Codechef has helped me improve a lot. I love to be participating every month.I would be happy with a 10 day long contest as am a beginner. For me , I need time to think about each problem , analyse , code and get accepted. Further we are working in some software firm. A 10 day long contest will give me good time to spend than 7 day contest.

Codechef has helped me improve a lot. I love to be participating every month.I would be happy with a 10 day long contest as am a beginner. For me , I need time to think about each problem , analyse , code and get accepted. Further we are working in some software firm. A 10 day long contest will give me good time to spend than 7 day contest.

Codechef has helped me improve a lot. I love to be participating every month.I would be happy with a 10 day long contest as am a beginner. For me , I need time to think about each problem , analyse , code and get accepted. Further we are working in some software firm. A 10 day long contest will give me good time to spend than 7 day contest.

Codechef has helped me improve a lot. I love to be participating every month.
I would be happy with a 10 day long contest as am a beginner. For me , I need time to think about each problem , analyse , code and get accepted. Further we are working in some software firm. A 10 day long contest will give me good time to spend than 7 day contest.

Thanks for listening and considering our feedback.I also agree with Praveen, long version of contest helps me to choose my time for the contest conveniently and with each month I learn something new.Initially ,tackling difficult problem was the one which i afraid most , but with CodeChef now atleast my viewpoint has changed from 'difficult problem' to 'every prbolem is solvable, just need to give some time to it' .Shorter version of contest has time factor with it and I can't learn any new concept during the contest.Though I can learn any new concept after the contest but it makes me feel very good ,once I use that concept to solve a problem in running contest. and also there are lot of shorter contest already running.With next contest CodeChef will complete a Year successfully. Congrats and best wishes for future contest

Thanks for listening and considering our feedback.

I also agree with Praveen, long version of contest helps me to choose my time for the contest conveniently and with each month I learn something new.Initially ,tackling difficult problem was the one which i afraid most , but with CodeChef now atleast my viewpoint has changed from 'difficult problem' to 'every prbolem is solvable, just need to give some time to it' .

Shorter version of contest has time factor with it and I can't learn any new concept during the contest.Though I can learn any new concept after the contest but it makes me feel very good ,once I use that concept to solve a problem in running contest. and also there are lot of shorter contest already running.

With next contest CodeChef will complete a Year successfully. Congrats and best wishes for future contest

well YOU GUYS are doing a fantastic jobWELL DONE ……….but here'sone more suggestion:-)instead of providing JUST SOLUTION for 10 days contest PROBLEM WHY NOT ???you prepare something like CONTEST EDITORIAL as TOPCODER 's MATCH EDITORIALso that PEOPLE can UNDERSTAND which PROBLEM requires which ALGORITHMS

well YOU GUYS are doing a fantastic jobWELL DONE ……….but here'sone more suggestion:-)instead of providing JUST SOLUTION for 10 days contest PROBLEM WHY NOT ???you prepare something like CONTEST EDITORIAL as TOPCODER 's MATCH EDITORIALso that PEOPLE can UNDERSTAND which PROBLEM requires which ALGORITHMS

well YOU GUYS are doing a fantastic jobWELL DONE ……….but here'sone more suggestion:-)instead of providing JUST SOLUTION for 10 days contest PROBLEM WHY NOT ???you prepare something like CONTEST EDITORIAL as TOPCODER 's MATCH EDITORIALso that PEOPLE can UNDERSTAND which PROBLEM requires which ALGORITHMS

well YOU GUYS are doing a fantastic job
WELL DONE ……….

but here's
one more suggestion

instead of providing JUST SOLUTION for 10 days contest PROBLEM

WHY NOT ???

you prepare something like CONTEST EDITORIAL as TOPCODER 's MATCH EDITORIAL
so that PEOPLE can UNDERSTAND which PROBLEM requires which ALGORITHMS

Regarding having all test cases in one file – I'm not sure that is the best idea, unless you set the time limit to be (number of test cases)*(time to process the worst case). But if half the test cases were small, that would give people a lot more time on the large ones than you want.The standard way of doing this would be to make sure you don't combine small/large test cases in the same file. So you could have 20 small test cases in one file, 20 large test cases in another file, and you set the time limit to be 20*(worst-case time for a test). Obviously that gives people an enormous amount of time to process the small test cases; but those cases will be testing validity, not speed.That way if users know there can be 20 test cases, and a time limit of 2 seconds, then they should be aiming for a worst-case 0.1 seconds per test, as opposed to wondering if a solution with a worst-case of 5 seconds per test will pass because 18 of the test cases are tiny.

Regarding having all test cases in one file – I'm not sure that is the best idea, unless you set the time limit to be (number of test cases)*(time to process the worst case). But if half the test cases were small, that would give people a lot more time on the large ones than you want.The standard way of doing this would be to make sure you don't combine small/large test cases in the same file. So you could have 20 small test cases in one file, 20 large test cases in another file, and you set the time limit to be 20*(worst-case time for a test). Obviously that gives people an enormous amount of time to process the small test cases; but those cases will be testing validity, not speed.That way if users know there can be 20 test cases, and a time limit of 2 seconds, then they should be aiming for a worst-case 0.1 seconds per test, as opposed to wondering if a solution with a worst-case of 5 seconds per test will pass because 18 of the test cases are tiny.

Regarding having all test cases in one file – I'm not sure that is the best idea, unless you set the time limit to be (number of test cases)*(time to process the worst case). But if half the test cases were small, that would give people a lot more time on the large ones than you want.

The standard way of doing this would be to make sure you don't combine small/large test cases in the same file. So you could have 20 small test cases in one file, 20 large test cases in another file, and you set the time limit to be 20*(worst-case time for a test). Obviously that gives people an enormous amount of time to process the small test cases; but those cases will be testing validity, not speed.

That way if users know there can be 20 test cases, and a time limit of 2 seconds, then they should be aiming for a worst-case 0.1 seconds per test, as opposed to wondering if a solution with a worst-case of 5 seconds per test will pass because 18 of the test cases are tiny.

You are right. It is very hard to read through and understand others' code even if the code is commented. I hope CodeChef will also explain the algorithm used in each problem.

You are right. It is very hard to read through and understand others' code even if the code is commented. I hope CodeChef will also explain the algorithm used in each problem.

+1

+1

+1

+1

+1

+1

I think it would be great if the Codechef admins like Aniruddha could write the editorials. I'm quite sure they would do a good job. I liked his video Tutorial for that BFS problem 9-puzzle

I think it would be great if the Codechef admins like Aniruddha could write the editorials. I'm quite sure they would do a good job. I liked his video Tutorial for that BFS problem 9-puzzle

Kill the monthly format. We as students dont have all 10 days to compete. I prefer topcoder to codechef, and you can see they have way more contestants and a stronger community than Codechef. This is despite the fact that Codechef gives out so much in prize money ( I read they gave away 18 lacs ) and so much expenditure behind it. I agree I'm not somebody who would win, but since i want to learn new things, I dont have the patience to wait 10 days and try something which requires entirely new concepts.For e.g. I didnt know anything about dynamic programming and graph theory when I did my first Codechef contest. But Codechef never helped me learn that ( during the 7 days or even after that ), I learnt much more from the Topcoder editorials.I hope you learn from your mistakes.

Kill the monthly format. We as students dont have all 10 days to compete. I prefer topcoder to codechef, and you can see they have way more contestants and a stronger community than Codechef. This is despite the fact that Codechef gives out so much in prize money ( I read they gave away 18 lacs ) and so much expenditure behind it.
I agree I'm not somebody who would win, but since i want to learn new things, I dont have the patience to wait 10 days and try something which requires entirely new concepts.
For e.g. I didnt know anything about dynamic programming and graph theory when I did my first Codechef contest. But Codechef never helped me learn that ( during the 7 days or even after that ), I learnt much more from the Topcoder editorials.

I hope you learn from your mistakes.

Hey Pankaj,We understand the need for shorter contests. But, we are still passionate about providing the community the longer version as we strongly believe there are many programmers who have benefitted from it and will benefit from it going further. To be frank I belong to one of those category and honestly I missed one such platform during my college days.I agree that as students and professionals have various other activities to take care of, they cannot sit through the 10 days to keep a watch as to what else they do need to do to win. I am slightly more biased towards learning over winning. Pardon me for that. And also probably they can still devote whatever time they have to solve as many problems as they want/can. However, we do understand the pressing need of the shorter version and are working towards it to see the feasibility towards conducting the same. Once again, really appreciate your feedback. We value it a lot .

Hey Pankaj,We understand the need for shorter contests. But, we are still passionate about providing the community the longer version as we strongly believe there are many programmers who have benefitted from it and will benefit from it going further. To be frank I belong to one of those category and honestly I missed one such platform during my college days.I agree that as students and professionals have various other activities to take care of, they cannot sit through the 10 days to keep a watch as to what else they do need to do to win. I am slightly more biased towards learning over winning. Pardon me for that. And also probably they can still devote whatever time they have to solve as many problems as they want/can. However, we do understand the pressing need of the shorter version and are working towards it to see the feasibility towards conducting the same. Once again, really appreciate your feedback. We value it a lot .

Hey Pankaj,We understand the need for shorter contests. But, we are still passionate about providing the community the longer version as we strongly believe there are many programmers who have benefitted from it and will benefit from it going further. To be frank I belong to one of those category and honestly I missed one such platform during my college days.I agree that as students and professionals have various other activities to take care of, they cannot sit through the 10 days to keep a watch as to what else they do need to do to win. I am slightly more biased towards learning over winning. Pardon me for that. And also probably they can still devote whatever time they have to solve as many problems as they want/can. However, we do understand the pressing need of the shorter version and are working towards it to see the feasibility towards conducting the same. Once again, really appreciate your feedback. We value it a lot .

Hey Pankaj,

We understand the need for shorter contests. But, we are still passionate about providing the community the longer version as we strongly believe there are many programmers who have benefitted from it and will benefit from it going further. To be frank I belong to one of those category and honestly I missed one such platform during my college days.

I agree that as students and professionals have various other activities to take care of, they cannot sit through the 10 days to keep a watch as to what else they do need to do to win. I am slightly more biased towards learning over winning. Pardon me for that. And also probably they can still devote whatever time they have to solve as many problems as they want/can.

However, we do understand the pressing need of the shorter version and are working towards it to see the feasibility towards conducting the same.

Once again, really appreciate your feedback. We value it a lot .

I'm a well-wisher of Codechef. I'm very happy that such initiative has been taken to improve the programming scene in India and I greatly appreciate your efforts. However I have some suggestions backed by some valid facts that I have noticed because of my experience with many programming contests. I'm really surprised why Codechef hasn't become as popular as many other competitions / onlines judges that are present.I have just kept some facts.http://pastebin.com/pcXDyBHRAs you can clearly see for majority of the contestants, shorter contests tend to be more popular. Shorter contests tend to take up less time of the contestants and are thus more attractive for people with busy schedules. Also the current prize structure allows for very few winners, that may also be a reason for low-participation. The prize structure must be changed and allow upto 15-20 winners per month ( even if it would mean reducing the largest prize ). Also I think that the reason that the current marathon style format is used is because you want more people ( from different time-zones ) to be able to compete. However I think it is not working that well. I suggest this 1. Keep shorter contests ( 3-5 hours ) and keep 2-3 of them per month with fewer prizes.Alternatively you may consider awarding assured Codechef merchandise to top 15-20 contestants.The current format is highly competitive and allows very few people to win and that too after involving a lot of effort. This clearly seems to be a deterrent. Many previous winners have only appeared once on Codechef and since then have not competed here.I also feel that the current format is often annoying. The problem-setters usually do not provide feedback and sometimes it takes days to rectify an error in the problem. I think this is also quite annoying and may drive away many users.I hope you are not offended by this constructive criticism. I only wish to convey my thoughts and suggestions and would be happy if you could implement these.

I'm a well-wisher of Codechef. I'm very happy that such initiative has been taken to improve the programming scene in India and I greatly appreciate your efforts. However I have some suggestions backed by some valid facts that I have noticed because of my experience with many programming contests. I'm really surprised why Codechef hasn't become as popular as many other competitions / onlines judges that are present.

I have just kept some facts.

http://pastebin.com/pcXDyBHR

As you can clearly see for majority of the contestants, shorter contests tend to be more popular. Shorter contests tend to take up less time of the contestants and are thus more attractive for people with busy schedules. Also the current prize structure allows for very few winners, that may also be a reason for low-participation. The prize structure must be changed and allow upto 15-20 winners per month ( even if it would mean reducing the largest prize ). Also I think that the reason that the current marathon style format is used is because you want more people ( from different time-zones ) to be able to compete. However I think it is not working that well. I suggest this

1. Keep shorter contests ( 3-5 hours ) and keep 2-3 of them per month with fewer prizes.

Alternatively you may consider awarding assured Codechef merchandise to top 15-20 contestants.

The current format is highly competitive and allows very few people to win and that too after involving a lot of effort. This clearly seems to be a deterrent. Many previous winners have only appeared once on Codechef and since then have not competed here.

I also feel that the current format is often annoying. The problem-setters usually do not provide feedback and sometimes it takes days to rectify an error in the problem. I think this is also quite annoying and may drive away many users.

I hope you are not offended by this constructive criticism. I only wish to convey my thoughts and suggestions and would be happy if you could implement these.

“As you can clearly see for majority of the contestants, shorter contests tend to be more popular.”You can hardly draw such conclusions from such a narrow data set. There are so many differences between TopCoder and CodeChef, and you've only attributed one of them. I could just as easily argue that people prefer contests with no prize, or that people prefer contests on websites that start with the letter T. TopCoder has been doing SRMs for almost 9 years, it really shouldn't come as a surprise that it's more popular.Here's a fact maybe you didn't know: TopCoder pays its SRM problem writers and testers more each month than CodeChef pays each month in prizes. So the price we pay for having prizes is less polished problems.You yourself mention that “it takes days to rectify an error in the problem”, yet suggest contests only last 3-5 hours. I think you've outlined the precise reason why contests need to stay as long as they are. How could we have 3-5 hour contests if it takes days to rectify errors? TopCoder can have short contests because it is extremely rare for there to be an error.

“As you can clearly see for majority of the contestants, shorter contests tend to be more popular.”You can hardly draw such conclusions from such a narrow data set. There are so many differences between TopCoder and CodeChef, and you've only attributed one of them. I could just as easily argue that people prefer contests with no prize, or that people prefer contests on websites that start with the letter T. TopCoder has been doing SRMs for almost 9 years, it really shouldn't come as a surprise that it's more popular.Here's a fact maybe you didn't know: TopCoder pays its SRM problem writers and testers more each month than CodeChef pays each month in prizes. So the price we pay for having prizes is less polished problems.You yourself mention that “it takes days to rectify an error in the problem”, yet suggest contests only last 3-5 hours. I think you've outlined the precise reason why contests need to stay as long as they are. How could we have 3-5 hour contests if it takes days to rectify errors? TopCoder can have short contests because it is extremely rare for there to be an error.

“As you can clearly see for majority of the contestants, shorter contests tend to be more popular.”You can hardly draw such conclusions from such a narrow data set. There are so many differences between TopCoder and CodeChef, and you've only attributed one of them. I could just as easily argue that people prefer contests with no prize, or that people prefer contests on websites that start with the letter T. TopCoder has been doing SRMs for almost 9 years, it really shouldn't come as a surprise that it's more popular.Here's a fact maybe you didn't know: TopCoder pays its SRM problem writers and testers more each month than CodeChef pays each month in prizes. So the price we pay for having prizes is less polished problems.You yourself mention that “it takes days to rectify an error in the problem”, yet suggest contests only last 3-5 hours. I think you've outlined the precise reason why contests need to stay as long as they are. How could we have 3-5 hour contests if it takes days to rectify errors? TopCoder can have short contests because it is extremely rare for there to be an error.

“As you can clearly see for majority of the contestants, shorter contests tend to be more popular.”

You can hardly draw such conclusions from such a narrow data set. There are so many differences between TopCoder and CodeChef, and you've only attributed one of them. I could just as easily argue that people prefer contests with no prize, or that people prefer contests on websites that start with the letter T. TopCoder has been doing SRMs for almost 9 years, it really shouldn't come as a surprise that it's more popular.

Here's a fact maybe you didn't know: TopCoder pays its SRM problem writers and testers more each month than CodeChef pays each month in prizes. So the price we pay for having prizes is less polished problems.

You yourself mention that “it takes days to rectify an error in the problem”, yet suggest contests only last 3-5 hours. I think you've outlined the precise reason why contests need to stay as long as they are. How could we have 3-5 hour contests if it takes days to rectify errors? TopCoder can have short contests because it is extremely rare for there to be an error.

And the biggest point; short contests mean they won't be global contests. If you run a contest lasting only a few hours, heaps of countries in the world wouldn't be able to compete because they'd be asleep. If you want to run a global competition, it must be a longer contest.

And the biggest point; short contests mean they won't be global contests. If you run a contest lasting only a few hours, heaps of countries in the world wouldn't be able to compete because they'd be asleep. If you want to run a global competition, it must be a longer contest.

And the biggest point; short contests mean they won't be global contests. If you run a contest lasting only a few hours, heaps of countries in the world wouldn't be able to compete because they'd be asleep. If you want to run a global competition, it must be a longer contest.

And the biggest point; short contests mean they won't be global contests. If you run a contest lasting only a few hours, heaps of countries in the world wouldn't be able to compete because they'd be asleep. If you want to run a global competition, it must be a longer contest.

And the biggest point; short contests mean they won't be global contests. If you run a contest lasting only a few hours, heaps of countries in the world wouldn't be able to compete because they'd be asleep. If you want to run a global competition, it must be a longer contest.

And the biggest point; short contests mean they won't be global contests. If you run a contest lasting only a few hours, heaps of countries in the world wouldn't be able to compete because they'd be asleep. If you want to run a global competition, it must be a longer contest.

And the biggest point; short contests mean they won't be global contests. If you run a contest lasting only a few hours, heaps of countries in the world wouldn't be able to compete because they'd be asleep. If you want to run a global competition, it must be a longer contest.

Do you wish to imply that Codechef problems are inherently erroneous? I see many independently organised Indian OPCs ( though not all ) which have very correct sets. Example being the ones conducted by IIT-Madras, IIIT-Hyderabad and NIT-Trichy, there may certainly be others, but I dont recollect them.

Do you wish to imply that Codechef problems are inherently erroneous? I see many independently organised Indian OPCs ( though not all ) which have very correct sets. Example being the ones conducted by IIT-Madras, IIIT-Hyderabad and NIT-Trichy, there may certainly be others, but I dont recollect them.

Do you wish to imply that Codechef problems are inherently erroneous? I see many independently organised Indian OPCs ( though not all ) which have very correct sets. Example being the ones conducted by IIT-Madras, IIIT-Hyderabad and NIT-Trichy, there may certainly be others, but I dont recollect them.

Do you wish to imply that Codechef problems are inherently erroneous? I see many independently organised Indian OPCs ( though not all ) which have very correct sets. Example being the ones conducted by IIT-Madras, IIIT-Hyderabad and NIT-Trichy, there may certainly be others, but I dont recollect them.

Whats wrong if the contest times are rotated amongst a 8:00 15:00 21:00 IST with a strong preference to week-ends. If you still have time-zone issues it's a little absurd.Anyways I don't have much hope of a shorter contest from Codechef, they seem to be very dogmatic about their current format. I have just offered my suggestions ( which are futile in many senses ).

Whats wrong if the contest times are rotated amongst a 8:00 15:00 21:00 IST with a strong preference to week-ends. If you still have time-zone issues it's a little absurd.Anyways I don't have much hope of a shorter contest from Codechef, they seem to be very dogmatic about their current format. I have just offered my suggestions ( which are futile in many senses ).

Whats wrong if the contest times are rotated amongst a 8:00 15:00 21:00 IST with a strong preference to week-ends. If you still have time-zone issues it's a little absurd.

Anyways I don't have much hope of a shorter contest from Codechef, they seem to be very dogmatic about their current format. I have just offered my suggestions ( which are futile in many senses ).

Hey Pankaj,Probably we have failed to make you realize how much we value your feedback. In fact the more critical it is, the more it lets us know where to improve. They can never be futile. And we really appreciate the effort you have taken to press forward your point. We realize and have been telling you too that there has been a compelling need for the shorter versions of the contest. They provide a different set of challenges which the longer format does not. Also, we need to understand that to conduct the same there are separate set of challenges for us too. And we need to prepare for that, which in turn takes time. Right now we have quite a few things to iron out but we do have that point in our To-do list. Can only request you to not loose faith in us. Bear with us for some more time. Dogmatism is the last thing that can take us to where we want to go and that is where you all want us to see .

Hey Pankaj,

Probably we have failed to make you realize how much we value your feedback. In fact the more critical it is, the more it lets us know where to improve. They can never be futile. And we really appreciate the effort you have taken to press forward your point.

We realize and have been telling you too that there has been a compelling need for the shorter versions of the contest. They provide a different set of challenges which the longer format does not.

Also, we need to understand that to conduct the same there are separate set of challenges for us too. And we need to prepare for that, which in turn takes time. Right now we have quite a few things to iron out but we do have that point in our To-do list.

Can only request you to not loose faith in us. Bear with us for some more time. Dogmatism is the last thing that can take us to where we want to go and that is where you all want us to see .

Sure. If you rotate them through different times, then I'd be able to compete in a third of them. This way, I get to compete in all of them. Of course I don't want to miss out on 2/3 of all contests.

Sure. If you rotate them through different times, then I'd be able to compete in a third of them. This way, I get to compete in all of them. Of course I don't want to miss out on 2/3 of all contests.

Sure. If you rotate them through different times, then I'd be able to compete in a third of them. This way, I get to compete in all of them. Of course I don't want to miss out on 2/3 of all contests.

(Sadly, for this exact reason, I can't compete at Topcoder at all. The contest held in my timezone is in the middle of the afternoon when I'm working.)

(Sadly, for this exact reason, I can't compete at Topcoder at all. The contest held in my timezone is in the middle of the afternoon when I'm working.)

(Sadly, for this exact reason, I can't compete at Topcoder at all. The contest held in my timezone is in the middle of the afternoon when I'm working.)