Re: printing floats, bigfloats, fortran floats



>
> >> I think the problem is with the printing. Format controls are
> >> wonderful for producing human-readable syntax, but many controls such
> >> as ~g do not necessarily preserve print/read consistency.

Currently (and previously), printing of floating-point numbers is
> controlled by fpprec, so we won't have read/print consistency.


The key point here is "human-readable syntax". Ordinary 2d display in Maxima
has no reason to use re-readable numbers, since there is no reader for that
form! Where this matters is in file printing either in Lisp format or in
string format (e.g. string(1.0/3)).

fpprec is controlling many quite different things, and needs to be
rethought. Even in the case of bfloats, it isn't quite the right thing. Just
because you're calculating to 30 decimal figures doesn't mean you want to
see 30 decimal figures. And even bfloat doesn't have a decimal output form
which guarantees re-readability (though of course the internal Lisp form
works....)

Sorry for the rushed notes....

-s