Re: [boost] math constants, including their intervals

From: Guillaume Melquiond (gmelquio_at_[hidden])
Date: 2002-10-03 15:22:36

On Thu, 3 Oct 2002, Paul A. Bristow wrote:

> I have updated the documentation on math constants to include especially
> information on the smallest intervals containing these constants so as to
> fit in with the interval package recently reviewed.
> 
> URL to Boost files section attached.
> 
> I am still working on producing a few sample constants for review.
> 
> Paul

Your ``paper'' is quite interesting. And since you did understand quite
well the problems we encountered in designing the Interval library, I'm
eager to use such a math constants library.

I would just like to say that your paragraph 'Should calculations be
exactly the same on all compilers and platforms?' is a bit wrong. Indeed,
using the precision bits of the control word of the x86 to ensure the
extended double precision format isn't used is only a partial solution,
because these bits reduce the precision of the mantissa, but not the
precision of the exponent. So the only complete solution is to force the
intermediate results in memory (it's the method used by the Interval
library).

Guillaume

PS: Starting with this same paragraph, there is a lot of problems with the
links in your html file. They don't stop, they continue till the end of
the text blocks. My navigator is Mozilla-based, so a lot of people may
encounter the same problem.