04 April 2007, 12:16 PM
Hi there, noke!
Firstly, you model is looking pretty damn good in terms of topology.
The shoulders are definately a tricky area and in fact one in which I'm having some problems myself. Dont have any pictures here to show you, but the main thing you want to do when you join the polys together is to get the edgeflow from her breasts to go up over the shoulder and across the shoulder blade on the back.
In terms of anatomy, the bottom of the pectoral muscle meets at the armpit and the front of the pecs sorta flows over the shoulder. So thats more-or-less what you want to aim for.
With regard to your main question of how to approach modelling with good topology....
...thats also tricky to answer. ;)
It comes down to a lot of personal preference, and I for example prefer starting off with a cube with some divisions and building the torso from crotch to neck with that.
I try to keep this stage VERY low poly because that way you dont commit yourself to any shape to much, and it makes it much easier to change the way your polys are split.
In fact in general, at this early stage dont add extra geometry unless its for a good reason. eg, If you're adding a new edgeloop because you have a tri or n-gon, first see if you can rearrange surrounding polys to fix your problem.
Then, i extrude the arms and legs out about 6 times (shoulder, mid-arm upper, before elbow, elbow, after elbow, mid lower arm, wrist).
I've found that thats generally good enough to work on the low poly character, cos smoothing will give you the extra geometry for deformation.
This is a handy Thread thats dicussing Body Topology and also discussing whether one should always aim for quads, or whether its ok to have tris and n-gons in your low poly model.
Dont know if thats helped much, but Private Message me and I'll hook you up with some images at least.
04 April 2007, 08:27 AM
Well, my advice to you is, yanno, good topology is good, but good final results are better. No one is gonna give you a hard time about tri's if your models are freakin amazing. Look at some stuff stahlbergh does. He has quite a good bit of tris in his models, but i mean, look at the end result. Its always good to keep good topology in mind, but the final result is always what you are going for in the end.
04 April 2007, 05:33 PM
thanks guy ...i'll keep on working with character model .. i think it's the hardest thing to model
practise make perfect ...like many of you guys once told me
04 April 2007, 05:33 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.