transferring to different deformers

[ThE_JacO]

Save the weights as a preset, delete the envelope operator, delete or rename the nulls, rename the bones to what the nulls were called, apply envelope operator selecting the bones, load the weights.

Weight presets are namesXvertex based, if the nulls are one per bone all you need is re-enveloping with names matching.

Btw why do you want to remove the nulls?
Unless you have some particular reasons, enveloping to an interface and not directly to what's animated is a much better procedure than enveloping directly to the bones.

[Chirone]

well... i'm not exactly familiar with the XSI workflow for rigging and animating, so i don't know what's considered good practice and what's not

basically, i just thought that a mesh should just be weighted and skinned to the skeleton(or lattice or other deformers) that it should be constrained to

also, when i move a null in this particular model random verticies move around too... like a few verts on the head move when the leg is rotated.

i thought that enveloping it to a skeleton would fix that... but come to think of it, it might not, since the weight is still there...

also child nulls aren't at a fixed length from their parent nulls, so its possible to deform the mesh in odd ways... i figured since a skeleton has such constraints it would be better to use the skeleton instead of the nulls

i will have to try your suggestion out tomorrow


can you explain the 'why' of your last sentence? i'm not sure i quite follow the reasoning yet.

[ThE_JacO]

well... i'm not exactly familiar with the XSI workflow for rigging and animating, so i don't know what's considered good practice and what's not
Regardless of application skinnig to animation controls is considered bad practice, because anything you might later on need to do to it will require re-weighting if lucky, or even re-enveloping if not.

also, when i move a null in this particular model random verticies move around too... like a few verts on the head move when the leg is rotated.

i thought that enveloping it to a skeleton would fix that... but come to think of it, it might not, since the weight is still there...
Correct, it's badly weighted, swapping what it's weighted to will not fix it, you have to clean up the weights.

also child nulls aren't at a fixed length from their parent nulls, so its possible to deform the mesh in odd ways... i figured since a skeleton has such constraints it would be better to use the skeleton instead of the nulls
Not sure I understand what you mean there.
A null child of a bone, or pose constrained to it, with no offset is for all intents and purposes, when it comes to enveloping/skinning, exactly the same of the bone itself.

can you explain the 'why' of your last sentence? i'm not sure i quite follow the reasoning yet.
Separating things by functionality, when it involves no overhead of work, usually allows for easier modifications.

[Chirone]

Not sure I understand what you mean there.
A null child of a bone, or pose constrained to it, with no offset is for all intents and purposes, when it comes to enveloping/skinning, exactly the same of the bone itself.

the mesh has a heirachy of nulls inside it. so when one moves all the child nulls move. but if a child null moves (say i move the null that is the deformer of the elbow) then it can potentially deform the mesh (if i translate the elbow by mistake or if i was meant to select another null with it then the arm is no longer in proportion)

how do you suggest about getting this model rigged to the skeleton then? should i parent the skeleton to the nulls?

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.