improvement vander() routine

[Christoph Dalitz]

Dear Octave maintainers,

while palying with the vander() routine, I have observed that its code
(octave 2.1.35) uses two loops for the costruction of the matrix.

The attached code myvander.m cuts down the construction time considerably,
as it only uses one loop:

        octave> t=cputime(); vander(0.9*ones(1,100)); t=cputime()-t
        t = 0.49000
        octave> t=cputime(); myvander(0.9*ones(1,100)); t=cputime()-t
        t = 0.01000

Maybe this could be included in a future octave version.

Christoph Dalitz

PS:
If this is not the right list for sending patches, please let me know
where to send it instead.

myvander.m
Description: Binary data