Hi all
I’ve developed a plugin for maya that implements the dual quaternion skinning technique shown here. The main feature of this deformer is that only the deformed parts are recomputed (and not all the mesh like usually) and this speeds up the process.
If you are interested here are the sources and the compiled plugin (only for linux64 and maya 2009)
http://code.google.com/p/dqdeformer/
Any relation to this?
Or what are the differences between yours and the linked one? If any?
My version is quite faster: for the mathematical operation it doesn’t use the maya library (witch is really slow) and it recomputes only the part of the mesh that is interessed by the deformation.
A small video that shows the differences between dual quaternions and maya skinCluster:
http://www.vimeo.com/2971775
This looks very interesting. A Maya 2008 version would be greatly appreciated!
Cheers!
Hey Michele,
Is it possible to turn the DQ solver on and off [u]per joint[/u] similar to XSI 7.0's DQ ICE compound (example video [here](http://webrel2.softimage.com/open/video/xsi7/02_Deformation_Dual_Quaternion_Skinning.flv))? A major limitation of Guido DeHaan's plugin was that the DQ solver effected the entire skeleton, thus preventing you from using a mix of solver types depending on the situation (knee vs. shoulder deformation).
+1 vote for a Maya 2009, Win x64 compile :)
Thanks for sharing your work with us!
-mental :surprised
I’ve compiled versions for 2008 and 2009, 32bit and 64bit. These are untested by me. Not much of a rigger 
Micle, where should I put them? Would you host them on your site?
i loaded the x64 2009 plugin compiled from ticket1 (thx by the way), it load without errors, but after the “createSceneSkinClusterDq;” command errors occurr, unknown nodes 
in 2008 it works fine
btw. how does this plugin compare to the DQ_skinning in maya2011 ???
i could somehow imagine , that it is faster than the “latest invention of autodesk” … on the orher side , any chance for CUDA support ( GPU_skinclusters … )
In de video it looks like at some frame when the DQ box is doing a bicept flex, the algorithm goes wack and twists the mesh… What causes this ?
In de video it looks like at some frame when the DQ box is doing a bicept flex…
you mean at ~0:26?
I reckon he keyframed (in time) two scenarios. one for the twist (1-140 f), the other for the bulge (behind that) and he is scrubbing from frame 1 on.
Oh there is another at 0:30…
what I would miss badly with this Plu-In, is the lacking opportunity to update weights via the component editor:-(
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.