[Bf-python] Python -- Orientatoin problems.

[Gilbert, Joseph]

I don't remember the whole issue, it's been a while now, but the rolls
are not being propagated correctly from parent to child. I don't
remember exactly y but it had something to do with the fact that bones
are stored as axis/angles but they are drawn (and parented) using the
bone's matrix, and the rolls are being set incorrectly in the children
because of this. It has to do with the decomposition of a parented child
matrix into the correct axis/angle I think. Or something like that lol.

Also at issues with the armature module is its lack of IK support, its
poor ability to deal with the rest pose, and it simplistic capacity for
adding key frames (which is more of a hack). It works, but just. (I can
be critical here :) Add to that the way that NLA works (which is truly
bizarre).

However, previously there was NO support for dealing with armatures and
key frames at all (other than the ability to 'wrap' them) however these
2 modules still have many issues have yet to be resolved and this is one
of them (a BIG one of them). The internal Bone/Armature code is 9*d
awful which adds to the troubles of writing a good python wrapper. 

-----Original Message-----
From: bf-python-bounces at projects.blender.org
[mailto:bf-python-bounces at projects.blender.org] On Behalf Of Ken Hughes
Sent: Tuesday, June 07, 2005 1:54 AM
To: Blender Foundation Python list
Subject: Re: [Bf-python] Python -- Orientatoin problems.

I think I found the problem for bug #2230, but can't think clearly
enough
to find a solution (if someone can read the armature code, please
explain!).

Looks like getRestMatrix() is relying on the bone's roll variable to
calculate
the rotation, but roll is always 0.  In most of the armature code I've
read
seems they calculate roll then reset to 0; can't see where it's being
used.
I put a printf in make_boneMatrixvr() -- called by Bone_getRestMatrix()
--
to print roll and it's always zero.

Nick Lawson wrote:
> It could be related to
>
http://projects.blender.org/tracker/index.php?func=detail&aid=2230&group
_id=
> 9&atid=125
> 
> In which the quat and loc of bones is not outputting correctly via
python,
> even in 'bonespace'.
> 
> -Nick
> 
> 
> -----Original Message-----
> From: bf-python-bounces at projects.blender.org
> [mailto:bf-python-bounces at projects.blender.org] On Behalf Of Kenney
Phillis
> Jr
> Sent: Monday, June 06, 2005 3:08 PM
> To: bf-python at projects.blender.org
> Subject: [Bf-python] Python -- Orientatoin problems.
> 
> with most models i make the origin for the bones is off....
> picture of the offset issue of blender: 
> http://img65.echo.cx/img65/3514/offsetissue3ck.png
> ik picture issue:
http://img138.echo.cx/my.php?image=standardikoutput9nn.png
> 
> the two images point out two different issues... you can look at my 
> attached script for debugging purposes, but the issue is that the
world 
> space and bone space quats don't show negative values on rotation..
> 
> _______________________________________________
> Bf-python mailing list
> Bf-python at projects.blender.org
> http://projects.blender.org/mailman/listinfo/bf-python
> 

_______________________________________________
Bf-python mailing list
Bf-python at projects.blender.org
http://projects.blender.org/mailman/listinfo/bf-python