I am wondering why the developers of CGAL would make such an
assumption if it were indeed useful to be able to change the values of
the final row to something other than 0. So my question to Len is what
is your transformation trying to do?
Basically I am trying to increase the scale of the object as Z
increases, similarly to a perspective transformation. The
transformation is not affine, but i see no reason why the CGAL library
shouldn't let you do arbitrary transformations. I am guessing they are
assuming the typical use case is affine transformations only.
On Thu, May 12, 2011 at 9:13 PM, Giles Bathgate
<[hidden email]> wrote: