3
$\begingroup$

Parent offset in Blender is difficult to understand, because there are obscured values that determine the outcome of user operations. See this question.

Are there any tools/addons to help with this? For example, a replacement parenting/reparenting function, or even a tool to simply display the internal values, to aid in understanding what is going on?

Secondary question, is there a stated or obvious reason for this seemingly intentional operational obfuscation? Is it something worth my time in seeking to drive change or offer to help change?

If you've worked with it for a while, and it now makes perfect sense to you, let me remind you of where others might be coming from. Here's a sample .blend I'm currently looking at:

enter image description here

All of these are simple, visibly identical plain cube meshes with identical rotation, and the object origin at the mesh center of each.

For starters, note that both D and C report translations of 0, 0, 0, and yet are in different locations, both in world space AND relative to their parent. How odd, I think. However, the real confusion is when I try to actually make parenting changes.

We know that there are various operations to unparent, keep transform, clear parent inverse, and set parent. However despite my goal of transferring the child relationship of D from parent A to parent B, nothing results in a destination of H. Various operations result in G, F, and E.

That means that if, historically I've moved something, I might be building objects with defects that cannot be measured through the UI. I had just this scenario recently. Rotations and multiple duplicates were involved, and I had to redo lots of work to get things properly aligned.

For me it was a cautionary tale of "be very paranoid of where things were historically", but that doesn't seem like an ideal lesson for me to learn. Or, maybe I'm just doing something else wrong.

$\endgroup$
2
$\begingroup$

I'm not use that I understand the question. But it seems to me that instead of using the parenting tool, you might want to look into an interactive and non destructive workflow using constraints like Copy location, Child Of and Copy Transforms You'll have very clear control on the influence and is quite easy to create complex parent-children hierarchies.

enter image description here

$\endgroup$
  • $\begingroup$ Thank you. I would love to use a workflow with this clarity. This works great for animation and representing relationship with clear calculable translations and weights. However, when the design calls for actually directly placing elements in the 3D workspace (using a mouse or 3D input tool, for example), creating multiple constraints and adjusting sliders for what should be a more "physical" operation isn't efficient. For example, reworking/replacing lug nuts parented to a template wheel object at some point in design after wheels are placed. $\endgroup$ – shannon Jan 1 '15 at 21:46
  • $\begingroup$ I know there's a lot of cubes/examples in my question. I tried in my initial question to clarify what I was trying to do that prompted the question (which is still my actual question: an attempt to get a handle parent/child transforms). I think it's a reasonable modelling scenario and so makes the question relevant to others. I wanted to transfer a child from one parent to another and retain the dependent transform. This arose because I added a child embellishment object to a template parent that derives multiple other objects in my scene. Moving a copy of D (say, H) from A to B, in my image. $\endgroup$ – shannon Jan 1 '15 at 21:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.