RE: How to populate a set of isosurfaces by giving a set of (x,y,z,value)

On Thu, 26 Sep 2002, Chien-Ting Chen wrote:

> I merge all volumes into one volume ( (x,y,z)->(val1, val2, val3, ...
> valN) ), but
> each value still shelter others in an order


Iso-surfaces aren't a good approach for this. I recommend
resampling your data to a FlatField with a Linear3DSet for
its domain, then using volume rendering. Multiple nested
transparent iso-surfaces look bad, even when all transparency
bugs have been fixed.

Good luck,

> -----Original Message-----
> From: owner-visad-list@xxxxxxxxxxxxx
> [mailto:owner-visad-list@xxxxxxxxxxxxx]On Behalf Of Curtis Rueden
> Sent: Wednesday, September 25, 2002 4:00 PM
> To: cchen@xxxxxxxxxx
> Cc: visad-list@xxxxxxxxxxxxx
> Subject: RE: How to populate a set of isosurfaces by giving a set of
> (x,y,z,value)
> Hi Tim,
> This is probably a manifestation of the Java3D transparency bug.
> Is there any way you can combine your volume data into one volume?
> If the volumes have the same domain set, you could try averaging
> the range values at each sample.
> -Curtis
> At 05:18 PM 9/25/2002, you wrote:
> >Thanks for the help!!!
> >
> >I encountered the other problem that when I display a set of volumes (
> >(x,y,z)->(concentration) ) and
> >map the "concentration" to Display.RGBA. To make a "foggy" effect in low
> >concentration values,
> >I set the RGBA table for each volume. When I displayed multiple sets (
> >multiple maps of RGBA ),
> >the transparency didn't work well:
> >    Transparent area in a set shelters other other volume set data. It
> looks
> >like a front invisible object cull out a back object,
> >but I exepct it should not cull out any object in the back if it is
> >"transparent".
> >
> >Do you have any solution for it?
> >
> >Thanks again,
> >
> >Tim