Upgrade of Timus test server happened on autumn 2008. I became several times faster, but very few time limits were changed. This problem is one that needs time limit to be lowered. Because with 1s TL and new servers some brute force solutions get AC - solutions that would never get AC on old testing server.

I suggest TL=0.2s considering that some languages (Java and C#) need about 0.1s to start program.

I solved this problem by probabilistic approach and brute-force. No TLE, just stupid WA :)

need help me please....

Edited by author 01.05.2009 17:09

Edited by author 01.05.2009 17:09

ofcourse i've tried it on my computer
it runs that program just fine
i've tried submitting my solution
3 times. why is it giving me CE and WA

can anyone explain it?

Is there anyone who's got AC after getting WA#6 and then rectifying his code. Any tricky test case.

Edited by author 28.11.2007 21:39

All codes listed below works good on my computer (and they give correct answers), but when submitting I get:

Code #1:
---------------------------------------------------------
#include<stdio.h>
#include<iostream.h>

int main()
{
  long long n;
  cin>>n;

  long long wynik;
  wynik=(n*(n+1)*(n+2))/2;

  cout<<wynik<<endl;
  return 0;
}
----------------------------------------------------------
Compilation error

Code #2: removed iostream.h and replaced cin and cout with
scanf("%lld",%n); and printf("%lld\n",wynik);

WA on test 5

Code #3: same as #2, but with
printf("%0.lf\n, (double)wynik); instead

WA on test 1
-----------------------------------------------------------

All those programs work and give correct results on my computer. What's wrong here? Please help.

I solve this task with DP, but wa#12... Please give me some useful tests. Thanks.

Just use The Wave algorithm..

program Project1;
{$APPTYPE CONSOLE}
var
 sym : char;
 a, b : AnsiString;
 la, lb, i : qword;
 l, m, n : byte;
 c : array of byte;
begin
  read(sym);
  repeat
  a:=a+sym;
  read(sym);
  until (sym=' ');
  readln(b);
  la:=length(a);
  lb:=length(b);
  if la>lb then setlength(c, la) else setlength(c, lb);
  i:=0;
  n:=0;
  repeat
    i:=i+1;
    if la>0 then l:=ord(a[la])-48 else l:=0;
    if lb>0 then m:=ord(b[lb])-48 else m:=0;
    c[i]:=l+m+n;
    n:=c[i] div 10;
    c[i]:=c[i] mod 10;
    if la>0 then la:=la-1;
    if lb>0 then lb:=lb-1;
  until (la<=0) and (lb<=0) and (n=0);
  while (i>0) do begin
  write(c[i]);
  i:=i-1;
  end;
end.

=)
      (=

"Сначала даны начальная и конечная позиции на доске в стандартной шахматной нотации"

Standart chess notaion is http://www.google.ru/imglanding?q=%D1%88%D0%B0%D1%85%D0%BC%D0%B0%D1%82%D1%8B&imgurl=http://www.chess-master.net/images/play_chess.gif&imgrefurl=http://www.chess-master.net/articles/9.html&h=350&w=400&sz=58&tbnid=1oNeUJIo8F_rpM:&tbnh=109&tbnw=124&prev=/images%3Fq%3D%25D1%2588%25D0%25B0%25D1%2585%25D0%25BC%25D0%25B0%25D1%2582%25D1%258B&usg=__rCBt0C2kIGUCdnfufCJ5tdd96cw=&ei=blB-S6fEOZLj-QaRmPH8Cw&sa=X&oi=image_result&resnum=5&ct=image&ved=0CCIQ9QEwBA&start=0#tbnid=1oNeUJIo8F_rpM&start=0

but in this problem row numbered from top to bottom

sorry  my English is bad

Edited by author 19.02.2010 14:04

Edited by author 18.02.2010 20:17

Why I got MLE#3? I'm using 1 int[M][3] matrix to store the edges, 3 short[N+1] arrays for Union-find, and an additional short[N] to store the resulting spanning tree.

Bug in qsort.

Edited by author 18.02.2010 15:19

why it is wrong?

"An arithmetic expression in D++ is always opened by "(", is closed by ")" and may contain any of the following symbols: "=+-*/0123456789)(""

arithmetic expression is opened by '(',may contain '*'and closed by ')'.so "(*)" can seen as a right  arithmetic expression.

anyone can answer me??

Dear Team,
It is really very good support by you.
But please, look at the problem 1001-in java.
There might be problem in compiler-there is no one who solved this question in java.
I will include my answer too.:

import java.util.Scanner;
public class second{
public static void main(String[] args) {
Scanner s = new Scanner(System.in);

double[] arr2=new double[200000];

int i=0,k=0;
do{
arr2[k]=s.nextDouble();
k++;
} while(s.hasNext());

for(i=k;i>0;i--){
System.out.printf("%.4f\n",Math.sqrt(arr2[i-1]));
}
}
}

My solution use emulation with some heuristic conditions (for avoiding deadlock). Solution is o(n).

Some test:
1
output
3
0 2 1
--------
2
output
8
1 3 4 2 0 1 3 2
--------
3
output
15
2 4 5 3 1 0 2 4 6 5 3 1 2 4 3
--------
4
output
24
3 5 6 4 2 1 3 5 7 8 6 4 2 0 1 3 5 7 6 4 2 3 5 4
--------
5
output
35
4 6 7 5 3 2 4 6 8 9 7 5 3 1 0 2 4 6 8 10 9 7 5 3 1 2 4 6 8 7 5 3 4 6 5

    Please tell me.
    Thanks.

#include <iostream>
using namespace std;
#include <math.h>

struct Point
{
    long double x;
    long double y;
};
struct Point2
{
    int number;
    long double angle;
};
void insertSort(Point2 ** a, long size) {
  Point2 * x;
  long i, j;

  for ( i=0; i < size; i++) {  // цикл проходов, i - номер прохода
    x = a[i];

    // поиск места элемента в готовой последовательности
    for ( j=i-1; j>=0 && (a[j]->angle < x->angle); j--)
      a[j+1] = a[j];      // сдвигаем элемент направо, пока не дошли

    // место найдено, вставить элемент
    a[j+1] = x;
  }
}
/*
int compare( const void *arg1, const void *arg2 )
{
   if(((Point2*)arg1)->angle>((Point2*)arg2)->angle)
       return -1;
   else
        return 1;

}*/
int main()
{

    int n;
    cin>>n;
    Point * arr = new Point[n];
    long double min_a=0;
    int min_ai;
    for(int i = 0;i<n;++i)
    {
        cin>>arr[i].x;
        cin>>arr[i].y;
        if((!i)||arr[i].y<min_a)
        {
            min_a = arr[i].y;
            min_ai = i;
        }
    }
    cout<<min_ai+1<<" ";
    Point2 ** arr2 = new Point2*[n-1];
    int j = 0;
    long double xx = arr[min_ai].x;
    long double yy = arr[min_ai].y;
    for(int i = 0;i<n;++i)
    {
        if(i!=min_ai)
        {
            arr2[j] = new Point2;
            arr2[j]->number = i;
            arr2[j]->angle =  (arr[i].x - xx)/sqrt(pow(arr[i].x-xx,2)+pow(arr[i].y-yy,2));
            ++j;

        }
    }
    //qsort(arr2,n-1,4,compare);
    insertSort(arr2,n-1);
    cout<<arr2[(n-1)/2]->number+1<<endl;
    return 0;
}

I don't know why, but qsort sorts them wrongly (I commented part that is qsort and changed it with simple O(n^2) sorting and got AC... But could tell me, what's so wrong with this qsort? I think I could get a better time with it.

The only email I found on this is site is timus_support@acm.timus.ru, which when I send an email to, I got mail return:

The original message was received at Sun, 24 Jan 2010 15:38:57 +0500
from mail-pw0-f57.google.com [209.85.160.57]

  ----- The following addresses had permanent fatal errors -----
judge_mailbox@busin.usu.ru
   (reason: 550 5.1.1 User unknown)
   (expanded from: <timus_support@acm.timus.ru>)

  ----- Transcript of session follows -----
550 5.1.1 judge_mailbox@busin.usu.ru... User unknown

Final-Recipient: RFC822; timus_support@acm.timus.ru
X-Actual-Recipient: RFC822; judge_mailbox@acm.busin.usu.ru
Action: failed
Status: 5.1.1
Diagnostic-Code: X-Unix; 550 5.1.1 User unknown
Last-Attempt-Date: Sun, 24 Jan 2010 15:39:05 +0500


---------- Forwarded Message ----------
From: Nat Pavasant <n...@gmail.com>
To: timus_support <timus_support@acm.timus.ru>
Date: Sun, 24 Jan 2010 17:38:44 +0700
Subject: ...