Java：对Wallis产品进行编码以找到pi

Question

I am trying to write a code to output pi using the Wallis product formula: https://en.wikipedia.org/wiki/Wallis_product

This is my code:

public class PI {
    public static void main(String[] args) {
      pI();
    }

    public static void pI () {

    double pi = 0;
    int a = 0;
    int b = 1;
    double denom = 0;
    double num = 0;
    int temp = 0;
    int count = 0;
    double halfpi = 1;


        while(true) {
          temp = b;
          b = temp + 2;
          denom = temp*b;

          a += 2;
          num = a*a; 

          halfpi *= (num/denom);
          count++;

          System.out.println(halfpi*2);                    
        }    
    }
}

I have found that the code works just fine until the loop exceeds around 33000 iterations. After this point it just prints: "-0.0":

-0.0
-0.0
-0.0
-0.0
-0.0
...

----jGRASP: process ended by user.

What could be causing this?

Answer 1

redFIVE is right. You are mixing int and double which causes problems. This would fix it

public class PI {
    public static void main(String[] args) {
      pI();
    }

    public static void pI () {

    double pi = 0;
    double a = 0;
    double b = 1;
    double denom = 0;
    double num = 0;
    double temp = 0;
    double count = 0;
    double halfpi = 1;


        while(true) {
          temp = b;
          b = temp + 2;
          denom = temp*b;

          a += 2;
          num = a*a; 

          halfpi *= (num/denom);
          count++;

          System.out.println(halfpi*2);                    
        }    
    }
}

Answer 2

The issue is because you are mixing INT's and Doubles. Your code will not produce any errors because this type of multiplication is fine however when you are assigning ints and then trying to assign them back to doubles you are losing everything except for the integer value before the decimal.

double pi = 0;
double a = 0;
double b = 1;
double denom = 0;
double num = 0;
double temp = 0;
double count = 0;
double halfpi = 1;

Answer 3

The ints are restricting the capablility to get all the decimal points so fixing it should work. Also you will run out of heap space very quickly with this program. Just saying