Specialization for CGAL_Homogeneous<integer> with floating-point filters added

This commit is contained in:
Sven Schönherr 1997-04-23 12:52:59 +00:00
parent d74d4c25c8
commit 7f6309d71f
2 changed files with 36 additions and 16 deletions

View File

@ -2,7 +2,7 @@
% The CGAL Reference Manual
% Section: 2D Smallest Enclosing Circle
% -----------------------------------------------------------------------------
% file : spec/Min_circle_2.tex
% file : Library/spec/Min_circle_2.tex
% author: Bernd Gärtner, Sven Schönherr (sven@inf.fu-berlin.de)
% $Id$
% =============================================================================
@ -201,6 +201,16 @@ linear time, but substantially less than computing the new smallest
enclosing circle from scratch. For the member function \ccStyle{reserve}
see the container \ccStyle{vector} from STL~\cite{STL}.
We provide a specialization with homogeneous representation and
numbertype \ccStyle{integer} that uses floating-point filter for the
sign calculations in the predicates (see~\cite{MehlhornNaeher94}).
This decreaes computation time by a factor of four compared to the
generic implementation. If both \ccStyle{<CGAL/Integer.h>} and
\ccStyle{<CGAL/Homogeneous.h>} are included before
\ccStyle{CGAL/Min_circle_2.h>} then the specialization for
\ccStyle{CGAL_Min_circle_2<CGAL_Homogeneous<integer>>} is used instead
of the generic implementation.
\ccExample

View File

@ -2,7 +2,7 @@
% The CGAL Reference Manual
% Section: 2D Smallest Enclosing Circle
% -----------------------------------------------------------------------------
% file : spec/Min_circle_2.tex
% file : Library/spec/Min_circle_2.tex
% author: Bernd Gärtner, Sven Schönherr (sven@inf.fu-berlin.de)
% $Id$
% =============================================================================
@ -201,6 +201,16 @@ linear time, but substantially less than computing the new smallest
enclosing circle from scratch. For the member function \ccStyle{reserve}
see the container \ccStyle{vector} from STL~\cite{STL}.
We provide a specialization with homogeneous representation and
numbertype \ccStyle{integer} that uses floating-point filter for the
sign calculations in the predicates (see~\cite{MehlhornNaeher94}).
This decreaes computation time by a factor of four compared to the
generic implementation. If both \ccStyle{<CGAL/Integer.h>} and
\ccStyle{<CGAL/Homogeneous.h>} are included before
\ccStyle{CGAL/Min_circle_2.h>} then the specialization for
\ccStyle{CGAL_Min_circle_2<CGAL_Homogeneous<integer>>} is used instead
of the generic implementation.
\ccExample