Java Double Trouble

Question

Possible Duplicate:
Moving decimal places over in a double

Code :

    public class Teste {
        public static void main(String[] args) {
            System.out.println("Exponent loop:");
            for(int i=1; i<10; ++i)
               System.out.println(Math.pow(0.1, i));

            System.out.println("\nMultiplication loop:");
            double a = 1;
            for(int i=1; i<10; ++i){
               a *= 0.1;
               System.out.println(a);
            }
        }
    }

Output:

Exponent loop:

0.1
0.010000000000000002   
0.0010000000000000002    
1.0000000000000002E-4    
1.0000000000000003E-5    
1.0000000000000004E-6    
1.0000000000000004E-7    
1.0000000000000004E-8    
1.0000000000000005E-9

Multiplication loop:

0.1    
0.010000000000000002    
0.0010000000000000002    
1.0000000000000003E-4    
1.0000000000000004E-5    
1.0000000000000004E-6    
1.0000000000000005E-7    
1.0000000000000005E-8    
1.0000000000000005E-9

Duplicate of many questions. Is there one standard answer to this question? — Reinstate Monica -- notmaynard, Jan 29 '13 at 21:45
@AndreasGrapentin 1.0000000000000002E-4 != 1.0000000000000003E-4 — Luiggi Mendoza, Jan 29 '13 at 21:45
`pow` doesn't perform iterative multiplication, so it's no surprise that sometimes the result is slightly different. — Daniel Fischer, Jan 29 '13 at 21:48
Don't use `pow` if you can avoid it. It's expensive and can produce significant rounding errors. — Peter Lawrey, Jan 29 '13 at 21:49
@LuiggiMendoza: perhaps better than saying the results look "correct", we should say that the results look "as expected". — Hovercraft Full Of Eels, Jan 29 '13 at 21:49
Dear OP, I suggest you read this, so understand what's going on behind the scenes of your code: http://en.wikipedia.org/wiki/Floating_point — Andreas Grapentin, Jan 29 '13 at 21:50
Instead of multiplying by `0.1` and cannot be represented exactly, try dividing by 10 instead (10 can be represented exactly) — Peter Lawrey, Jan 29 '13 at 21:50
@JuniorWate Perhaps you should explain why this is not possible ;) — Peter Lawrey, Jan 29 '13 at 21:51
@JuniorWate The link I posted contains everything you need to know. 0.1 * 0.1 is internally **rounded** to 0.0100[...]2 and when you print it, you see the rounded value — Andreas Grapentin, Jan 29 '13 at 21:51
Try running `System.out.println("0.1 is actually " + new BigDecimal(0.1)); System.out.println("0.1 * 0.1 is actually " + new BigDecimal(0.1).pow(2)); System.out.println(" and the closest double value is " + new BigDecimal(0.1).pow(2).doubleValue());` — Peter Lawrey, Jan 29 '13 at 21:54

Java Double Trouble

0 Answers0