I mean for short contests. I know that LC of Codechef intends users to learn new techniques. There is never an overload during Long Challenge, and if there is, it is like who cares, it is a 10 DAY Challenge. For short contests time is a major issue. Wrong submissions too, for ACM-ICPC based contests.
I am sorry, but I guess you need to understand the difference between a 10 day long contest and a 3 hour long contest, that too with time being a factor in the rankings.
Hi @anon84548374, sorry for that. But this contest was actually intended as a onsite event at our college( which is actually happening) and we then extended it to codechef. And too on saturday night we are already having August lunchtime. To avoid that clash and also conduct our onsite online event this date and time was decided. And well I hope contest is favoured over classes. 
You can reasonably estimate the runtime of a piece of code except if it’s heuristic, then it’s one of the “trash algorithms” I mentioned
Im not saying no tests are given. A good strategy to test algorithms’ correctness is to use a brute force solution first, so you dont have to waste time coding data structutres, then if no WAs happen start coding the correct algorithm.
10 - 15 tests are reasonable enough for a code to be expected to be correct for general tests. If it failed after either you most likely didnt consider edge cases and more time consuming tests, which is on you.
1 Like
I literally said adhoc probs !!
they do not hav brute force soln
you either get AC or WA
Will it be based on data-structures and algorithms(cp), or any random stuff not related to cp ?
Lunch time will be unrated. Lol.
2 Likes
Are you sure ??? 
What you are saying, is basically that our solution won’t run on ANY test case and will give the verdict after the contest. That is not my point. My point is that suppose we have 40 test cases for a problem, during the contest it will work on 8 to 10 of them and the rest after the contest. Thus there will be a less overload on the server. Please understand my point and then provide a rebuttal for that.
It won’t. When I participate it is rated.
Not sure bro. Was joking. Btw July lunchtime was unrated.
whats the pt of verdit if it can change after contest ?
I completely understood ur pt but u don’t understand why there are multiple test cases
why do u think there are multiple test cases?
coz to cover edge cases
Hence whether u get verdict on some or none of it doesn’t matter coz it can change due to remaining test cases
You will understand where you went wrong and rectify it next time.
Codechef doesn’t have a problem with any edge test case (or it rarely does). Secondly, you will always add edge test cases into the pretests.
For you, lets conduct a poll, on how many people think that there is an edge test case in the pretests.
The question is do you think that edge test cases are inserted in pretests?
- Yes
- No
0 voters
That happens only when a person applies some wrong algorithm(to be formal). You can always see the editorial and come up with a solution. And also understnad which test case prevented you from getting AC.
As a datapoint, Hackerrank use a similar system to that proposed by @aryan12 for their “Week of Code”, except that the “additional” testcases for a Problem are run every 24 hours after a Problem is revealed (so the penaity for getting AC on the preliminary tests but then later failing some of the “additional” tests is effectively a reduction in the maximum score you can get for that Problem). It seems to work quite well, though is nerve-wracking 
1 Like
That’s quite a good solution to be very honest. But the problem lies in between the short contests. But still, the suggestion was quite good. 
1 Like
Ya, so on that basis, div2 will be RATED and div1 will be UNRATED 
2 Likes
You don’t seem to be able to comprehend that the pretests are only there to provide you with a reasonable expectation of the correctness of algorithm. Also Codechef sucks with test cases anyway, edge cases are barely covered so it doesn’t even matter that much.
Verdicts doesn’t matter? Ok, your thinking. If others think it matters and works, then your opinion doesn’t matter in the slightest.
what a childish debate u r doing?
what is the use of extra test cases if all edge cases are covered in pretests ?
1 Like
if codechef sucks then why r u here ?
and I never meant that verdict shld not be provided
i was saying that providing verdict on some test cases is meaningless
yeh topic kaha se kaha tak ja raha hai…
betr someone closes this topic right now
2 Likes