Hello everyone,

We are back with the winner-list of this month’s algorithm challenge. For the first time we have ties at 2 places in the rank-list for India π

**Winners :**

**Top 5 (India):**

1st – Harpreet Singh (5.557)

2nd – Satyam Shekhar (5.116)

3rd – Gaurav Agarwal (5.000)

3th – Anshuman Singh (5.000)

5th – Anil (4.116)

5th – Pradeep B (4.116)

**Top 5 (US):**

1st – Wenyu Cao (6.919)

2nd – pmnox (6.902)

3rd – Josh Metzler (6.284)

4th – Balakrishnan Varadarajan (6.127)

5th – Chris Narrikkattu (5.000)

** Stephen Merriman requires a special mention for having secured the first place over-all π **

**Statistics:**

All problems are now available in the practice area and the solutions are public. The test cases for the problems are now available. As the test cases for the problem Mosaic are really huge, we are uploading two sets of test cases. The test cases here are for problems other than Mosaic, while the entire set of test cases including those for the problem Mosaic are available here.

A few interesting statistics for the contest follow :

Length of Contest |
Unique Visitors |
Unique Participants |
Total Number of Submissions |
Percentage of user who have solved at least one problem |

10 days | 10,482 | 219 | 2950 | 22% |

Country | Total Participants | Average Score per User |

IN | 171 | 0.310 |

US | 22 | 1.879 |

Rest of World | 26 | 2.206 |

**Feedback:**

We removed the submission limit on the challenge problem this time as the problems were considerably tough compared to the ones in the previous contests. We agree that some of the problems had a very strict time limit and this issue will be addressed in the future contests.

Also, if you have any kind of feedback related to the contest or otherwise, feel free to comment.

Regards,

Aniruddha.

Thanks to CodeChef and the problem author(s) for the hard problems. I found them to be much more difficult on average in this contest compared to the previous contests, and I enjoyed having to struggle to come up with solutions.

That said, I worry that the difficulty might discourage some participants. Would CodeChef consider copying an aspect of the Google Code Jams and having both small and large inputs for all problems? The small inputs would have the constraints set to make the problem much easier and be worth 0.2 points (or something similar).

Thanks to CodeChef and the problem author(s) for the hard problems. I found them to be much more difficult on average in this contest compared to the previous contests, and I enjoyed having to struggle to come up with solutions.

That said, I worry that the difficulty might discourage some participants. Would CodeChef consider copying an aspect of the Google Code Jams and having both small and large inputs for all problems? The small inputs would have the constraints set to make the problem much easier and be worth 0.2 points (or something similar).

For submitted solution I am getting an error that solution is not public.

For submitted solution I am getting an error that solution is not public.

@Josh

I wonder how many people working at Codechef can solve these problems.

@Josh

I wonder how many people working at Codechef can solve these problems.

Hats off to the codechef team for setting such hard problems! I did not have enough time this month to compete.

I guess I am gradually get a grip of programming after 3 months of solving these wonderful problems ! π

Hats off to the codechef team for setting such hard problems! I did not have enough time this month to compete.

I guess I am gradually get a grip of programming after 3 months of solving these wonderful problems ! π

The solutions are still not accessible π

It shows a msg “Solutions are not public”.

The solutions are still not accessible π

It shows a msg “Solutions are not public”.

Yes, it was a very nice set of problems. It was just a shame that the challenge problem was rather wasted by the fact that reading the input took up virtually all of the time limit (or all of it if you used scanf). Now that we’ve had this complaint several contests in a row I’m sure it will be better next time π

One other point – as Tomasz Czajka mentioned about the Divide and conquer problem from the August contest (http://www.codechef.com/AUG09/problems/F6/), a judge needs to perform accurate computations, not approximate ones. That caused major problems for that problem, and it wasn’t very fun spending ages on Mosaic before realising that it was using a precision of 1e-6. There was absolutely no need to use approximate computations here, but if in the future this is to be done it needs to be mentioned clearly in the problem statement.

Yes, it was a very nice set of problems. It was just a shame that the challenge problem was rather wasted by the fact that reading the input took up virtually all of the time limit (or all of it if you used scanf). Now that we’ve had this complaint several contests in a row I’m sure it will be better next time π

One other point – as Tomasz Czajka mentioned about the Divide and conquer problem from the August contest (http://www.codechef.com/AUG09/problems/F6/), a judge needs to perform accurate computations, not approximate ones. That caused major problems for that problem, and it wasn’t very fun spending ages on Mosaic before realising that it was using a precision of 1e-6. There was absolutely no need to use approximate computations here, but if in the future this is to be done it needs to be mentioned clearly in the problem statement.

The solutions are now public. Sorry for the delay.

@Stephen We will try to have clearer problem statements next time onwards as far as precision is concerned. π

@Josh Yes, this time the problems were considerably tougher and I agree that the high difficulty level discourages some participants . We will try to have simpler problems from next time π

Used to think that I can solve tough problems easily! But no longer! Repeated TLE’s for the very first problem (Bowling Game) put me off.

Please make the test cases available.

Used to think that I can solve tough problems easily! But no longer! Repeated TLE’s for the very first problem (Bowling Game) put me off.

Please make the test cases available.

The test cases have been uploaded.

Some interface changes I’d love to have:

1) Immediate feedback on the memory consumed and time taken for each submission that I make. Preferably when the “correct answer/wrong answer” is displayed.

2) When I view “My profile”, I’d like to see “View solution” links to each correct latest submission I’d made. Something like a list of Problem name, “View my solution”.

Some interface changes I’d love to have:

1) Immediate feedback on the memory consumed and time taken for each submission that I make. Preferably when the “correct answer/wrong answer” is displayed.

2) When I view “My profile”, I’d like to see “View solution” links to each correct latest submission I’d made. Something like a list of Problem name, “View my solution”.

@JoshI wonder how many people working at Codechef can solve these problems.

@Josh

I wonder how many people working at Codechef can solve these problems.