isolating a special case

Dushan Mitrovich dushanm at spinn.net
Wed Dec 5 21:26:34 CST 2007


What is the best way to isolate a special case?  I have a large collec-
tion of surface patches contributing fields at an observation point.
But that obs. point may be at one of the patches, in which case that
patch's contribution has an analytic expression (otherwise it would be
singular).  The results are put into a matrix whose indices represent
the surface.  I know how to this using the approach

    fieldmatrix = (range<0.1)*selffield + (0.1<=range)*otherfield

where 'range' is the same size matrix as 'fieldmatrix', giving the
distances between patch and obs. point.  But then I'd have to compute
'selffield' in all the positions where it's not needed.  Is there a
more efficient way of doing this?

- Dushan Mitrovich



More information about the Help-octave mailing list