107530155
There's still time.

>>107530224
I didn't expect such a hard problem on a monday, by the time I had written the parsing part I had to leave for work, and I didn't get around to it later in the day.
It always happens that the days accumulate and I prefer solving the same problem at the same time as other anons, so I can post my code and discuss solutions. Half of the fun is in that, so going back to old problems always feels like a slog.

>>107530247
fair enough. the keyword is union-find if you change your mind.

>>107530262
Thanks, added it to my notes. it seemed doable, just tedious.
I think the problem that got to me was day 10. I spent the entire evening on that one, and while I found an approximation approach that works on the test input, it doesn't find a solution for most of the lines in the real input (instead returning 0), and nobody in the thread was able to explain their solution to me, so I saved some of anon's non-import solutions for later review...
day 11 I'm sure I can figure out what the optimization is, my brain just hasn't been working since day 10.

>>107530155
>day 8
>day 11 part 2
these should be relatively straightforward, unless you've never done unions or dp before

>day 10 part 2
if you make an entire linear algebra solver for this you're stronger than 98% of the coders in this thread
most of /g/ and almost all of reddit used z3 which is just import solution
there's also that one divide by 2 way that someone on reddit made and then someone on 4chan posted here

>day 12
there is a quirk of the input that makes the problem way easier than what eric is actually asking you to do

I think just the OOB check left to write on my PS2 day 9.

>>107530329
>day 10
Part 1
Pressing a button twice does nothing, so the minimal solution must have each button pressed 0 or 1 times. Try every combination of 0 or 1 presses of each button.

Part 2
Each button is pressed either an even or odd number of times, therefore every solution can be broken down into an even number of presses of each button, plus an extra 0 or 1 presses. If an optimal solution for some value has an even number of presses of each button, then half the number of presses of each button is an optimal solution for half the value. Try every combination of 0 or 1 presses of each button, and if all the remaining values for the targets are even, recurse on half the value.

>>107529938
thank you so much o7

>>107530000
holy kino overload
bless you anon

>>107530329
barely anyone in here groked it but everyone ofc pretends they did. it's not that the solution is necessarily very difficult to implement, it's just highly unlikely that most anons would be familiar with the concepts involved at anything but the highest of levels, the majority of the board going "aha!" is laughable.

>>107530000
did you make one of these last year by any chance?

>>107529659
>>107529938
Merry Christmas, anon! See you next year!

File: __hakurei_reimu_touhou_an(...).png (492 KB, 609x869)

492 KB PNG

Popping in at the end of the last day for one final reminder: if you'd like to help make puzzles or get reminded, please drop by obonofcode.com!

I'll get started with the puzzles mid-January or so, and I'll also be in the #aocg IRC from time to time. Hope to see you all in the summer before saying hi next Christmas!

Merry Christmas to all of you! You're always the highlight of my holiday!

>>107530727
Yeah, I think the catbox is still alive https://catbox.moe/c/zvkms4

>>107530770
awesome stuff

>>107530160
Nope, not crazy at all. Filtered once, filtered always, filtered eternally.

>>107530828
I bet you felt really good posting that.

>>107530866
Bet you seethed posting yours

>>107530965
bet both of you faggots got massive butt plugs up your asses

>>107530828
>>107530866
>>107530965
What no more December puzzles does to a mf

Now what lmao

File: aoc-day-01-sloppa.jpg (1.04 MB, 1408x768)

1.04 MB JPG

bonus sloppa: Day 01

Advent of Sloppa Index (Gemini 3 Pro Image Preview)

Day 08 >>107486885
Day 09 >>107497012
Day 10 >>107510101
Day 11 >>107521253
Day 11 Alt >>107520952
Day 12 OP >>107523907
Day 12 >>107529102
Day 01 >>107531411

Day 13 status?
https://poal.me/pw4kr9

In this post, I want to thank Eric, the Advent of Code team, and the entire community.

Last year (2024) was my first year participating, and it turned out to be incredibly inspiring for me. At the time, I had a problem: for some reason I couldn't dedicate enough time to self-learning and working on my pet projects, work felt like it was draining all of my energy. But after almost a month of solving problems every day, I decided that this practice should continue.

Over the course of the entire year, I learned something every single day, solved problems from different areas of computer science, and broadened my horizons. I worked on my pet projects without exceptions, even on weekends. Sometimes it was hard, sometimes easier, but after a year I feel a huge amount of progress, which gives me the motivation to keep going.

As for 2025, I really liked the 12-day format. During this period, you don't have time to get tired, and overall it feels like the contest flies by in one breath. This year had a lot of interesting and great problems.

My favorite was Day 4, I even tried to solve it on the GPU. In the end, the performance was about the same as on the CPU, but maybe I just need to improve my GPU programming skills

The most controversial day for me was probably Day 10. After several hours of struggling with Part 2, I decided to check to see how others solved it, and I was surprised that many people used Z3. For me, it felt like the problem shifted from programming to math, though I might be wrong.

Once again, thanks to the Advent of Code team for the wonderful and inspiring problems, and for the great weeks I got to spend doing what I love. Thanks to the community for all the inspiring visualizations, solutions, and discussions. All of this pushes us to become better and grow.

Thank you all so much, happy holidays, and see you next year!

File: aoc10.png (1.1 MB, 1280x9265)

1.1 MB PNG

chat it finally worked day 10 part 2 completed

>>107531679
Good job, Anon.

I think I have made a mess of my day 9 ASM solution.
It's getting kinda convoluted and I'm not even keeping track of what's going on.

>>107531602
I appreciate the salt mining. Remarks directed at the redditor OP.

>it felt like the problem shifted from programming to math
Mathlets who are so openly proud about it are such detestable creatures.
>All of this pushes us to become better and grow.
Kill yourself.

>Thank you all so much, happy holidays, and see you next year!
It's Merry Christmas, fuck you.

>>107532283
Jesus wouldn't want you to be so bitter this close to his birthday, anon.

>didn't write a single BFS all year
I feel empty

>>107531602
Holy reddit
And even the spacing matches
Obvious joke post but still

Thinking about it, was Day 12 intended by Eric to be anti-AI? there's no way claude would think (if you can call AI "thinking" lol) of going the easy way out and not actually solving the actual satisfiability problem

>>107532440
this was my initial thought as well, but I don't think it holds because it confuses humans too

>>107532447
*confuses humans who use AI
ftfy

>>107532440
filtering AI on.... the very last puzzle? Doesn't make any sense

File: 5rnslq.png (3 KB, 230x219)

3 KB PNG

>>107532423

>>107532064
Are you writing your own hashmap in assembly? I did day 9 in riscv but I already had a hashmap I wrote for 2023

>>107532306
>Defending millenial sop slop.
Here's the link, go be inspired by it.
https://old.reddit.com/r/adventofcode/comments/1pknsnk/thanks/

>>107532423
Sometimes it's hard to distinguish bteween parody, chat gpt, or real reddit, but this is authetic.

Eric tricked us, there's day 13 after all!
https://adventofcode.com/2025/day/13

File: 1646254439200.png (7 KB, 400x400)

7 KB PNG

I used all my energy on completing d10 and have no energy for making my sloppy d12 broot smart

File: day09emu.webm (159 KB, 662x583)

159 KB WEBM

Turns out if I only try to make rectangles from half the points with my current code, the answer pops out.
So let's just call that pruning and end it there.

File: day09.png (1.32 MB, 2202x3443)

1.32 MB PNG

>>107532917
I am so done with this.

>>107532917
Nice dude. I love your work. Very impressive

Day 10 was by far my favorite. Part 1 was basic dynamic programming, but part 2 was such a bliss. It unlocked memories of playing with vector spaces in college, which was something I really enjoyed but not really use much anymore. Basically my strategy at first was to expand the set of buttons into component vectors and then compute the decomposition of the joltage vector by them. (Un?)fortunately, since we have more vectors than needed to span the vector space, and since we have a minimization problem, it's just Linear Programming problem and not really linear algebra, so it was solved in a few very satisfying lines of scipy-powered python.

Day 11 was also easy, but it's still fun just because everything with graphs is fun. Please more graphs next years! I love graphs! I don't mind easy problems if they're graph-shaped !

But, honestly, the best part of this whole adventure has been lurking on this sub. It felt good to be a part of a like-minded community and to read all of your quirky (in the most positive sense) approaches, pretty visualizations, and your hilarious memes! I was seriously considering dropping out on the first week and stayed thanks to y'all! Thanks to every poster, commenter and mod <3. And of course, big thanks to Eric Wastl for making this possible !

Merry xmas to all of you!

>>107532917
>>107532937
How do you compile, link, and run this on your hardware?

File: 20251213_123455190.webm (1.4 MB, 853x480)

1.4 MB WEBM

>>107532937

>>107532917
>>107532937
Based. I don't really have much to add to your posts, but it's impressive that you keep it up.

>>107532806
Holy shit guys there's a day 99 https://adventofcode.com/2025/day/99

>>107532963
hello fellow 172 enjoyer

>>107532970
Day 13 is UP!
https://adventofcode.com/2015/day/13

>>107532952
The assembler I use, naken_asm, handles the assembling and linking. Then I just push the elf over the network with ps2client running on the PS2.

>>107532937
so can that thing import z3 or are you filtered now

Eric Wastl?
More like Eric Washedl

So we're all in agreement that Eric phoned it in this year?

>>107532937
The most annoying part of this, and likely of my own doing, is that with every point you have two values, x and y, and when you have a line you now have four values. And when you want to check an intersection now you have eight values, and I'm trying to shuffle these around in registers and having to remember and check which register has which points x/y value and I probably should have made use of RAM more often.

>>107532988
>read puzzle
>start thinking about edge cases
>get really confused

>look at my solution
>broot every permutation

10 years of aoc has rotted my brain. Just Broot.

Today was ruined for me because this (warning: day 12 spoiler!) post showed up in my feed.

I'm not subbed to here. Reddit's algorithm threw it on my feed because I visited the sub a couple of times.

This year was really fun, but having the last day instantly spoiled kind of left a sour taste in my mouth, because it seems like a really fun day to figure out on your own.

Please, mods, could we enforce more spoiler shielding next year? Some of the memes just spill the tea. Which is fine, but those posts really shouldn't have any chance of appearing on anyone's feed without some guard-clause such as a spoiler tag.

And yes, I know, it's safer to completely stay off Reddit, but I didn't have much time for AoC today. I went to work in the morning, and was just browsing some memes on my way back home from work. I think it's fair that I wasn't expecting to be spoiled by getting the answer shoved in my face.
---
What a fucking faggot.

>>107532998
I suppose I could try writing a z3 implementation in assembly....

>>107533040
>miss showtime
>get filtered
>cry about it

>>107533032
your stack?

>>107533031
Totally. I think an AI generated puzzle could have been better.

File: IMG_5062.jpg (29 KB, 520x476)

29 KB JPG

I guess it’s time to go back to a year I haven’t done before and see if I can do it all the way without these threads.

>>107533117
>coming to the thread before you have both stars
filtered

>>107533117
Good luck anon. I already did all the previous years. I guess I will try the weird dragonduck stuff on everybody codes

>>107533040
he's right doe

File: day4.png (92 KB, 1424x922)

92 KB PNG

rewrote the ones I used import solution for
>day 4