[fluxus] skinning and pfuncs

[Dave Griffiths]

>>The implementation I've done is the simple classic approach - I suppose
>>the idea is that using pfuncs we can have as many approaches to this sort
>>of thing as we want (not locked into one way).
>>
>>
> nice, still have to play with it, but as a side question what are you
> planning with these on fluxus?

For skinning - I just want a simple way to mangle stuff for livecoding :)

>>The DualQuaternion system looks very quick to try out - I'll give it a go
>>at some point.
>>
> cool.
>
>> The next thing I want to look at though, is how to make
>>dealing with scenegraph subtrees easier - some skeleton utility type
>>things.
>>
>>
>>
> this might not be so complicated, normal parenting stuff might do it for
> fks, for iks the subject is a bit more involved i think and dont think
> its worthy the effort, depends on the first side question answer.

I'm only interested in the forward kinematics at the moment, it's all much
more basic than what you're thinking :)

I was thinking it's a pain to have to build two skeletons by parenting for
the bind pose and animated pose separately - which led me to think we
should be able to copy them, which means something like (build-deep-copy)
which would copy a primitive and all the children of it's node
recursively. Then we'd need to find out the id's of all the children - so
we'd need a (children) or something for a primitive, to return all the
primitives parented to it... or maybe this (build-deep-copy) returns a
tree list of primitive id's it's generated.

But some form of (children) might be good anyway, as currently there is no
way to navigate the scenegraph.

cheers,

dave