Open GoogleCodeExporter opened 9 years ago
Issue 265 has been merged into this issue.
Original comment by miller.lucas
on 4 Dec 2011 at 6:32
An important comment to issue 265 that should be made here:
The constant shape from the exported Alembic file needs to replace the
corresponding shape in the scene. I was thinking that the old shape should be
deleted and replaced with the constant shape, but this may have unfortunate
side effects for certain workflows (constraints?). Maybe we would only do this
if the topology were different.
Original comment by miller.lucas
on 4 Dec 2011 at 6:33
If you delete the old shape you'd also loose any custom attributes from the
original shape. Why not just transfer the vertex position?
Original comment by davidwo...@gmail.com
on 4 Dec 2011 at 6:54
That's what we'll do if the topology is the same. For other cases we may have
to delete the shape. (or warn and skip)
Original comment by miller.lucas
on 5 Dec 2011 at 1:38
Original comment by miller.lucas
on 9 Dec 2011 at 5:21
MFnMesh::createInPlace may help, but a function like this doesn't appear to
exist for all the shape types.
Original comment by miller.lucas
on 1 Jun 2012 at 5:46
[deleted comment]
Why not connecting directly the constant shape from the alembic node, as it's
done with animated shapes ?
Original comment by sebastie...@gmail.com
on 10 Jul 2014 at 9:21
That is likely what needs to happen, as other solutions have had serious
drawbacks.
I do wonder if there is any performance implications for having a bunch of
connections to constant shapes, or suddenly having a bunch of Alembic nodes in
your scene for archives that were once all static.
Original comment by miller.lucas
on 10 Jul 2014 at 4:36
Theorically the inmesh outmesh type connections are quite light, but coming
from the alembic node I don't know if it would impact on how maya refreshes the
shapes. I think it's a matter of dirty/clean info of the DAGs ? If the
connection stays clean, it would not refresh I think.
Original comment by sebastie...@gmail.com
on 10 Jul 2014 at 4:50
Original issue reported on code.google.com by
miller.lucas
on 8 Nov 2011 at 9:33