There might be something more to this concrete scene, because it seems that Corona does not ignore point cached geometry at all under certain conditions.
Can anyone confirm this, please? I’ll investigate further…
Ah, no, my bad, the motion blur IS NOT THERE, it was being created by transformation (atop the point cached geom, I’m animating the object’s transform as well), so that is being calculated, but the PT is being IGNORED alltogether.
Seems to work fine here, though I am not sure if I am doing this the same way as you did. Picture and scene in attachment.
Can you try using mesher on the point cached object and see if it gets proper MB?
I was told it was because the changing topology of my mesh. Not sure about that, I’d have to double check that as I’m subdividing heavily the mesh based on smoothing groups, dynamically, so that might be the real problem here. Which would be a huge problem for me as I need this “paper like” look.
Hmm how did you do this? When I am trying to save point cache file for object with animation where number of faces changes it says “point count changed while recording” or “cache file is not a valid file”.
Tried enabling MB for the geometry with modifiers, got a freeze right now, will try to submit minidump.
Confirmed. Motion blur will not work when using “smooth” modifier if the smoothing will change dynamically. It will work fine if set to 0 or 180 (a value where nothing changes during the animation). Attaching the file if someone wants to take a look. Not sure if fixable. smoothing-mb.zip (36.5 KB)
The problem is that the normals could still come out garbled. I will try to look to see if there is a way to fix this one, but general fix (for any change) is still far away.