Thanks Eric, I think that this is the best option, infact I'm using the same workflow to drive a Facerobot rig without having a C3D file, and its working properly ( custom rig that drives another rig ), but for syflex I was hoping to do a 4 bone rig directly on the mesh ( to save some time ) but I guess having two rigs ( one with constrained nulls and the other one being the actual rig with effectors ) its the best solution, since I would just run the simulation and bake it to the other rig
Thanks :) 2013/10/14 Eric Turman <[email protected]> > Hi Nicolas, > > You need to use a source and target solution. The source has the > simulation on it and the null deformers are constrained to it; think of the > source as if it were a rig. The nulls in turn deform a second mesh, the > target which is point weighted. The target and its deforming nulls are > what is exported to the game engine. > > Cheers, > > -=Eric Turman > > > On Mon, Oct 14, 2013 at 8:10 AM, Nicolas Esposito <[email protected]>wrote: > >> Hi Oliver, >> Thanks for the answer >> >> More or less the simulation will be an approximation of what the >> simulation will be, so not 100% as the syflex simulation ( otherwise I'll >> have thousands of effectors for each poly/vertex ), but something that close >> >> What I'm worried about ( need to test if it works ) is that once I did >> the simulation, and I add the nulls ( effectors ), if I paint the envelopes >> on the mesh I will loose the simulation on the cloth... >> I mean, envelopes ovverride the simulation? Or I can still apply the >> pointcache to the cloth even if my mesh is enveloped? >> >> >> 2013/10/14 olivier jeannel <[email protected]> >> >>> Hello there, >>> I'm not familiar with games pipes, so maybe I misunderstand your problem. >>> If you put chain bones you'll need to add as many bones (or effectors) >>> as there are points on your mesh. >>> Otherwhise you'll only get a rough approximation of the cloth sim. Plus >>> you'll need to paint some enveloppe weights. >>> Anyway if this what you're after, IMHO, a simple point constraint of the >>> Null effector should work, and then a Plot of the bones. >>> Imho, shapes or a cage of nulls will be more predictable (if pointcache >>> is not an option). >>> >>> Le 12/10/2013 12:52, Nicolas Esposito a écrit : >>> >>> Hello, >>> >>> I'm working on a character that will be exported into a game engine ( >>> UDK ), but since SI doesn't have an Apex plugin to export the data I need >>> something to simulate the cloth behaviour >>> >>> So, first I'm simulating the cloth during the animation on my >>> character, that has a Species rig applied ( face + body ). >>> >>> What I would like to do is to build a bone chain to be able to export >>> the complete rig into UDK >>> >>> Since the function of the bone rig is to drive the mesh in my case I >>> need to do the opposite, so the simulated syflex mesh will drive the bones >>> position/rotation, and add a keyframe for each movement, in order to bake >>> all the animation into a single rig (* since UDK for animation accept >>> only rig animations*...and shapes, but I would like to use a rig ) >>> >>> Consider that my character wear a coat, so after the simulation I was >>> thinking to build 4 chains ( 2 on the back and 2 on the front ) and then >>> let the syflex simulation drive the bones position/rotation >>> >>> I hope that is clear what I'm trying to do >>> >>> So, is there a way to achieve something like that? >>> >>> Nicolas >>> >>> >>> >> > > > -- > > > > > -=T=- >
