Sorry for forcing you to lead me by the hand(nose) through this...

I'm much better at this stuff than I am at Perl- and I've benefited plenty from your posts over many years. One recommendation while you're doing your trial runs-- since they're so expensive, save all your F(thetas) and hash them later. Unless your fields are really pathological they're likely to come in handy later (unless the problem setup is such that you're always getting a new field geometry)