Difference Between HyperNURBS and subD's?

[ThirdEye]

Maya has two types of subdivision surfaces:

1) smooth proxy: more or less the same as Hypernurbs, but less flexible (no weighting a la CINEMA 4D, no possibility to toggle the same mesh between smoothed and low poly, it creates a new mesh, no isolines editing)

2) subdivision surfaces: they're claimed to be the Pixar's implementation (personally i doubt it), they are hierarchical (multilevel of subdivision) and they're subdivided at rendertime if you use Renderman afaik

[RorrKonn]

If ya bending a human HyperNURBS acts alot more like skin u can weight n make smart skin n all.

RorrKonn
http://www.atomic-3d.com (http://www.atomic-3d.com)
TS 6, C4D 9, zBrush 2.

[kuui]

3rdeye is right.

_real_ SDS are pixar's invention and they are the only type of geometry along with NURBS that renderman handles very very fast.

[Rev9]

Does this mean that real subD's are not the equivalent of the HyperNURBS object? IOW, I cannot take a C4D HyperNURBS cage mesh and in Maya convert directly to subD's?

Best
BT

[ThirdEye]

you can, but there are really no advantages of so called "real" subds over traditional ones, in fact handling them is quite slower

[kuui]

you'll only need real SubD's if you dont like NURBS but want to render efficiently with PRman.

[Rev9]

you'll only need real SubD's if you dont like NURBS but want to render efficiently with PRman.


This was what I was wondering about.. The workflow would be ->Model HyperNURBS cage in C4D as mesh so it looks like you want it to be-->Export as .fbx (assuming animation)-->import to Maya and convert to subD's-->apply materials and render in PR Renderman?

Will the subD model in Maya look the same as the HyperNurbs model in C4D?

Does PR Renderman render out the subD surface the same way as it does NURBS? By this I mean that at render time PRRM breaks each pixel into a micropoly for a NURBS surface having infinite smoothness regardless of how close the camera gets..with no poly edges..is this so with subD's?

Best
BT

[LucentDreams]

transfer the low rez cage alone, the apply subdivision surfaces to the low res model in maya if you need.

[shakes]

transfer the low rez cage alone, the apply subdivision surfaces to the low res model in maya if you need.

I've got a friend who uses 3dsmax, and my cinema models export perfectly to him using this method.(as wavefront/ .obj for max)

[JIII]

There really isn't a reason to export with HN because it should convert it into all polys, a HN object converted into polys is basically uneditable.

[LucentDreams]

I've got a friend who uses 3dsmax, and my cinema models export perfectly to him using this method.(as wavefront/ .obj for max)

the issue isn't what format to move, he's concerned about having the model inside an HN move over to Maya and use maya's subdivision surfaces (like having a model with mesh smooth in max) Problem is obj and other formats only carry the model not subdivision modifiers so you need to transport (using OBJ or FBX or whatever) the lowres model, from on to the other and reapply using the other applications version of a subdivision modifier.

[Rev9]

the issue isn't what format to move, he's concerned about having the model inside an HN move over to Maya and use maya's subdivision surfaces (like having a model with mesh smooth in max) Problem is obj and other formats only carry the model not subdivision modifiers so you need to transport (using OBJ or FBX or whatever) the lowres model, from on to the other and reapply using the other applications version of a subdivision modifier.


Yes. I knew I would not be taking the HyperNURBS object but just the poly lowrez cage into Maya to get it into PR Renderman. I was concerned about how this cage translated into subD's in Maya and whether it would be able to render them with the speed it is known for for this type of geometry. I believe I have my answer. Thanks folks..

best
BT

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.