On 5/24/10, Leo Butler <l.butler at ed.ac.uk> wrote:
> < Here's yet another idea: let g(x, y) := if numberp(x) and numberp(y)
> < then f(x, y) else 'g(x, y) and then try plot3d(abs(g(x, y)), ...).
> < Is that any improvement?
>
> Based on a trace of g, it appears that this puts plot3d into an infinite
> loop. I'm inclined to think that there are at least 2 issues here.
Yeah, that wasn't such a good solution, although, to be fair,
an ordinary user could hardly be expected to know.
The problem is that plot3d calls the function with symbolic
arguments once before going into numerical calculations
as part of an attempt to determine if the function to be
plotted depends on any variables other than the ones specified
in the plot options (or something like that). Maybe that isn't
really necessary.
best
Robert Dodier