Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Advent of Code 2024

adventofcode.com

Post reply on HN
Print view View original

101–110 of 580 posts

Re: Advent of Code 2024

#101

Post by rich_sasha »

zwnow wrote: post #60
rich_sasha wrote: post #57

Earlier quoted context omitted.

Without condoning cheating, I am impressed with the automation aspect of it. 9 seconds sounds more or less like the inference time of the LLM, so this must have been automated. Login at midnight + lots of C&P may not have done it. Perhaps there is a scope for an alternative AoC type competition aimed at AI submissions... ...though of course that would be experimenting to get us all out of work. Hmm.

If real life problems were as easy and defined as AoC problems we might be able to be replaced at some point. I highly doubt you can replace software devs otherwise. Who else is going to take the blame for software issues?

"Write me a snippet that does X" is a step behind "figure out how to log into this page, download the data, write a snippet that gives the right answer to the sample data, then run it on the real thing and submit the output to the text box".

Re: Advent of Code 2024

#102

Post by andix »

wiseowise wrote: post #23
tmountain wrote: post #3

It feels like AI will shape a lot of the submissions for these types of coding “traditions”. But, I guess that’s just the world we live in.

All the people I know already participate on private boards. AI enshitification will only speed up the process of moving there.

It also doesn't make any sense for most of the people to compete with the geniuses on the public leader board. It's like signing up for the Olympics as an amateur athlete.

Re: Advent of Code 2024

#103

Post by davidgl »

andrelaszlo wrote: post #94
ducdetronquito wrote: post #77

I will try to do it with F# and Gleam this year, but like every year I won’t have time (and brain) to do more than 10/12 days ^^ For the pythonists around here, give F# a try: it can feels very close to scripting and it has a wonderful REPL too :)

How's linux support? :)

Perfect, run loads of f# and c# on Linux, and have for years

Re: Advent of Code 2024

#104

Post by thinkingemote »

Almondsetat wrote: post #19
arjvik wrote: post #2

9 seconds to get both stars is absolutely insane - there had to be some AI assistance here. Come to think of it, a pipeline that feeds the problem text into an LLM to generate a solution and automatically runs it on the input and attempts to submit the solution, doing this N times in parallel, could certainly solve the first few days' problem in 9 seconds.

Caring about the leaderboards is the problem. Are you (impersonal) seriously doing AoC for clout or something?

If people are concerned about leaderboards there are private leaderboards: https://adventofcode.com/2024/leaderboard/private

Personally I don't do it to compete, I just like puzzles.

Re: Advent of Code 2024

#105

Post by SleepyMyroslav »

Am I alone in thinking that measuring time to get answer is the worst possible metric? I have not participated because of that (yet).

If there is a community for those who use other rules to compare actual solutions instead of answers I would be interested to hear about it.

I am coming from low level C++ gamedev side so I understand that most people here use different tools to solve different problems.

Re: Advent of Code 2024

#108

Post by MattRix »

Barrin92 wrote: post #82
tmountain wrote: post #3

It feels like AI will shape a lot of the submissions for these types of coding “traditions”. But, I guess that’s just the world we live in.

The global leaderboard is so fast that any AI assistance would literally slow them down, here's one of the guys who tends to score highly solving today's puzzle. ( https://youtu.be/ym1ae-vBy6g ), and on the more complicated days that's even more pronounced because anyone who is even somewhat decent doesn't need to ask chatgpt how to write Dijkstra. Obviously if you're doing it recreationally you can cheat with AI but…

The thing is that the AI can read a puzzle faster than a human can. If someone put any effort towards an AI-based setup, it would easily beat human competitiors (well, up until the point the puzzles got too difficult for it to solve).

Re: Advent of Code 2024

#109

Post by 0xEF »

setopt wrote: post #100

The tiny thin gray font on black background is impossible to read on a mobile device :(

Color aside they also ignore a number of solutions for font resizing that follow the users accessibility settings. I can think of at least three easy ways to do this with html/css alone, and yet hear we are; a site that is unreadable to me.

I love what Advent of Code does, but when your site is all text, there's just no excuse to not let the user resize it by default.

Re: Advent of Code 2024

#110

Post by dingensundso »

SleepyMyroslav wrote: post #105

Am I alone in thinking that measuring time to get answer is the worst possible metric? I have not participated because of that (yet). If there is a community for those who use other rules to compare actual solutions instead of answers I would be interested to hear about it. I am coming from low level C++ gamedev side so I understand that most people here use different tools to solve different problems.

I think most people don't participate for the competition. I did it several times to increase my coding skills, have fun or get more practice in a new programming language.
Post reply on HN