> > 1. Numerical Recipes is not a superior reference for numerical methods.
> > I have received several testimonials about defects therein. There are
> > parts of it that do work. See http://math.jpl.nasa.gov/nr
>
> Fully agree. Have had several poor and some very bad experiences with NR.
> People must learn to use NETLIB and GAMS first.
While there are some valid objections to NR, there are also some urban
legends, some rumours still floating about which have been corrected
years ago etc.
> > 2. Bulirsch-Stoer is not a superior method. Predictor-corrector methods
> > are superior, if evaluation of the derivative is expensive.
>
> Thank you for saying it loud and clear! This misconception (about
> Bulirsch-Stoer method) must be another one emerging from the NR book.
I once had an application where Bulirsch-Stoer turned out to be much
quicker than Runge-Kutta. It depends on the problem.
Numerical Recipes might be a good place to look first, especially since
one has the text and the code together, and all in one volume. Note
also that there is also a F90 version. However, it shouldn't be the
LAST place one looks as well.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|