Re: Arithmetic with durations

[Simon Wright <simon@pogner.demon.co.uk>]

Larry Coon <lmcoon@home.com> writes:

>   total_real_time: constant duration := 1000.0;
>   total_sim_time: constant duration := 100.0;
>   real_delay_time: constant duration := 25.0;
>   sim_delay_time: duration;
[...]
>   -- Convert the delay from real time to simulation time
>   -- by multiplying it by the ratio of total simulation
>   -- time to total real time.  Here is where it gives the
>   -- error.
>   sim_delay_time := real_delay_time * total_sim_time / total_real_time;

If you write

  sim_delay_time :=
      real_delay_time * duration (total_sim_time / total_real_time); -- A

or

  sim_delay_time :=
      real_delay_time * duration'(total_sim_time / total_real_time); -- B

you should do better (works fine here, 3.09 on Linux).

The relevant section of the LRM appears to be 4.5.5(18) et seq:
===========
(18)   Multiplication and division between any two fixed point types are
provided by the following two predefined operators:

(19)
       function "*"(Left, Right : universal_fixed) return universal_fixed
       function "/"(Left, Right : universal_fixed) return universal_fixed

                                   Legality Rules

(20) The above two fixed-fixed multiplying operators shall not be used
in a context where the expected type for the result is itself
universal_fixed -- the context has to identify some other numeric type
to which the result is to be converted, either explicitly or
implicitly.
============
The AARM says:
============
20.a Discussion: The small of universal_fixed is infinitesimal; no
loss of precision is permitted.  However, fixed-fixed division is
impractical to implement when an exact result is required, and
multiplication will sometimes result in unanticipated overflows in
such circumstances, so we require an explicit conversion to be
inserted in expressions like A * B * C if A, B, and C are each of some
fixed point type.
============

I'm far from an expert in this area, we don't deal much with numbers,
but I think that my version (A) means "evaluate a universal_fixed
result and convert it to duration" while (B) means "evaluate as a
duration".

I suspect that choosing appropriate intermediate fixed-point types is
less than straightforward?? (I remember fun with fixed-point types in
Coral 66, our compiler was almost guaranteed to return 0 unless you
really understood what was happening)
 
-- 
Simon Wright                        Work Email: simon.j.wright@gecm.com
GEC-Marconi Radar & Defence Systems            Voice: +44(0)1705-701778
Command & Information Systems Divsion            FAX: +44(0)1705-701800