Egads. OK, our parent company MathEngine has a collision library which is in
pre-release right now. MathEngine is still open for
ideas of what to do with this library ( so make suggestions! ).
I agree that I think that the collision library we're using is extremely cool but of
course it may not be suitable for some people
who want something that is completely free. Alternatively it should be possible to
take RAPID and make a Java version of it - I
know it has been done by somebody. I also understand that the J3D team is looking at
implementing more accurate collision.
In any case, please send me mail if you'd like a full copy of the collision library.
This includes documentation and examples and
serves as a better base than the version integrated directly into our game. I don't
want to misrepresent the two years of effort
these people put into the collision library by having you use the version in the game
as your sole reference. I'll pass this onto
the people who own that effort and hopefully they can help you out.
Briefly the collision system lets you define collidable surfaces, either polygonal or
based on primitives. You submit those
collidables to a "set" and the job of the collision engine is to maintain pairwise
relationships between two "sets" of potential
collidants. It turns out that properties like friction etcetera only exist in
pairwise relationships and are not uniquely
properties of either collidable. Also it turns out that it is inconvenient to
directly express which pairs collide, that in fact
one wants a level of abstraction to express that certain "types" of objects collide.
In this way objects can come and go from a
simulation environment and participate in appropriate collisions. Finally, an
important part of the work in defining collision in
combination with dynamics is to provide the right kind of information back to
dynamics. In our case it isn't good enough to know
that collisions occur, or even where on the polygon surface they occur, we also need
to know the collision depth. We also had
certain expectations such as that the collision library would exploit the interframe
coherence present in our simulation.
In my mind I see three important factors in a product - the visualization, the
dynamics and the collision. These are orthogonal to
each other and form a space within which other parts of the project can be arranged.
Cheers!
- Andy
-----Original Message-----
From: Robert Vidar Bjarnason <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Wednesday, November 10, 1999 2:05 PM
Subject: Re: [JAVA3D] Fast collision detection routine
>Hey,
>
>Sunny Leung wrote:
>> Are there any robust and efficient collision detection
>> routines (such as VClip, SOLID, etc) written for the
>> Java 3D environment?
>
>Check out MathEngine @ http://www.mathengine.com/ and
>http://www.openmediagroup.com/.
>
>Super impressive collision, gravity and physical
>simulations with good J3D support and Open Source
>examples.
>
>Warm regards,
>Robert Vidar Bjarnason
>[EMAIL PROTECTED]
>
>===========================================================================
>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".
