Καλησπέρα, Codeforces!
I am very excited to invite you to participate in Codeforces Round 1063 (Div. 2), which will take place on Nov/10/2025 18:05 (Moscow time). Please note the unusual starting time (30 minutes after the standard starting time).
You will be given 5 problems and 2 hours to solve them. At least one problem will be interactive, so please make sure to read the guide for interactive problems before the contest. This round will be rated for all participants with rating below 2100. The problems were authored by me, and prepared by Proof_by_QED and me. We are also excited to announce that:
This will be the first rated round to feature a communication problem.
Make sure that you are familiar with the problem format of communication problems. Recently, Testing Round 20 (Unrated, Communication Problems) was held in order to introduce this new type of problems. Please read the blog on communcation problems and familiarize yourself with these problems. Note that it is the first time we are experimenting with communication problems in a rated round, so please keep this in mind if you choose to participate.
Additional Note: In this round, hacks will be disabled on problems A,B,C and we will have pretests=systests. Hacks will be enabled on D,E as usual.
This round would not be possible without the help of the following people:
- Proof_by_QED for his very good (and first) coordination, and satyam343 for helping us both make this round better;
- Alexdat2000 for russian statement translation;
- The testers: Dominater069, A_G, Andreasyan, __baozii__, rui_er, _istil, nifeshe, Intellegent, fishy15, efishel, wuhudsm, AksLolCoding, FetFot, mzen, temporary1, turska, Timosh, cry, rangerscowboys, -firefly-, Friedrich, madlogic, pupilemus, sokratisi, pop, Lilypad, Edeeva, cowthecow, coder3000, macaquedev, Teto, chromate00, Non-origination, Stathis, pianapolataa, radaiosm7, beaten_by_ai, Vladosiya, Banis, AG-88301, Godgift42, streidi, ALnQ417, Ashrayy;
- MikeMirzayanov and KAN for Polygon and Codeforces platforms.
Score distribution: $$$500 - 1000 - 1250 - (1500 - 1250) - 2750$$$.
Winners:
Div.1+2:
Div.2:
as a tester, dont miss out on the first round with a communication problem
orz
communication
as a tester, i communicated
how to communicate? I do cp...
Only 5 problems? wow.
There is a problem divided into 2 subtasks, according to the score distribution
As a tester, I struggled with communication
Isn't it strange that Benq doesn't participate in contests so as not to lower his rating? He does so much. Thanks again for the great contest ;)))
the contest is for div 2 even if he participates he would be unrated :/
I mean Div.1+2:
As a tester this is a fairly fun round, actually.
As a tester, I communicated with __baozii__ during testing.
As a participant, by solving the communication problem, I will be able to make some friends.
excited for the new kind of problem, hope I can communicate well !!!
btw, my guess is D1 and D2 are communication problems with different amounts of data you can communicate :D
As a participant, I'm mostly excited about the communication problem.
I feel that coding theory now being a core part of compprog will help with the "problems are too standard now" complaint I've seen again and again; excited to see how Div.2 participants handle what will be uncharted territory for many.
Can you communicate with problem writers and other people for this problem then?
Read this
As a participant that knows this participant,
this is a joke he's aware what these problems are.
Instead he will communicate with AI for this problem. /j
Lance0 said that since I'm the president of my school's AI club (whose logo is also in my profile picture...)
It's called Tomorrow Club, and we do a bunch of cool AI/Statistics stuff. Here's our social media pages:
You can communicate with problem writers using the "Ask a question" button
as a random guy, i feel really excited :DD
You again...... Wait to be hacked(probably).
UPD:
Damn, I have some things to do, you're lucky.
I will not submit D or E hehehe :DD
Agrrrr...., i can wait to other contest
As a participant, I recommend participating in this round because the problems by SpyrosAliv are really cool!
What is a communication problem ?
Think of it as two bots talking to each other to solve a task. You write both bots.
First communication problem in a rated round? Time to finally talk… to my compiler!
Interactive + communication = the round will talk back when I get WA
gotta love getting destroyed by communication problems
Oh no I have to unleash my communication skills
Thank you for organizing! I’m excited to try the communication problem. Good luck to everyone!
orz
Hope I could arrive at 1700.
Gyrosi round
As a tester, I'm not sure how to communicate effectively.
Can we communicate, bro?
Ruler always communicates smoothly
.
As a participant, I struggled with communication, but Codeforces will help me fix it.
it's interesting to try a new type of problem for codeforces!
Super Excited for the communication problem :)
Disable hacks on first few problems will also make it harder for those who intend to drop rating harshly. Maybe we will never see new negative rating accounts then? XD
Hope I reach expert again!!!
Guide me, I have never given any codeforces contest before,I have been practicing on Leetcode and can solve 2-3 problems in contest, Should I enter this contest? Or div 3?
Try this contest , u will be able to solve first 3 problems within 1:30 hrs max.
But before attending , try atleast one problem of interactive type and communication type at codeforces.
For interactive ,
https://mirror.codeforces.com/problemset/problem/1807/E
For communication ,
https://mirror.codeforces.com/contest/2168/problem/A1
I can't wait for this match.
Excited for the contest !
Hopefully, I can learn something new after this contest.
i hope +delta from this contest
All the best to every participants,
HOPE, I reach Expert today.
Problem B – “Supremacy”!
cooked
got stuck for finding pairs of l and r in b :(
but were you like prove / check that we can do in 5 steps before finding
landrnah i was also checking if we were all=ble to cover all ones in less than eq to 5 ops
yeah .. ok
I was thinking that sometimes authors play tricks in problem B ..
so maybe there is a solution to do it in like 2-3 moves every time... but I was not able to prove anything like that
I thought 4 moves would be enough, but it gave me a wrong answer on test 1. Then I realized one more operation should be added between p₁ and pₙ.
awesome and congrats for +ve delta
can you please share basic idea of what you doing in
Cwithbinary search.. I upsolved it using editorial's 2 pointer but would love to know more ideaslike how u quickly check reachability ?
Once you move down to the second row, you can’t go back up.For most problems with all pairs l<=r we can fix one endpoint — here, I fix L. For a fixed L the value of R is monotonic , if L-R is reachable then L-(R+1) is also reachable.So, I perform a binary search on R to find the minimum valid R for each L.To check if a given R is valid I run another binary search that determines how far we can go — the maximum right position from the upper grid and the maximum left position from the lower grid using this mid value.
ok, thanks for sharing your idea !!!
11k submission ==> 3k submission we can a long way
me
me for the second time
What is WA test 6 at D1?
I was also having the same issue looks like we have used more queries
if (!(cnt <= max(300, (n / 2) + 2))) while (true);when I do this it gave me tle.kitchenForcesbecause I got cooked byB....spent too much time on
B.. no more time for other problems..good bye
expertrank ...and I would like to
no thanksauthors for including communication problem LMAO !!communication problem was last problem, which most of the div2 participant won't even reach and read LOL
B are just tripping man
Ok buddy
sauce of your profile pic?
HSR Evernight
How to find l and r in b ?
hint: try to make the max 1's in the string that you can regardless of the input. then check the condition.
Actually I tried this way but it doesn't work. Like I already find the max and min in the array and then from 2<=i<=n-1 if arr[i] falls between min and max then I set a[i]=1. But it doesn't work.
I too thought the same but if you look carefully in the question they mentioned for each i btw l and r, so we need to check for each i in btw them.
solution always exists when in x ,x[1]!=1,x[n]!=1,x[pos where p[i] is n]!=1 and x[pos where p[i] is 1]!=1, as for soln find pos of 1 and n in the permutation, let it be a and b. now you can guarantee a sol'n using only these two position. think how?
You can always use 5 operations to create the string s (if it is possible to create s).
1 — (index of element 1)
1 — (index of element n)
n — (index of element 1)
n — (index of element n)
min((index of element 1), (index of element n)) — max((index of element 1), (index of element n))
This always ensures that all the elements other than 1, n, and elements which are at corner are marked as 1.
this intuition is so legit
no way !!!!!!!! .. so cool.
I started thinking in this direction but dropped it .... aaahh !!!!!
I think authors assumed that people will be able to see this solution that is why this problem is B
Problem B just cooked everyone, from newbies to experts
B — the worst problem
What's the point of codeforces? Idk how many people can relate bt I started cf after a long time and I came back because I loved the excitement of being able to solve challenging problems. I also wasn't able to do B, bt I am still going to try and solve it after the contest. If we are not able to solve something, that just means there's room for improvement! If atleast 1 of us was able to do it, that means its solvable and there is something exciting waiting for us at the end of the struggle to solve a problem(a better and sharper you).
Thanks contest preparers, I can't even fathom how you come up with a new problem.
PS: Saw your stats, you have a crazy cf dedication!
Yayyyyy, I actually managed to solve B without looking at any editorial or any hints! Love for CF++
Can we stop with Bs the "notice some niche property" problems? This is second round in a row that does this
Nooooo just solved $$$B$$$ 1 min after contest ended :(
make the verdict better/more informative for interactive problems, you could mention
Number of queries limit exceededinstead ofWrong answer on pretest XThe kind of verdicts are fixed, and you cannot see checker comment but in the sample.
Downvote for B...
no please don't ... lol !!!
As a D2 solver, I have no idea what D1 intended is even; just skipped directly to D2
suffering from success !!!
After almost 3 months continuous rating drop, finally I had a well-performed round. Though I can't say I have acutally improved my programming ability, but this is definitely one of the happiest moments for the past few months.
I'm not sad or angry at all, because I know that a setter puts in a lot of hard work to create a contest. Sometimes, it even takes 2–2.5 years.
The problems are completely out of order of difficulty, and the checker for Problem B was also wrong. I highly recommend this contest be unrated.
Sorry for B being harder than usual
However, checker being wrong was caught fast and affected less than 5 people.
B literally shook my confidence
The correlation between the ranks this round and the initial ratings is probably the lowest it's been in recent memory. It absolutely fails to reflect the participants' true skill.
Unrated is really better, maybe?
Please try having better opinions
I've always loved participating in your contests but I'm extremely sorry to say that this was one of the worst contests I've ever participated in :(
I'm sorry about that, I would encourage you to read the editorial. In my opinion the ideas used to solve B is nice, and the only fault really was not putting a problem between A and B. Don't hate a problem just because it was hard for you.
Apart from the temporary stress and rating loss, I'm happy that I can learn something from these problems!
Maybe problem B had a bit of randomness in distinguishing participants? At least judging from the standings, quite a few people solved C but not B. From the editorial, B is indeed simpler. I hold a neutral view toward such problems appearing in a difficulty-ordered contest, but it's clear that this contest had no sufficient reason to be unrated.
Wangzai I love you!!!!!!!!!
B is much harder than C.
And E's solution is so Ad-hoc.
please share idea of
Cnvm... editorial is here !!!
Can you tell me the approach to B. Mine was — I tried this way but it doesn't work. Like I already find the max and min in the array and then from 2<=i<=n-1 if arr[i] falls between min and max then I set a[i]=1. But it doesn't work.
https://mirror.codeforces.com/blog/entry/148125?#comment-1324669
please unrate...it is so difficult
this contest makes "having alt-accounts" logical
Are problem B tests correct ?
I believe so, do you have any reason to believe that it isn't?
Can you tell me the approach to B. Mine was — I tried this way but it doesn't work. Like I already find the max and min in the array and then from 2<=i<=n-1 if arr[i] falls between min and max then I set a[i]=1. But it doesn't work.
The editorial is posted, you can refer there for solution and example code. If your approach is correct and you still are getting wrong answer, you likely implemented something wrong.
Yeah I got that But everyone seems to done the same approach but mine was to create the max 1's string possible but doesn't;t work out I just want to know what is wrong with.
Then you probably implemented something wrong, if your approach is correct.
https://mirror.codeforces.com/contest/2163/submission/348360732
Have a look at it and please tell me where I am wrong
I think they are also not correct. I wonder how this solution passes " [thisone] "(https://mirror.codeforces.com/contest/2163/submission/348303770).
on this test 1 10 7 8 9 10 3 4 5 6 1 2 0000000000
in the ranges it also gives l = 1, r = 9, yet the problem said that for every i such that l < i < r, min(pl,pr) < pi <max(pl,pr). And note that in the test , 10 which is the maximum is included in the range 1 9
Read the problem statement again. If x[i] = 0 then s[i] is allowed to be either 0 or 1. So if the given binary string is all 0, any set of at most 5 valid operations is correct.
Thank you so much!!!!!, I have seen how I was not reading the problem well
348349302
It gets TLE even though it clearly shouldn’t. I wasted 20–30 minutes just because I thought it was a bug in the code, and it was actually a WA.
Yep, sorry about that. You were probably reading from a closed stream because of a WA. Perhaps we should have participants read in 1 or -1 (depending on if the answer is correct) before proceeding to the next case.
One of the best contests in a while imo (might be biased due to solving E)
348365179 I need your guidance qaq I know I should explain my thinking, but the steps to prove it are too long, and the submission hasn't opened yet.
Third time in a row where b is harder than c. Whats going on.
Problem B always requires an interesting solution.
For problem B .
Can someone tell what case i am missing ?
I got WA on TC 522 .
*``
~~~~~~~~~~~~~~include <bits/stdc++.h>
using namespace std;
define ll long long int
int main() { int t; cin >> t; while (t--) { int n; cin >> n; vector arr(n);
int p1 = 0, pn = 0; for (int i = 0; i < n; i++) { cin >> arr[i]; if (arr[i] == 1) p1 = i; if (arr[i] == n) pn = i; } string s; cin >> s; bool f = true; for (int i = 0; i < n; i++) { if (s[i] == '1' && (i == 0 || i == n - 1 || arr[i] == 1 || arr[i] == n)) { f = false; break; } } if (!f) cout << -1 << endl; else { vector<pair<int, int>> ans; int ls = p1; priority_queue<int, vector<int>, greater<int>> pq; for (int i = p1 + 1; i < n; i++) { while (!pq.empty() && arr[i] > arr[pq.top()]) { int z = pq.top(); pq.pop(); ls = i; } if (s[i] == '1') pq.push(i); } if (ls != p1) ans.push_back({p1, ls}); priority_queue<int, vector<int>, greater<int>> pq2; ls = p1; for (int i = p1; i >= 0; i--) { while (!pq2.empty() && arr[i] > arr[pq2.top()]) { int z = pq2.top(); pq2.pop(); ls = i; } if (s[i] == '1') pq2.push(i); } if (ls != p1) ans.push_back({ls, p1}); ls = pn; priority_queue<int> pq3; for (int i = pn + 1; i < n; i++) { while (!pq3.empty() && arr[i] < arr[pq3.top()]) { int z = pq3.top(); pq3.pop(); ls = i; } if (s[i] == '1') pq3.push(i); } if (ls != pn) ans.push_back({pn, ls}); priority_queue<int> pq4; ls = pn; for (int i = pn; i >= 0; i--) { while (!pq4.empty() && arr[i] < arr[pq4.top()]) { int z = pq4.top(); pq4.pop(); ls = i; } if (s[i] == '1') pq4.push(i); } if (ls != pn) ans.push_back({ls, pn}); cout << ans.size() << endl; for (auto &p : ans) cout << p.first + 1 << " " << p.second + 1 << endl; } }} ```~~~~~~~
wait for the case
13 2 3 1 4 5 6 7 8 9 10 12 14 7 10
0 0 1 0 1 1 1 1 1 1 1 1 0 1 0
ur output ->4,1
4,14 12,1 12,14
in this we miss out for 12 as 1,4 does not cover the number 12 4,14 does not cover 12 number and 1,12 does not cover 12 number 12,14 does not cover 12 number.
If you have any doubt message me.
Screencast with commentary
Seriously couldn't solve E in 30 minutes. Incredible problem, well done!
I think problem E is unsuitable to be honest, the problem statement is kind of nonsensical and just having to transmit 1 bit with some contrived conditions on what you are allowed to send, while the solution has basically nothing to do with the connectivity.
Well, yes, I don't really like the solution. But the fact that it is possible is amazing, and I think the problem statement is beautiful. This actually feels like a different kind of problem, all my experience in cp didn't help at all. It would be frustrating in a rated contest (and it is a bit frustrating even now), but I appreciate the novelty.
For me it is one of the problems which are cool because the solution is dumb (like when I first heard of linear preprocess and constant query RMQ/LCA).
for D1, I misread $$$max(300, n/2 + 2)$$$ as $$$min(300, n/2 + 2)$$$, and I actually solved it with that wrong constraint, lol.
Does anyone find the language of problems difficult?
B was actually awesome!
It's surprising how so many people (including me) found it hard to come up with such a simple idea during contest
2163D1 - Diadrash (Easy Version) Weak D1 Testcases?
My solution ends up making about
n/2 + log(n)queries in the worst case, but it still got accepted. I actually came up with this idea during the contest but spent a lot of time trying to find a better approach because I thought it might exceed the max query limit. After the system testing, I got AC with the same idea.Submission Link
Failing Testcase (requires 5012 queries)
It shows "hacked" verdict
yep, someone uphacked it using the provided testcase above:3
I became pupil by solving A and C : )
Problem B ;)
"Even though the problem statement was a bit unclear, I still really liked problem B. It actually took me over an hour to solve it, haha."
Thank you for the interesting contest, the tasks were really interesting!
Hello
Hello, I received a message that my submission 348341053 for problem 2163C coincided with others. I want to clarify that I wrote this code myself during the contest. I can provide my rough work, notes, or explain my thought process to show that I was the original author. This is my first such case. I would appreciate it if my submission could be reviewed. Thank you.
Subject: Request for Review of Similarity Warning –348315939 Codeforces Round 1063 (Div. 2) Dear Codeforces Team, I recently received a warning regarding similarity between my submission and those of other participant. I would like to clarify that I wrote my solution entirely by myself. I sincerely respect the Codeforces rules and would never intentionally violate them. The B question is constructed in such a way that there is a easy few lines of code solution and comparing small codes is the same as putting a plagiarism check for printing "hello world".I can provide anything you might need to validate myself.I kindly request you to review this warning once again and consider removing it from my account. This affects my profile integrity, and I assure you that I always participate honestly and fairly. Thank you for your time, understanding, and for maintaining a fair competitive environment.
Hello. I just received a plagiarism warning for my submission [348361441] for problem 2163C. I want to clarify that this is my own original code, written entirely by myself. I have never shared my code with anyone, and I don’t even know the other user mentioned (Ash_26). I am from Iran, 16 years old, and a student, while that user is from India and a university student, so there is absolutely no possible connection between us. I always participate fairly and independently. I even wrote a blog post about anti-cheating on Codeforces because I truly care about honesty in competitions. Please check again — this is a false positive. It’s really disappointing to receive a strike after working hard to reach Pupil. I kindly ask you to recheck the detection results manually. Thank you very much for your time and understanding.
I am his teacher. He was in the school computer lab during the contest and I personally confirm that he wrote all his code himself without copying from any source. SpyrosAliv
Hello, I received a plagiarism warning for my submission for Problem 2163C. I want to clarify that I did not copy anyone’s code and did not share my solution to anyone. My submission might look similar because the problem has a very direct and unique solution pattern (prefix/suffix mins and maxes with a priority queue). The user mentioned (Terr0r) if from Egypt while I am from India, please observe that there is no possibility of sharing solutions. Many independent implementations can appear similar when the logic itself is standard. I always code on my own and never share or copy solutions. Please kindly review my case, I can share my reasoning steps or drafts if required. Thank you for your time and understanding.
— vennelaa_
Hi, I received a plagiarism warning for my submission [348317500] on Problem 2163B, which was marked as coinciding with other submissions.
I would like to clarify that I did not copy or share my code with anyone. The logic for this problem is quite simple, so similar solutions may have appeared coincidentally.
I kindly request the team to please recheck the similarity and review my case once more.
— vivek2711
orz
Wow communication problem i like it >_<
Tks
is there any way to see complete test case? 349253870