Comments
0

Thank you very much, that was the problem!

0

Strange, I'm doing the same, yet I can't get past test #2. I made random tester and checker just to find that my code easily passes a few million tests. It seems I am missing something, can somebody help me with tests/suggestions?

The code is 8417893.
On fmm → c++ STL vector is too slow, 13 years ago
+13

One vector is not that slow. However, in your stl-vector code you allocate a million vectors! What happens then? Well, for each vector are allocated 3 pointers (begin, end and storage) as well as some elements for initial storage (this size doubles every time when reached). When you have a single vector that overhead is not that much, but when you have a million of them, it is terrifying.

Note: The same holds for other STL data structures as well — for example allocating a million priority_queues results in the same overhead.

On Zlobober → CEOI 2012, 14 years ago
0

I terribly sorry, but can you also upload your solutions from Day1. Thanks in advance :) And, as I understood from the Russian version of the thread, congratulations for your top scores in both days, a really amazing achievement!
On Alex_KPR → Codeforces Beta Round #58, 16 years ago
+4
Oh, that "Confirm" button is sooo tempting.. 
On Alex_KPR → Codeforces Beta Round #58, 16 years ago
0
I have an idea=) Give all people the opportunity to choose whether their rating will be rated or not (although it is difficult to choose a default option - rate or not) and then make new rankings only with the people who want to be rated. After that calculate new ratings based on that performance. 

In this way the element of knowing approximately where you would be after sys tests is removed. 
+1
Thanks for this great contest. I liked the problems, but I didn't like that I spent at least 15 minutes until I guessed what the fifth task is about. One of the inputs was wrong which made it even harder. Anyway, I'd be interested to see the official solutions:)

So only the first 3 win prizes?
0
I'm not that good at maths and although I knew little Fermat's theorem before, using it in that way was something new to me. 

But what if the module wasn't a primer number? In the contest I used dp to calculate the combination and my approach isn't depended by the module. I'm curious if there is a way (maybe involving math) different than this if the module isn't prime?
On nevidomy → Codeforces Beta Round #53, 16 years ago
+30
Really nice problemset:) Thank you. It is great to see here other fans of Family guy:P
On the_saint → Member photo change day(s), 16 years ago
0
I am stunned by the fact that the entire site is maintained and being updated by only one person. Because of his efforts the site keeps growing and growing with every single day. The last thing I want to do is to give him extra work.

But how much work from the admins is really needed to make this possible? We can choose the question in this way: the best rated (+/-) from the comments of a given blog post for this event. Everybody who wants can change his/her photo so we don't need admins help here. The only thing they can help with is by making the event a bit more official by putting the post in a more visible place.

PS. It took me a week or so to notice what your handle really means:D
On philolo1 → Just a suggestion, 16 years ago
+3
I really like that idea, but such feature will be useful only if it shows how many problems somebody has solved in an actual contest (and there should be two different rankings for each division). There are many other stats which can be interesting (like "Best hackers", "Coders with most points from solving", "Average coding speed", etc.)

Nevertheless, it would be great if you can view every problem you have solved. In that way keeping track of your practice will be much easier than now.
0
Successful hacking attempt +100