Wow, what a great response! Thanks Guys. I'll give this new implementation a shot.
I guess it might make sense longer term to plug into the netlib implementation by Cody: this does seem to be used as a standard. However, there may be all sorts of difficult knock on effects from that! I'll let you know if I come up with any issues. Neil On 4 March 2010 06:30, Søren Hauberg <[email protected]> wrote: > ons, 03 03 2010 kl. 18:28 +0000, skrev Neil Lawrence: >> The implementation of erfcx in the octave-specfun package is simply >> >> >> exp(x^2)*erfc(x) > > Could you try the attached implementation instead? It seems to behave > like the Matlab implementation, but as I don't have a real example to > work on, I have just tried some arbitrary numbers as my test set. > > If it works for you, I'll commit it to the 'specfun' package. > > Thanks, > Søren > ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Octave-dev mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/octave-dev
