Hello, Codefoolers!
We, AlperenT, flamestorm, prvocislo, temporary1, ToxicPie9, and -1e11, would like to invite everyone on Codeforces to April Fools Day Contest 2025! The 13th April Fools Day Contest will take place on 01.04.2025 17:35 (Московское время). This is a joke competition in which solving the problem is often easier than figuring out what the actual task is.
In this round, you will be given $$$n$$$ problems where $$$\sum\limits_{x=1}^{\infty} \frac{1}{x^n}$$$ converges and 2 hours to solve them. The contest will use extended ICPC rules (no hacks, the standings are decided by the number of solved problems and penalty time earned on them). You can submit solutions in any language allowed by Codeforces, unless the problem says otherwise.
Please note that the round is unrated and that the penalty for a wrong submission is 10 minutes.
To get an idea of what the contest will look like, you can check out the contests of the past years: 2012, 2013, 2014, 2016, 2017, 2018, 2019, 2020, 2021, 2022, 2023, 2024
Thanks to our precious testers (they exist, trust me):
Please do not bother sending clarification requests in the contest as we obviously do not enjoy sending funny responses like we did in the last two years. It is definitely not the most fun part of April Fools.
Lastly, we have a video editorial this year (Russian captions available)!
Good luck, and have fun!
Edit 1: Editorial is out!
Edit 2:
Congratulations to the winners:
And the first solves on each problem:
- A: vladimirfilip
- B: Auchenai01
- C: Dark (after AprilFoolsClairvoyant)
- D: koishi.komeiji
- E: jiangly
- F: tiom4eg
- G: BucketPotato
- H: N/A (but circle977 got closest with every word except 10th)
- I: SeptaCube
- J: fishy15 (initially conqueror_of_tourist, but got hacked after the contest)
yearly "flamestorm orz"
Dont check the previous revision
"Please do not bother sending clarification requests in the contest as we obviously do not enjoy sending funny responses like we did in the last two years. It is definitely not the most fun part of April Fools."
My lab supervisor must be mad if he knows I’m spending so much time preparing the problems. Give me contribution.
💊
I definitely didn't join the authors just to answer the boring clarifications...
How do I get better at April fools?
Think outside the box.
box outside the think
think the box outside
box think the outside
outside the box think
the think outside box
think the outside box
get out
outside think the box
Think outside the problem statement
Think outside the "box"
Box thinks inside us
dont think, get inside the box
Get the box inside the think
get out side the box
bro gave the hint of problem B lol
i shat in the box
Think outside the 16th column.
Also think outside the 0st column for the editorial's code is negative number.
AlperenT orz :)
AlperenT orz
AlperenT orz :)
AlperenT orz
AlperenT orz
-1 problem? Everybody knows that this series converges to $$$-\frac{1}{12}$$$
It could be $$$3.141+2.718i$$$ problems as well :D
The series doesn't converge. This is an abuse of notation brought on by the analytic continuation of the Riemann Zeta Function, which normally is well defined only for values pf n respecting Re(n)>1. As such, the only valid assumption that can be taken is that there is strictly more than one real problem in the problem set.
Yeah, more than 1 real problem so like $$$\sqrt{2}$$$ problems for example
Shhh... Do not ask how, but I have link to the solutions. You can find them here.
Warning: Be careful for being scammed.
For those who always ignore warnings, Error: Be careful for being scammed.
at least don't use the most popular link, every body knows it ends with gxcq.
is it rated?
At least it's said to be not.
What if that was the April Fools joke by the person who wrote this
As one of the testers, the problems are very interesting!!!
april fools! i didn't test :(
A video editorial premieres as the contest starts... Bro's up to something
oh man, now I am curious what that video is going to be
Do you think the style of the most problems to be:
Based problems:
Based problems:
Based problems:
Can i believe the "unrated" thing?
if it is a genuine question, yes it is unrated,
please make a lot of wrong submissions to try various ideas to have more fun
love first April contest ,cause this is only one contest where i can solve only one problem
If every round was like this, there would be no more AI cheaters!
no more cheaters and no more problem statement
Can we make it rated for a day? :pleading_face:
my favourite contest of the year!
So there will be $$$n=-1$$$ problems because $$$1+2+3+4+\cdots = -\frac{1}{12}$$$?
right
There are actually $$$-4$$$ problems, since $$$1+16+81+... = 0$$$ if used the same zeta expansion logic as $$$1+2+3+4...$$$. Also, why no one mentions the harmonic sequence?
No harmonic sequence but we were talking about problem b in the forces round and...
Expected
Are these problems AI proof?
these problems are really tricky for me, very curious to see what AI is able to pick up
Looking forward to it very much!
That's a great work
what
fools are fooling the fools as the fools used to fool the fools cz fools don't know that fools fool fools everytime.
I just remembered “Are You a Procrastinator?” in the past year
This contest is the best gift for my birth day
your birth was a joke
even my grandpa didn't belived when i was born
oh man .... lololol.
Rated ?
It was fun being a tester. UwU :)
Really excited for this
April Fools !
is it rated ?????????????????????
April Fools contests are really fun, especially the out of the box thinking required to solve problems such as 1812G - Colour Vision and 1952I - Dark Matter. Its a nice break from the usual contests. Is it possible for there to be more contests like it, perhaps every 3-4 months?
there is only so much unpaid labor can do
What about paid labor?
Finally a rated contest
hope we all become legendary Grandmaster
Is that means n>1 ? Hope my Math is right :D
Luogu also have Fools Day Contest that is LGR-(-25)! Come and have Fun!!!
Now that the contest is over, we can discuss the problems. I didn't solve A, any hints?
You must use divide and conquer in combination with segment tree beats and heavy light decomposition. You may think this is too slow, but you can use bitset to optimize it.
can it be solved for higher constraints like $$$n \lt 2×10^5$$$?
it is slow even with bitset.. now i use python.. that will make u understand why its slow xd... so upon all that, i added DSU data structure... helped me optimize it... while adding sliding window inside the segment tree... from O(N^5) to O((N^2)*gcd(n,m)*log(max(n,m)))
Another way is to use a bitset of bitsets, if that doesn't work, a bitset of bitsets of bitsets should do the trick!
I solved it using online algorithm for 4D half-hyperspace intersection. It wasn't obvious, but you can transform input data in such a way that after calculating intersection, the result is always a strictly convex 4D hypergon. All what remained to do — check if the hypergon you got is pentachorizable (possible to divide hypergon into non-intersecting pentachores — 4D analogue of tetrahedrons).
Took me 10 mins to implement — I just copy-pasted 4D half-hyperspace intersection online algorithm from my library.
can someone tell me how the difficultly for these problems are
like should I solve them in order, or I should try to read all problems .. .
are they only tricky in terms of "out of box" thinking or they are hard in implementation / Algorithm knowledge as well
try previous rounds
yes, I am planning to start early for this round and give a quick look at last years questions before start.
I guess there will be a problem related to $$$(20+25)^2=2025$$$.
As a tester, I can confirm that there are questions.
Eagerly hoping chatgpt o3-mini / deepseek R1 will not able to solve these problems.
Wow, what a cool April Fool's Day event! Many thanks to the organizers for putting on such a fun and unforgettable event. I can't wait to see what surprises they have in store for us in the future!
is 2nd just generating random number unless ur luck works and random answer doesnt match any previous answers ?
i am so dumb
I waited for at least one successful submission and then binary searched on the status page filter (AC, test >= x, binsearch x)
wow looks like my dao is far inferior
The problem didn't specify that it had to be within columns 1–16, so I chose column 17, and it just fell right through.
whoever dms me answer will become billionaire
Queue is fked
why is queue so slow, it is just ruining the fun of trying random things :P
Fix the queue I can't do shit
Hawk Tuah Button ---------->
any website that gives accurate result for reverse engineering photo location?
how to J? I thought we were supposed to print the first and second half of authors of the contest but failed :(
I was printing their names written in 3 words (from youtube description)
looking at the AC submissions we had to print 'Nauuo and Votes' and 'Nauuo and Chess' but whyyy
The editorial is the perfect example of walking through the solution.
i did it
legend
orz
patience is key in life. Orz
Box the outside think!
what is the ans of problem c? :(
I observed a running code to see how far the test cases go, 143 is the answer. Insane problem haha
A less painful cheat through it: binary search the status filter with "test >= x" until no accepted is found. Works only after 1st AC.
interesting thing is no one ever guessed 143 until 80th minute. there were like around 10000 submissions before first AC.
Discrete/Independent/Uncoordinated attempts would be more erratic than you think.
I think we should be collectively ashamed. There were over 10,000 contestants. If each of us had guessed a random number between 1 and 1000 then one of us would have gotten the answer, and then the rest of us could have copied the solution from the test filter page.
But we were too uncoordinated!
Morals of the story: "Workers of the world, unite!" 🫡
who go the first AC tho ?
A user named AprilFoolsClairvoyant, who registered during the contest, only submitted this problem, and their pfp is the image in announcement 2 (read: the organizers).
Even the problem gives different text for every input. Hilarious.
Go to status and binary search the filter "Test" in the right panel "Status Filter".
everyone who couldn't solve C is the reason why they are single.
whats the idea behind 143? I still cant get it
Just had to keep an eye on the status page and observe which test a solution is running on. 143 is the highest it went.
TheScrasse should know
"Note: I'll forgive you for lying this time, but you better tell me the right answer next time I ask!"
this sentence at the end of the problem explains it, like something a girlfriend would ask or smth.
I like the detail that just like how we have 10 fingers and use base 10, the Balikons have 9 fingers (3 hands with 3 fingers each, as can be seen in the photo) and use base 9.
Also they use {l, sh, t} prefix based on which hand, and then {a, e, on} suffix for the fingers of each hand.
And the digits of their script are based on a 3x3 grid:
etc.
Best Video Editorial Ever :)
i might be like a tiny bit stupid
For F you change the language to Russian to see the other 2/3. Wow
I was thinking B was just like a gamble game (luck based), you keep playing until you win , lol.
no, it can take shortcuts
good veryhappy
The best April Fools Day Contest I've ever participated.
Fun fact for Problem F:
I wrote a Competitive Companion parser for personal use, and for some reason, I filtered all non-ASCII characters for the file name.
After struggling for the rest part of the statement, I eventually noticed the unusual file name of problem F:
F__f__rblm.cpp. The missing letters (a, o, e, and P) were actually Cyrillic characters ("а", "о", "е", and "Р") that had been filtered by my parser.Having fun while being fooled, the trick on D and F is insane.
First experience on April's fool contest comes out great!
Happy April Fools day!
Where is problem J ?How did 2 participants solved that problem?
READ editorial
April fools contests feel like magic