c000be
02-24-2009, 09:25 PM
painting subDs' is turning out to be a real pain, and i'm going to try one more time, but from what i'm reading, animating them is just as slow.
if i convert the model i have at my site to a poly, uniform/poly count using level 1, the model turns out to have 16,000 - 17,000 faces, half as heavy as i've seen in the hyperreal series for some of those models [the subD has 5648 faces originally]. It follows the smoothness patterns of the subDs, but i might have to add some vertices to the fingers, and take out some from the eye area, but that's it. [subD model - http://c000be.cgsociety.org/gallery/555991/
(http://c000be.cgsociety.org/gallery/555991/)
i read on one forum thread that using a poly model conversion as a wrap deformer can make pixels fly if the vertices aren't all the same [the TD guy was from WETA, so i believe him], and i originally wanted to use a subD so that i could use less vertices in areas where i wouldn't need them and more in areas where i would. now that subD's are turning out to be way more uncooperative than i expected, i'm finding i'm not alone in figuring this out. Also, i'm not using blendshapes - a few influence objects for breathing that will be manipulated by hidden rib joints on the spine [atleast that's what i've got so far] and the lip sync will be on-the-fly.
any opinions on this that i haven't read about, am missing the point on, etc.? Thanks.
if i convert the model i have at my site to a poly, uniform/poly count using level 1, the model turns out to have 16,000 - 17,000 faces, half as heavy as i've seen in the hyperreal series for some of those models [the subD has 5648 faces originally]. It follows the smoothness patterns of the subDs, but i might have to add some vertices to the fingers, and take out some from the eye area, but that's it. [subD model - http://c000be.cgsociety.org/gallery/555991/
(http://c000be.cgsociety.org/gallery/555991/)
i read on one forum thread that using a poly model conversion as a wrap deformer can make pixels fly if the vertices aren't all the same [the TD guy was from WETA, so i believe him], and i originally wanted to use a subD so that i could use less vertices in areas where i wouldn't need them and more in areas where i would. now that subD's are turning out to be way more uncooperative than i expected, i'm finding i'm not alone in figuring this out. Also, i'm not using blendshapes - a few influence objects for breathing that will be manipulated by hidden rib joints on the spine [atleast that's what i've got so far] and the lip sync will be on-the-fly.
any opinions on this that i haven't read about, am missing the point on, etc.? Thanks.