C compiler bug (floating point arithmetic)?

Question

#include<stdio.h>

int main()
{
    double fract=0;
    int tmp;

    //scanf("%lf",&fract);
    fract=0.312;
    printf("%lf",fract);
    printf("\n\n");
    while(fract>0){
        fract*=(double)10;
        printf("%d ",(int)fract);
        fract-=(int)fract;
    }
    getch();
    return 0;
}

this code shoud have an output of: 312

but somehing isn't going right.. i'm using devcpp 4.9.9.2 compiler...

Answer 1

Kernighan & Plauger say, in their old but classic book "The Elements of Programming Style", that:

• A wise old programmer once said "floating point numbers are like little piles of sand; every time you move one, you lose a little sand and gain a little dirt".

They also say:

• 10 * 0.1 is hardly ever 1.0

Both sayings point out that floating point arithmetic is not precise.

Note that some modern CPUs (IBM PPC) have IEEE 754:2008 decimal floating point arithmetic built-in. If using the correct types, then your calculation will be exact.

Answer 2

Floating-point arithmetic is confusing, and not guaranteed to behave intuitively.

Here's a good reference document: What Every Computer Scientist Should Know About Floating-Point Arithmetic. It's a long document, because it's a complicated problem.

In summary: Don't use floating-point values if you are relying on exact values.

Answer 3

So you multiplied 0.3119999999999999999895916591441391574335284531116485595703125 by 1000 and truncated it and got 311? I don't see where the problem is.