home *** CD-ROM | disk | FTP | other *** search
- Path: newsbf02.news.aol.com!not-for-mail
- From: bigdumjock@aol.com (BigDumJock)
- Newsgroups: comp.lang.c++
- Subject: Re: Comparison M++ v. math.h++ ?
- Date: 14 Jan 1996 15:48:43 -0500
- Organization: America Online, Inc. (1-800-827-6364)
- Sender: root@newsbf02.news.aol.com
- Message-ID: <4dbq7b$3gq@newsbf02.news.aol.com>
- References: <4cq5am$ac4@metz.une.edu.au>
- Reply-To: bigdumjock@aol.com (BigDumJock)
- NNTP-Posting-Host: newsbf02.mail.aol.com
-
- > Does anyone know any reviews comparing the features of the matrix-based
- > math class libraries -
- > M++ from Dyad Software
- > Math.H++ from Rogue Wave.
-
-
- Check out the Jan '96 issue of The C++ Report. Cay Horstmann reviews M++,
- Math.H++ and Visual Numerics' IMSL Math Module.
-
- The results are mixed -- no single library stands out, if I may interpret
- his conclusions. Math.h++ has a solid implementation, but is not as
- comprehensive as the other two. M++ has relatively complicated class
- structures for arrays, matrics & vectors, but has useful functions for
- signal processing applications in particular. And the IMSL library is
- comprehensive, but its implementation is poor (C++ code apparently does
- some no-nos like undocumented aliasing, no implicit conversions between
- matrix types, and has the potential for ambiguous overloading).
-