It looks like you're new here. If you want to get involved, click one of these buttons!
I have tried to update my BVHReader so that it is compatible with 2.6 (due to the deprecation of the Animation ext, especially the Clip object). The changes seem quite simple (although docs are a bit lacking), just exchange Clip by AnimClip.
When I did that, suddenly I got double transformations in the results. I can hack a solution by storing the ReferencePose for each bone in local instead of global space. It 'looks good', but it does not feel right.
I have tried to isolate things, to see where this change is coming from, but so far no luck.
I could assert that, after evaluating a AnimClip ClipPose and perfom a reset() all deformer Xfos are zeroed out (see attatched canvas file). Apart from that I could not find any other changes in behavior in isolation.
Maybe it is something I am doing wrong, but I though it was worth to post in case someone else is having similar problems.
©Copyright 2017 Fabric Software Inc. All rights reserved. | Privacy