Re: Fixed vs float and precision and conversions

["Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>]

On 08/07/2020 20:10, Björn Lundin wrote:
> Den 2020-07-08 kl. 20:08, skrev Björn Lundin:
> 
> The main reason for asking was to see if I got the whole concept
> of fixed types wrong or not.

Fixed-point is conceptually a scaled integer. You should deal with it 
accordingly. [It could be a bit surprising in Ada where conversion to 
integer rounds. In most languages conversion to integer truncates]

> I did expect
> 'You should do this or that one-liner' as Niklas proposed.
> I did not get that to work though

What do you mean it does not work? Of course T'Round works for 
fixed-point types. Simplifying it for positive numbers only T'Round is 
roughly same as:

    T (X + T'Small / 2.0)

As a side note. Consider the problem of splitting an interval into 
same-width subintervals. If each subinterval is denoted by its center, 
you have a problem with the margins of the whole interval as they do not 
denote any center. Furthermore if zero is included that again poses a 
question whether it should be in a center of a subinterval (which is 
then both negative and positive!) or at the margin of two subintervals 
(and thus not represented as a center!). It is a difficult problem which 
does not have universal or even intuitive solution. Thus do not expect 
one-liners.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de