How I solved E: open Minecraft and

Well, I'm not a high level contestant like you, but recently I had a break of 3844 days between two codeforces contests, 10 years and 6 months! Lol I participated in ACM ICPC world finals in 2013, did some contests around that time and never opened codeforces again until recently :P

As a tester,

I see a lot of hate in the comments. I hope it doesn't discourage the author from making more contests. I personally found ALL (div1) the problems very interesting!

If you view the problem as: "construct the smallest possible final strip such that you can create a 'walk' on the strip that corresponds to the original string", then you can see that any repeated 00 or 11 is 'suboptimal'. From there there's only one possible string you can create and you can simulate it.

Of course, formally proving that the final string cannot contain any 00s or 11s is harder (and I eventually did it using the same solution as the editorial) however for my original submission I just used 'proof by AC' after it passed the samples.

bro came back and placed #3

As a problemsetter, hope you have fun!

Div.2 will be as hard as Div.2, and Div.1 will be as hard as Div.1 :)

As another problemsetter, Good Luck.

Just curious: how this complaint is related to todays contest? As it seems to me D is the only problem of that kind today. But A, B, C and (probably)E have pretty short implementation so D is compensated by this, no?

+1 jdurie i think you did a good job.

Cauchico orz

I might be biased, but I find it impossible to appreciate Div1 D. I guess messy caseworks are not interesting to most participants, so problem setters might consider using fewer such problems.

Loved the problems.
Thanks for the round!

I thought each of ACD is quite nice (despite losing 15 rating) and B is ok Are you sure youre not biased due to losing rating?

Hope to get postive delta!

All images for the problem and editorial were created in Minecraft lol

Two boring implementation problems in 1D and 1E. Worst round of this year.

Skill issue

Problems that, as far as I'm aware, are "actually ad-hoc":

1. https://dmoj.ca/problem/dmopc21c9p4
2. https://dmoj.ca/problem/egoi23p8
3. https://dmoj.ca/problem/dmopc21c7p5

I agree that the "ad-hoc" label is very frequently misused, but that does not mean that such problems don't exist at all.

okay, true

Reyna breaks this record and has 1313 days.

If you are failing to come up with good problems, try not to propose a contest next time

Div1 B / Div2 D seems quite similar to https://atcoder.jp/contests/DEGwer2023/tasks/1202Contest_j :)

Please suggest other options when the authors did not want to prove solutions.

For me, the funniest thing was — "The sufficiency of this condition is obvious, and the necessity is left as an exercise to the reader" which I gave in the blog as a basic example

Cauchico orz

conqueror_of_tourist could you share how you managed to solve C in just 4 minutes? Did you see the problem before? It's hard to believe it could be that easy

There is a very simple solution for F (one without any data structures).

I would say there are some ad-hoc problems indeed (lazy to list but not negligible) but I mostly agree to your claims. Sometimes people can't do induction and say how genius this ad-hoc problem blah blah, I've seen this shit too many times. If they can't do induction they should prly go study DP first.

On point 2 I'm not sure, I think those "adhoc" problems are fine, but then why ds problems are not fine if they do? Probably because in ds problem those techniques are important and you should actually care, whereas "adhoc" problem they are kinda shit and nobody have to care. A weird sort of underdogma, but I can see why someone would want to shape contest that way... kind of?

(disclaimer: this is not related to today's contest and in fact I didn't read anything except F today)

Cauchico orz

Div1C is a much easier version of problem 1394E. I know that it is obviously an overkill to use the solution of that problem, but this obviously brought some unfairness to this round. Added that B was also originally an atcoder problem, this round just can't be called perfect.

I myself have done both the atcoder problem and 1394E beforehand, so I can quickly think of both solutions (though the one for C was the more complicated one). It was my fault to make a mistake when implementing and messing everything up, but I did feel that seeing these problems will indeed make solving them in the contest easier.

I'm not blaming the author or anything because the problems were not intentionally copied so that's OK. But maybe next time just devote more time in the contest? 1394E is right from codeforces and many high rated users have done it. If more testers were involved or if only the author could google a bit more about this folding process, situations like this could definitely be avoided.

Hope for high-quality problems!

Cauchico orz

Cauchico orz

The sponsor is not available.

Send this after you become an IGM/LGM to farm contribution.

Cauchico orz

Why are bounds so tight in D1E? My solution only uses 348800, do you think it is a good practice to set bound to 350000 and also make your solution fail?

That's why I can't get to GM. They think different

Cauchico orz

I am curious about why this contest changed into div1,div2. I remember it was div1+div2 until yesterday.

Cauchico orz

As an author of the contest, problems are great and challenging.

IOI rule brings more strategic choices. Remember that your task is to maximize your score.

Good luck & have fun!

He increased your iq ,as you won't get scammed again....He was genuine:)

I think you will get upvoted if you mention the fact that you are also a team member of us.

Can we just talk about how disgustingly simple the code for D2E — Folding Strip looks??

I also want to point out many math concepts you see that are simple to describe and code (think pigeonhole, gcd, basic combo formulas, etc.), are only able to be used in beginner problems because experienced setters have seen them so many times they think it's accessible.

Yet, such ideas are not inherently any more beginner friendly to someone just starting cp than many algorithms you see in an intro CS class (except maybe being a few less lines to code). We just slowly sway the community to only adjust to seeing some types of ideas more than others, and then say these ideas are easier because more have seen. Why do we want to keep people contained to only seeing boring ideas that don't actually have much programming?

Why do you think someone skilled would want to contact noob at topic with no incentive for them?

Just try to search some public competitive programming discord and discuss there.

I was at world finals last week so I trained a little and found I might still be able to do some competitive programming

And then low level participants get this false information that they are just plain not as good at logic, while there are also many hidden prerequisite ideas that make a problem easier.

Yes, a lot of tricks will turn problems easier, and that will allow you to get more time for harder problems, but if you can't simply know every trick that will be om every problem. Strong contestants will know a lot of tricks but theyll figure out the ones they don't know. It is super common for two people to solve a problem while one of them says that it was an application of an idea for another problem and the other one just came up with that idea on the spot.

The difference isnt that the low level participants are "bad at logic", they're bad at the logic that regularly appears in programming competitions. As you do more problems you learn which things to look out for and become accostumed to making arguments that result in fast algorithms. I don't think that is "learning hidden prerequisites", thats getting better.

Cauchico orz

sorry, too mad last night, apologize to the authors and everyone else affected by the hate i sent

really should reflect on my contest strategies and problem solving skills now

As a tester, I tested

Totally agree. I enjoy the contest iff coordinator == antontrygubO_o

Feeling that other coordinators don't know what a good contest should be like.

I had a construction for 1E that used at most $$$4 \cdot 10^6$$$ operations instead of $$$4 \cdot 10^5$$$ :( though I guess that's where the difficulty of the problem is meant to come from.

We already have a div1 drought, why do you want to make it worse?

Another problem — unclear codes to the tasks. Besides the fact that solutions are without proofs, codes to the tasks are also bad. What's the problem to write clear, understandable code?

I always understand Tourist's or Jiangly's codes better than the ones in editorial

Hi, thank you for reporting.

I tried to check the interactor and found one possible reason.

In your solution, for the case N > 3, you are assuming that the interactor will reply with CORRECT after the second query and immediately read the value of the next N. However, the interactor still replied with INCORRECT after your second query followed by a numeric string, so the next value of N that you read can be any arbitrary number. Then, you print a string of length N to the interactor, which may be an arbitrarily large string.

The interactor uses cin to read a string in a line, so if you send a very large string to the interactor, it may raise the memory problem. I can prevent this thing from happening in the interactor by limiting the number of characters to read in a line, but doing any changes to the interactor without further testing can be risky, so I decided to leave it as of now.

I suggest you properly handle the CORRECT and INCORRECT replies by the interactor. We apologize for the imperfect interactor.

Can somebody explain their solution for problem G?

literally one of the best contest statements easy to understand not annoying respect for problem setters

I might be the unluckiest man ever to get 6 second places in all ABCs while never getting the first place :( Why are these Japanese so fast :cry:

antontrygubO_o please restart coordinating rounds :)

OK I understand that. I can now accept all my accounts except for this one getting banned.

To be honest, I was too childish before so that I don't relize the seriousness of violating the rule.

I am here to show my sincerest apology and promise I will never use multiaccounts.

And thank you Rvess, you correct me from a wrong path.

I don't have a strong opinion on this and am just asking out of curiosity, but do you think a coincidence like this has a real impact on the standings? That problem had 67/104 submissions, and I feel like there are very few people whose performance on 1B increased because of that past problem.

Good joke.

orz wyrqwq

Mortals maintaining their consistency to get good at Codeforces, meanwhile bro, a respected Immortal placing #3 after ruining consistency.

As a tester, I comment after purp4ever for a living.

I don't think the Div 2 portion of the contest was particularly bad anyway.

As a tester I don't know what to comment.

I also did an unofficial contest as an author. link

Proofs take big time, and other things such as developing tasks and communicating with the coordinator (my friends spent a year on the official round) take a huge amount of time. Because of this, I do not condemn the authors.

This blog is like a recommendation because this problem is very common. I understand that most authors do not do the contest for money, but this problem of editorial articles can also be eliminated :)

The break was long, but the newer performance wasn't good... T.T

As a tester,I also tested :)

GLHF!

Pretty sure half of the people who read the problem saw the powers of 2 thing, but how to apply it wasn't immediately obvious.

Task statements in English are unavailable

Cauchico orz

As a person who read three body problem, hope to solve three problems.

Guessforces

Literally just guess greedy works in C and get +30 rating orz

max = 281150 sol : 258515976

I am rui_er fan!

Unfortunately I can't participate, but GL&HF to you!

Brother, I also have a nice counting problem: count all strings of n bits which have exactly k bits equal to 1. My very clever solution: generate all bit strings and count the number of bits for all strings. Use $$$dp[i]$$$ = number of strings with k bits of 1. Complexity: $$$O(2^n * n)$$$.

(It is not really a counting problem when you have to generate by brute-force every possible combination. (By the way, for my "bit counting problem", the answer is "n choose k", which can be calculated in O(n) instead of O(2^n * n)).)

Definitely not. They should ban 6 your accounts except the original one.

I'm sorry if my messages seem harsh and aggressive to you. I just baited you for response (tbh it was a mistake from your side responding that way)

I messaged MikeMirzayanov and BledDest. I'm not even sure they will read my messages. Do they even care? That's why we see so many empty profiles with 10 contests and 2500> elo.

Agreed. It will also be cool, if there is a discussion section per problem, where people can share their insights, observations, proofs or intuitions, just like Luogu.

OMG ruler round

She always come back

So sad that I can't participate this contest, GLHF

You are biased towards this particular problem style.

(fwiw i found this at least median contest quality, no problem is too stupid, none is my favorite but it at least feels well thought.)

just two dijkstras: d(1,k)+d(k,n) ?

I was 1599 yesterday lol

It's hard to come up with a problem that is 100% original, ofc there will be things that will repeat between problems. The point of adhoc is that it is reasonable to come up with what you need in contest if you have enough skill, not that seeing a similar problem beforehand won't help.

For example, I don't see how this problem fits your philosophy for example: https://codeforces.com/problemset/problem/1817/D . I've shown it to an IPhO silver who barely did cp and he eventually solved it. You can't tell me it's because he seen a lot of those before, lol.

From personal experience, the people who are better adhoc arent the ones that know all the "hidden tricks", those tend to be people who are good at stuff like ds and try to apply the same logic to adhoc (which can work to some extent). They test something inherintly different and different people have different levels of affinity to it.

antontrygubO_o please restart coordinating rounds :)

1966B: the hidden technique is considering invariants: what are properties that the operations can never change? E.g. if an edge of the matrix is same-color, then no operation can ever change the color of any cell on that edge. There are many problems like that.

I can certainly say that the process of solving that problem was almost formulaic for me. "Answer is obviously yes if some corners have opposite colors" -> "What happens if none are" -> "What can never change" -> "Ok, solved".

I also participated in "today"s round after 1016 days.

I would not call D ad-hoc. It is type of trick "brute force small case that works and hope it's extendable by repeating same pattern" which is very common. I think more specifically you even see this idea particularly applied to building up pattern of digits.

I think the reason that particularly feels trollish is the statement gives hopes to more interesting generic insights in number theory, but instead it reduces to such boring reused idea.

We are getting closer to the top -1 contributor with this blogs

It is not hard to teach something you are good at but no longer care. It is not uncommon to come back to what you know and have fleeting feelings of wanting fulfillment from helping others.

Seems like it can be solved with mincostflow approach, there are two units which go from source to start and end vertices and two units can be transfered from vertex pk to target, you need to find min cost max flow while the cost of transfering one unit between vertices is 1, except source and target vertices. (Sorry for bad english)

As we all know, cute rui_er is a kind of ruler.

minecraft is useable to create edtiorial picture of D1E.