Mesh Deform bind locks Mesh Deform target object #49388

New Issue

William Hensley · 2016-09-18T16:45:46+02:00

William Hensley commented

2016-09-18 16:45:46 +02:00

Broken: 2.78 RC, 0e5089c
Worked: 2.76, da81227

In the past, it was possible to change the target of a mesh deform after binding the mesh. This was useful for many tasks, but, in particular, for binding to meshes with displace modifiers on them, by creating a copy, applying the modifier on the copy, binding to the copy, then switching the target to the original.

This is still possible through python. It's just that the interface settings are locked.

I'm not sure whether this is an oversight or an intentional change, but it's definitely a workflow slowdown.

Broken: 2.78 RC, 0e5089c Worked: 2.76, da81227 In the past, it was possible to change the target of a mesh deform after binding the mesh. This was useful for many tasks, but, in particular, for binding to meshes with displace modifiers on them, by creating a copy, applying the modifier on the copy, binding to the copy, then switching the target to the original. This is still possible through python. It's just that the interface settings are locked. I'm not sure whether this is an oversight or an intentional change, but it's definitely a workflow slowdown.

William Hensley commented

2016-09-18 16:45:46 +02:00

Changed status to: 'Open'

William Hensley commented

2016-09-18 16:45:46 +02:00

Added subscriber: @Arcayer

Sergey Sharybin commented

2016-09-20 11:16:47 +02:00

Added subscribers: @brita, @mont29, @Sergey

Inês Almeida was assigned by Sergey Sharybin

2016-09-20 11:16:47 +02:00

Sergey Sharybin commented

2016-09-20 11:16:47 +02:00

This was changed by @brita in 22f2405. It indeed makes sense to see binding options, but the commit message does not explain why other setting became disabled instead of inactive.

On another hand it;s quite weird situation to change options after binding. it's a straight way to some disaster, so kinda makes sense to prevent that.

@mont29, @brita any strong opinion here?

This was changed by @brita in 22f2405. It indeed makes sense to see binding options, but the commit message does not explain why other setting became disabled instead of inactive. On another hand it;s quite weird situation to change options after binding. it's a straight way to some disaster, so kinda makes sense to prevent that. @mont29, @brita any strong opinion here?

Bastien Montagne commented

2016-09-20 11:48:32 +02:00

I’m not sure to see the point in changing target after binding? Explanations of the report are rather… unclear to me.

Changing mesh after binding is supported by the modifier, it just does nothing anymore if number of vertices changed iirc (and otherwise, you just keep using computed binding data, which will give strange things if you are using a different mesh, that’s all).

So no, no strong opinion, I’d rather stay with new behavior though, unless we can get examples or real needs of ability to change target after binding.

Note that this is technically not a bug, code is behaving as it should, and change was intentional afaik.

I’m not sure to see the point in changing target after binding? Explanations of the report are rather… unclear to me. Changing mesh after binding is supported by the modifier, it just does nothing anymore if number of vertices changed iirc (and otherwise, you just keep using computed binding data, which will give strange things if you are using a different mesh, that’s all). So no, no strong opinion, I’d rather stay with new behavior though, unless we can get examples or real needs of ability to change target after binding. Note that this is technically not a bug, code is behaving as it should, and change was intentional afaik.

Alexander Gavrilov commented

2016-09-20 12:04:41 +02:00

Added subscriber: @angavrilov

Alexander Gavrilov commented

2016-09-20 12:04:41 +02:00

Can you test if it now works if you just bind to the mesh without applying the displace modifier? I have reported the fact that it didn't work properly almost a year ago as #46944, and it is supposed to have been fixed.

William Hensley commented

2016-09-20 18:30:29 +02:00

Binding to mesh deform does work correctly in the most recent build.

After testing a number of my meshes, the ones that present the largest issue are the physics meshes with large numbers of internal faces-

untitledC.blend

Also, in some cases the physics mesh is missing faces, or is otherwise non-manifold. Basically, any case where faces are added/removed for binding.

Binding to mesh deform does work correctly in the most recent build. After testing a number of my meshes, the ones that present the largest issue are the physics meshes with large numbers of internal faces- [untitledC.blend](https://archive.blender.org/developer/F365224/untitledC.blend) Also, in some cases the physics mesh is missing faces, or is otherwise non-manifold. Basically, any case where faces are added/removed for binding.

Bastien Montagne commented

2016-09-30 12:41:16 +02:00

Changed status from 'Open' to: 'Archived'

Bastien Montagne closed this issue

2016-09-30 12:41:16 +02:00

Bastien Montagne commented

2016-09-30 12:41:16 +02:00

Oki, so think we can archive this now? Thanks for the report anyway.

Sign in to join this conversation.

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

Mesh Deform bind locks Mesh Deform target object #49388