Round 1C editorial is in a state that can be called "ready to publish". I will probably still do some more fixes and so and so
This was a interesting round. according to practice rooms I would have gotten around 690 points. 250 took me longer than it took to your average high rated coder, that is because it occurred me to use binary search and it took some extra time to code. 500 was easier, I had an overflow bug but I still passed, later I was able to find proof that the implementation is such that the overflow cannot happen.
Contrary to my expectations, this match had the hardest problem out of the 3 round 1s. 1000 was quite hard. Even though I could think of a solution in time for my 75 minutes coding phase simulation, It took me a couple of hours to finish the code. And a couple of more hours and external to understand how to come up with those simpler-looking approaches. This is the reason I needed a day instead of half a day to solve finish this match's editorial.
Both 500 and 1000 are interesting in that experience could have actually been a liability instead of an asset. In 500, I think many coders may have failed a challenge or two trying to challenge a solution that seems to overflow but doesn't. In 1000, I think that solving HyperKnight before had an effect in me that lead me to a complicated solution instead of trying harder to find simplifications. Sometimes our experience behaves more like preconception than experience.
I still remember that SRM in which the division 2 version of a problem had larger constraints than the division 1 version. Because people in division 1 version would take those small constraints as a hint to implement a complicated exponential time solution. Meanwhile, division 2 coders easily noticed the two-lines greedy solution (and most likely didn&apot; prove it).-----------------------------
About SRM 572's editorial, I really wouldn't hold my breath :(. Still trying to solve that division 1 hard. But remember that you can see the editorial preview for the other problems here.