NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.

To learn about what's going on, see About the Archive Site.

RE: partially buried contours

Interesting suggestion, Craig, but to work on a point by
point basis it has to be implemented in the 3-D graphics
API. You may want to make your suggestion to the Java3D
developers.

Bill

On Tue, 7 Jan 2003 Craig_Barrington@xxxxxx wrote:

> Hi Bill and Donna,
>
> I'm just a casual observer; somewhat interested in Visad.
> I'm not sure who the developers are, or if they are "listening", but this is
> mainly to them.
>
> I'm just guessing that the logic that Visad uses to determine if the contour
> lines are visible needs to be changed to be much simpler as follows:
>
> Visad must have an algorithm that determines whether a point on the surface
> is visible.
> If that point is (would be) visible, then a contour line on the surface at
> that point is visible (instead of the point).  That is the basic idea.  i.e.
> The surface is drawn, and then the contours are drawn, using the same
> algorithm to decide on visibility as the surface itself.
>
> Well, as I said, that's the basic idea I wanted to suggest to the Visad
> developers.  It might save everyone a lot of futzing around with ad hoc
> hacks (Hey, how about "ad hacks"?).
>
> Craig Barrington
>
> Fluor Federal Services
> Richland, WA
> 509-373-0946
> Craig_Barrington@xxxxxx
>
>
> -----Original Message-----
> From: Bill Hibbard [mailto:billh@xxxxxxxxxxxxx]
> Sent: Tuesday, January 07, 2003 12:24 PM
> To: Donna L Gresh
> Cc: visad-list
> Subject: Re: partially buried contours
>
>
> Hi Donna,
>
> I did a couple experiments and found something you can do, but
> its a bit of a hack. You can raise the contours away from the
> surface, if you can decide on a fixed direction for them to be
> raised. Say you decide to raise the contours by 0.02 in the
> positive ZAxis direction. Ideally, you could construct a
> ConstantMap of 0.02 to ZAxisOffset and include it in the call
> to DisplayImpl.addReference(). However, the system doesn't
> seem to pick up the ConstantMap to ZAxisOffset. So instead,
> replace your ScalarMap to ZAxis with an identical ScalarMap
> to ZAxisOffset, and use a ConstantMap to ZAxis in place of
> the ConstantMap to ZAxisOffset. Ugly, but it worked in a test
> I ran.
>
> Note that for 2-D displays, the system adds an implicit offset
> to raise line graphics above images and trianlges.
>
> Good luck,
> Bill
>
> On Tue, 7 Jan 2003, Donna L Gresh wrote:
>
> > I am adding IsoContours over a surface in three-space (an analog to
> > discussion in tutorial 3.7 of "IsoContours over image") and find that the
> > contours are partially "buried' in the surface. It is more or less obvious
> > depending on viewing angle. In my days of using Data Explorer there was a
> > way to add "fuzz" to the contours to allow them to "float" a bit above the
> > background object. Is there something equivalent in VisAD? Apologies if
> > it's documented and I just didn't see it; I did look in the javadoc and
> the
> > discussion.
> >
> > Here's a picture to show what I mean:
> > (See attached file: buriedcontours.jpg)
> >
> > Donna L. Gresh, Ph.D.
> > IBM T.J. Watson Research Center
> > (914) 945-2472
> > http://www.research.ibm.com/people/g/donnagresh
> > gresh@xxxxxxxxxx
>
>


  • 2003 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the visad archives: