0
$\begingroup$

I was following this tutorial for making a 3d avatar from scratch, and found that there was an artefact on the nose of my model. The modifiers used are the mirror modifier along the x axis and the subdivision modifier.

0 subdivisions: enter image description here 1 subdivisions: enter image description here

Can anyone explain why this artifact is occurring and what I can do to remove it?

$\endgroup$
2
  • $\begingroup$ Do you have clipping turned on on your mirror modifier? It looks like the vertex just up from the nose tip is "not connected" to the mirrored side. $\endgroup$ Commented Jun 22, 2020 at 0:13
  • $\begingroup$ Unfortunately, my mirror modifier's clipping is turned on. $\endgroup$ Commented Jun 22, 2020 at 19:01

2 Answers 2

1
$\begingroup$

Catmull-Clark subdivision works well-- meaning, it makes smooth meshes-- under certain circumstances. One of those circumstances is roughly planar quads. Here, your quads are most definitely not-planar. How can we tell? The biggest clue is that you have two quads that share three vertices. That should never happen; adjoining quads should only ever share two vertices (ideally, for a 3- or 4-pole) or one vertex (for a 5-pole). You have a pair of strongly concave faces.

We can fix it just by insetting a new face in those two, then scaling that new face down in it's normal Z axis to make it more planar:

enter image description here

This does come at the cost of creating non-planar 3-poles, which are to be avoided for C-C subdivision, but are not as serious of a problem.

But also, poles can be moved to better locations; the more planar the neighborhood of the pole, the better. Techniques for moving poles are dealt with elsewhere so I won't cover them in this answer.

$\endgroup$
0
$\begingroup$

i had the same problem a few months back, my solution to this was checking for duplicate vertex, check if a face on the inside of the mesh is preventing the merge, and recalculate normals, that or your vertex isnt aligned to 0 properly.

most of the time i fixed it by recalculating normals with Shift N hope it helps

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .