This is a wee discussion abotu a modelling method I've been playing with.
We've all seen new modellers use the cylinder extrude method, where all the limbs of a charcter look like tubes, and the polys look like a bunch of squares.
I've been working on a sort of tutorial to help people break that habit, by looking at the REAL for of what they are building.
I've found that a great method is to visuale the object as a series of cross sections, then extrued into those shapes, rather than extruding the limb as a cylinder.
For some people, it may be easier to actually build those cross sections, either as guide objects, or as actual geometry which can then be bridged to create the complete object.
Does any of this make any sense?
Replies
If this is a charcter modelling tutorial you are thinking of doing im all for it. its been awhile since I attempted a a charcter model of any kind and although my modelling has improved greatily in recent months I still haven't found a comfortable methoid of making organic meshes...so any new methods or techneques is a welcome thing from my point of view.
as for whether i understand what you are saying....I could do with an example or a drawing taht shows an example of these cross-sections.
Good luck with the tut if you decide to proceed either way.
john
An example is that a humans wrist is not a circular crosssection - its a rounded off rectangle. Halfway up the forearm is more penagon shaped. The entire arm has no circular sections, they are more like rounded off rectangles, but they twist around - the upper arms and lower arm are at 90 degrees to each other.
It sounds like it might be similar to Scott Ruggels' tutorial, except using extrude instead of patches.
I'm curious to see how you weld your sections together, since you'll have different edge counts where pieces meet.
Max7 has a really nice bridge feature that connects 2 set of open edges.
I guess with your method you avoid the Surface Tools unpleasantness of all those connecting splines, and you avoid the whole tri/quad-sided patch limitation...
I'm curious if you're finding yours to be faster/easier than the Silo/Modo approach.
It's hard to put down without actually building a model that way.
It's just simple things, like the calves of a person when wearing jeans actaully often hae a triangular cross section, simply due to the way the demin folds itself. Have a look at your legs next time you are walking.
I'm trying to think of a method to get people to look at the rough shapes of those masses, build the masses, then refine the model.