CGAL users discussion list

[Daniel Duque Campayo <>]

Dear Monique,

> I am interested in learning more about the way the problems arise in
> applications. What do you mean by "kernel for toroidal conditions"?

I just vaguely recall these conditions could be implemented at the kernel 
level, by redifining distance between two points. I might be wrong, and it 
certainly does not matter to me how the conditions are implemented: a valid 
triangulation would certainly do the trick.

About how the problem arises in my field (molecular simulation): it is very 
simple. One is usually interested in the behaviour of a huge number of 
molecules. Since that is impossible, one takes as many as one can handle, but 
what does one do with the "borders"? The usual answer is to endow the system 
with toroidal boundary conditions, so the borders are not there at all --- 
the system is still finite, but at least there are no borders to care about. 
These conditions are often called "periodic boundary conditions" in my field.
There is some info (not much, this is a growing project) in
http://www.sklogwiki.org/SklogWiki/index.php/Boundary_conditions

> http://www-sop.inria.fr/geometrica/team/Monique.Teillaud/other-geometries/

Nice. Thanks very much.

Best,

Daniel

-- 
Daniel Duque
http://rincon.uam.es/dir?cw=950067138671875