The API for Java 3D Collision support is mostly found in the behavior
wakeup condition documentation. The collision support in Java 3D was
intended to provide aynchronous notification of collision events. It can
handle multiple collision "triggers" with multiple collision "reports".
Inside Java 3D, we currently use a spacially organized tree structure for
all of our geometric operations. So, a collision test is fairly quick,
although it dependes upon how well the tree is balanced, how many objects
are moving, how many collision "triggers" are pending, etc...
It is fairly hard to predict how well the built in Java 3D collision system
will work for your application, because there are so many variations. In
general though, the Java 3D collision system was designed for an application
like the one you describe.
A lot of the negativity surrounding the Java 3D collision system stems
from the fact that most applications discussed here have not been able
to use an asychronous notification collision system. They needed a
collision avoidance system, which Java 3D does not currently supply in
the context of collision detection. Our picking API may be used in the
context of collision avoidance, and has been done by many applications
here.
Doug Twilleager
Sun Microsystems
>MIME-Version: 1.0
>Subject: Re: [JAVA3D] Java 3D's Collision Detection
>To: [EMAIL PROTECTED]
>
>Hi Justin & all others interested;
>
>Can't a person get the source for Java 3D & if so, what would be the
>rational for hiding what the collision detection algorithm does? I'm a
>robotics engineer and for my applications IT DOES MATTER what the collision
>detection algorithm does. At a given time I'll need to be able to check
>whether any of the moving entities in an environment are colliding with any
>other entities (moving or not moving) in the environment in an efficient and
>intelligent way. What I've read so far sounds more like the collision
>detection is oriented toward a single entity navigating the environment
>(although I qualify this by saying that I haven't spent any time looking at
>the actual API ... & I actually don't want to). Sun should let developers
>know (at least from a functional point of view) what the algorithm can and
>can't do. If it simply implements a brute force algorithm, or an algorithm
>that only optimizes for a single (or a few) objects that move in the
>environment it is simply useless for my application. If companies are to
>consider Java 3D seriously, we have to be given the information that will
>allow us not to waste a whole lot of time trying to settle basic questions.
>
>I guess from what I've read here that if I decide to use Java 3D for my
>application, I'll probably be stuck writing my own collision detection code.
>Has anybody actually looked at the Sun code for collision detection? ...Or
>is it not available?.
>
>That's all for now,
>
>Karen Dias,
>ISE
>
>
>-----Original Message-----
>From: Justin Couch [mailto:[EMAIL PROTECTED]]
>Sent: Tuesday, February 27, 2001 2:03 AM
>To: [EMAIL PROTECTED]
>Subject: Re: [JAVA3D] Java 3D's Collision Dectection
>
>
>Karen Dias wrote:
>
>> detection for complex worlds. I briefly looked on the net trying to
>> determine exactly what Java 3D does, didn't find anything significant.
>> Which collision detection algorithm does it implement and where can I read
>> about it?
>
>Sun's standard answer to this, and many other specification questions,
>is "we don't want to tell you because we might change it - and it
>doesn't really matter anyway". Naturally, many of us are very unhappy at
>this stance - there is not even a simple abstract architecuture document
>of how the API conceptually works. Very poor work IMNSHO.
>
>Basically, the general opinion of the list is that J3D's collision
>detection sucks for anything but trivial applications. Most people doing
>real work implement their own collision detection using the picking
>system.
>
>--
>Justin Couch Author, Java Hacker
>http://www.vlc.com.au/~justin/ Java 3D FAQ Maintainer
>http://www.j3d.org/ J3D.org The Java 3D Community Site
>-------------------------------------------------------------------
>"Humanism is dead. Animals think, feel; so do machines now.
>Neither man nor woman is the measure of all things. Every organism
>processes data according to its domain, its environment; you, with
>all your brains, would be useless in a mouse's universe..."
> - Greg Bear, Slant
>-------------------------------------------------------------------
>
>===========================================================================
>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".
