Mesh Topology HELP!!

[Fede]

Ok, so i gave it another shot after checking some links on edge loops.

Can anyone confirm if i did the changes correclty or if they aren't right yet.

Thanx

Link to changes (http://www.psend.com/users/fredster/mesh2.htm)

[actarusprocyon]

A:First of all,on this one,you got a Six edged vertex,you will need to dissolve/remove one of them and add and reconnect,maybe add an edgeloop.

Another of the things I do to eliminate triangles is to first delete it and extrude one of the edge that makes it,then reconnect to surrounding vertice keeoping all quads and maximum 5 edged vertice.

Basically,once you apply this you should have many problems doing the rest.

[actarusprocyon]

About the changes,it actually depends on what results you're expecting from the sub-D mesh,if that's what you want.

Usually,being more of a left-brainer,I toggle between subd and cage a few times until I'm satisfied with the shapes.

[Fede]

Thanx for the reply.

Well, the smoothed mesh works fine, and is it will not deform it will work but i'm not just trying to get the model done i'm trying to learn how to model in a good way and i want it to be done right.

i was wondering if my changes were a form of "correct topoology" and if seen as good or bad modeling and if there are better ways of sorting them out.

Also i want to know what would be the right way of going from a high density part of a mesh to a low density part. what is the middle point supposed to look like?

Thanx

[Equinoxx]

not the most elegant solution, but it works to keep them quads.

hard to give a right fix without really knowing what the model is and what it's gonna look like.

also, based on these views alone, it looks like a dense mesh for level-0

hope it helps you out somewhat, if not, let me know and i'll have a more indepth look tonight when i'm at home.

Cheers.

- Qui

[lildragon]

Fede Equinoxx had the solution I would've used. Also yeah there's updates on the Kar stuff happening alot lately and going onto May 04, sign up on the mail list or check the site.

-lild

[Fede]

Thanx for the replies guys i really apreciate it and i know how busy you all are.

I'm just wondering now from Equinoxx's reply is my base cage too dense and how much must i bring it down by? I think the reason why it is so dense is because i was trying to keep all quads by having all the deges align.

I think the bottom is more dense that the top half, so i'll try clean that out.

here is a link to what i'm trying to get to with other grabs of my initial mesh.

Link to concept and other mesh shots (http://www.psend.com/users/fredster/concept.htm)

Thanx for all our help guys.

[Equinoxx]

Again, a quick fix, but I think you can safely collapse the co-colored loops into one loop, assuming you're gonna use meshsmooth at the end to make it a nice rounded and smooth model.

might be 2 of those blue lines instead of one, but thats a matter of trying.

[Fede]

Equinoxx, that is great thank you so much for all the help.

I'm going to reduce it as much as i can maintaining the shape and good smoothing quality.

I'll be using the nurms subdivision which i assume has the same end result as Meshsmooth or i'll add a meshsmooth operator ontop if needed.

Sorry for this silly question, but when you say cage 0 do you intend the most basic mesh with out smoothing and say cage 1 would still be with out smoothing but more detail? and so on and so forth?
Obviously as you add more detail to the cage it's going to get denser right?

Thanx again, i'll go home a bit more confident now.

[Equinoxx]

level-0 means pre-meshsmooth/nurms

and yes, it does get denser by adding more detail, but for such a 'non-detailed' carbody it has imo to many subdivisions/edgeloops.

[Fede]

Got it, thanx.
I'm just itching to jump on it now.

Hope to have something tomorrow or thursday latest.

Cheers

[Fede]

Ok so i'm back, but satesfied with the progress. I can't stress how much the help received has really contributed to my modeling skills.

So here is the base mesh now:

Link to changed mesh (http://www.psend.com/users/fredster/changes.htm)

I think it's nice and clean and i've started detailing it already and hopefully the next time i have a decent post it will be when it's complete.

Just a few more questions, i have attached a pic with the changes Equinoxx suggested which worked great but what if i only have one edge like in the example i made, how do i get that to blend correctly into the topology?

Also in the bottom pick i tried to illustrate where i see some funny smoothing even though there are only quads and no tris or ngons. why does that happen is it a polar and is it bad?

Link to example:
example link (http://www.psend.com/users/fredster/help.htm)

Also is there a place i can learn how to remove tris, ngons and other non quad issues or just trial and error till i learn 100%?

Thanx for all the help guys.

[thorn3d]

Originally posted by Fede
Also is there a place i can learn how to remove tris, ngons and other non quad issues or just trial and error till i learn 100%?


It's mostly trial/error learning, as each mesh is different. The key is to learn to predict how what you're doing now will result in, 2-3 steps from now... almost like playing chess; you're always thinking several moves ahead.

thorn

[BnE]

"Also is there a place i can learn how to remove tris, ngons and other non quad issues or just trial and error till i learn 100%?"

go there :
http://www.suture.net/tutorials/modeling/page09.shtml

it's about lightwave but it's not really software specific. since LW doesn't handle ngons and doesn't smooth well tris, the technics he explains are extremely usefull.... browse the pages and you'll find what your looking for. Click on the pictures on the right, most of the time, they are videos...

hope it'll help....

see ya.

[Fede]

Thans for the replies guys, much appreciated.

BnE, i think that looks exactly like what i'm looking for especially the tip on how to end edge loops and also the fixing cut quads section.

Again, thanx and later.

[CGTalk Moderation]

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.