[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gnuplot-info-beta
Subject:    Re: tessalations
From:       Michael Cohen <mic109 () rsphy1 ! anu ! edu ! au>
Date:       2000-01-26 22:45:48
[Download RAW message or body]

On Thu, 27 Jan 2000, you wrote:
> This is, of course, a special case of what one might like to do
> with a color map, correct? In fact, I would probably argue that
> while it may be one of the more common applications, it's often
> one of the less useful. The reason is that coloring by z-value,
> or height above the x-y plane is essentially redundant information.

It would be redundant if you could see the surface in 3D, but on a computer
screen you can only see a 2D representation. This means that if there is no
color variation across the surface (i.e. the whole surface is the same color)
you can only see a sillouhette. This makes it very hard to get any depth
information. Besides a common usage of plotting is to obtain a final
"non-rotatable" print. (eps or whatever). One way to achieve this color
variation is by using color-dependant of z (colormap), another is by
superimposition a second function (ala texture mapping) on the surface, as you
suggested. Either methods _must_ be used to get a viewable filled polygon plot.
(Of course wireframe plots have the grid to give a depth cue). Colormap is most
common.

> So, (and I know this has been discussed before) if, in general,
> one could represent the color as some f(x,y) at each vertex, he
> might choose the simple case of f(x,y) = z, or he might choose
> some other function. Next, he should be able to decide on how
> to represent those colors on the graph--either average the values
> at each vertex (color-per-face), or interpolate between them.
> In the case of the latter, tesselation makes sense for all the
> reasons Michael eloquently describes.

Sounds good to me.

	Michael

[[[[ unsubscribe from info-gnuplot-beta via majordomo@dartmouth.edu ]]]]

[prev in list] [next in list] [prev in thread] [next in thread]