Donna L Gresh:
|> How are Screen objects intended to work in DX? In particular, I'm
|> interested in what SCREEN_WORLD and SCREEN_STATIONARY Screen objects are
|> supposed to do. The docs don't describe this in detail.
|> you're right it's confusing.
|
|I believe that for most applications, what you want is SCREEN_STATIONARY.
|That's the one that glues things to the screen (like captions and colorbars
|for example). SCREEN_WORLD is more appropriate for text glyphs, which are
|supposed to sit at a particular spot in 3D space, but the text is supposed
|to always face the viewer.
Thanks for the clarification, Donna. I have two short follow-ups:
a) how should occlusion work between WORLD screen objects and other
objects, and
b) should WORLD screen objects scale with the world or stay a fixed
size on the screen?
Thanks,
Randy
P.S. In more detail...
What I'm really looking for is a way to put "markers" (think billboards) in
the dataset at world locations, markers which always rotate to face the
viewer but which scale with the data and can be occluded by the data. That
is, the signs appear to be "in the data": as you zoom in the signs get
bigger, but as you rotate around the signs always face you.
This sounds like SCREEN_WORLD Screen objects, except:
1) occlusion results between WORLD Screen objects and other world
objects are inconsistent (see my dx-users post)
2) WORLD Screen objects are not only rotated to face the viewer, they
are rescaled to maintain a fixed size on the screen
I suppose my questions could be summed up as follows: are WORLD Screen
objects really "in the world", or should they just look like fixed-size
post-it notes stuck on the screen at the right spots over the data. I can
see uses for both.
--
Randall Hopper (mailto:[EMAIL PROTECTED])
Lockheed Martin Operation Support
EPA Scientific Visualization Center
US EPA N127-01; RTP, NC 27711
