Comparing floats

[saneman]

I have a function:

int F(double a)
{
if (a = =1.0)
{
return 22;
}
return 44;
}


When I call F(1.0) it returns 22. But if I call F from a loop like:

double a = 0.0;
for (int j = 0; j <= 1000; j++)
{
std::cout << F(a) <<std::endl;
a = a + 0.001;
}

It returns 44 instead when 'a' reaches 1 instead of 22!

After a few hours of messing with my brains I found a website where they say
that comparing floats in C++ is not guranteed to return an expected result.
Instead something like this should be used:

if (fabs(result - expectedResult) < 0.00001)

But before I rewrite all my code that compare floats I would like to hear if
anyone has any better suggestion.

 

[Gennaro Prota]

I have a function:

int F(double a)
{
if (a = =1.0)
{
return 22;
}
return 44;
}


When I call F(1.0) it returns 22. But if I call F from a loop like:

double a = 0.0;
for (int j = 0; j <= 1000; j++)
{
std::cout << F(a) <<std::endl;
a = a + 0.001;
}

It returns 44 instead when 'a' reaches 1 instead of 22!

After a few hours of messing with my brains I found a website where they say
that comparing floats in C++ is not guranteed to return an expected result.
Instead something like this should be used:

if (fabs(result - expectedResult) < 0.00001)

It depends on what you want to do. Most times it is wrong as well. But
this isn't C++ specific; for starters, study the paper "What Every
Computer Scientist Should Know About Floating-Point Arithmetic", by
David Goldberg.

 

[saneman]

This is because:

1) When internally represented in binary, fractional amounts are
irrational.

2) Addition accumulates roundoff errors.

Add the following to your inner loop:

std::cout << a << std::endl;

You'll see what's happening.

I have tried doing std::cout << a << std::endl; but it just prints the
numbers from 0.001 to 1 as expected.