On Wed, 2005-19-10 at 18:10 -0600, Robert Dodier wrote:
> Robert,
>
> I find the following (adapted from your example) is executed without
> complaint, and yields a reasonable-looking graphic in Maxima 5.9.2 (Windows).
>
> plot2d ([%e^x, x, [parametric, t, log(t), [t, 1e-10, 10]]], [x, -10,
> 10], [y, -10, 10], [gnuplot_preamble, "set size ratio -1"], [nticks,
> 100]);
Of course, I should have thought of ensuring that t does not start at 0.
I am too use to Mathematica which must simply skip over points where the
function is undefined in parametric plots.
> notes:
>
> (1) I don't know of a way to plot functions with different ranges
> without resorting to the parametric option. That seems like a generally
> useful capability, maybe we should consider adding it.
Being able to use different domains in one plot2d command would be handy
for piecwise functions and for plotting things like the square root of x
which plots as 0 for x negative, presumably the real part is plotted? Is
there a way to plot only real values as opposed to the real part?
Setting realonly to true does not help.
> (2) Gnuplot has to be told explicitly to make the aspect ratio equal to 1.
> That seems like a misfeature to me. Rats.
I don't mind having to tell it to use an aspect ratio of 1 but it would
be nice if it was a simpler looking option for the plot2d command.
> (3) Parametric plotting doesn't exercise the adaptive sampling algorithm
> (although it should) so I increased nticks explicitly.
Yes, that is unfortunate, adaptive sample for parametric plots would be
nice.
Thanks for the help Robert, looks good in Linux also.
--
Dr. Robert J. Jerrard, Professor of Mathematics,
Concordia University College of Alberta,
7128 Ada Blvd., Edmonton, Alberta, T5B 4E4, Canada.
Phone: (780) 479-9291, Fax: (780) 474-1933.