Re: casting types

[Dennis Lee Bieber <wlfraed@ix.netcom.com>]

On Wed, 01 Oct 2014 19:38:22 +0200, Dirk Heinrichs
<dirk.heinrichs@altum.de> declaimed the following:

{responding to Stribor40}

>Stribor40 wrote:
>
>> i would like to represent all numbers in interval [0, 1]
>
>Well, according to what I've learned in mathematics, you can't. Neither in 
>real life, nor in any programming language. You need to define some 
>constraint, like number of digits after the decimal point.
>
	And on most machines these days -- you can expect the hardware to
support single precision (32-bit) with ~7 significant digits, double
precision (64-bit) with ~15 significant digits... And only within the
floating point processor, maybe IEEE 80-bit (downconverted when the final
result is returned to the user).

	Note that "significant digits" does not take into account the decimal
point...

	12300000.0, 1.23, 0.00000123

all have 3 significant digits if they are /floating point/ values (they are
all 1.23E<some exponent>)


	Fixed point types are closer to integers with an implied decimal point.
-- 
	Wulfraed                 Dennis Lee Bieber         AF6VN
    wlfraed@ix.netcom.com    HTTP://wlfraed.home.netcom.com/