Once a fbm applied, cloth don't fit well

mh69arq

Hi all!

Modeling a cloth i found a problem when i try to make it conform to all basic shapes for genesis...

First, i try to explain my problem by schema (because of my englih), and don't show exactly what it is, what cloth or what part of body simply because i want to understand where is a problem, not resolving it by just an easy cheat for such cloth or other.

fig.a
1. base model (blue)
2. model posed (up the red line, the figure will be static)
3. model with FBM applied (in my case FBMheavy)

fig.b
1. cloth modeled on base model (yellow)
2. cloth modeled on "FBMmodel" (yellow)

without being fitted, the morph works and it's in position, and all seems correct.

fig.c
1. cloth fit to the model... ok
2. cloth fit to the model + FBMmodel applied... PROBLEM!

fig.d
Looking bones, the different alignement is causing the problem... at least is that i think...

fig.e
i resolve part of the problem adding an ERC to my FBM cloth... it works until i rotate the joint at the base.

i was looking the tutorials again and again, and don't find how to resolve this and i don't understand where is exactly the problem.

sorry for my english, i hope you can understand my problem. :/

Richard Haseltine

Ideally you should model around the base shape for the Transfer utility to work, but you can model around a morph as long as there is no scaling or translation linked to the morph and then set the shape you modelled around as the Source Shape in the Transfer Utility - select Morph for shape, then select your morph - the click Show Options and check Reverse Source Shape from Target. You can then use Morph Loader Pro to load the original shape as a morph with the same name (name, not label - you can obtain this from Parameter Settings by going to the gear icon at the end of the morph's slider) and setting Overwrite existing to Deltas Only.

mh69arq

Hi Richard,
i forget to tell that i've already use the 'transer utility', works well for a majority of basic fbm genesis (originals). But for some reasons, and that is my problem, for some fbm like FBMHeavy that don't work. It's surely due of a scaling or translation in that FBM like you said, then i think about same problem with other figure like victoria5... :/

I managed to make my morph (FBMHeavy) works Adding Some ERC after using a 'transfer rigging'. the result was like fig.e, that's what i want, the cloth Correctly Posed and morph but for some reason, and just in only one joint (one that is at the base of the part that moves and that remaining static fig.e). If i plan to not move my figure, it's ok. but i would at least can move my figure at this point fig.e

Is there a possibilty adding an ERC with the FBMheavy created for my cloth and keep that's bones positions? i mean, after fit, if i select genesis, change my shape to Heavy, the bones keep the position of the ERC added... and not changed by the FBMHeavy of the genesis that apparently leads all after the cloth fitted.

hope you can understand what i mean... i think it's not very clear...

thanks.

Richard Haseltine

I know Heavy is quite a tricky morph to support, which is why it often isn't supported, but it isn't affecting any scales as far as I can see and the joint centre adjustments should be handled by DS (though it is true that it does change the orientation of the chest and abdomen bones - that may be the issue).

kitakoredaz

It seems more simple problem,, not about FBMheavy used rigs rotation (but not change poze)

even though the FBM do not rotate rig, when you make morph for pozed figure, it should happen.

The porlbem of OP is, fig b.

he said cloth modeled on FBMheavy. but no.
he modeld for FBMheavy + pozed shape. (rotate rigs for pozing)

the rotaiton when pozing rig, and the rotation when set joint for zero pose
to fit morphed mesh shape has diffeernt meaning.

if OP try to make morph shape obj, which fit to current pozing shape,
OP can only apply delta from current pozing and morphed shape.
and usually plan to set ERC for rotattion of the rig for pozing.
so that, when rotate rig, the delta morph work well,
and may plan to set ERC for FBMheavy too.
to work it when apply FBMheavy only.

Then actually he did so. then it work for the pozing.

But it only work well for the pozing (or the node rotation of the axis, he pozed)
but ERC keep step value , so that sometimes it is not so smooth transition.
you may need many point for one rotation. or you may need many anotehr ERC morph for each rotation.
(so that, we do not plan it usually, I sometimes try them but it can never perfect fit clothing for all pozing)

OP seems try to use the obj made for pozed shape with FBMheavy,
to overwrite FBMheavy.

but it is delta morph from zero poze sero shape. so that it can not work as he expected.

kitakoredaz

To simplyfy what I mean,

1. I model obj for FBMheavy shape.
2. use TU to set rigs from genesis with FBMheavy
with reverse defomation, it may cause some problem for zero shape genesis.
3. usually reshape the obj for zero shape genesis again.
4 use TU to set rigs from genesis zero shape.
5, but next when I apply FBMheavy to genesis, the clothing seems not fit well.
6, but I have already made good shape which fit to FBMheavy zero pozing, by step1
then load the obj to overwrite FBMheavy (which auto generated from genesis)
7, now my clothing fit for zero shape genesis,
when I change genesis shape, my handmade morph(FBMheavy) overwrite original,
then the clothing fit to FBMheavy without problem

(As rechard said already, it do not use scale, or translation rig for zero shape,
so that there is no problem)

8 But,, when I pozing genesis with FBM, it cause some problem my clothing
it is weight map problem. though current rigs of clothing and genesis rig position, is
perfectly same, (it controlled genesis rigs, and at start point, it transfer genesis rigs)

9 I may try as same as OP did. it means, make delta morph from current shape and current pozing.
but it can not be used as FBMheavy. it need to work with FBMheavy shape.

my conclusion is, OP need to make for zero pozing with FBMheavy. at start point.
then adjust shape with each rig rotation when pozing.

mh69arq

Hi kitakoredaz, and thanks for the explanation, sorry to respond late but i wasn't at home...you seem to have understood the problem... following your steps, i see my problem with an other point of vue... now it's resolved, but i am still in confusion, it's hard to deal with bones (positions, rotations... values can change with some erc and morphs), then i resolve it with cheat and the situation is still no clear...

i've just reorient -without use of TU this time, and not precisly...- the bone once fitted with the FBM... and just erc this change... it's seem to work properly, but it's like a cheat because having precise values i think that the good solution is somewhere else. For the moment i can deal with that but i will study this situation again to have the exact step to do.

thanks again for the help!

kitakoredaz

I do not know it is right or not,
but I think the reason why these problem happen,,,

When we make morph for pozing clothing, if the rigs do not move at all, it must work as we expected.
but as you know when clothing fit to the Actor, the rigs controlled by genesis with FBMs shape rigs.
not default genesis rig. and these FBMs usually change genesis rig position.

Complex point is,, even though , rig moved by FBMs, if clothing keep zero poze,
there is no influence from weight map. so that we can overwrite generated morph with zero poze.

but,, if the clothing are pozed, (that means rotate already)
your clothing shape, already have been influenced by weight map from current rigs rotation.
untill it fit to Actor.

then when clothing fit to the Actor, the rig jump to another place.
and pivot point of node move. it means the influence of weight map change too.:roll:
I think it is criminal.

to remove problem,, I have no other way,,
1 to make morph with zero poze, it work when we hope to overwrite generated morph.
2 simply make delta morph for clothing, keep fit to Actor and current poze.
(it means, rig position never change, and we do not consider weight map influence)