Fergus Murray wrote:
> What I'm doing involves a Java program (applet, for preference) controlling
> a bunch of 3D figures as per a stream coming in - in its current
> incarnation,
[snip]
> should be easier than a lower-level language to use for humanoid animation.
This sounds like you really should be able to check out the H-anim spec
work for VRML as this should do everything you need. I believe under the
latest set of work they have a bunch of canned behaviours and morphs
between behaviors (ie runninng, walking etc).
> My final deadline for this project is mid-September, so I have two months
> to completely finish it - but that includes a certain amount of writing
> about it, so it would be good to get most of the technical work out of the
> way a while before then.
>
> Do you think it's likely that the Xj3D work will resolve all (or most) of
> the problems associated with the EAI? That is - will it work properly with
> a Java2 VM? Will it be faster than what we have now? Will it stop
> crashing Netscape all the time?
If you use the current Xj3D code then everything would exist within a
single "applet" instance so yes it would be completely Java2 capable.
I've done a bunch of tests with Java3D applets using the java Plugin and
it works nicely. I have not yet tried using Xj3D within an applet so I
can give you an answer there. I think it should work, but that's only an
educated guess.
As for solving your problems in time, I really doubt it. We do not have
functional proto support in the Xj3D v2 code yet. EAI 2 support is most
likely the next thing to be done because we can bolt it on very quickly
given the current architecture (my current estimate is at most a week's
work as it is mainly cut & paste). Proto support is going to take a fair
bit longer than that. It's about 50% complete, but that's the easy part.
The hard part is the other 50%. If you want to use H-anim characters,
PROTO support is mandatory.
If you were to go back to Sun's original code, that will support protos
and hanim. However, it does not implement the EAI (in either form). Last
time I tested it, Hanim 1.0 worked but 1.1 was dodgy at best, however
there are reports floating around that some more work has been done to
get it working much better. However, that codebase is no longer
supported and is being replaced by our current work.
Your problem going with pure Java3D right now is the general lack of
knowledge in the community regarding character animation. There are a
few people playing around with it, but all have very different
approaches and there is no generally acccepted "this is the best/right
way to do it" yet. So, if you go down the all J3D path you'll be blazing
your own trail. I can't help you, as I'm not a character animator, but
there are a few others here that are (particularly a couple of guys
doing game development. Catch John Wright or Dave Yazel next time they
post).
So, right at the moment, I think the best summary is that you are stuck
between a rock and a hard place given your requirements and timeframe.
(yes, you can quote me to your lecturer on that!).
--
Justin Couch http://www.vlc.com.au/~justin/
Freelance Java Consultant http://www.yumetech.com/
Author, Java 3D FAQ Maintainer http://www.j3d.org/
-------------------------------------------------------------------
"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".
