I agree that overlays are very useful but more difficult than they
perhaps should be to implement in Java 3D (along with several other UI
functionality). If you are an overlay afficianado then the ideal
solution would be to simply designate an object as only being
displayable in overlay through the API, regardless of whether the
object is positioned relative to the view space (display overlay) or
remains positioned and oriented as if in the world (world overlay).
(Taking this one step farther, it would be nice to also designate that
an object would only be rendered in a specific view in support of
multiple displays, whether or not the objects are overlaid). Until
then, the complexity of dealing with overlays depends a whole lot on
how much of the problem you need to deal with.
The framework described in the book addresses a number of the more
subtle issues that arise when you start dealing with pseudo-overlay
(i.e. geometric-only forms of overlay involving scaling down the
overlaid object and moving it close to the view, which makes world
overlay very easy). Such issues include: Z overlap order of overlaid
objects, absolute and relative positioning of display objects in the
display space, size independence relative to display size (useful for
keeping text readable), and several subtleties required to do all this
other stuff while keeping overlaid objects small and close to the
view, which keeps them away from world objects and makes multiple
views more usable.
For simple applications all of this may be overkill. For more
sophisticated applications it may not be enough. Using direct overlay
(graphics2d) avoids some of the problems of pseudo-overlay, but it
also makes doing world overlay a bit more complicated. Even with
direct overlay, however, many of the issues described above remain
because they are inherent to overlay in general, and not to the
specific technique used to perform it.
--jon
> Date: Mon, 21 Aug 2000 08:21:39 -0400
> From: David <[EMAIL PROTECTED]>
> Subject: Re: Building Overlays, lets figure this out :)
>
> Yes I purchased that book and found it interesting. Yes it is possible to
> build geometry and keep it within view and facing the user, but its really
> like using a hammer when a screwdriver is called for. The question is
> whether or not the graphics2d interface is working as designed. Everyone
> seems to indicate that they cannot get it to work with acceptable
> performance, but my understanding was that this was Sun's answer to the
> overlay problem. In fact in reading through the e-mail archive there are
> several instances where sun's engineers responded that there would be
> overlay support. In fact I read somewhere on their site a list of features
> for 1.2 included support for overlays. I am assuming that this means the
> graphics2d handle.
>
> So the options are:
> 1-Overlays should be coded using the graphcis2d interface handle
> 2-Overlay support has not yet been implemented
> 3-Overlay support is handled by attaching geometry to the view side of the
> scene graph
>
> If the answer is 1, then it is either broken or I am using it incorrectly.
>
> Dave Yazel
--
__________________ JMB and Associates, Inc. ___________________
Jon Barrilleaux 3800 Lake Shore Ave. Consulting for 3D
[EMAIL PROTECTED] Oakland, CA 94610 Web Applications
510.444.4370 voc & Tech Industry
510.444.0231 fax www.jmbaai.com Management
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff JAVA3D-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
