c000be
06-08-2007, 07:35 PM
[see/click thumbnail below]
I'm converting/modifying models to poly to subD for final animation. In order to get good, texture applicable meshes for the subDs, I have to convert the nurbs to poly, attach everything in poly for one half of the model while doing a rougher but accurate modeling of the nurbs using the orignal nurbs model as a mold [put on a layer on 'reference', place poly over it and just pull vertices into position], and then after that convert to subD to get back to the original Nurbs smoothness using the same method with the original nurbs model again.
this is really the most painless and most accurate way I found to maintain the organic quality while keeping count of vertices/patches. Basically, keeping it as close to the original nurbs counts as possible. Of course, to apply the materials to an attached mesh, that needs some adjustment:
In the pictures below are the original hand model, the conversion to the first poly with a bit of vertices pulling to get the checkerboard texture a little more evened out in the middle, and almost completely evened out atleast horizontally on the right.
My question is, I'm trying to keep from having to do too much UV vertices pulling by having a good wireframe up front. Given that the palms have to match the fingers vertically in terms of lining up [the hands get used a lot in the animation, that's why I'm basing the rest of the model starting with them], should I have pulled the vertices horizontally as far over as I did? This is an organic model, and I don't want any unnecessary problems. I will have influence objects under the skin in the obviously bulgy parts. When this model is converted to subD, what you see now would be 0 level, and there would be 2x the amount of vertices for the 1 level to match the original nurbs, so you can see why I'm trying to be a sparse as possible.
thanks for any help, and I hope I've explained this adquately, not too much.
I'm converting/modifying models to poly to subD for final animation. In order to get good, texture applicable meshes for the subDs, I have to convert the nurbs to poly, attach everything in poly for one half of the model while doing a rougher but accurate modeling of the nurbs using the orignal nurbs model as a mold [put on a layer on 'reference', place poly over it and just pull vertices into position], and then after that convert to subD to get back to the original Nurbs smoothness using the same method with the original nurbs model again.
this is really the most painless and most accurate way I found to maintain the organic quality while keeping count of vertices/patches. Basically, keeping it as close to the original nurbs counts as possible. Of course, to apply the materials to an attached mesh, that needs some adjustment:
In the pictures below are the original hand model, the conversion to the first poly with a bit of vertices pulling to get the checkerboard texture a little more evened out in the middle, and almost completely evened out atleast horizontally on the right.
My question is, I'm trying to keep from having to do too much UV vertices pulling by having a good wireframe up front. Given that the palms have to match the fingers vertically in terms of lining up [the hands get used a lot in the animation, that's why I'm basing the rest of the model starting with them], should I have pulled the vertices horizontally as far over as I did? This is an organic model, and I don't want any unnecessary problems. I will have influence objects under the skin in the obviously bulgy parts. When this model is converted to subD, what you see now would be 0 level, and there would be 2x the amount of vertices for the 1 level to match the original nurbs, so you can see why I'm trying to be a sparse as possible.
thanks for any help, and I hope I've explained this adquately, not too much.