Re: how compiler decide the range of real numbers in C...

Glynn Clements <[email protected]> Thu, 30 Jul 2009 18:46:53 +0100

Newsgroups	org.kernel.vger.linux-c-programming
Message-ID	<[email protected]>

RAM_LOCK wrote:

> Say 16 bit compiler (for ex Turbo C) uses 2's compliment to find the range of
> signed integer.
> Say for 16 bit n=16. Now applying 2's compliment equation : -2^(n-1)  to
> 2^(n-1) -1
> so for n=16 : The range is -32768 to 32767
> 
> This is ok. The question arises when it tell about real constant's range.
> For 16 bit compiler the range of real constant is : -3.4e38 to 3.4e38
> 
> How they arrived to this real constant range?
> 
> Can any one share the mathematical calculation behind it?

	http://en.wikipedia.org/wiki/IEEE_754-1985

-- 
Glynn Clements <[email protected]>