@Aniruddha: I understand the need for a tie breaker problem – I was just hoping that maybe this shorter competition would be the one to break the streak of competitions that are decided by the tie breaker.

@Aniruddha: I understand the need for a tie breaker problem – I was just hoping that maybe this shorter competition would be the one to break the streak of competitions that are decided by the tie breaker.

The major reason for having only one tie-breaker was that in the past few contests, it has always been the tie breaker that decides the winners

The major reason for having only one tie-breaker was that in the past few contests, it has always been the tie breaker that decides the winners

I actually liked the fact that the contest had only problem — gives a different flavor than the longer contest.

Other than that, I agree with other about the I/O issues and would like to see longer time limits, so that I/O is not a big factor.

I actually liked the fact that the contest had only problem — gives a different flavor than the longer contest.

Other than that, I agree with other about the I/O issues and would like to see longer time limits, so that I/O is not a big factor.

It’s feedback time now :

1) Although we cannot do anything about it, it is not appreciating to find out that I/O plays a major role in acing the problems. It took quite some time for me to see that my cin, cout was causing timeouts even without doing any processing. I started using printf, scanf’s, but that seems not the best way apparently. So probably I/O could reduce to a level where it’s not significant.

2) For any problem, under the “successful submissions for the problem”, I’d like to see the “View solution” next to each entry. It’s time consuming to dig out the correct solutions from ‘recent activity for this problem’.

3) Though there have been explanations, I still probably don’t understand the time taken info. Time taken is supposed to be the time taken to execute for all test cases. But then I wonder if it’s true that some have solved in 1.69 secs while others have taken as much as 21 seconds for Aug mini.

4) I feel the way the solutions are ranked to a particular problem is misleading. Sudoku solver is ranked based on time taken to execute whereas the total score (aka penalty) is what counts in the final ranking. Shouldn’t the solutions be ranked according to the defined criteria?

5) Josh Metzler has hinted at having more problems for the mini challenge. I’d suggest keeping it just one or two for a duration of 2 days (it’s a mini challenge after all!) so that if I don’t have time to participate in the monthly ones, I will at least get to participate in the mini’s.

Concluding note :
I immensely appreciate this initiative by DirectI and wish Codechef becomes THE place for “programmers” to hang around.

It’s feedback time now :

1) Although we cannot do anything about it, it is not appreciating to find out that I/O plays a major role in acing the problems. It took quite some time for me to see that my cin, cout was causing timeouts even without doing any processing. I started using printf, scanf’s, but that seems not the best way apparently. So probably I/O could reduce to a level where it’s not significant.

2) For any problem, under the “successful submissions for the problem”, I’d like to see the “View solution” next to each entry. It’s time consuming to dig out the correct solutions from ‘recent activity for this problem’.

3) Though there have been explanations, I still probably don’t understand the time taken info. Time taken is supposed to be the time taken to execute for all test cases. But then I wonder if it’s true that some have solved in 1.69 secs while others have taken as much as 21 seconds for Aug mini.

4) I feel the way the solutions are ranked to a particular problem is misleading. Sudoku solver is ranked based on time taken to execute whereas the total score (aka penalty) is what counts in the final ranking. Shouldn’t the solutions be ranked according to the defined criteria?

5) Josh Metzler has hinted at having more problems for the mini challenge. I’d suggest keeping it just one or two for a duration of 2 days (it’s a mini challenge after all!) so that if I don’t have time to participate in the monthly ones, I will at least get to participate in the mini’s.

Concluding note :
I immensely appreciate this initiative by DirectI and wish Codechef becomes THE place for “programmers” to hang around.

That said, I agree that a longer time limit for tie-breaker problems would be useful. I was able to get the I/O time down quite a bit by buffering input and output on my own, but using scanf() and printf() would probably have left very little time for actually doing any processing. Having less I/O for tie-breaker problems is a possibility, but it shouldn’t be too small or people will do datamining on it.

That said, I agree that a longer time limit for tie-breaker problems would be useful. I was able to get the I/O time down quite a bit by buffering input and output on my own, but using scanf() and printf() would probably have left very little time for actually doing any processing. Having less I/O for tie-breaker problems is a possibility, but it shouldn’t be too small or people will do datamining on it.