Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
Several weightmaps on hip such as bulgeLeft_rThigh_jointz has one verticle(v 129) on the center. The symmetrical weightmap doesn't have it. And I found it meaningless to look for more of such problems.
Besides verticles and morphs, weightmaps are also not reliable. How does DAZ create a model? I believe the best way is to make half of it and mirror the other. Why does DAZ use a not only incorrect but also slow way?
Aiko 6 is a morph (so can't cut in half and mirror) for Gensis 2 Female so weight maps aren't the issue as other characters based off G2F don't show this behavior.
Morph can be mirrored. Open the cr2 file with a text editor, you will see the deformation values for each verticle, and the index of the verticles are the same as the original G2 model. It should be easy for a famous company like DAZ to do it automatically by writing a program. Even as an amateur, I repaired the the obj and morphs half automatically by processing the data through Excel/VBA.
I don't believe the weightmap issue only happens for A6 unless whoever shows me that the weightmap of hip part doesn't have the rubbish verticle I mentioned for another G2F-based model. Even though the problems in A6 are obvious if the mesh mode is turned on or you have a look into the cr2-file, few people realized that.
A6 doesn't have a Poser cr2 file.
As I already explained in the 1st page, Daz studio has cr2 exporting function and a selectable format option for poser. Either the asymmetry problems exist in the original data for models, or Daz studio is a bug-creator.
The mesh distortion is in A6 in DAZ Studio but not the base G2F nor any character I dial up from the body morphs. Morphs do not change the weight mapping of the base. DAZ created the DSON Importer because cr2 export did not work very well with the Genesis line so problems with converted figures is IMO opinion not relevant. There is a problem with A6, it can't be the weight maps so it must the JCMs or similar correction morphs
"The mesh distortion is in A6 in DAZ Studio but not the base G2F" -- I already confirmed this in the 1st page. I said G2 obj is symmetrical.
"Morphs do not change the weight mapping of the base." -- That's why I said recently that the asymmetry problem is not only for A6 but in the way DAZ creates models.
"DAZ created the DSON Importer because cr2 export did not work very well with the Genesis line so problems with converted figures is IMO opinion not relevant." -- For older models before G3 which poser is supposed to support, I don't find any problems caused by poser. Problems stay in the original data(poser before 2014 doesn't have subdivision like DSON importer but this has nothing to do with the problems we are discussing). The difference between using cr2 and DSON importer is that while using the importer you are less sensitive to potential problems because you can't see the original data. You don't find problems in other G2 models because they are not so obvious like in A6, just like you don't find any problems for Newton's law in normal speed and scale. Or you just ignored them like others did for A6.
Now even in the original G2 I found the rubbish data which is very unlikely to be created by poser from nothing. That means the way of DAZ creating models is problematic. That's the cause of serious problems in A6 and not so serious problems in many other morphs.
As an example, in actor hip, jointX rThigh_jointx uses
weightMap 1489059464
{
numbVerts 833
v 129 0.009781
v 432 0.085527
v 433 0.023285
v 437 0.166629
v 439 0.022507
v 440 0.005707
v 449 0.008667
v 453 0.020279
v 457 0.043137
v 458 0.013230
v 460 0.044160
v 461 0.003845
v 462 0.030747
v 463 0.099489
v 464 0.081483
v 465 0.010758
.......
Veticles with index number larger than 430 are all in the right side. Index number smaller no greater than 430 denote verticles on the left except number 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 103, 127, 129, 167, 168, 177, 182, 291, 292, 297, 303, 307, 323 which are on the center line. The first value in this weight map is v 129, the only verticle not on the right. The weightmap for jointX lThigh_jointx has one less value and that's v 129. For a symmetrical structure, this point should be in both thighs or in neither.