2
$\begingroup$

A visible seam is observable during SubD operations on a boolean cut.

Prior to applying the Boolean I created supporting edge loops which are not approximately* in line, they are exactly* in line, ie. mesh topology looks so good it brings a tear to the eye.

Despite this, for the past 2 weeks this issue persists. Weighted normals and proximity loops help the overall shape, but only minimize this particular effect. A blend file should be attached. Picture illustrates issues.

Issue Edge Norms Vert Norms

Is anyone able to identify this issue, and what the solution is?

$\endgroup$
2
  • $\begingroup$ Did u try increasing mean crease near that area? $\endgroup$ – JacksonPro Nov 23 '20 at 13:25
  • $\begingroup$ I did yes, thank you though. Most of those solutions only minimized the issue, apparently it was bad topo! Took me by surprise as everything was quaded rather nice, but the answer I've marked below shows the version which seems to work. I think the gist is to really drive home a full perimeter edge loop around any of these sort of things. $\endgroup$ – John Kowski Nov 23 '20 at 23:11
2
$\begingroup$

Your topology is not good, if we enable the Subdivision Surface modifier On Cage display option, here is what it gives, the central vertex is completely stretched, which gives the bad result:

enter image description here

If you choose this topology it will work fine:

enter image description here

$\endgroup$
1
  • $\begingroup$ So it is! I drummed up a different solution, but all of them seem to rely on having a proximity edge loop between the edge proper and the rest of the mesh to sort out the issues. I didn't realize that was, or could really, be an issue until today! Thank you. $\endgroup$ – John Kowski Nov 23 '20 at 19:35
3
$\begingroup$

Merge by distance M>B and change the threshold until you've removed enough vertices to minimize the seam.

Then add an edge loop to support the geometry in front of that area.

See the gif below for a visual:

enter image description here


Note that this will fix the issue for now, but in the future, you should consider better topology, and not strictly rely on booleans for modeling.

topology guides is a good resource to begin with, and there are many questions here on the site that will lead in the right direction also.

$\endgroup$
1
  • $\begingroup$ Hi! Thank you for the suggestion. I didn't use a boolean, but it was an easier way of describing the process. I find it odd though, I created and positioned loops to be perfectly perpendicular to the project, I'll need to look at my topo again to see what specifically was the issue there maybe? $\endgroup$ – John Kowski Nov 23 '20 at 19:32
1
$\begingroup$

Here are a few solutions:

  1. bevel around the area selected as shown below

enter image description here

  1. add an edge loop and move it the direction as shown below

enter image description here

$\endgroup$
1
  • $\begingroup$ Hi there, thanks for this! After posting this question those were some the things I tried, and they all helped indeed! But it seems the issue was with the way I set up topo here, I thought I had it set but no dice! Cheers $\endgroup$ – John Kowski Nov 23 '20 at 19:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.