PDA

View Full Version : Sub-D Modeling...Help, Please!

chrisrezende
03-24-2010, 11:38 PM
Hello everyone!

I'm starting with sub-d modeling and I'm in serious trouble already. When I studied a little of subdivision modeling theory last year, I could figure three common ways of modeling with this technique. Now that I'm delving myself into the practice, I really don't know what way I should follow. Thoses ways are listed below.

1) Artist models an object, adding major forms and details then subdivides the mesh to the desired level (i.e. 2 levels of subdivision) and sticks with it for unwrapping, animation and rendering.

2) Artist models an object, adding major forms and details then subdivides the mesh to level 1. Then he/she works on the object at that level, adding more detail and removing unnecessary edge loops. Then he/she subdivides the mesh to level 2 and sticks with it for unwrapping, animation and rendering.

3) Artist models an object, adding major forms and details then subdivides the mesh to level 1. Then he/she works on the object, adding more detail and removing unnecessary edge loops. Then he/she subdivides the mesh to level 2 and works on the object again, removing unecessary edge loops. Then he/she takes it to unwrapping, animation and rendering.

Honestly, I'm pretty worried about ending up with too many unnecessary loops/polygons on my models. So, which way is better? Which way you guys use?

Thanks in advance.
- Christiano
Kien
03-25-2010, 02:25 AM
I always thought sub-div was something you did as a last finish after you had all shape done. Or you had the mesh in shown in it's sub-dived shape all the time and worked from that. The later is troublesome with selection in 3ds max.
ezekiel66
03-25-2010, 10:35 AM
You seem to have misunderstood something here. "Subdiv modeling" and "subdividing" the mesh is not the same thing. You can model using subd surface without subdividing the mesh once, modeling with subd surfaces has - although the terminology is somewhat misleading - nothing to do with subdividing the mesh (unless you are talking about the subd surfaces in Maya which actually can have multiple subdivision levels. I never used this feature much though).

Subd modeling is basically the same as poly modeling. You build your model with polys, then convert it to a subd surface - or simply press the TAB key (Modo) or the PgUP/PgDown keys (XSI) to switch back and forth between the poly model and the subd surface.

The poly count of the subd cage remains the same as your original poly model after converting it to a subd surface.

Model your character or whatever using polys, convert it to a subd surface to check what it looks like subdivided/ smoothed out, switch back to poly mode and refine/ e.g. add edge loops where you want harder edges and so on. Create all polys yourself, add edge loops only where needed. This way theres no risk to end up with more polys than you want.
CapnPanic
03-26-2010, 05:53 PM
Assuming you are poly modeling with the intent to render as a subd, then I don't think any of your three options are really the best way to go about it. I generally just refine the areas that need it as I go, adding edge loops, etc, as opposed to subdividing the whole mesh. I'd much rather spend my efforts adding the edges rather than having to spend my time removing a bunch of extra loops.

But, I also do much more hard-surface modeling than I do characters, so perhaps a character modeler would prefer all those extra loops? For what character work I have done in the past, I still like the model to be lighter overall and only have the edge loops that I need.
CGTalk Moderation
03-26-2010, 05:53 PM
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.