RE: [boost] number<64> proposal

From: Paul A. Bristow (boost_at_[hidden])
Date: 2003-08-20 10:20:55

You might also like to look at the highly refined NTL by Victor shoup at
www.shoup.net/ntl

This has been used to calculate very accurate math constants, for example.

And provides a fairly simple way of providing both C++ quad (128-bit) and
arbitrarily higher precision too.

However I suspect that the requirement for 'infinite' precision FP is limited
and that this and/or other packages will meet most peoples rather special needs.

Boostification is not a trivial task I have discovered :-(

Paul

Paul A Bristow, Prizet Farmhouse, Kendal, Cumbria, LA8 8AB  UK
[phone]   Mobile [phone]
mailto:pbristow_at_[hidden]

| -----Original Message-----
| From: boost-bounces_at_[hidden]
| [mailto:boost-bounces_at_[hidden]]On Behalf Of Philippe A. Bouchard
| Sent: Tuesday, August 19, 2003 12:14 PM
| To: boost_at_[hidden]
| Subject: [boost] number<64> proposal
|
|
| Hi there,
|
|     Like I mentionned before, it would be great for Boost to have an
| infinite precision floating point number.  You can see an example here:
| http://members.lycos.co.uk/keithmbriggs/doubledouble.html.  Of course, the
| doubles could be replaced by long doubles, a muldiv() could be easily added,
| the exponents could be increased as well, etc.
|
|     Maybe someone can ask how to calculate trigonometric functions with less
| precise versions and so on.  It would be really great for Boost to have this
| user-defined precision because I am pretty sure research centers will be
| interested.
|
|
|
| Philippe
|
|
|
| _______________________________________________
| Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
|
|