super incomprehensible, in general the tests are about some kind of accuracy, you need to make 100 parcels and try different epsilons everywhere to pass these tests, it pissed me off

hypercircle does not interfere with the thread

1 2 0 0 0 0 0 0
3 3 0 0 0 0 0 0
0 0 0 0 0 0 0 0
2

2.24

New tests have been added to the problem which exploit a common mistake when using acos function. Hint: try to calculate acos(sqrt(3.0) * sqrt(3.0) - 4.0) in your programming language. 269 authors (58%) with accepted solutions have been challenged.

this task is the same as 1075, but with a different number of dimensions(it doesn't really matter)

I submitted some algo which on my computer gives correct result for sample input given in problem statement and I got WA in test #1. Then I tried this code:

public class Problem_1285 {
    public static void main(String[] args) {
        System.out.println("1.00");
    }
}

...and got the same WA #1. What's going on???

There are a lot of tests (i know two: 4 and 9), in which r <= 0. Please, change it!

1075 with same algorithm have got AC.
Can somebody help me? I need some tests or hints...

Is this test correct?

1000 0 0 0 0 0 0 0
-1000 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
999

answer:
3138.51

I got WA37 in 1285
and AC on 1075
I use not diffrens algos, only different input type.

I think correct add test and for 1075 too.

Can anybody explain me, where am I wrong?

Definitions:
ac, ab, bc - corresponding lengths
acb - angle between (ca) and (cb)

Such code:
...
acb = acos((ac*ac + bc*bc - ab*ab) / 2.0 / ac / bc);
..

gets AC.

Such code:
...
double spv(point x1, point x2, point y1, point y2){
    int i;
    double res = 0;
    for (i=1; i<=n; i++){
        res += (x2.coord[i] - x1.coord[i]) * (y2.coord[i] - y1.coord[i]);
    }
    return res;
}
...
acb = acos(spv(c,a,c,b)/ac/bc);...
...

gets WA25.

I had always thought that these two fragments must yield the same result. Why don't they?

Are you sure that your tests are ok?
Does the first test coincide with the example?
I have WA1.

Фрагмент кода при проверки лежит ли путь через сферу.
Почему то вещественная проверка проходит, а целочисленная нет. хотя в long long все должно влазить. Извините, что на русском, но писать на английском я совсем не могу)
    int i;
    long long a=0,b=0;
    for (i=0;i<maxd;i++)
    {
        a+=sqr(p2.x[i]-p1.x[i]);
        b+=(p2.x[i]-p1.x[i])*(x[i]-p1.x[i]);
    }
    if (b<0||b>a)
        return false;
//-AC-----------------------------------
    double t=(double)b/a;
    double rr=0;
    for (i=0;i<maxd;i++)
        rr+=sqr((p2.x[i]-p1.x[i])*t+p1.x[i]-x[i]);
    rr=sqrt(rr);
    return rr<r;
//-WA7-----------------------------------
    long long t=0;
    for (i=0;i<maxd;i++)
        t+=sqr((p2.x[i]-p1.x[i])*b+a*p1.x[i]-a*x[i]);
    return r*r*a*a>t;
}
//В задаче 1075 тоже самое только WA на 12 тесте.


Edited by author 31.07.2007 12:53

Tests 5, 9 and 10 were incorrect -> fixed.
Tests were too weak -> some new tests have been added.

Problem has been rejudged.

Although problem statement says r > 0 and AC, BC > r,

1. It's possible that r == 0
2. It's possible that AC == r or BC == r

I' ve got Crash (FLT_DIVIDE_BY_ZERO).
I added one condition:
If r = 0 then
  WriteLn(d:0:2)
and get AC.
But in the task we can read:
"The radius R is integer positive number"

Please!!!