Привет!
Codeforces Round 481 (Div. 3) начнётся в 13.05.2018 12:05 (Московское время).
Это будет второй в истории Codeforces раунд для третьего дивизиона. В этом раунде будет 7 задач, которые подобраны по сложности так, чтобы составить интересное соревнование для участников с рейтингами до 1600. Наверное, участникам из первого дивизиона они будут совсем не интересны, а для 1600-1899 покажутся простыми. Однако все желающие, чей рейтинг 1600 и выше могут зарегистрироваться на раунд вне конкурса.
Раунд пройдет по правилам образовательных раундов. Таким образом, во время раунда задачи будут тестироваться на предварительных тестах, а после раунда будет 12-ти часовая фаза открытых взломов.
Напоминаем, что в таблицу официальных результатов попадут только достоверные участники третьего дивизиона. Как написано по ссылке — это вынужденная мера для борьбы с неспортивным поведением. Для квалификации в качестве достоверного участника третьего дивизиона надо:
- принять участие не менее чем в двух рейтинговых раундах (и решить в каждом из них хотя бы одну задачу),
- не иметь в рейтинге точку 1900 или выше.
Независимо от того являетесь вы достоверными участниками третьего дивизиона или нет, если ваш рейтинг менее 1600, то раунд для вас будет рейтинговым.
Задачи для вас подготовили я и fcspartakm. От души спасибо тестерам AGrigorii, BigBag, nhho и Sert!
Удачи!
UPD 1: Спасибо за участие. Опубликован разбор задач.
UPD 2: Поздравляем победителей! Топ-5 (официальные результаты):
Really!
I didn't Say That I made this Picture I just post it
I hope less "dfs and similar" and "dp" and more "implementation" . thanks :D
You're probably the first person to ever say that
Also, some of the easier problems with those tags aren't very different from implementation problems.
okay , I will try to solve them thanks :D
Don't have that kind of attitude. You learn much more from graph, dfs and DP problems rather than implementation problems. Learning is more important than getting AC :)
Why do you think that dfs is simpler than 2x2x2 Rubik cube rotation? (there was such a problem on CF)
can u provide link to that problem or name please
http://codeforces.me/contest/887/problem/C
Are contests of div 3 always same educational rounds?
Are contests of div 3 always same educational?
there are some reasons to start with rules of Educational Rounds: Link
will it be possible to do the problems after the contest has ended?
You can always practice problems in the problem set after contests.
you can! awesome thanks for the reply
Hope it won't be Readforces round! :D
Why the unusual time? Finally on a weekend, but in the middle of the night :( :(
The time is too bad !!
I am a Chinese, the time is really good in China! It is 17:05, I can race who can solve problem faster with my friends!
so good luck : ) , don't let anyone hack you!
What about Educational Round ? Will that round continue or Div3 is the replacement ?
No, there will be educational rounds
Last weekend was also a div3 contest. That would be great if we could alternate between div3 and div2 contests on the weekends!
And Div1 was never to be seen again... :(
Why do you guys each time override the rules and create ambiguity, by saying "it's only for trusted participants", then, "regardless of whether you're trusted or not, if your rating < 1600, it'll be rated for you". Why do you even bother to mention "trusted users" or that it's to combat unsporting behaviour? In fact, it's not!
???????? Trusted participants are in the official standings table and it's rated for everyone currently under 1600
"Why do you guys each time override the rules"
They don't. last (and first) time was absolutely the same rule.
"Why do you even bother to mention "trusted users" "
Because it's about being included in the official standings table.
"In fact, it's not!"
In fact, it is, lol.
You have no idea of what I'm saying! Read the first sentence once again till the end.
"Read the problem statement more carefully"
Can you tell me where is the “official standing” which only includes trusted users? I see everyone gets rated including unrated ones and the ranking takes them into account.
Official standings table dosen't mean people who gets rated. Official standings table is a public table, which shows standings of the contestants who took part in the round. And from this table, because of new rules, they remove untrusted users. So they become invisible for public and there is no point in participating for the high places, like most does.
Can you share the "official standing" link for the last div3 contest? I can see everyone in "common standings", and in the rating changes, I see "those" unrated ones included. So where is that "official standing" which has only the trusted users? Please share it for the round #480
You're welcome (use right click — "open picture in the new tab").
Am I the only one who can still see some unrated accounts in the official standings??
Well, for some reason I now see them too. But, "Milhous", who actually won contest, still not here, which means for some reason this two guys suddenly become visible. Maybe bug, maybe intentional.
Yeah I think it makes some sense to be like this since hacking is still going on. Remove them from the standings table after hacking finishes; otherwise, their solutions are less likely to get hacked.
We are talking about round 479, which ended a long time ago, so.
Oh whoops you're right, my bad. Yeah, they shouldn't appear,
I got your point exactly. Why don't you reread me instead?
I can’t “reread” you sorry )))
Just to say that by saying overriding rules I didn’t mean the rules are different from the previous div3 contest, but I meant that he first says that it’s rated only for trusted users (excludes unrated users), and then says that if you are < 1600, then you will be included. This is my question.
"but I meant that he first says that it’s rated only for trusted users (excludes unrated users)"
Where? I think you are the one who needs to reread...
a. Remember that only the trusted participants of the third division will be included in the official standings table. As it is written by link, this is a compulsory measure for combating unsporting behavior. To qualify as a trusted participants of the third division, you must:
b. Regardless of whether you are a trusted participant of the third division or not, if your rating is less than 1600, then the round will be rated for you.
doesn't b. override a. condition? Either I'm missing very basic info, or it's confusing..
a. portion does not say anything about rated status of the contest.
Share that "official standings" for round #480 which considers only trusted users. I can see all the unrated participants show up in the standing.
Shared it higher. But why are you asking round 480 btw? This rule works only for div3 round, not div2.
Yeah I noticed, thanks for that. Seems like I never realised "show unofficial" checkbox and it was left checked=true so I wasn't able to get what the statement above was saying. But I finally got it and can sleep comfortably :) Thanks for your time!
Не иметь в рейтинге точку 1900 или выше значит, что если у участника когда-либо было 1900 или выше, то он недостоверный участник?
Although I rarely miss a CF round , I guess I will miss tomorrow's contest because I have to attend a national contest at that time :-(
Last one was Round #466
ops .. i can not join you . .
we will have our university acm
albaath contest
so ... i am very sad ..
How can I be 'trusted participants'.
I meant: I already take part in more than two rated rounds and I also have rating less than 1900.
But when I register, it alert like I'm not rated in this round.
This round is for 1600 below
OK. Thanks.
Я не понял: в чем смысл достоверного участия, если рейтинг для недостоверных участников все равно пересчитывается?
Достоверные участники отображены в официальной таблице результатов
И что, это типа круто?
Ну по мнению организаторов, самого Майка ну или непонятно кого еще — видимо да. Считается, что люди с высоким рейтингом создают новые аккаунты и участвуют в соревнованих для более слабеньких, чтобы выиграть их(ну и поднять свое чсв, наверное). И собственно, Майк и ко. считают, что если таких людей не будут показывать в таблице достоверных участников, то и действовать они так не будут.
С другой стороны понижается чсв топа1. Открываем его профиль и видим на графике 3-е место.
Oh!! Div-3 saved my coding career!!
Too Hard... 42mins to solve all problems...
Сколько даётся за успешный взлом и сколько снимается за неуспешный?
В соответствии с правилами образовательных раундов, успешные и неуспешные взломы напрямую не влияют на результат.
Тогда почему рейтинги меняются? Я был 56м, а теперь 45й? Разве это не из-за того что людей повзламывали?
Именно, это из-за того, что кого-то взломали (или убрали из рейтинговой таблицы). А не из-за того, что кто-то взломал удачно, либо неудачно.
То есть взлом никак не влияет на результат взломщика, но влияет на результат цели в случае успешного взлома (взломанная задача считается нерешённой).
Спасибо.
I feel that the difficulty of the contest should be a little more than it currently is
Pre-system testing, 250+ trusted participants solved all the questions, which is around 10% of all the trusted participants. This never happens in Div2 or Div1, so I feel that even for Div3 standards, the whole problemset might be on the easier side. (Maybe the last 2 questions should be harder?)
Agreed. Although having speed be part of the contest is inherently part of competitive programming, somebody who is very good at solving problems but codes slower should still be able to achieve a fairly high score on the leaderboard.
Having the problems be too easy also introduces a lot of variance on the leaderboard.
I solved E F G each in about 10 minutes but in the last 30 minutes I couldn't still solve D... I think D is harder than the last 3 problems
What is test case 7 for D? I couldnt pass it :(
Actually, I solved D, I was talking about E. D is not that hard.
There is a simple trick in D. You can find it. Try again to find that trick.
D is not harder...U r just a noob...cant see easy things lol
At least he took part in the contest
I'm not able to attempt hacking any problems. It says on the right that I am doing the contest as practice, although I participated in the round and am under 1600. Can someone help?
When the editorial will be added?
After system testing.
Test case 7 for E anyone? Is it legit to ask for test cases now?
А почему чужой код нельзя посмотреть во время фазы взломов? Или это пока временно?
Дай участникам повзламывать самим себя :)
Я посмотрю в чём дело чуть позже, сейчас идет награждение победителей внутренней олимпиады Саратовского университета.Исправил!hi everyone . I am a new codeforces user. I don't have any information about hacking. Can someone give me information about hacking
Click the user's solution twitch time to open his/her solution.
Then click the hack option. And figure out such a case that case will
Give an output that output does not match with correct output.
If don't match with correct answer hack will be successful otherwise hacked will be
unsuccessful.
Solved all problems in both of the Div.3 contest. I think it should be a little bit harder :D
At the same time, Div 3 isn't meant for you though, since it isn't even rated for you.
At the same time, old div2 was not meant for candidate masters but they struggle to solve all the 5 problems.
I did it for like 2.5 hours. Problems are very easy there. It's more like div4.
can anybody tell me why this is wrong on pretest 3 in problem G it seems correct to me . 2 2 2 1 1 3 4 0 4 4
Exam 3 starts later so you can't prepare for exam 3 earlier...
You can't study for an exam until the questions get released.
In this case, the questions for Exam 3 get released on day 8, and you're trying to study for them on day 6.
okay got it ..thanks
You can't start preparing for the test before s_i, which is in this case 8.
open hacking phase is running
But I can't open others codes why?
Thanks, I'll fix it soon. But now there is awarding ceremony of internal contest in Saratov State University.Fixed!thanks for fast response
I can only hack my code. Can anyone tell me what's happening?
UPD: It's fixed now!
I have participated in more than 2 rated contest and my rating is less than 1600 but still I am not able to view other's solution and try to hack them.Why is this happening.
Почему нельзя взломать кроме самого себя? Я из див2. регистрировал вне конкурса.
Мне кажется задача D была сложнее нежели E, F. Не знаю пройдет ли системное тестирование, но я попытался решить задачу другим способом. Вот мое решение. http://codeforces.me/contest/978/submission/38182546
В задаче D при попытках взлома получаются "неизвестные вердикты".
can someone explain the 4 th test case of E 2 10 -1 2 what should be the output??
Can someone look at these 2 solutions of C: 38161183 38162269.
The first has complexity O(k*k) the second O(k*logk) and yet the second takes 1.5 s more?? What am I missing?
38161183 is not O(k^2), it looks more like O(m) to me (m being the number of letters): p is not reset after the inner loop, it just catches up, at worst in n inner iterations per outer iteration,
but in constant time on average per inner iteration (n/m inner/outer iterations).Edit: please ignore the last brainfart. It is how Photon_ says :)
I didn't read the queries were in increasing order :D
Complexities are actually O(n+m) and O(m*lgn).
On solution #1 it uses 2 pointers, so in worst case, you will traverse whole array a over all m queries hence the running time is O(n+m)
On solution #2 it uses Binary Search over whole array. In worst case all queries will be in the start of array a so this solution will take O(logn) for each binary search hence the total running time will be O(m*logn)
You can improve second solution to binary search not in [start, end] but in [last_pos, end]
please explain the 4th case of E please please please
2 10 -1 2
why us the output 9?
There can be [1, 9] people on the bus before the first stop; if there were 10, after the second stop there are 11, and there has to be at least one before first stop.
there can ve 0 also ?
[0,w] inclusive , right?
No because the first time 1 person gets out.
it could be 1 to 9. it can't be zero because after the first bus stop -1 remains. it can't be 10 because after the second bus stop there are 11 person in the bus and its more than limit (w).
ohh my my , ok thanx , got that.
I think this guy (__123456__) is cheating he used special conditions to hack his codes
hacking doesnt give any extra points.
i know but maybe there is announcement for the top hackers & this will be unfair for the real hackers
I think it could've been more interesting or challenging if the duration were 2 hours (not 2.5 hr).
Why ProblemC need binary search? b_j is in increasing order.
Got "unexpected verdict" in hack 448105, 448122 and 448145.
Can any one help me in Problem C ? I think my code is correct but can't understand why it fails.
LINK : http://codeforces.me/contest/978/submission/38163975
change MAX = 100010 to MAX = 200010.
Is it only me or everyone else is also facing problem of latex Check this link...This is how my problem page looks
Link : https://ibb.co/c4iQwJ
Interestingly, my penalty in both (all :)) of the div.3 contests is 413.
It's weird, my Java code passed if I use int, and it failed at test case 5 if I use Boolean and Integer :|
It's not. You should learn how operator == works for objects.
The testcases are wrong. The test 21 on problem E is the first one that is wrong. The answer should be 1000000000, not 1000000001.
The testcases are wrong. The test 21 on problem E is the first one that is wrong. The answer should be 1000000000, not 1000000001.
Nope. There are 1 million and 1 possible starting values. [0, 1 million] inclusive on both ends.
1000000001 is correct.
I wonder if the ranking will be updated?
Just curious :D
I am not able to understand my rating change. I solved 2 out of 3 problems correctly (A went wrong due to a silly mistake and B, C were correct), but the rating still went down instead of going up?
Your rating is dependent on how better or worse you did compared to other participants. Losing rating means you were expected(in relation with other participants) to get a better rank than what you got.
You might be comparing your results with the earlier div 2 rounds where in general you would have gained rating on solving 2 questions, but this was an easier round than the previous div 2 rounds and contestants with lesser rating than you did better which caused you to lose rating.
It seems that my hacks on problem F (with n = 76297, ID 448348 and two others) weren't added to the final tests. Very nice.
Where's rating changes? Any announcement about it?
UPD: Fixed!
The rating I gained in this contest is now gone.What happened?
Where did my rating go after this contest? UPD. Rating returned
Soemthing strange is with ratings of participants..
ZаVтра раZбор будет?