[R] shading in image()

[Angel]

The rgl package offers 3-D rendering (and Gouraud shading) capabilities
through an OpenGL.
I don't know why it is not in cran, I found it at:
http://wsopuppenkiste.wiso.uni-goettingen.de/~dadler/rgl/
In any case that's not what I wanted (although I still have to explore in
more depth this package).
My question was (naive me!) more simple! I was just thinking of an
interpolation function in R, maybe not very efficient but it'd do the job
for me.
Although the idea of linking blender is nice, it might be "too
sophisticated"  for simple things in 2-D.
Thanks,
Angel

----- Original Message -----
From: "Ross Ihaka" <ihaka at r-project.org>
To: "Barry Rowlingson" <B.Rowlingson at lancaster.ac.uk>
Cc: <r-help at stat.math.ethz.ch>; "Angel" <angel_lul at hotmail.com>
Sent: Monday, August 04, 2003 9:52 PM
Subject: Re: [R] shading in image()


> Barry Rowlingson wrote:
> > Ross Ihaka wrote:
> >
> >> If anyone knows how to Gouraud, Phong or other smooth shading method
> >> portably in a vector system I'd be keen to hear about it.
> >>
> >
> >  Why bloat R with something like that? All we need is a way of linking R
> > with Blender and we have a fully-GPL statistics and 3-d rendering
> > facility. Blender is scriptable in python, so you could read data files
> > from Blender, or possibly one could write an R graphics device that
> > produced Blender-compatible data files. Blender
> >
> >  Eventually you'll be able to walk through your data in 3-d....
>
> Its actually not that big, and it's useful for things other than 3d.
> Interpolation could be used for things like gradient colour fills for
> plot backgrunds or bars in barplots, drawing light spectra, etc.
>
> I've much about with some of this using existing primitives, and its
> quite hard work to get the effects I want.
>
> --
> Ross Ihaka                         Email:  ihaka at r-project.org
> The R Project and R Foundation
>
>