Weak test cases in BINFUN

general
, ,
1

A bruteforce approach for N less than 1000 and an incorrect logic (Take the maximum element for 1st string and iterate over array once to find the 2nd string) for rest of the test cases works fine in Binary Fun.
Take a Look at my submission
https://www.codechef.com/viewsolution/36021312
[A contradictory test case which fails in O(n) Logic mentioned is]:
1
3
507 513 722
Expected output: 256518
O(n) Algo output: 213807
[just FYI] I randomly checked 2 other solutions (do not wanna mention), both of them used the same logic to get AC in all the test cases. So, I’m pretty sure there are many others.

16 Likes
2

oh damn! Does that really work for big testcases? I used it for all the testcases and got WA. It should be WA for others too.

2 Likes
3

This is codechef
Nothing is going to happen, I am damn sure.

6 Likes
4

Yes, I too checked few solutions and all used this logic. Isn’t this kind of issue with testcases happening quite frequently.

5

This is really heartbreaking. I spent almost two hours torturing myself over why my code wasn’t getting accepted. This sorta thing just makes me wanna give up.

3 Likes
6

#SHITTYTESTCASES forever

7

more than 10% of accepted solutions are like that itself.

2 Likes
11

But, why does “this” make you want to give up?
Either ways, you couldn’t solve the problem at hand! The best you can expect at the moment, is to get the incorrect AC solutions get rejected again, so that the ratings are recalculated, and you get a bit more positive delta than you did the first time!
But, do you really think that having that bigger positive delta/ or getting the wrong solutions rejected worth it, so much that you are willing to give up on cp for this?

1 Like
12

I meant giving up on CodeChef but you’re right, it’s no reason.

2 Likes
13

Dear @admin,

Situations like this ought to be avoided. This incorrectly gives a sense of successful submission of solution/logic to users whose approach was not covering all the cases. It also affects ratings of other users and decreases user trust on the contests hosted by the platform.

14

Le admin- Your suggestion has been automatically ignored