CGTalk - Low-Mid poly Character edgeloops, best practices?

View Full Version : Low-Mid poly Character edgeloops, best practices?

ChrisDeKitchen

01-07-2008, 06:59 PM

Hi, I've seen several experienced people say that edgeloops/polyloops should follow the general flow of a characters muscles when it comes to subD modeling, but does this also apply to more low/mid-poly models? (10000-15000 tris) Models that mainly are intended for game purposes.

It seems to me like if you don't follow the flow of muscles etc on a low/mid poly character you'll be able to save a lot of polygons... However, what I'm uncertain of is what is best for animation? When there's not as many polygons to go around could it actually be better not to follow the flow of muscles etc in order to get a better result out skinning/rigging/animation? I'm mainly thinking of characters that are human-like since that's where it's easiest for the average person to spot problems.

I'd love to hear some proffessionals thoughts on this, or if someone had a wireframe screen of some great looking animated games character that would be awesome.

Rainroom

01-08-2008, 07:04 AM

i dont understand how you can get great results in skinning/animation if you dont follow the muscle anatomy correctly with edgeloops :shrug:.

NikLG

01-08-2008, 10:24 AM

Firstly, from my point of view / experience, I wouldn't say that 10000 - 15000 polys was 'low to mid'. Mid maybe, not low by any means....

Anyway, you could make a modle that doesn't follow the flow of real topology ( muscle areas and joints ) but it wouldn't look as good as one that does, and it wouldn't animate as well either. So it's not generally a good idea, unless your are making a really low poly character ( 500 polys or something ).
You can actually SAVE polys by following the topology properly because the basic shapes and joint areas would be in the correct place so it would read well enough to support less polys....

ChrisDeKitchen

01-08-2008, 09:08 PM

Well, I wasn't thinking of completely abandoning topology but more out of a detail perspective. Say in the neck for example, the most visible muscle runs diagonal from the back of the head down to the top of the chest. I have a hard time seeing that it could actually look better to follow those muscles with geometry when animated, unless you have a great deal of polygons in that area. But I duno, I never tried it, that's why I'm asking... I suppose I could make a rig and try different approaches. I don't really know what polycount I'll end up at, at this point... I'll simply do whatever looks good, while still keeping it optimized. Best guess, 7000 tris perhaps. Maybe it's a good workflow to rig your character while still at the modeling stage so you know how it will animate before you get to texturing and UVmapping etc.

CGTalk Moderation

01-08-2008, 09:08 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.