hello, thanks for the reply, are you telling me to take off all the positions of all the geos and then evaluate in other program?
what about gem_listinfo? i could take of all the positions an save them in a text file? and which other software should i use to make this calculations? I just checked there is a Ode version that run in pd , called pyode, i think ode works fine for rigid body simulations, do you think this is a good option in pd?
On Tue, 3 Mar 2009 10:01:03 -0500 (EST), "Mathieu Bouchard" matju@artengine.ca said:
On Tue, 3 Mar 2009, simeon cosik wrote:
Hello, how can i evaluate 3d structures generated in puredata to know if it can support itself if i build it in reality? for example imagine that each structure is conformed by a group of boxes one over the others made with "repeat",i would like to add gravity and weight to all the boxes so i can i calculate the threshold that i need to maintain in order to avoid the collapse of my structure. Is this can be done with pmpd? Any idea?
You would need to remake your patch to also work without GEM, or find a way to download the geos from GEM.
One way to make such a "dual-output" patch, is to make it as an abstraction, taking an argument that selects the library it is going to use. suppose your patch is called mymodel.pd:
[mymodel] would use GEM [mymodel gemdump/] would use a library called gemdump/
Then within that patch, you would prefix every geo and every geo transformation by $1, such as [$1rotateXYZ]. After that you'd have to make a somewhat complex abstraction library that could dump all the coordinates.
Actually, this pattern (of switching libraries) is very useful, but in this case, it may be too hard, so if I were you, I'd first look for something that can download the geos from the video card, and if I don't find any, I'd make that GEM emulation library that would give me all vertices and edges.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal, Québec