RE: Problem with Display.ZAxis Using Different Colors for same Z

  • To: "Kenneth Evans, Jr." <evans@xxxxxxxxxxx>
  • Subject: RE: Problem with Display.ZAxis Using Different Colors for same Z
  • From: Bill Hibbard <billh@xxxxxxxxxxxxx>
  • Date: Wed, 28 Jun 2006 07:21:56 -0500 (CDT)
Ken,

>     I tried that with 1, 2, 3 (and 100 ;-).  I could not see any difference.
> They all looked like the default.  I also tried 0, and it looked like
> gmc.setTextureEnable(false); that is, shaded.

I looked at the code in VisAD, and there is logic to
clamp CurvedSize to be at least 2 and no more than
the texture size / 32. And there is more recent logic
to split large textures (such as your 1000 x 1000)
into several smaller textures, which will affect this
clamping.

If you are energetic, you might put some print statements
into the buildCurvedTexture() method at the end of the
visad/ShadowFunctionOrSetType.java file, to print the
value of curved_size before and after the clamping statement:

  curved_size = Math.max(2, Math.min(curved_size, size / 32));

You might also print the value of size, which is the size
of the smaller texture patches.

Bill

> -----Original Message-----
> ...
> Yeah, for your 1000 x 1000 data set you are creating and
> rendering 2,000,000 triangles. The dirty look comes from
> the shading applied to the surface (as I recall, we
> disabled shading with texture maps).
>
> Tom Rink suggests that rather than disabling texture
> mapping, you might try:
>
>   GraphicsModeControl mode = display.getGraphicsModeControl();
>   mode.setCurvedSize(1);
>
> Or for a compromise, try 2 or 3 instead of 1. These
> will give you a texture map with finer sub-sampling
> (with 1, I think there is no sub-sampling at all).
>
> Good luck,
> Bill
>
>

==============================================================================
To unsubscribe visad, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html
==============================================================================


  • 2006 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the visad archives: