Re: math function accuracy -- was Re: text2interval again /
On 2013-03-12 07:20:40 -0500, Ralph Baker Kearfott wrote:
> Vincent,
>
> On 03/11/2013 08:16 PM, Vincent Lefevre wrote:
> >Also note that a P1788 implementation can provide several versions
> >of a function, e.g. one tightest, one accurate and one valid. So
> >depending on his speed/accuracy compromise, a user could choose
> >the best version for his application.
>
> When did we decide that, and where is it stated?
I think this was the idea between the different accuracy modes.
AFAIK, the standard doesn't say that there shall be a unique version
of each function. Anyway an implementation could provide 2 functions
sin_accurate (corresponding to the requirement of the standard) and
sin_valid as an implementation-defined function.
--
Vincent Lefèvre <vincent@xxxxxxxxxx> - Web: <http://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)