hi,

someone may come up with something simpler, but one idea is to make
the Rubbersheet output into a 2D field with the Z heights as computed
by Rubbersheet as the data value, then you can use it in Map with your
2D scattered points and the dimensions will be fine.  then you can make
the scattered points into 3D and finally put them into Autoglyph.

(disclaimer: i didn't test this - my apologies if i've missed a step)

Rubbersheet(2d_data) - Mark(positions) - Compute([a.z]) - Replace(2d_data)

Compute outputs a field where the data component contains the Z values
generated by Rubbersheet, and they are inserted back as data into the
original 2d field which went into Rubbersheet.

then for your scattered points, assuming the positions are 2D (if not,
use Mark - Compute([a.x,a.y]) - Unmark to make them so) then:

Map(scattered, replace_output) - Mark(positions) - Compute([a.x, a.y, b]) - 
Unmark - Autoglyph            \ ---------------- /

the output of map contains your scattered points with the Z heights in the
data component, so you need to get that into the positions component.  i'm
not sure my ascii art will make sense - the inputs to Compute are first the
output of Mark, and second the output of Map.

nancy


> Hi!
>
> I need to map z-coordinates from a non-regular grid (output of
> rubbersheet) to the z-coordinates of an n-body (unconnected)
> set (to be displayed via autoglyph as spheres). I.e. I want to
> have the glyphs be vertically aligned with the rubbersheet surface.
>
> just doing
>
> rubbersheet - mark[positions] \
>                                 compute - unmark
> autoglyph - mark[positions]   /
>
> does not work (Bad parameter: inputs must be of the same length / input 1
> not matching the master (input 0)).
>
> Is there a way to extract (sample) the rubbersheet at the positions of
> the n-body set? I tried to use lookup for this purpose but did not
> succeed (well, because I do not have integer indexes for the lookup), also
> map does not work because rubber-sheet is 3d, but the connections are
> only 2d.
>
> Any Ideas? Richard.
>
> --
> Richard Guenther <[EMAIL PROTECTED]>
> WWW: http://www.anatom.uni-tuebingen.de/~richi/
> The GLAME Project: http://www.glame.de/

Reply via email to