Topic: blender 2.6x vertex weights and bone envelopes?

[ratty redemption]

hi, i'm having issues with the deforming of what should be a simple rig and mesh.

in the first image i've managed to get the falloff working as i would ideally have for all deforming bones, but only by selecting the same vertex group in the armature modifier as the selected bone is using.

if i try to use any other vertex group, like for example, one that includes all the verts in the mesh with the weights set to 1, then the falloff stops working for the bone specific groups, and in this case of the bone being rotated 45 degrees, the whole vertex group is rotated which is confusing me why the gradient weights aren't being used?

if i try to use just bone envelopes for deforming they produce the same undesirable results, ie will only use falloff if i enable the specific vertex group in the modifier. no combination of the two methods seems to work for all bones at the same time, but i admit i might be missing something obvious. i hope this isn't another bug?

[ratty redemption]

in an attempt to find a solution to this i registered over at blender artists forums. batfinger was able to help me with better understanding the armature modifier panel, but hasn't yet commented on the falloff not working. i've also done more tests today with a new blend file. here is a link to the blender artists forum where i go into more detail, and post the latest test images:

http://blenderartists.org/forum/showthread.php?263688-blender-2.63-possible-vertex-weights-bug

edit: apparently my latest post over there hasn't been made public yet, but hopefully will within the next hour i would assume.

[kat]

Make sure your bone group names are coupled (the same) otherwise you may get deformation though envelopes but not bones (envelopes attach/pair to vertexes slight differently iirc, don't really use them myself so you'll need to check). You'll also need to check whether you have just a single bone/single vertex pairing, hat would explain the rotational issue - unless the group has more than one bone assigned it will rotate 100% for one bone, 50% for two and so on (assuming standard weighting).

[ratty redemption]

thanks kat, and i've checked my vertex group pairing. i have one deforming bone, with one group containing several weight painted verts. i'm still confused though, what is the point of the weighting if all verts in that group are going to be rotated or translated 100% ie not use falloff?

ot: i'm really not impressed with waiting over 12 hrs for my last posts on blender artists to be made public. are they really that fan boyish that they can't stand any constructive criticism of blender? if so i wasted my time joining their site.

[kat]

When you're using groups and bone pairings 'falloff' isn't really attributable, you do that through pairing to multiple bones and/or reducing the influence weighting each bone has on a vertex (so two bones paired to a single vertex influence it 50/50 - unless otherwise weight-painted). So to answer your original question, if you use envelopes (which I believe use fall-off) and bone weighting you're potentially creating a set-up where it's possible for one mechanism of control to counteract the other. This basically means you have to be careful with your assignments to make sure you're not overlapping. You'll need to do some reading up on using envelopes though, I don't use them myself as a main production tool so can't expressly say with any firmness this is what's causing the problem you've encountered.

Re. the BA posting wait: it's an anti-spam thing.. the wait and/or manual first-post check is supposed to limit the amount of spam registrations that sign up to post hit-n-run messages.

[ratty redemption]

thanks kat, and i think i understand. just to clarify i only tested envelopes a few times but i don't enjoy working with them so haven't been using them since. what workflow would you recommend for me to produce the results of one deforming bone having a falloff effect on the verts in it's group?

re the ba site, it shouldn't take 12+ hrs to determine if a post is spam or not. don't they have staff working in shifts? or do they only work during our day time hours? i can't help but suspect this is more like a draconian form of censorship.

[kat]

You might have to do some 'clever' bone pairings or weight-painting to get the desired effect. This does of course depend on what you're trying to do as a deformation.

Re. BA again.. it's not censorship, heh, just forum admins not knowing how to deal with spam properly.

[ratty redemption]

thanks kat, and now i know it's not a bug and more of a feature, i think i can work around that. what i'm attempting to do is analogous to stretching and distorting uv's in a similar way to 2d morphing and warping apps, and i guess, similar to pinning uv's in blender's image/uv editor, i might be able to use additional bones to help create the falloff effect from the main deforming bone. i'll try that this evening and report back.

re ba. my follow up posts finally did get approved, but only after i complained to wes, the site owner. and considering the registration yesterday was an unnecessarily stressful ordeal, just to get my username and account details accepted, i can't say i'm happy with the service they are providing at the moment. maybe that will improve over time, but for now i'm a lot more comfortable hanging out here on katsbits, where i know, you will always reply to users in need of help, as soon as you get a chance.

[ratty redemption]

a couple more guys over at the ba forums have posted and one of them, larry phillips, explained further about how the weights are divided between two or more bones, so that is definitely what i needed to work with. i'm doing some more tests but it seems like i'm back on track thanks to you guys, which is a huge relief as i've been struggling to understand this for about a week.

[ratty redemption]

here's an image of my now functioning test. it was simple once i understood what i was doing.