Re: Implement the "unum" representation in D ?

[Francesco Cattoglio]

On Friday, 21 February 2014 at 19:12:39 UTC, Frustrated wrote:
Simply not true. Maple, for example, uses constants and can
compute using constants.
You are mixing symbolic calculus and numerical computations. The 
two are completely unrelated.
Basically the benefit of this(and potential) outweigh the cost 
of
1 bit out of 64-bits.
Unless I'm completely mistaken, what you are proposing is 
basically creating a tagged union type. It's pretty much 
unrelated from the "unum" proposed by the PDF.