You are not logged in. Please login at to post your questions!


How did you solve the problems in today's paper ? What's the expected cutoff ?

The first problem was easier while the second one involved some math and big modular arithmetic. I solved just the first problem. Will I possibly qualify ? Also, how did you solve the problems ?

asked 31 Jan '15, 12:42

nibnalin's gravatar image

accept rate: 0%

About the cutoff, I think it will be around 130-140 this time :'(

(31 Jan '15, 13:37) ketanhwr6★

I honestly doubt it's going to be more than 125 (all subtasks but the last on each problem). It's always been 100 in the past. You overestimate how much others are getting

(31 Jan '15, 14:19) superty3★

I do hope so. Expecting 130. You?

(31 Jan '15, 14:37) idraumr0★

200 unless I made stupid mistakes

(31 Jan '15, 14:44) superty3★

Ah, I blooped on problem 1. Curse me.

(31 Jan '15, 14:48) idraumr0★

I won't be getting more than 100, does that mean I'd fail to qualify ? :( (if only I'd studied Maths more)

(31 Jan '15, 14:49) nibnalin6★

@superty That is soo cool! Kudos!

(31 Jan '15, 14:49) sandy9992★

I doubt the cutoff is less than 100.

Edit: If even idraumr only got 130 I doubt the cutoff is more than that.

(31 Jan '15, 14:52) superty3★

Haha, you flatter me. To be fair, I know a few getting in the upper 100s, and am myself scared of my chances.

I made disgustingly silly mistakes while attempting problem 1.

(31 Jan '15, 14:57) idraumr0★

25-30 people will make it to camp, you know of only a few people getting higher than you. I don't know of anybody (excluding myself). I honestly wouldn't worry if I were you.

(31 Jan '15, 15:05) superty3★

(Now that I've a physical keyboard to answer with.)

I started with the second question, roughly following the same track as @superty. For the first question, I quickly coded the solution for subtask 3, being as it was the easiest with maximum marks. Then I started thinking of a "complete" solution, and once I realized the track, started implementing it.

I couldn't do that in time though, so I got 130 total. I was silly because I didn't realize I can solve subtasks 1,2,3 by checking which of the assumption holds true. A few of my friends did that, scoring 160. Hope that's not the cutoff.

(31 Jan '15, 16:53) idraumr0★

I intended to do subtasks 1, 2, 3 that way, but I got the full solution after I did the first subtask and then I implemented the complete solution.

After that I went back to the second problem and did the supposedly O(N^2) solution.

(31 Jan '15, 17:00) superty3★

Heh; that didn't even strike me as a possibility during the exam. I'm not sure what to think of myself.

(31 Jan '15, 17:02) idraumr0★

Well, we all overlook things. I don't know.

Any clue when the results are supposed to be out?

(31 Jan '15, 17:05) superty3★

I asked---in a week.

(31 Jan '15, 17:07) idraumr0★

Thank you.

(31 Jan '15, 17:12) superty3★

@neo1tech9_7 How much did you get?

(31 Jan '15, 19:24) superty3★

@superty 200 if I pass all the test cases.

(31 Jan '15, 19:46) neo1tech9_76★
showing 5 of 18 show all

I saw an email today with subject, "INOI 2015 Evaluation". I lost a heartbeat. Then I realised that it was just a general email -_-


answered 08 Feb '15, 21:03

ketanhwr's gravatar image

accept rate: 15%

Yeah, same. A better title would have been "Update on INOI 2015 Result Date" or something.

(08 Feb '15, 21:08) superty3★


(08 Feb '15, 23:49) animesh_f ♦6★

Heh, exactly; I thought they'd e-mailed me my scores. sigh

(09 Feb '15, 16:30) idraumr0★

It's thursday already. When will they release the result?

(12 Feb '15, 13:36) ketanhwr6★

Hopefully Friday, Saturday or Sunday.

(12 Feb '15, 14:23) superty3★

Or next week.

(12 Feb '15, 15:41) idraumr0★

Yeah, but hopefully this week.

(12 Feb '15, 15:45) superty3★
showing 5 of 7 show all

"[The cut-off is] Likely to be 100. Will have an announcement soon."


answered 13 Feb '15, 16:29

idraumr's gravatar image

accept rate: 45%

Good, good.

(13 Feb '15, 18:10) superty3★

Nothing yet.

(13 Feb '15, 20:24) nibnalin6★

Soon might be 2 days.

(13 Feb '15, 20:26) superty3★

And how right we both were, @superty. :D

(14 Feb '15, 13:10) idraumr0★

Well, I wasn't right, I was talking about the announcement, not the final result. Technically the announcement came today. (announcement of your score)

(14 Feb '15, 13:32) superty3★

Ah, I meant the announcement of the final list.

(14 Feb '15, 13:38) idraumr0★

When is the final result coming?

(16 Feb '15, 13:19) ketanhwr6★
showing 5 of 7 show all

So, I scored 100. How much did you score?

If anyone finds a discrepancy in their score, please tell here too. That would imply a re-evaluation of the cutoff.


answered 14 Feb '15, 12:35

idraumr's gravatar image

accept rate: 45%

well, unfortunately im getting 60, not the claimed 100. it appears my code trying to access out of bounds elements in subtask4, anddue to just that, ill be unable to make it to TC this year, unless of curse cutoff falls to 60... :( the worst day of my life...

(14 Feb '15, 12:59) nibnalin6★

110 :) Got exactly what I expected!

(14 Feb '15, 13:02) ketanhwr6★

@nibnalin, no issues, brother. You've got more years to try again. The cut-off is likely to be 100, unless a re-evaluation happens for several, in which case it will likely increase.

@ketanhwr, yayy! We'll meet at IOITC! :)

I couldn't score that 30 marks because I forgot the case where a[i] and a[i + 1] would be maximum, for a [i, i + 1] sequence. sigh

(14 Feb '15, 13:10) idraumr0★

I ended up getting 100 as well -.-

(14 Feb '15, 13:32) superty3★

Oh, elaborate? What did you do wrong?

(14 Feb '15, 13:38) idraumr0★

Wait a few minutes while I confirm what I know is the only mistake

(14 Feb '15, 13:38) superty3★

I got 130 instead of 200 because of a stupid initialization mistake. Thank god everyone made equally stupid mistakes :P

(14 Feb '15, 13:41) neo1tech9_76★

Yeah haha.

(14 Feb '15, 13:58) superty3★

Initially I had:

int curmin = a[1];

int cur = a[1];

This wasn't working out so I changed a[1] to -a[1] + b[1] but forgot to change curmin -_-

One or two cases in each subtask are wrong because of this

int curmin = a[1];

int cur = -a[1] + b[1];

(14 Feb '15, 14:00) superty3★

Ouch. I hope someday this happens with an online judge, or maybe an offline one with all the test cases with output hashed and present there.

(14 Feb '15, 14:03) idraumr0★

The cutoff better be 100

(14 Feb '15, 14:05) superty3★

I think the cutoff is, as of now, 100. It would change if a sufficient number of people ask for re-evaluation of their programs, and they then score greater than 100 (and previously they scored < 100).

(14 Feb '15, 14:08) idraumr0★

I had ll curmax = ans; changing it to ll curmax = A[n - 1]; got me 100

(14 Feb '15, 14:18) neo1tech9_76★

My score is 110 .
It was 75 on the grader.
The time limit is 8 seconds for JAVA and my code works within that time , (7.5 seconds) in the worst case N=150000 , but it gives some weird Wall Clock Time Limit Exceeded error instead of Execution Timed out . They said they checked my algorithm and its OK so my score is 110. Hopefully cut-off won't increase!

(14 Feb '15, 14:20) animesh_f ♦6★

I think the wall clock thing means you took more time on I/O. At least, that's what it means on wcipeg

(14 Feb '15, 14:26) superty3★

Java I/O is really slow. I can't help it. I can't memorize a huge template for INOI.

(14 Feb '15, 14:29) animesh_f ♦6★

Yeah that's probably why they gave you the 110.

(14 Feb '15, 14:30) superty3★

Just to be clear, had they already given you 110, or did they only re-evaluate once you told them that your solution works within the time limit?

(14 Feb '15, 14:39) idraumr0★

"They said they checked my algorithm and its OK" I'm guessing they looked at his program and probably realized it timed out only due to Java I/O and gave him the score later.

(14 Feb '15, 14:41) superty3★

@superty, probably, or maybe they had already looked at the algorithm earlier and given him 110. That would conform with our assumption that they actually read most of the programs.

(14 Feb '15, 14:45) idraumr0★

When I checked on the grader it was 75 so I immediately emailed them . They replied that they had checked and it was OK. They also said , "As things stand , I should qualify." . So I guess my score is 110.

(14 Feb '15, 14:50) animesh_f ♦6★
showing 5 of 21 show all

How many of you are getting above 110, please comment. Hope not many :D


answered 31 Jan '15, 17:33

jayaganesh's gravatar image

accept rate: 0%

See above answer, please.

(31 Jan '15, 17:35) idraumr0★

I am pretty sure about these 21 people who are expecting above 100. Rajat De
Sameer Gulati
Siddhant Bansal
Daksh Anand
Tanay Kothari
Malvika Joshi (duh)
Mriganka Basu Roy Chowdhury
Sagnik Majumder
Jaya Ganesh (you)
Shikhin Sethi
Saarthak Sachdeva
Arjun P
Vikram Nitin
Ketan Gupta
Aditya Puranik
Debjit Paria
Kartikay Shandil
Nalin Bhardwaj
Sandhya Saravanan
Marshall Dev
Abhinav Baweja
I think cut off will be around 110 or more this time.

(31 Jan '15, 20:31) animesh_f ♦6★

Where did you get this list?

(31 Jan '15, 20:40) superty3★

I've asked all these people. :) :D
And there might be more also. I expect around 35 people to get more than or equal to 100 this year. Then it might get interesting.

(31 Jan '15, 20:42) animesh_f ♦6★

sandy has an answer on this very question saying that she didn't implement the second question and simultaneously asking about the last subtask of the first question...

(31 Jan '15, 20:43) superty3★

Okay . This is considering only the codechef group of people and the IOI and ACM fb group , there will be more people . Lets keep our fingers crossed. :) :D

(31 Jan '15, 20:47) animesh_f ♦6★

OMG Please please remove my name from the list. Whatever I have posted, I never did in the exam. And I brute forced the 1st one. I doubt if I'll even get more than 10! I should learn to manage my time better I guess.

(31 Jan '15, 21:01) sandy9992★

Thejas went to camp last year so I guess it's reasonable to assume he got at least 100.

(31 Jan '15, 21:21) superty3★

There are 4 people who I have not included in the list and qualfified for IOITC last year. If you add them and remove Sandy999 , we have 24 people expecting above or equal to 100. :P

(31 Jan '15, 22:04) animesh_f ♦6★

Interesting; I've hopes high, as of now. Keep us posted, you well connected folks! :)

(31 Jan '15, 23:37) idraumr0★

Updated List:.
Rajat De Sameer Gulati
Siddhant Bansal
Daksh Anand
Tanay Kothari
Malvika Joshi (duh)
Mriganka Basu Roy Chowdhury
Sagnik Majumder
Jaya Ganesh (you)
Shikhin Sethi
Saarthak Sachdeva
Arjun P
Vikram Nitin
Ketan Gupta
Debjit Paria
Kartikay Shandil
Nalin Bhardwaj
A Reddy
Abhinav Baweja
Thejas R
I think cut off will be around 110 or more this time.

(01 Feb '15, 19:09) animesh_f ♦6★

The maximum number of people that have qualified for training camp in an year till now are 29. Let's see what happens this time.

(01 Feb '15, 20:01) ketanhwr6★

I think if they can set the cutoff such that 24 people get, or set it such than 29 people get, I think they will choose the 24 option.

29 probably happened because the next option must have been like 19 people.

(01 Feb '15, 21:53) superty3★

We only have 20 names for those with >= 100; I don't know why everyone's anticipating a higher cut-off.

(01 Feb '15, 22:11) idraumr0★

They are kind people.
They will select 29 if all get >=100. At ZCO many got 200 . Still they set cutoff = 100 and 67 people qualified. :)

(01 Feb '15, 23:33) animesh_f ♦6★

Maybe you are right. I actually have no clue, I was just speculating.

(02 Feb '15, 00:03) superty3★

Any updates? :-)

(03 Feb '15, 15:02) idraumr0★

Any update with < 100 is really appreciated :)

(03 Feb '15, 16:37) ketanhwr6★

I'm unsure of my own score, and might get 100 or 130. sighs :-(

(03 Feb '15, 17:10) idraumr0★

Why are you now unsure of getting 130?

(03 Feb '15, 17:11) superty3★

I'd prefer to discuss that on a PM; drop me an e-mail, @superty?

(03 Feb '15, 17:20) idraumr0★

@idraumr What is the error?

(03 Feb '15, 17:51) ketanhwr6★

24, did you count 'Manas Kumar Verma'?

(04 Feb '15, 20:24) idraumr0★

Yes. Did you count sandy?

Edit: Right, I counted number of lines and Rajat De is on the same line as Sameer Gulati.

(04 Feb '15, 21:00) superty3★

Add Rahul Chabbra to the list, also, Parth Dhar who sat very close to me said he solved entire 2nd question correctly, so now there are 26 ppl on the list...

(08 Feb '15, 12:29) nibnalin6★

That's not true, the count of 24 includes them both.

(09 Feb '15, 12:24) idraumr0★

How so? It was 24 already, those two are new entries to the list.

(09 Feb '15, 12:36) superty3★

No. There are 20 people on the list. Then Manas, Rahul, Parth, and Shagun.

(09 Feb '15, 12:39) idraumr0★

Ah, you're right. I didn't realise those two are the same from before. I guess I misread your comment as the list already including them (I just checked with the list in this comments thread).

That was dumb, haha.

(09 Feb '15, 13:54) superty3★

Will the camp be before or after JEE advanced?

(09 Feb '15, 15:29) ketanhwr6★

Last two years it was May 1-9, so before.

(09 Feb '15, 16:16) superty3★

If I'm not wrong, the IOI schedule has changed (by a significant number of dates)? The camp might see a change accordingly.

Too premature to worry about camp dates, anyway---at least till the list is out.

(09 Feb '15, 16:23) idraumr0★

Oh, I didn't know that. Yeah, apparently it's two weeks later than IOI 2014.

(09 Feb '15, 16:28) superty3★

I think they set the date so that no exams clash with the camp.

(09 Feb '15, 16:42) ketanhwr6★

Yes. If it clashes with JEE a lot of people wouldn't come.

(09 Feb '15, 19:28) superty3★

Exactly. Let's see. The result might come tmrw or day after tmrw.

(09 Feb '15, 21:25) ketanhwr6★

Doubt it. Probably at 10 PM on Sunday since they said "the coming week".

(09 Feb '15, 22:18) superty3★

We've got an audience of 1.0k, for this thread. I am quite tempted to think our list of 24 is comprehensive, and 100 would be the cut-off.

(09 Feb '15, 22:27) idraumr0★

It's not unique views. Less than 250 people even have access to this site.

I alone probably account for over 50 views.

(09 Feb '15, 22:32) superty3★

@superty, and 200 people could appear for INOI.

Actually, can the admin give stats as to how many unique users gave their handles for this subforum?

(09 Feb '15, 22:37) idraumr0★

I cited the 250 people thing to prove that it wasn't unique views, and because 200 people could appear for INOI.

(09 Feb '15, 23:04) superty3★
showing 5 of 42 show all

Ok guys, as all of you would have seen your scores, and most of you finalised it, lets try to approximate the cutoff:

Please choose your score: INOI score

(vote only after you have finalised your score)


answered 14 Feb '15, 19:39

codelegend's gravatar image

accept rate: 0%

edited 14 Feb '15, 19:41

I don't think we need this. The cut-off is most probably going to be 100.

The result will be out tomorrow, anyway.

(14 Feb '15, 19:55) idraumr0★

But still. Why not check it?!

(14 Feb '15, 19:56) ketanhwr6★

@idraumr the result will be out on Monday.

(14 Feb '15, 19:56) ketanhwr6★

D'oh. Sorry, totally thought tomorrow is Monday.

I've voted, for what it's worth.

(14 Feb '15, 19:58) idraumr0★

I know it is not at utmost necessity, but an estimate is definitely going to be good. We can also know the score distribution this year.

(14 Feb '15, 20:01) codelegend6★

Omg, 10 people got 200!!!! And 19 people getting >100! So cutoff probably is 110 or so....

(17 Feb '15, 21:38) codelegend6★
showing 5 of 6 show all

Same here ,the second problem was a bit tricky. Felt a bit sad . :( . Could have done better.


answered 31 Jan '15, 13:03

kartikay101's gravatar image

accept rate: 0%

Ya, once I heard the logic, it felt but-obvious, but then, the past is past...

(31 Jan '15, 13:17) nibnalin6★

About the 2nd problem, I was trying something like this, but I couldn't implement in the exam (so no use basically). Only now, I got clarity in what exactly I was thinking then.

Number of non periodic strings for 'n' = 2^n - (Sum of Number of non periodic strings for each factor of n other than n)

If it's wrong please feel free to disprove it.

Also, how did you solve the first problem? I could only think of an O(N^2) algorithm which would pass the first sub task.

I calculated the cumulative sum of b array at each index. Then I had a loop with i running from 1 to N and an inner loop with j running from i+1 to N and similarly a loop with i running from N to 1 and inner loop with j running from i-1 to 1 and accordingly updated maximum based on Ssum[i,j]. And another loop to update maxi based on the value of a[i] alone. Won't pass, I am sure.

I spent around 1 hour debugging my first problem's code, even though it was a useless solution only to find I had interchanged a + and - somewhere in between. And I was messing up the pow function in 2nd one.

Could have done heaps better. My mind seems to work better after the exam, I guess plenty of timed contests is the way to go for me ;)


answered 31 Jan '15, 13:18

sandy999's gravatar image

accept rate: 10%

edited 31 Jan '15, 13:27

I initially did some inclusion exclusion nonsense for the second problem, but that turns out to be totally wrong. Wasted 1.5 hours doing that.

Finally I decided to at least do the 65 mark subtask with what I thought was an O(N^2) algorithm. Your DP is correct as far as I know. You have to find all the divisors for all the divisors of N. I assumed this takes N^2 time, but apparently not - it worked for all subtasks.

I even checked by running my program on all possible inputs (n = 1 to 150000) and it was taking maybe 1 second to run for 100 n's even close to 150000.

(31 Jan '15, 14:18) superty3★

It's right? But I didn't implement it. I wasted time on the 1st one, 1st subtask. How bad is that?

(31 Jan '15, 14:28) sandy9992★

You have two years to go, don't worry

(31 Jan '15, 14:31) superty3★

That's exactly how I did it; it's not O(N^2) because the number of factors a number has is very small. See bounds on omega (n).

(31 Jan '15, 14:38) idraumr0★

Yeah I assumed it was something to that effect but I wasn't really sure. I assumed that some number might exist for which the time taken is quite high, but nope.

(31 Jan '15, 14:46) superty3★

:=( Will score 30 in the 1st problem... believe me, I Implemened d r8 soln. But mh soln. Was not working gor d last tedtcase of d last 2 subtasks cos I didnt make my negative infinity small 3nough. Im feeling kinda depressed. However ive got 3 more chances.... btw is there any chance of ctoff being less than 30 or 10....I know I sound stupid. ...

(31 Jan '15, 17:34) arpanb83★

Sorry, it's probably not going to be less than 100.

Better luck next time I guess.

(31 Jan '15, 17:39) superty3★
showing 5 of 7 show all

I did 1st one. It was a nice DP. First calculate the maximum sum possible for i less than j using Kadane's algorithm (had to modify it a bit). For i greater than j, precalculate the maximum sums possible and traverse the array one. Overall complexity O(n).

Couldn't solve the 2nd problem completely. Solved first subtask using Inclusion Exclusion Principle. Since w = v^n. Therefore |w| = n|v| where |s| denotes the length of string s. So for n>=2, find all factors of |w| and use IEP to find all the possible periodic strings. Now in the end subtract this value from 2^n.

Another answer: Here we will consider '1' as a prime number. Now the total periodic strings will be: 2^p1 + 2^p2 +... - 2x(k-1) where k are the number of prime factors of n (including 1 and excluding n). SO the answer will be 2^n - (2^p1 + 2^p2 + ... -2x(k-1))


answered 31 Jan '15, 13:25

ketanhwr's gravatar image

accept rate: 15%

edited 31 Jan '15, 13:26

As far as the cutoff is concerned ,I think it will be >=125, considering most of the students solving the first question correctly ,and the second question ..... duh .... say most with subtask 1 correct.

Could have been selected , if had may be 1/2 hrs more ... but yeah .. past is past.. :(


answered 31 Jan '15, 14:30

kartikay101's gravatar image

accept rate: 0%


Am I the only one who thought that 1 problem was harder than the second one ?

(31 Jan '15, 14:52) neo1tech9_76★

Nope. I thought the same. Problem 1 is definitely harder IMO.

(31 Jan '15, 14:53) superty3★

Did anyone feel that the sample testdata for Subtask 3 (and maybe 4 also) of Problem 2 was incorrect?

I think one of them at least is easy to prove wrong.

Input: 1260 99999989
Output: 3323612

(2^1260) mod 99999989 is 3007916. So their answer is more than 2^n, not possible. I know its mod, but there's no way that the amount you have to subtract is 99684293, that's way too large.

Really crossing my fingers that those answers were wrong, I attempted only Problem 2!


answered 31 Jan '15, 14:31

vikramnitin9's gravatar image

accept rate: 0%

@sandy999 Yeah, that was more or less the way I did it, except I grouped the factors by prime numbers and took away all the strings of that length, not just non-periodic strings.

(31 Jan '15, 14:41) vikramnitin92★

sorry but I don't think they are wrong at least my program gave correct answers on all the sample test data

(31 Jan '15, 14:52) neo1tech9_76★

Yeah I don't think they're wrong.

(31 Jan '15, 14:53) superty3★

@vikramnitin9 If you take away all strings of that length, you'd end up taking away strings like 00...000 and 11...11 more than once!

(31 Jan '15, 14:58) sandy9992★

Yeah, so subtract 2 after raising every prime factor except 1. Those are the only two repetitions, I think. Am I right?

(31 Jan '15, 15:04) vikramnitin92★

You're right I guess. Which anyway means that you're taking the non periodic strings only (I don't know why you said you're taking away all strings of that length), considering that 1 gives 2 non periodic strings.

(31 Jan '15, 15:20) sandy9992★

Are you saying that the number of periodic strings of length 1260 is less than 10^8?

There are 2^630 strings of length 630, and thus there are at least that many periodic strings of length 1260.

(31 Jan '15, 15:27) superty3★

Just to clarify, as an example, what would be the correct approach if n=12? 12 = 1 x 4 x 3, so my approach would be (2^12)-((2^1) + (2^4 -2) + (2^3 -2))

So for 1260 = 1 x 4 x 9 x 5 x 7, I subtract (2^1)+(2^4 -2), etc

I have a vague feeling that there's something wrong with this, though.

(31 Jan '15, 15:27) vikramnitin92★

It's correct, but the way I reached it is different:

2^12 - ((2^1) + (2^2 - 2^1) + (2^4 - (2^2 - 2^1) - 2^1) + (2^3 - 2^1))

(31 Jan '15, 15:35) superty3★

So are the two ways exactly equivalent? In your method, do you also include (2^6 -2) ? And @sandy999, I did that intentionally, to combine both in one step. Eg: I wanted to take away 1010 and 0101 also. These are patterns for 2, but they can be merged with 4.

(31 Jan '15, 15:42) vikramnitin92★

Yeah, didn't notice that.

(31 Jan '15, 15:45) sandy9992★

I don't think so, I think they happen to be equivalent in this case.

(31 Jan '15, 15:46) superty3★

My method must be wrong, because like you mentioned above, for 1260, I don't take 2^630, only its individual factors.

(31 Jan '15, 15:47) vikramnitin92★

Do you mean prime factors? If so, that's definitely wrong. If not, 630 is a factor of 1260.

(31 Jan '15, 15:49) superty3★

I think I get what you mean. So you would include 2^6 as well in the 2^12 example, right?

(31 Jan '15, 15:51) vikramnitin92★

Oh yeah you are right, I was wrong in that comment. Wow two people upvoted that lol.

The answer is:

2^12 - ((2^1) + (2^2 - 2^1) + (2^4 - (2^2 - 2^1) - 2^1) + (2^3 - 2^1) + (2^6 - (2^3 - 2^1) - (2^2 - 2^1) + 2^1))

(31 Jan '15, 15:54) superty3★
showing 5 of 16 show all

I'm surprised there wasn't a graph question, I was convinced that there was going to be one graph and one DP.


answered 31 Jan '15, 15:12

superty's gravatar image

accept rate: 31%


Same! I wish there wasn't a departure from the pattern, but sigh.

(31 Jan '15, 15:18) idraumr0★

last year was comparitively cakewalk. One question was direct Floyd warshall and the 1st question was difficult, but not impossible...

(31 Jan '15, 15:49) nibnalin6★

The first question last year was, in my opinion, easier than both of the questions this year. It was definitely easier to qualify though, because the cutoff was only one question.

(31 Jan '15, 15:52) superty3★

On the other hand, we've better programmers appearing this year...

(31 Jan '15, 15:55) idraumr0★

I did take that into account, I still think it was harder last year.

(31 Jan '15, 15:56) superty3★

according to me last year's first question was definitely harder then the second question this year and It totally depends on the person whether or not it was harder then the first question this year. last year's recurrence was easy where as coding part was difficult whereas it was opposite in this year's first problem

(31 Jan '15, 16:08) neo1tech9_76★

@superty, today being the day I spend drowned in anxiety as to whether I'd qualify, I can only hope that's true and I qualify. :-)

(31 Jan '15, 16:09) idraumr0★

I disagree that last year's recurrence was easy, I would argue that the recurrence was difficult but the implementation was easy. This year both were hard... oh. Yeah maybe this year was harder. One would expect it to get harder as time goes on anyways.

(31 Jan '15, 16:16) superty3★

No Highwaybypass was a tad easier than both the problem s

(31 Jan '15, 18:10) arpanb83★

Okay, maybe I undercompensated

(31 Jan '15, 18:15) superty3★

Heh; so last year, poor skills + poor questions = 100 cut-off. This year, good skills + good questions = (!).

(31 Jan '15, 18:48) idraumr0★

The average skill level increases only very slightly every year. Most of the people who did well last year aren't participating this year, and we won't be participating next year. The people who did badly this year will do better next year. The number of people doing well doesn't change much from year to year.

(31 Jan '15, 18:53) superty3★
showing 5 of 12 show all

I'm just wondering, do they actually check the code, or just judge with testcases? I came up with an algorithm for the second question, similar to @sandy999's but I was anxious about the O(N^2) complexity (and it was terribly slow, was taking more than 3 seconds for an n=150000 input, which was probably some flaw in my implementation, considering that it worked for some others), so I precomputed the number of non-periodic strings for all prime numbers till 387 and put them in a lookup table. I'm hoping precomputation doesn't count against my score, since the code was working fine for all sample cases.


answered 31 Jan '15, 16:53

hackpert's gravatar image

accept rate: 0%

edited 31 Jan '15, 16:54


I don't think that's officially wrong (no rule says you can't pre-compute). You'll want to mail the organisers to be double-sure. It's still somewhat a "bad" way to solve the problem, given that you obviously did something wrong.

(31 Jan '15, 16:56) idraumr0★

Given that we've more or less a week to the results, does everyone want to participate in a quick collaboration to figure out the cut-off?

Just simply note in a comment the number of people you think will score >= 100, including yourself. Please don't give answers with duplicates.

I know of myself scoring (probably) 130. (A few more, but those are guessimates.)



answered 31 Jan '15, 17:16

idraumr's gravatar image

accept rate: 45%


I'm the only person I know of, so that makes 2 + the few people you know of getting ~160.

(31 Jan '15, 17:18) superty3★

Shagun , Parth of dps rkp and me would also be most probably scoring above or equal to 100.


answered 04 Feb '15, 15:02

rahulchhabra07's gravatar image

accept rate: 0%

edited 04 Feb '15, 15:06

That brings us to 24. Shagun has been to IOITC last year.


answered 04 Feb '15, 15:06

rahulchhabra07's gravatar image

accept rate: 0%

edited 04 Feb '15, 18:38

when can we expect the results?


answered 05 Feb '15, 10:45

jayaganesh's gravatar image

accept rate: 0%

I was told within a week (from 31st Jan).

(05 Feb '15, 11:51) idraumr0★

The result must be coming in a day or so.

(06 Feb '15, 11:18) ketanhwr6★

It seems it's going to take a few more days.

(06 Feb '15, 12:02) superty3★

Why? Did you ask them? They aren't replying to my mail. Last year, the result came in exactly 2 weeks.

(06 Feb '15, 13:47) ketanhwr6★

@idraumr asked them.

(06 Feb '15, 13:51) superty3★

Sigh; someone should figure out what takes them so long.

(06 Feb '15, 14:01) idraumr0★

Waiting sucks. Ask them to tell your score if yours have been checked.

(06 Feb '15, 14:13) ketanhwr6★

I don't think they would be willing to do that. Someone who's gone to camp in previous years might have a better idea of what's going on.

(06 Feb '15, 14:19) superty3★

Yeah. Anyone?

(06 Feb '15, 14:23) ketanhwr6★

I asked them my score, regardless. Let's see how they respond. Did you ask for the same, @ketanhwr?

Although, yes, can someone who has more knowledge of how the process goes educate me as to why this takes an insane amount of time?

(06 Feb '15, 14:51) idraumr0★

@idraumr no I didn't, but am planning to if the results don't come out by sunday.

(06 Feb '15, 16:06) ketanhwr6★

Guys it's sunday today, anyone got any updates ?

(08 Feb '15, 12:33) nibnalin6★
showing 5 of 12 show all

Will the qualifiers be informed by e-mail( as they did in case of ZCO )?


answered 06 Feb '15, 17:48

anupam_datta's gravatar image

accept rate: 7%

I presume so; how much are you expecting?

(06 Feb '15, 17:54) idraumr0★

Do they tell us the scores?


answered 10 Feb '15, 14:57

data2000's gravatar image

accept rate: 0%

Yes they email the feedback given by the grader along with the scores.

(10 Feb '15, 20:06) animesh_f ♦6★

cut off looks to be <100 :)


answered 11 Feb '15, 11:24

jayaganesh's gravatar image

accept rate: 0%

How do you know?

(11 Feb '15, 11:49) idraumr0★

And how does she know?

(11 Feb '15, 12:58) idraumr0★

I imagine she has better contact with the organizers than anyone else.

(11 Feb '15, 13:18) superty3★

Is this another troll? Because I am not liking these jokes.

(11 Feb '15, 13:34) ketanhwr6★

Lol stop the cut off discussion . We will get to know that anyway in a few days.

(11 Feb '15, 14:06) animesh_f ♦6★

when can we actually expect the results? 2dy or tomorrow?

(13 Feb '15, 07:26) jayaganesh3★

Today most probably. They said "We have done a preliminary scan and hope to make an announcement in a day or so."

(13 Feb '15, 08:02) ketanhwr6★

They said that yesterday.

(13 Feb '15, 08:02) ketanhwr6★

And any word on the cutoff?

(13 Feb '15, 10:36) idraumr0★

Nopes. @idraumr mail them asking whether the result will come today or not. I might have irritated them a lot by now :P Also, what did they reply when you asked them about your score?

(13 Feb '15, 11:08) ketanhwr6★

if any body has asked them their score and got a reply, pls tell me, i also wanna ask. Also I wont mind asking for cutoff in my mail. Also any news on when the official results would be published?

(13 Feb '15, 12:06) jayaganesh3★

@ketanhwr, I essentially got told I've irritated them enough. No word on my score, so I'm not sending anymore mails.

(13 Feb '15, 12:29) idraumr0★

It is taking forever and scrambling my brain ;_;

(13 Feb '15, 12:50) ketanhwr6★

@idraumr , who told, u irritated them? also when r they planning to give the results? by today?

(13 Feb '15, 14:33) jayaganesh3★
showing 5 of 14 show all

Did u send a mail??


answered 13 Feb '15, 17:52

rahulchhabra07's gravatar image

accept rate: 0%

btw @idraumr, when did u get that mail? so most probably results by 2dy?


answered 13 Feb '15, 19:36

jayaganesh's gravatar image

accept rate: 0%

1606 hrs is when I got it.

(13 Feb '15, 19:47) idraumr0★

if we ask, will he tell our score by mail?

(13 Feb '15, 20:32) jayaganesh3★

Don't think so. He already denied idraumr and I think someone else as well.

(13 Feb '15, 20:34) superty3★

Did anyone else get a couple of random TLEs in Special Sums because of slow I/O? I rather stupidly forgot to set stdio sync to false :/. By any chance, are they considering that for reevaluation?


answered 14 Feb '15, 14:57

hackpert's gravatar image

accept rate: 0%

They are. Slow I/O shouldn't lead to any less marks, as long as your algorithm is correct.

(14 Feb '15, 14:58) idraumr0★

They checked my algorithm for P2 .They found it "OK" so they increased my score to 110. I had the same problem. :)

(14 Feb '15, 15:01) animesh_f ♦6★

@idraumr , @animesh_f Thanks! I currently have a 10 :( because I messed up problem 2 completely by mixing up two variable names. Wrote them an email, hoping for a 100 at the very least.

(14 Feb '15, 15:03) hackpert1★

Please tell us if they give you any marks for your second problem.

(14 Feb '15, 15:06) superty3★

Sure. But I doubt they would, since my code for the second problem is pretty much unreadable with the pre-computations and stuff.

(14 Feb '15, 15:12) hackpert1★

Well, just in case they do.

(14 Feb '15, 15:13) superty3★

I didn't use any fast I/O but still got 100 in special sums.

(14 Feb '15, 18:16) ketanhwr6★

Yeah, once I fixed my mistake I got 100. cout/cin. No fast IO.

(14 Feb '15, 18:24) superty3★
showing 5 of 8 show all

In periodic string, in JAVA, it is showing TLE(execution timed out(wall clock time limit exceeded)) and the execution time is 5.6s and 5.8s for the first two input data of subtask 4.But when I ran it on my computer each of those two data sets are taking approximately 1s. Will I be given marks for subtask 4?

PS: I used System.currentTimeMillis() for timing.I think there is something wrong with Special Sum also so I am waiting for test data to be downloaded(currently showing 5 hours left) before I finally write for reevaluation.


answered 14 Feb '15, 17:37

divyansh13's gravatar image

accept rate: 0%

By the looks of the previous answer, this is a common issue.

Do tell us if you get more marks on reevaluation.

(14 Feb '15, 18:09) idraumr0★

@idraumr For periodic string, my solution was "deemed to be" 100/100. I am currently waiting for my response of special sum.

(14 Feb '15, 18:54) divyansh134★

I had the exact same issue with the same problem (I also code in JAVA) and I wrote to them. They checked my algorithm and they made it 100 after seeing it. Slow I/O error is unfair for JAVA :P

(14 Feb '15, 19:11) animesh_f ♦6★

I had the same 'wall clock time limit exceeded' error as animesh_f (I used Java too). I've written to them for re-evaluation. I'm yet to get my result. :-(


answered 16 Feb '15, 14:51

data2000's gravatar image

accept rate: 0%

edited 16 Feb '15, 15:00

I'm interested to know how you guys approached the SPECIALSUMS problem...

Can anybody share how you did it? (which algorithm?, what logic?, etc.)


answered 16 Feb '15, 16:13

data2000's gravatar image

accept rate: 0%

Results are out! All the best to all those who qualified. May we have a strong competition at IOITC to get into IOI :)


answered 18 Feb '15, 07:39

jayaganesh's gravatar image

accept rate: 0%

Congrats man! See you at the camp :)

(18 Feb '15, 08:07) ketanhwr6★

Congrats everyone! :)

(18 Feb '15, 10:36) superty3★

Btw, will anyone appear for JEE? Can they keep the camp after JEE? If anyone is appearing for it, please mail them. Considering the fact that IOI is pretty late this year, so it won't be a problem.

(18 Feb '15, 10:53) ketanhwr6★

@ketanhwr, I suppose that will not be possible, as the team has to finish all formalities for travel atleast a month ahead. So, they will try to host the camp (and subsequently select the team) as soon as possible (ie. the starting of May). And the camp has been before JEE (or JEE Advanced) for the past two years.

You can ask them to have the camp later, but there is very little chance for that.

(18 Feb '15, 13:44) codelegend6★

I did an email and the response was that it is impossible to conduct the camp before JEE.Will the class 12th guys still attend the camp ? I'm also a class 12th student and not quite sure about attending it.

(19 Feb '15, 21:06) divyansh134★

What was the cutoff?


answered 18 Feb '15, 10:18

anupam_datta's gravatar image

accept rate: 7%

100 I think.

(18 Feb '15, 10:52) ketanhwr6★

What happens at the camp?

This was the first year I applied for INOI and didn't qualify for IOITC.

Am curious to know..


answered 20 Feb '15, 13:29

sakshi_garg's gravatar image

accept rate: 0%

They teach us with important algos. And then there are tests for 3 days where they select the Indian team for IOI.

(20 Feb '15, 14:05) ketanhwr6★
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here



Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text]( "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported
  • mathemetical formulas in Latex between $ symbol

Question tags:


question asked: 31 Jan '15, 12:42

question was seen: 13,367 times

last updated: 20 Feb '15, 14:05