On Sunday, October 16, 2016 at 5:34:51 AM UTC-4, Emmanuel Charpentier wrote:
>
> Playing fast and loose with plot3d's options, a bit of brute-force
> experimentation (see enclosed PDF, the corresponding Jupyter notebook is
> way too large to be enclosed), showed that, among other undocumented
> features :
>
>
> - plot3d accepts undefined value of the function when one returns
> float("NaN") at those undefined points ;
> - plot3d accepts a tuple (coloring_function, colormap) as a color=
> argument, as documented for (some) 2D-plotting functions.
>
>
> This allows, among others to create (approximate) representations of
> complex functions. The price to pay is an atrociously slow initial
> rendering (further renderings of the same object, even transformed), are
> acceptably fast).
>
> Note that those tricks are incompatible with the adaptive=True option.
>
> This leads me to think that the scene-creation algorithm could benefit of
> a slight revision to take these features in account, and I'm seeking advice
> about that.
>
> In the interim, this post might help people looking for workarounds around
> partially undefined functions and/or coloring a 3D map.
>
> Shouldn't these features be somehow documented ?
>
>
Yes, if we knew about them! That surprises me a bit but I think it means
the 3d plotting code is more robust than we sometimes give it credit for.
--
You received this message because you are subscribed to the Google Groups
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
