kostka's blog

By kostka, 2 months ago, In English
 
 
 
 
  • Vote: I like it
  • +76
  • Vote: I do not like it

»
2 months ago, # |
  Vote: I like it +14 Vote: I do not like it

Thanks for the reminder, just completed the 2 sets for today, if anyone is interested in joining my room, they can use the code 1047751-f3a5bcf8

  • »
    »
    2 months ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Hello, how do I join rooms? Also what is a room?

    • »
      »
      »
      2 months ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      A room is a kind of a subset of global leaderboard, where the users get relative scores. You can join one by entering the code in the following input as shown,

»
2 months ago, # |
  Vote: I like it +5 Vote: I do not like it

as a NA east coaster, me when i sacrifice sleep to do aoc and fall asleep in class the next day >_<

  • »
    »
    2 months ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    I thought stuvy students already don't get any sleep

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

How does this compare to codeforces? Is this easier than leetcode?

  • »
    »
    2 months ago, # ^ |
      Vote: I like it +29 Vote: I do not like it

    it’s a combination of readforces and implementationforces based on past years

  • »
    »
    2 months ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Based on what I have observed in the past 2 seasons of AoC,

    Upto Day 5: LeetCode Easy
    Day 5-10: LeetCode Medium-Hard (or Codeforces 2000)
    Day 10-15: Codeforces Div 1
    Day 15-25: Never reached yet, will try to push myself this season

    • »
      »
      »
      2 months ago, # ^ |
        Vote: I like it +10 Vote: I do not like it

      Idk what problems you're talking about, but advent of code problems aren't difficult. I've completed 2 seasons and I'm < 1900 rated on codeforces. Just painful implementation problems and nothing more.

      • »
        »
        »
        »
        2 months ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        ^^^

        no way days 5-10 is 2000 rated worthy and 10-15 is div 1. Most problems are essentially just brute force but the harder part is understanding the statement, so i don’t know if you’re incorporating the language barrier factor as well :v

      • »
        »
        »
        »
        2 months ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        How come painful implementation come under 1900, I don't understand, and last year I didn't gave it seriously, so the experience can be very subjective to me, I would say.

        • »
          »
          »
          »
          »
          2 months ago, # ^ |
            Vote: I like it +2 Vote: I do not like it

          I said I have less than 1900 rating on codeforces, I wasn't talking of the advent of code problems. Yeah, maybe you should look at the problems again, they aren't as difficult as you think :)

»
2 months ago, # |
  Vote: I like it +14 Vote: I do not like it

There's a leaderboard hosted by eldar/gojira which seems to have persisted from last year. I don't see an obvious way to invite, so I'll defer to him and/or someone can dig up the old link if they're desperate (all 9 of us, 9!!). For clarity's sake, last year vrintle did 7 days, some of them only 1 star, and all below day 10.

About difficulty: relative to cf, 'implementationforces' and 'gridforces' complaints are fair. The authors aren't neck/hip deep in cp, so their idea of a problem set with both good variety and range of difficulty will look a little bit weird if your main exposure to programming is cp/algorithms. Programmers of a certain age probably started because of games: so basic sim physics of projectiles/collisions or jumping/platforms might be underweighted as familiar whereas they'd probably be out of place here. There's definitely more geometry, but it's mostly kept to right angles (until it isn't). Again, some subset of coders will be more comfortable with stuff like composing matrices of 3D transforms than others... but stuff like that has come up, ever. There's usually some element of 'write a parser' or 'novel file format' too.

So if you can turn off the part of your brain that will notice the same method melting away multiple problems (e.g. enough to win a fb hacker cup shirt this year), AoC might be a good palate cleanse and/or excuse to try a new language. I got into it in 2020 because I hadn't seen a non-work community for coding in forever, so the attached reddit discussions were some sort of breath of fresh air (sharing ideas! visualizations! tortured golf solves...?). Then I ended up here (too).

Buuuut not to end on a dig -- for 2020 I only cracked 3-digit global rank 4x but after a year of some-level-of cp I was able to do so for the majority of days in 2021. If you're hardcore out to leaderboard for 2-digit global ranks, you'll likely need to do things I didn't: gitgud (obvi), skim statements effectively (e.g. skip to samples for input format, guess well about which standard problem is being invoked), automate getting data and submitting answers, and/or use pre-written code. I tryharded at some level below that for those 2 years on the notion that I had to be awake at the time anyway (midnight usa east). I will not be doing so this year or ever again in all likelihood (see downward slope of right butt cheek of my rating graph for late 2021).

ONE MOAR THing... as a heads up -- there was a problem where cp assumptions backfired, like it was reasonable to assume it was still in the easy days, so some people further assumed it was a standard dp problem where you could only move down and right. It wasn't intended to be, but the issue was that it happened to be randomly true depending on which data set you ended up with. I can't fathom liking dp that much, but I can see how that might've been annoying if you got bitten by that. The other extreme is people who have pre-written d/bfs code who sleepwalked through it, of course. Either way, hope y'all find what you're looking for, however you go about doing so.

  • »
    »
    2 months ago, # ^ |
      Vote: I like it +4 Vote: I do not like it

    I also had no idea about competitive programming until I participated in AoC 2020. I barely remember I read some comment in the Reddit discussion about some top Codeforces participant, so I decided to search what was that Codeforces thing. I keep participating there even though I find the problems rather boring, but I'm grateful for whoever mentioned Codeforces over there.

  • »
    »
    2 months ago, # ^ |
      Vote: I like it +7 Vote: I do not like it

    Heh, love the writeup and definitely agree with most points. The code for my "all 9 of us!" room is 1669977-95eae9da btw.

»
6 weeks ago, # |
  Vote: I like it +3 Vote: I do not like it

Thanks to noticing this announcement I joined somewhere in the middle of the second day. This wasn't too late and the whole event turned out to be fun.

The day 1-8 puzzles were all trivial (they required zero problem solving skills, some reading comprehension skills and some implementation skills). I guess, some people from codeforces probably dropped out during these days just because they got bored too early.

The second part of the day 9 looked more like a competitive programming problem. And the second part of the day 11 too.

I think that the day 10 puzzle was probably the most unusual/uncomfortable for competitive programmers, because it's basically a simulation of a simple CPU and a simple display controller. Was there anyone, who found it challenging to even understand what needs to be done?

As for the puzzles difficulty, I only got stuck on the second part of the day 22. Didn't really solve it properly, just special cased something specifically for my input in the end. And only after getting back to it on day 25.