Yes. Thank you for the code. A very thorough set of calculations for the
pixel width using the various policies in your Canvas3d implementation.
Dave
-----Original Message-----
From: Kasparian, Raffi J. [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, June 12, 2001 1:47 PM
To: [EMAIL PROTECTED]
Subject: Re: [JAVA3D] Overlay Library: Implemented
Blaine, thanks for the code. With all the talk about overlays right now, I
don't understand why no one else has commented on it. Any way, I look
forward to playing with it in a couple of days.
Raffi
-----Original Message-----
From: Blaine Alex Bell [mailto:[EMAIL PROTECTED]]
Sent: Monday, June 11, 2001 4:25 PM
To: [EMAIL PROTECTED]
Subject: Re: [JAVA3D] Overlay Library: Implemented
Woops, forgot to put this test file with my last message. This program is
a simple example of how you can use the OverlayGroup. (simple
modification of HelloUniverse).
B
On Mon, 11 Jun 2001, Blaine Alex Bell wrote:
> Here is the Overlay Library that I have implemented. I have tried
> minimizing this implementation so it can be as flexible as possible.
> Two classes exist, both in the package edu.columbia.cs.cgui.j3d:
> FixedCanvas3D and OverlayGroup.
> FixedCanvas3D is interchangeable with Canvas3D, and NEEDS to be used if
> you want to use OverlayGroup. OverlayGroup allows you to place overlayed
> objects by having a Node that is in pixel coordinates (origin is in the
> bottom left hand corner of the Canvas), and gives you the
> flexibility to create a scenegraph below this group node to place the
> overlay objects.
>
> Some functionality that you might find useful:
>
> * Supports all View Model Parameters that are not part of head tracking.
> As noted in the documentation, you cannot currently do Overlays in a head
> tracked environment (Java3D limitation, hoping to be fixed in the near
> future).
>
> * Supports changing Z values for Overlays. This gets very interesting,
> especially in stereo since 2D overlays cannot do this. You do not need to
> use this functionality, you can also place it directly in front of the
> front clipping plane.
>
> * Works around the 'lag' problems introduced in the WakeUpOnElapsedFrame
> behavior because Canvas3D caches values from previous frame. Also, I
> provide new functions in FixedCanvas3D that should be in Canvas3D, such as
> transformations relative to the ViewPlatform.
>
> Please send be comments/suggestions. I know that this is a hot topic on
> this newsgroup, and I hope this helps everyone.
> Please let me know what you think!
> Soon coming: A View Model Demo Program to explain this crazy View Model!
>
> Blaine
>
> On Mon, 11 Jun 2001, Artur Biesiadowski wrote:
>
> > "Yazel, David J." wrote:
> >
> > > 17. No plans to implement resize because of the issus involved in
rebuilding
> > > the sub-textures. Would be easier to just make a new one. We might
support
> > > use of model-clips to clip in from a maximum size.
> >
> > I was trying to do overlay system some time ago and failed. Anyway, I've
> > come up with following interface - it might be an overkill if you don't
> > plan resize, but maybe you will like to reuse some idea.
> >
> > Overlay have specified xoffset, yoffset, width and height at creation
> > time. These are double values, with following meanings:
> >
> > xoffset -
> > if positive > 1, means distance of left edge of overlay from
left side
> > of window in pixels
> > if negative < -1, means distance of right edge of overlay from
right
> > side of window in pixels
> > if positive < 1, means how much free space should be used for
the left
> > side (fraction)
> > if negative > -1, means how much free space should be used for
the
> > right side (same as 1+negative)
> > yoffset - the same, just top and bottom
> >
> > width -
> > if positive > 1 means width of overlay in pixels
> > if negative < -1 means how much free width should be left on
screen
> > after substracting overlay width (in pixels)
> > if positive < 1 means width of overlay as fraction of entire
window
> > width
> > if negative > -1 means fraction of width which should be
unoccupied by
> > overlay (same as 1+negative)
> > height - same as width, just for height
> >
> > This would allow for very dynamic system, but unfortunately size of
> > texture would have to change with window resizes. It might be too much
> > for your needs, then please implement at least negative offsets - to
> > allow placing overlay relative to right and bottom edges (not only top
> > and left). Fractional offsets would be also nice (allows for example
> > centering with 0.5, 0.5 offsets).
> >
> > Artur
> >
> >
===========================================================================
> > 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".
> >
>
===========================================================================
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".
===========================================================================
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".
