[Bf-committers] SOC Project Ideas? .. not a really good one, but still

[bjornmose]

original subject:
Re: [Bf-committers] SOC Project Ideas?

One i can think of and i think oranger's would have liked very much if 
they had :
Baking Softbodies in parts ( of the timeline ) / Editing baked results ( 
and start simulation from that point in time )

This is not only because SB performs too bad for collisions and cloth ( 
which is all my fault ) but mainly to get better initial conditions.

Says:
At the moment the only relyable state of a SB object is the 'rest 
position' --> position of the vertices in edit mode + object modifiers 
with no initial velocities.

As discussed loooong time ago (when SBs were about to be coded), it 
would be very nice to start ( continue ) the simulation with a differnt 
set of 'initial conditions' --> positions and current speed vectors.

1rst step IMHO would be: (might be a SOC project on it's own)
Run the simulation for a defined period in time ( e.g with the fully 
rigged and posed character ) to let the (mass-spring) system stettle 
down to a new rest position other than in edit mode, "key" that to frame 
   "1" and start the simulation (baking) from there.

2nd step:(could be another SOC project but depends very much on the 1rst)
Is amost like the first, but run the time line while simulating ( --> 
character and environment are allowed to move this time ) until frame 
"n" is reached and make this state (pos. and vel.) the initial for frame "m"

3rd:
Treat baked SB as an editable proposal.
This will become a nightmare UI wise for zillion k vertices meshes .. so ..
But orange team tried to solve "proogs" jacked ends with a nice low 
vertices SB approach ( in lack of physic's  simulation {integrating ODE 
for that is still desired :) } available on armature/bone level ) which 
utterly failed ( because of poor SB implementation ). I did not see 
elepant's dream yet, but i'm pretty sure that basse threw some
hand animated bones in there ( which happly would have taken some 
"automation" from some physics engine if it were there .. sigh..).
Cutting it down:
Having a UI to fix glitches the simulation engine slipped in and 
restarting at as in 2nd stated.

Ahh.. hold your breath..
Unfortunatelly everything described here looks like working on the open 
'edit' and 'data structures' heart of blender  .. and needs deep cardiac 
knowledge .. (even i'm not sure if i have it)

..hum..i did the work writing it down and don't want to throw it away ..
may be i'd better change subject of e-mail

OTH .. may be zaz or genscher can do it to earn a few euros

BM


Chris Want wrote:
> It seems that most people think that having a
> project suggestion page is a good idea
> (tsk, I hate when people disagree with me).
> 
> These suggestions would not be manditory,
> only suggestions. On this page I will add some
> text that should scare off people that want
> to just copy/paste.
> 
> Last year we used the 'projects openings'
> page as our ideas page, but it's horribly
> out of date:
> 
> http://www.blender.org/cms/Project_Openings.589.0.html
> 
> So ... any suggestions as to what might be
> good student projects for the summer?
> 
> Chris
> 
> P.S. A preliminary working version of the content of the
> ideas page is here:
> 
> http://bebop.cns.ualberta.ca/~cwant/soc_page_wip.txt
> 
> 
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at projects.blender.org
> http://projects.blender.org/mailman/listinfo/bf-committers
> 
>