Hello,
I'll repost it here since it seemed to be in the wrong thread before.
Is Sub-D Modelling the recommended way to do game models?
For hardsurface stuff, is it not easier to just use whatever 3dsmax or other programs offer. Like Boolean, Bend, Lathe, attach primitives to one another until the overall shape looks good and then remesh the whole shabang in Zbrush/Mudbox/retopo with graphite tools?
It looks like it is pretty complicated to model elaborate forms with the Sub-D modeling technique although the topology is admittedly beautiful ... is it time-effective or is it even the recommended (or only) way to do?
Thank you,
Coffee
Replies
Of course you can use any modeling method you want to get the job done(especially for high-poly bake meshes) but, really, you are limiting your skillset by avoiding learning sub-d. Not just for the flexibility you will have modeling but for other reasons: understanding topology(very important), making life easier when UVing and skinning, being able to create optimised topology by understanding it, showing potential employers/clients that you know what you're doing, and basically being the difference between a good modeller and a below average one. Also, other techniques like opensubdiv and creasing rely on an understanding of sub-d fundamentals.
Deforming meshes(organised topology) such as characters obviously need suitable topology and here again is where the fundamentals of sub-d's will aid you.
There are so many modeling methods now that artists apply - from sculpting to CAD modeling(Fusion360) that often beginners may get confused when they see what they think is an easy option like booleans, and wonder why they should bother learning sub-d.
But I say learn it. Learn them all. Fill up your toolbox and you will simply be a better, and more productive, artist.
look at this, maybe it could help you.
http://wiki.polycount.com/wiki/Subd
you can do the high-poly however you like.
Sub-D, bool and dynamesh, maybe even CAD and dynamesh or simply turbosmooth. I use a combination of all of these. Sometimes you need to find a certain shape so you just bool and dynamesh to get a complex form really fast and try different approaches without getting headache about topology. Sometimes you know exactly what your going to model but you want to tweak it later so Sub-D would be the right choice.