Re: whole number constants

On Jul 21, 4:38=A0pm, nos...@[EMAIL PROTECTED]
 (Richard Maine) wrote:

> There is no such thing as "an equation of a specified kind". Kind is not
> a property of a whole equation. That's actually im****tant and
> occasionally confuses people. Each individual entity in an expression
> has a type and kind, as does each operation result.

I apologize for not using the right language terms here, specifically
I should have said "an expression with a result with a specified kind"

I understand that:

real(sp) x
real(dp) y
y =3D x+1

x has the kind of sp (1 is an integer), y is of kind dp

the expression could also be written

y =3D real(x+real(1,sp),dp)

>
> "Same value for mutiple use of an integer". I think I know what you are
> asking there, but I'm having to guess, as the words don't quite parse.
> There is one variant of the question that is quite subtle... but that
> variant doesn't apply to integers. Repeat after me, small integer values
> are exact.

I think this is the answer I was looking for.

>
> Also, I don't know what distinction this is that you are making between
> "machine level" and the real kind.

I think my confusion has been in the conversion of lower rank number
(say single to double precision).  If it is not a "floating point
number" the decimal conversion of the bianary produces non-zeros in
the digits 8-15 places to the right of the decimal point.  I
understand now that these are not extra bits but conversion digits.

>
> Finally, I must note that you specifically asked about whole numbers. It
> really does make a difference. Whole numbers of modest magnitude are
> represented exactly in every kind on every compiler that you will ever
> see. There aren't going to be any extra bits (except for zero bits) or
> lost precision, no matter what kind you are talking about.

This is specifically what I was asking about (the use and practice of
coding whole numbers, I am NOT confusing with non-whole numbers) and
you have very clearly answered.  Thanks!

>
> I'm thinking you need to read Goldberg's "What Every Computer Scientist
> Should Know About Floating-Point Arithmetic"
> (<http://docs.sun.com/source/806-3568/ncg_goldberg.html>),
but I'm
> afraid your confusions might actually be more fundamental thanthat.
> Still, reading it is unlikely to hurt.
>

Groan!!! you did that on purpose! ;)  And yes, my confusions are far
more fundamental.

c