Блог пользователя ahmed_aly

Автор ahmed_aly, 12 лет назад, По-английски

Something I noticed in many Codeforces contests, the less rated author makes harder problems. Did anyone else notice this?

I think this is because the low rated authors underestimate their problems, and maybe this is happening because they are trying to think like the high rated coders to make sure that the problems are not too easy for them, which makes the problems hard even for the high rated coders.

If you think this is happening, what do you think are the reasons for this?

  • Проголосовать: нравится
  • +90
  • Проголосовать: не нравится

»
12 лет назад, # |
  Проголосовать: нравится +94 Проголосовать: не нравится

I think low rated coders are better at making problems which are solved using a non-standard novel idea.

Usually high rated coders rely on the ideas used before while solving problems. And I think this has its effect on the problems a high rated coder makes. The problems are more likely to be solved with a combination of hard yet standard and known algorithmic ideas.

But low rated coders make problems with a less background of advanced algorithmic techniques. So the hard problems they set might need a unique, nice and not easy-to-find idea to be solved.

Sometimes in the class I receive some nice and new solutions and proofs from the students for the problems I had solved before in the standard way. This could be a proof on their different way of thinking.

Although this does not hold for all high rated coders. As an example I enjoy the problems authored by dolphinigle and rng_58 as challenging and new problems.

On the other hand another reason which might cause harder contests from low rated coders is that someone might try to add traps and corner cases to the problem to make it harder. Personally, and fortunately I don't see this kind of problems on Codeforces recently. :)

»
12 лет назад, # |
  Проголосовать: нравится +23 Проголосовать: не нравится

Before beginning to theorize about why this happens, I'd like data to first confirm it is actually happening. You know, a table sorted by rating and the average solved percentages of each problem. And probably a control from another site in the similar format.