kostka's blog

By kostka, 7 years ago, In English

Stats about hacks in Codeforces Round #263 (Div. 1) (quite late).

Stats

Problem Successful hacks Unsuccessful hacks Other Sum
461A - Appleman and Toastman 6 (8%) 23 (32%) 43 (60%) 72
461B - Appleman and Tree 0 (0%) 2 (33%) 4 (67%) 6
461C - Appleman and a Sheet of Paper 1 (100%) 0 (0%) 0 (0%) 1

Fastest hackers

Problem Time Hacker Defender Hack
461A - Appleman and Toastman 00:18:40 IWillBeRed thermal 110022
461C - Appleman and a Sheet of Paper 01:56:17 FatalEagle mareksom 111898

Congratulations for FatalEagle for one and only hack for problem C! It was really fatal hack for mareksom (in last 5 minutes).

Remarks

461A - Appleman and Toastman

In Div. 2 there were no hacks in this problem, but many solutions failed on the simple test (#34):

2
1 2

Some people had additional condition for this case and output sum of a0 and a1. Good thing is that only 4 more solutions failed during systests — all on this killer test.

461B - Appleman and Tree

Many solutions failed on test #19 and the problem was bad use of modulo. I described it in Div. 2.:

In such case:

x = a%m + b%m

sometimes x can be grater than m (for example for a = 6, b = 7 and m = 4), so we should use:

x = (a%m + b%m)%m

461C - Appleman and a Sheet of Paper

Just one hack, but there were more possibilities. Let's first look at this hack. Test was pretty simple:

** Generator source **
#include <bits/stdc++.h>

using namespace std;

int main()
{
    printf("100000 100000\n");
    for(int i=0; i<100000; i++)
        printf("2 0 99995\n");
    return 0;
}

Where was the problem? Let's look at mareksom's solutions: 7586989 and 7596313 — the first was one hacked and the second accepted (after contest). The only change was adding tree to deal with query no. 2. Many solutions without BIT or segment tree passed pretests, but didn't pass test #13.

461D - Appleman and Complicated Task

In nearly every solution which didn't pass system tests, the problem was incorrect checking if it is even possible to make one board fulfilling the condition, in particular, the condition from tutorial: "If both 2i and 2i + 1 are in the same set for any i, the answer is 0." was checked with some minor mistakes. Depends on where was the problem, test 19 or 25 was not allowing to pass.

Hackers

There were so little hacks that we can show all successful ones:

Hacker Stats Successful hacks Unsuccessful hacks
littlelittlehorse +1-0 A: 111344
mhadih +1-0 A: 110307
maozaozao +1-0 A: 110225
IWillBeRed +1-0 A: 110022
FatalEagle +1-0 C: 111898
Erich +1-0 A: 111146
skyxuan +1-1 A: 110666 A: 111766

Rooms, where something happened

Room #hacks Hackers
34 1 FatalEagle [1]
31 1 littlelittlehorse [1]
16 1 maozaozao [1]
15 1 skyxuan [1]
13 1 IWillBeRed [1]
8 1 mhadih [1]
6 1 Erich [1]

Countries which successful hacks

Country #hacks Hackers (with at least one successful hack)
China 2 skyxuan [1], Erich [1]
Ukraine 1 IWillBeRed [1]
Iran 1 mhadih [1]
Canada 1 FatalEagle [1]
Country #hacks / #hackers Hackers (with at least one successful hack)
Ukraine 1.00 IWillBeRed [1]
Iran 1.00 mhadih [1]
Canada 1.00 FatalEagle [1]
China 0.33 skyxuan [1], Erich [1]

Gap

Hack Hacker Defender
111898 FatalEagle [1972] mareksom [2278]
111146 Erich [1777] hiyot [1863]
110225 maozaozao [1701] alex4814 [1759]
110307 mhadih [1700] Fantasy-zwj [1747]
110022 IWillBeRed [1717] thermal [1750]
110666 skyxuan [1748] Boxer [1716]
111344 littlelittlehorse [2372] paulwang [2041]

As we can see, more common is hacking something with better rating. :)

Charts

Here should be graph.

Successful hacks over time

Here should be graph.

Thanks and again: sorry for all my mistakes, feel free to correct me. Other posts can be found here.

 
 
 
 
  • Vote: I like it
  • +52
  • Vote: I do not like it