Previous Thread adventofcode.com/ >Advent of Code is a series of small programming puzzles for a variety of skill levels. They are self-contained and are just as appropriate for an expert who wants to stay sharp as they are for a beginner who is just learning to code. Each puzzle calls upon different skills and has two parts that build on a theme. You need to register with an account but you can appear as anonymous.
While no longer necessary, here's the private leaderboard info: Private leaderboard join code: 368748-e33dcae3
People who are inactive for 5 days will be kicked, but other leaderboards can be made if desired.
Alternative leaderboard for those waiting for inactives to be kicked from the first: 208834-4a1a9383
Sweet calendar senpai I like how my day 25 podium design from last year was improved upon
Kevin Gomez
Call me a newfag but how do you guys even know that the guy browses Jow Forums and not just a random person who wants to use user
Joseph Robinson
Salty AF
Christopher Ross
He's posted here numerous times.
Angel Ross
Nice thanks for the confirmation, my first year actually following / doing AoC so i don't know much of the history
Christian Gutierrez
Private leaderboard and last year he'd frequently post his solutions right after the problems appeared (only someone high in the global leaderboards could do that)
Aiden Rivera
Why so mad?
Jayden Cook
It isn't in that cap, but the user also posted a link to last year's calendar. Probably pushed the wrong buttons.
David Miller
Ah, that would do it. topkek
Ryder Young
To be honest anyone who would get offended by the advent calendar deserves to choke on a dick anyway.
Jaxson Moore
Just checked their post history looking for that, and it turns out he also called the mod a faggot.
It was a big boi input on a problem where a typical implementation sufficient to solve the normal version of that puzzle would end up consuming like 30GB of memory, iirc after the user tried running it and it hit swap, he tried checking the SMART attributes or something on his drive and that program did bad things to it. Anecdotally, a few months ago I almost killed a RAID array by doing the same thing when whatever shitty program I was using at the time caused my entire system to hang, and upon rebooting my controller was convinced that all drives in the array were dead for a while.
Julian Collins
What is job market looking like for software developers? Are we going to see a massive fall in wages as the tech bubble bursts? Should I switch to something more stable like EE (from CS)?
Jayden Russell
so where's the final big-boy. someone make it. something crazy like 2gb of bigboi
So were you guys actually getting all of these done in a day or was that for the wizards only? I don't consider myself too retarded but a lot of them I certainly couldn't casually do in a day unless I dedicated all my time to it.
Kevin Rogers
A few minutes for the easy ones 20min to 1 hour for the hard ones Even more time when you don't find a solution
Jaxson Edwards
Congratulations Anonymous user #193354 I'll probably do a last ditch effort later to finish all my stars but I think this is as far as I can get pretty impressive, considering it's the first time I actually got to try all the problems in the competition Merry Christmas everyone
Adam Green
well fuck, maybe I am a retard then. I've only done day 1 and 15. 1 was obviously really easy but 15 I've found to be quite time consuming. Is that a particularly hard one or am I just slow?
James Hughes
Day 15 is probably the most time-consuming puzzle this year.
Cameron Hill
I consider myself fairly bad at programming, and they usually take me 15-50 minutes. The hardest ones can take me 2+ hours, and the one that took me the longest was the reverse engineering at 5 hours (I wasn't rushing to get it done though, could've gotten it down to 3-4 hours like last year). I didn't time day 15, I did it later in 2 sessions. Still think it took less time than the reverse engineering.
Connor Stewart
>tfw will never complete this shit I-I just d-don't have t-t-time...
Based. /our guy/ pulled it off again. Solving these problems with immense speed and great style without ever revealing his identity...He's the Internet Chad
I consider myself an average coder, but problems usually take me long because of focussing issues. I keep doing other shit like exploring other languages, tools, games, TV, etc. Easy ones don't take me long. Maybe 20 minutes, if I don't have to optimize the code for speed. The hard ones have taken me longer than a day, pretty much always because I missed a tiny detail in the directions, or took a bad approach and stuck too long at trying to make it work, instead of just throwing it out and do it different. I'm happy to see there are so many people doing shit like this so fast though. At my work I'm constantly frustrated at how bad my colleagues are, so this gives me the hope there have to be places with good programmers from whom I could learn shit.
Landon Ortiz
>Chad What if it's a girl?
Juan Rogers
Right, I'm in a similar position then. I struggle so much to figure out if I'm an absolute retard with programming or if I'm not actually that bad and just keep comparing myself to really skilled people.
Joshua Scott
I want to ____ 193354-chan!
Ryan Lee
>tfw no Jow Forums browsing super skilled programmer trap gf
Michael Nelson
KEK
Mason Flores
what a clusterfuck
Colton Thompson
Dumb question but which language does {:year 2018} represent?
Joshua Brooks
Clojure
Benjamin Clark
cheers m9
Parker Powell
To my shame I am featured on the final pic on two different days quite prominently.
The picture should be correct. As in: it visualizes the largest collection of bots with a collective overlap. But even the tool messes up when I try to have it render the intersection. Not that it would help that much, besides giving some visual support, and hopefully some numbers to aim at. I still need to code the solution, which I hopefully still will.
Leo Lee
My solution to that still seems like the biggest bullshit ever. I wonder if some weighted average of the points where you gave greater weight to smaller radiuses would give a close enough point you could just bruteforce from there
Camden Wilson
Day 15 is the 2nd most time-consuming AOC problem ever maurits.vdschee.nl/scatterplot/ (I don't count the 2015 leaderboard times because it probably had low sample size)
Oliver Price
I was thinking in the line of grabbing the 6 outlying points (top, bottom, east, west, north, south) of each bot's radius, and focus on the furthest away lying inner point, guessing that it has to be a defining point of the intersection. I have no idea if that line of thought has any merit, but it's all I can think of right now.
Cooper Robinson
I got my graph on day 6 and I'm the guy with the italian task manager pic in day 14 I think the day 11 background is also from me but I'm not too sure
William Ward
does Eric put out a complete solution set after the event is over? I'm curious what the official solution is for 23 part 2
while QUEUE is nonempty: REGION, SCORE = pop highest score entry from QUEUE if REGION's size is 1x1x1: /* For 1x1x1 regions, MAX_BOTS returns the actual number of BOTS_IN_RANGE for REGION. Because we always pop the highest-score region, we know that all regions we haven't checked have MAX_BOTS
Elijah Brooks
>Instead of checking every point in REGION to see if it's IN_RANGE, we need >to check only the point in REGION that's closest to BOT. If that point > isn't in range, then no points are in range. This is nuts and overly complicated for no good reason. If your region is just another nanobot then checking if there's a point in range is as simple as checking that the distance between the centers is within the sum of the radii.
Eli Thompson
Subdividing bot ranges is harder than subdividing AABBs
Cooper Jones
The best solution is to first find the largest complete subgraph of the graph where nodes are nanobots and edges are overlapping range After that the problem becomes pretty trivial. I already explained it here
Elijah Johnson
Do you think blogging about my AdventOfCode solutions will help me get a job?
Joseph Johnson
I don't think it will hurt. Even better put on a wig and livestream it next year: Eric is going to retweet you and you'll get a job for sure.
Chase Flores
just make a github repo for your solutions
Henry Myers
Can this solution handle that pairwise overlap between a set of bots does not guarantee that there is a point where all of them overlap?
Colton Rodriguez
No, it's not. I posted details a few threads ago.
Christopher Fisher
I'm pretty sure that for octahedrons like this every complete subgraph will have a nontrivial intersection. That's just my intuition, I haven't proved it
Robert Thompson
I just think it would be interesting to find out if the divide by million solution was the intended one or something else since this problem had so many different solutions
Isaiah Edwards
Your solution is really good. I originally wanted to solve it this way. No idea why it was ignored in the previous thread.
