talk.bindpose.com has been shut down on the 6th of March 2021. You can read a bit more about it here.

You are currently reading an archived copy of the latest content, with all the usernames deleted.

Maya has stopped working

by user66 posted 26-11-2017 | 2 comments

I've run an issue that has become very annoying. Occasionally when I try to open a file, I immediately get "Maya has stopped working" and the file seems to be corrupt. I try to go back to other incremental saves but they get the same error and they will not open.

I thought I had narrowed it down to removing the rig from a skinned object, without detaching the skin first. But I can't replicate the error in a simple scene. It only happens when I try to remove a complicated limb rig, which is making it difficult to track down the problem.

It's annoying because whatever I'm doing is not immediately noticeable, and I can work for hours, even saving multiple versions, without realizing anything is wrong. Only when I close the scene and open it again do I get the error.

I should note that this exact same process of removing the limb rigs worked fine in Maya 2013, and only causes this issue now that we use 2016.

Any ideas?

 

by user3 posted 27-11-2017

Are you using the latest 2016 update? I remember at university we used 16 at one point and it did seem to have quite a few bugs, but I am assuming with the updates most of these should be fixed.

 

by user66 posted 27-11-2017

2016 has been pretty buggy. I'm using service pack 6, but who knows what bugs remain.

I went through my process again and paid more attention to warnings, and sure enough I got this:

Weights could not be normalized due to locked influences. // Locked influences prevented enforcement of the max influences on some points. //

So I feel confident I know what's causing the corruption and hopefully can avoid it in future... but now I guess the question is whether I can salvage the files after this mistake has been made. I've opened the files in a text editor, but I'm not sure what to look for. Maybe deleting any mention of the skinCluster in question?