PDA

View Full Version : Catmull-Clark Madness!!!



Remoc
05-29-2014, 06:12 AM
Hi Lightwavers.

I have attached a .zip containing a model of eye-lashes and it's texture.
When I press [TAB] with Catmull-Clark set, this model goes nuts! ( see screen cap )

Can anyone advice me of how to fix this?
Or repost the lashes fixed ( subpatched with CClark )

Cheers,

Remoc

Remoc
05-29-2014, 06:27 AM
Ok, I fixed straight after I posted.

When I made the original right lash it was nicely working and subpatched Catmull-Clark.
I then used mirror X
This command mirrors but also auto merges points.
This cause the above madness.

Remoc

Snosrap
05-29-2014, 06:30 AM
Catmull-Clark is pretty much useless in LW. :)

Tobian
05-29-2014, 07:26 AM
In this case it's the auto merge which did it. You also can't apply mirror to any object which has edge weighting. Apply any edge weighting after mirroring.

Snosrap
05-29-2014, 07:06 PM
In this case it's the auto merge which did it. You also can't apply mirror to any object which has edge weighting. Apply any edge weighting after mirroring.Yep - but a real workflow hindrance. Wait a minute - Sensei is about to step in and tell us about his tools that don't break CC edge weighting. :)

hrgiger
05-29-2014, 07:33 PM
Catmull-Clark is pretty much useless in LW. :)

Well I wouldn't say that but unfortunately it's one of those things that was implemented in LightWave and seemingly forgotten. It has breakage issues as mentioned and its a bit under performing but edge weighting is useful. Subdivision of ngons on flatter surfaces can also be useful but I really hope they consider going with Open Subdiv in the future.

Snosrap
05-29-2014, 07:53 PM
but I really hope they consider going with Open Subdiv in the future. Yeah - that's fine by me too. :) I really don't care about what kind of algorithm they use - as long as it works. But why re-invent the wheel and it just makes so much sense for interopera'bility.