----- Original Message -----
From: "Carl Smotricz" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, August 29, 2000 12:03 PM
Subject: [JAVA3D] Windowed Canvas3D Performance?
> David Y. recently said that one of the reasons he dislikes showing a
> Canvas3D in a subpanel of a Swing GUI is that the 3D graphics would end up
in a
> window, which he believes causes a performance hit.
>
> I've pondered this for days before realizing that I wouldn't know how to
> display 3D in anything but a window. As a Component, don't I need to add
> Canvas3D to a Frame? And doesn't a Frame always have a title bar and a
border?
> And doesn't this mean my 3D canvas is not occupying the full screen and
thus
> not in full control of all screen space?
>
> So: Is there a way to display Canvas3D over the *entire* screen? How?
>
> Does anyone know for sure if there is a performance penalty for not
> grabbing the whole screen? If so, is there a way to avoid this penalty in
Java3D?
>
> Thank you for any enlightenment,
I'm pondering a corollary problem associated with simulation clocks. Does
my simulation 1.) use the j3d clock and work around (or through) the
glitches caused when the processor is busy doing high-priority windows
services or 2.) do I keep my own (smooth) simulation clock and just delay it
(rather than jump it forward as in (1.)) when the processor is off doing
other things?
If there IS a way to post a Canvas3D to the full screen, does that imply
that I can trap keyboard and mouse events myself? Would this mean that I
could use approach (1.) (which is much easier because
I'm certainly left with garbage-collection glitching, but that might be
easier to deal with than everything together.
Cheers,
Fred Klingener
Brock Engineering
===========================================================================
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".
