I want to discuss separately the negative feedback on the round 505.
I understand, that many participants are disappointed by solutions, which failed on system tests. In fact, the pretests in problems B, D and E turned out to be weak. The problem C also didn’t gone too smoothly, but I don’t see nothing critical.
Surely, this was a flaw in a work of author and coordinator. Unfortunately, such situations sometimes arise and it is difficult to avoid them at all. For these problems, the number of pretests and their type was not looking too weak.
Problem B: 9 pretests, there are small and large answers, two tests with answer -1, there are pretests with n = 1 and n = 2, there are four pretests with n = 150000.
Problem D: 14 pretests, among them manual tests and four different generators, few pretests with n = 700, majority of answers is ‘Yes’, but there are ‘No’ as well. In my opinion, too little pretests with ‘No’.
Problem E: 14 pretests. Yes, this problem on VK cup finals contained 10 pretests and caused many systests fails for participants. I have added 4 more tests to pretests from tests, which caused system tests fails for onsite participants. I was very surprised to see, that there were still so many fails after systests.
Summing up, the pretests turned out to be incomplete, but it is hard to say, that it was obvious defect by author or coordinator. Probably it is a combined effect from the problem specifics and the lack of experience of _kun_ as a coordinator.
I haven’t examined all the problems thoroughly, but still round seemed interesting to me. There were no serious fails with statements, bugs in tests and solutions. The system was also working smoothly, without large queue.
Please explain your negative feedback about the round. It will be very valuable to read a reasoned opinion from experienced participants.