[Cfp-interest 2787] Re: about FLT_TRUE_MIN and DBL_TRUE_MIN in 5.2.4.2.2p29 example

[Vincent Lefevre]

On 2023-06-27 18:50:46 -0700, Jim Thomas wrote:
> > On Apr 4, 2023, at 5:03 AM, Vincent Lefevre <vincent at vinc17.net> wrote:
> > 
> > N3096 5.2.4.2.2p29 is an example with FLT_TRUE_MIN and DBL_TRUE_MIN
> > definitions, in particular:
> > 
> >  FLT_TRUE_MIN  1.40129846E-45F // decimal constant
> >  DBL_TRUE_MIN  4.9406564584124654E-324 // decimal constant
> > 
> > But since these are the minimum positive subnormals, one could use
> > only one decimal digit.
> 
> Why? These are decimal constant intended to represent minimum
> positive subnormals in binary types.

Using numbers with 1 decimal digit only would also yield the
expected subnormal value, and this would be shorter.

-- 
Vincent Lefèvre <vincent at vinc17.net> - Web: <https://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)