How to avoid calculation mistakes with primitives

Question

The following routine is expected to return 1, but instead it returns 0.9999999999999999.

double sum = 0;
for(int i=0; i<10; i++){
    sum+=0.1;
}
System.out.println(sum);

The same mistake occurs also with other java mathematical operations.

Is there any way to avoid this issue?

Thank you in advance.

It's not a mistake. It is the nature of floating point operations. — meskobalazs, Jul 06 '16 at 14:34
Floating point arithmetic is inherently inaccurate. You could try with ```BigDecimal``` but that's a lot slower. — Jorn Vernee, Jul 06 '16 at 14:35

score 1 · Accepted Answer · answered Jul 06 '16 at 14:37

This is what's called a floating point error. It is natural and to be expected. The way to avoid it is to use variables having higher precision, such as BigDecimal. However BigDecimal is slower to perform calculations, so consider how important accuracy is important in your solution.

I'd recommend using BigDecimal despite any performance hits if these quantities represent currency. Its more important your calculation be correct than fast in that case.

How to avoid calculation mistakes with primitives

1 Answers1