Subtracting Errors Binary Format

Question

Im currently making an application that takes two numbers and subtracts them
Everything is going fine the only problem is that if I take two numbers, lets call them A and B and their subtraction C a error occurs.
For Example:
A = 5.99
B = 5.98
C = 0.009999 is the current result
I believe the error comes from the subtraction in binary format but how can i fix this binary error?

Answer 1

I believe at least one, and probably all, of those articles describe multiple approaches for dealing with this issue.
If you had read my previous comment or even part of any one of the articles, you would know that no, it is not possible to get float or double to exactly represent every possible real number.
Try this excercise. Imagine a base-2 decimal number system that's just like the base-10 system we're familiar with.
100.0 = 4
10.0 = 2
1. 0 = 1
0.1 = 1/2
0.01 = 1/4
0.001 = 1/8
etc.
Now, write a fractional number with that system that is exactly 1/10 in a finite number of bits. Notice how it is impossible, just as it is impossible to represent 1/3 exactly in a finite number of digits base-10.
Now, read the articles, and decide which approach is best suited to your needs.
Edited by: jverd on Oct 27, 2008 1:14 PM