C casting

[Loctan]

Can anyone tell me the difference between:

float answer = 3.0f;

float answer = (float)3.0;

I understand if you have a variable you can't append an "f" on it, you would have to cast it. Is the second one bad coding or not recommended? Will the compiler do anything different for each one? I always thought the first was sort of shorthand for the second, but I wasn't sure. The second one is called a "cast," but what is the first one called? Thanks!

[MacGyver]

I believe that the first one is natively taken as a float, and the 2nd one is taken as a double and then converted to a float. The first one is done at compile-time and the 2nd at runtime, unless the compiler is smart enough to optimize.

[Elysia]

I understand if you have a variable you can't append an "f" on it, you would have to cast it.

Yes, for numbers, you can append "f" to make it treated like a float. It cannot be done on variables; those must be cast.

[laserlight]

3.0f is a float literal, 3.0 is a double literal.

[BEN10]

i think as such there is no difference between 3.0f & (float)3.0 as both the statements do the same thing i.e treat 3.0 as a float.and yes 3.0f is stored as a float and 3.0 as a double.there sizes in memory will also differ.

[matsp]

i think as such there is no difference between 3.0f & (float)3.0 as both the statements do the same thing i.e treat 3.0 as a float.and yes 3.0f is stored as a float and 3.0 as a double.there sizes in memory will also differ.

Correct, the same value will end up in answer, but what is in memory that the answer gets loaded from may be different (assuming compiler doesn't optimize it - which is very likely in this case).

However, if we do something like this:

... 
   answer = 2.8f;

   ...

   if (answer == (float)2.8)
      printf("The same\n");

The result could possibly not be what you expect.

--
Mats