Easy way to fix bad skinning?

[dunkelzahn]

1. Don´t just move the forearm, move the clavicle as well.
2. Skinning is the icky part of rigging, so there´s no magic button here
3. If you have a low poly version of your mesh, try skinning it and then add Turbosmooth on top. Less vertices, less skinning, less headaches
4. Read the manual on skinning and on how to paint skin weights, this should give you some more insight
5. If you still have probelms with certain areas while animating, check out this tutorial : http://www.deep4d.com/pages_tuts/sk...le_deformer.htm (http://www.deep4d.com/pages_tuts/skin/skin_morp_angle_deformer.htm)


Have fun

Chris

[phoelix]

there is too many verts in that area, did you use the skin modifier below the turbosmooth?, if the mesh is like that it should be more simple, it's not necessary to have such complex topology in the shoulder if it doesnt have too many details.

when i'm skinning joints i use to select a loop of verts and set the weight with all of them together, that is a simple and effective way of doing it. In the shoulder the task is more difficult because the lower verts of the loops should have less influence by the arm, in this case i set the loops weights first and then start lowering the weights of the lower verts

[carlw]

The Turbosmooth isn't even on! This is just a poly-heavy model. So I should probably remove some edges from the shoulder area....but my issue is that the Morpher modifier I have will stop working if I change vert ID's.

Is there any way around this?

EDIT: I just realised I can use an edit poly modifier for this. So I'll give it a try now.

[phoelix]

its better to dont use poly modifiers to change the mesh above a morpher or others animated modifiers, thats CPU consuming and will affect your viewport performance...

[IkerCLoN]

I suggest you to forget about using envelopes when 'skinning'. It seems they can give you nice results in fewer times, but in the end I think you'll prefer to have absolute control of your weights from the very beginning.

For that area, try to use a 'tendon', 'muscle' or whatever you want to call it. This is quite old but it can give you some clues on a possible way to face the losing volume problem:

http://www.jhaywood.com/croiProjectArm05.htm

You should also try, apart from selecting vertices and give direct weight values to them, weight painting. It's very handy when dealing with non-loop areas, also with shoulders and hips.

Paul Neale made a DVD explaining the process of vertex weighting under 3DSMAX. If you are new to this discipline, you'll learn a lot with his explanations. I think it's a good investment! ;-)

[Ruramuq]

for sikinning, vertices weights depends on bones entirely, so the limits comes from the rig itself. but once you know how envelopes behave in that context, I prefer to use them, and then improve the rig, 'cause it becomes obvious what are the limits in animation, and perhaps try to fix later the weights, only for fast projects.
Personally I find painting weights too precarious and random. something that turns evident with high density models.
IMO most problems related to skinning are: bad rigging, bad position of bones

[carlw]

for sikinning, vertices weights depends on bones entirely, so the limits comes from the rig itself. but once you know how envelopes behave in that context, I prefer to use them, and then improve the rig, 'cause it becomes obvious what are the limits in animation, and perhaps try to fix later the weights, only for fast projects.
Personally I find painting weights too precarious and random. something that turns evident with high density models.
IMO most problems related to skinning are: bad rigging, bad position of bones

Yeah, it's hard for me to tell exactly where to place bones...and how far certain limits should be.

But at the moment it's getting better, I'm not painting any more, just slowly selecting verts and trialing different weights...it's slow but seems to be working OK.
I'll post some pics when the weighting is better.

[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.