Quoting Antonio Bleile <[EMAIL PROTECTED]>:

> Hi,
> 
> as there's still a lack of animation support in OpenSG,
> and as Dirk said in the past that it won't be trivial
> to add it to OpenSG (mainly because of clustering and
> multiple viewports), I thought it would be good to let
> it to a talented student :-) We'll probably have two
> students here writing their master thesis in collaboration
> with the local university, so I thought it could be
> very incentive and interesting for them to let them work on
> Open Source stuff. What do you think? Is anybody working
> on animation support already? Do you think it is worth
> a thesis? Or is it just a one week hack for a core
> developer? What are the main problems in animation
> for OpenSG? What kind of problems have to be resolved?
> Any references to existing systems where the problem has
> already been solved? How?

If they could research options and develop some mechanism in OpenSG to..

 1. Distribute time & update across objects (in clusters) generally
 2. Manage dependent data (Maya/VRML node-graph or simpler way)
 
. I would think they have their hands full. It might be a partially solved
problem (worth checking out) but a paper describing various methods, pros/cons
and results on an actual implementation (contributing to OpenSG) would IMHO
certainly be a good thesis project. It would contain both theory & practise,
which I think is a Good Thing(tm). It of course depends on the students' goals
and ambitions, whether they want an academic or applied thesis, however I think
both goals could fit under this umbrella.

> And if you have other suggestions for a master thesis
> within OpenSG, please let me know! Another thought of mine
> was to add Shadow Volumes to OpenSG, but I don't know
> if that's sufficiently "scientific/innovative" for a thesis,
> but if you add "soft" and "real-time", it might work out :)

Developing a GPGPU framework within OpenSG would be cool. Using the clustering
mechanism to distribute/gather calculations, etc, etc. :-D It could be two-fold,
GPGPU is one thing, a generic scatter/cather cluster calculation framework
(with/without OpenGL) is another.

Also, setting up a framework for deterministic replay (i.e. a changelist
save/replay function + random numbers & time + thread/cluster scheduling) would
rule, but that's perhaps not scientific enough. (I think there is a patch for
changelist save/restore on SF...)

Cheers,
/Marcus

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier.
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Opensg-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensg-users

Reply via email to