Deactivating position and rotation from IK-solver crashes Blender (ITaSC) #33176

New Issue

Tobias Oelgarte · 2012-11-14T21:34:04+01:00

Tobias Oelgarte commented

2012-11-14 21:34:04 +01:00

%%%I appended an example file with a selected bone that has an IK constraint. Since I'm using iTaSC i can deactivate "rotation" and "position" in the constraint panel. The problem is that Blender will always crash if i deselect both options (both on = fine, one on = fine, both of = crash). I can't reproduce this error with a more simple scenario. It seams to happen when two IK-Constraints deform the same chain (tree ik).%%%

Tobias Oelgarte commented

2012-11-14 21:34:04 +01:00

Changed status to: 'Open'

Tobias Oelgarte commented

2012-11-14 21:37:12 +01:00

%%%PS: iTaSC also behaves strangely. I would expect the bone behind (parent) the selected bone to move if i move one of the red bones to the left or right. But it shows no rotation at all. This might be the reason for the crash as well or both influence each other.%%%

Tobias Oelgarte commented

2012-11-14 21:40:44 +01:00

%%%Tried some more. The chain is always only as long as the shortest "chain length". In my example "seil-b" has a chain length of 3 and "plattform" has 1. Both together result in a chain length of 2 for "seil-b". That means that the IK constraint from "seil-b" has no influence on it's third bone "seil-a".%%%

Benoit Bolsee commented

2012-12-05 23:07:31 +01:00

%%%thanks for the sample file. I can reproduce the crash and I see where it's happening in the code. Will fix that soon.

Regarding the chain length issue, this is a known limitation. See this comment in the code:
Oh oh, there is already a chain starting from this channel and our chain is longer... Should handle this by moving the previous chain up to the beginning of our chain
// For now we just stop here
It is easy to work around this limitation: just make sure that all overlapping constraints have consistent chain length (i.e. use the same root bone) and iTaSC will behave as expected.
%%%

%%%thanks for the sample file. I can reproduce the crash and I see where it's happening in the code. Will fix that soon. Regarding the chain length issue, this is a known limitation. See this comment in the code: *Oh oh, there is already a chain starting from this channel and our chain is longer...* Should handle this by moving the previous chain up to the beginning of our chain // For now we just stop here It is easy to work around this limitation: just make sure that all overlapping constraints have consistent chain length (i.e. use the same root bone) and iTaSC will behave as expected. %%%

Benoit Bolsee commented

2012-12-06 23:25:05 +01:00

%%%Fix the crash in revision 52803
%%%

%%%Fix the crash in revision 52803 %%%

Thomas Dinges commented

2013-01-20 17:56:14 +01:00

%%%Closing this report as it has been fixed. %%%

Thomas Dinges commented

2013-01-20 17:56:14 +01:00

Changed status from 'Open' to: 'Resolved'

Thomas Dinges closed this issue

2013-01-20 17:56:14 +01:00

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

Deactivating position and rotation from IK-solver crashes Blender (ITaSC) #33176