From: mjs@hubcap.clemson.edu (M. J. Saltzman)
Newsgroups: sci.math
Subject: Re: >>n-dimensional Convex Hull algorithm?<<
Date: 5 Aug 96 15:27:29 GMT
mcg@mcg.math.waikato.ac.nz (Martin C. Glanvill) writes:
|Hi, I'm in (urgent) need of some C/C++ source for a n-dimensional Convex Hull
|algorithm.
|Failing n-dimensional Convex Hull alg, does anyone know of the existence of
|a C/C++ alg for 4-D Convex Hull generation.
|Any pointers or direction to literature also gladly accepted... I have three
|different algorithms for 3-D Convex Hull algs, but I'd rather not re-invent
|the wheel if n-dimensional algs already exist (and work).
Check out the Linear Programming FAQ at
http://www.skypoint.com/subscribers/ashbury/linear-programming-faq.html
--
Matthew Saltzman
Clemson University Math Sciences
mjs@clemson.edu
==============================================================================
From: orourke@grendel.csc.smith.edu (Joseph O'Rourke)
Newsgroups: sci.math,comp.graphics.algorithms
Subject: Re: >>n-dimensional Convex Hull algorithm?<<
Date: 6 Aug 1996 00:55:21 GMT
In article <4u3svr$ss2@thebes.waikato.ac.nz>,
Martin C. Glanvill wrote:
>Hi, I'm in (urgent) need of some C/C++ source for a n-dimensional Convex Hull
>algorithm.[...]
>Any pointers or direction to literature also gladly accepted... I have three
>different algorithms for 3-D Convex Hull algs, but I'd rather not re-invent
>the wheel if n-dimensional algs already exist (and work).
They exist and work and are available. See the comp.graphics.algorithms
FAQ. Re-inventing the wheel in this case would take you six months at the
minimum, so check out the several pieces of code via
http://www.geom.umn.edu/software/cglist
If you like what you find, complain to the US NSF for pulling the plug
on the Geometry Center.
==============================================================================
From: antoine.gautier@fsa.ulaval.ca (Antoine Gautier)
Newsgroups: sci.math
Subject: Re: Matrix Problem. HELP
Date: 31 Aug 1996 22:38:49 GMT
whebert@digital.net (Wyck Hebert) wrote:
> I need help with the following problem...
>
> "Find an N-by-N matrix of 0's and 1's with specified row, column, and
diagonal
> sums."
You can translate this into the set of constraints of an integer
linear problem (no objective function) and use any available code.
Expect to solve problems up to 50x50 in reasonnable time; In general,
your problem is NP-complete, unless some special structure can be
identified.
For more on integer problems, see e.g.
http://www.skypoint.com/~ashbury/linear-programming-faq.html
-------------------------------------------------------
Antoine Gautier (antoine.gautier@fsa.ulaval.ca)
-------------------------------------------------------
Professeur, De'partement OSD
Faculte' des Sciences de l'Administration
Universite' Laval
http://www.fsa.ulaval.ca/personnel/gautiera/