From: Matt J on
"Steven_Lord" <slord(a)mathworks.com> wrote in message <i39n74$8p2$1(a)fred.mathworks.com>...

> > I think it's also doubtful that native MATLAB routines like lsqcurvefit
> > itself would be written to invoke Parallel Computing Toolbox features.
>
> Why _wouldn't_ they?
>
> http://www.mathworks.com/access/helpdesk/help/toolbox/optim/ug/briutqn-1.html
=================

It's good to learn that they do, in fact. I didn't expect it to be the case partly because of the inter-Toolbox management hassles it might mean for software developers. Partly also, I wondered about the complications of nested parfor loops (and indeed your link does show that it is a rather tricky issue).