1
$\begingroup$

Is there a possibility to let a selection of vertices glide on their respective edges, so that they end up in a straight line? The edges upon which the vertices are to glide are not necessarily parallels. Please have a look at the attached picture.

enter image description here

You can get this simple example here:

background: The problem arises from "tidying" up a model that did not care about subdivision surface modelling at all. In order to do subdivision modeling I want to add support loops. By connecting vertices from outer and inner structures with the added loop I end up getting zic-zac-lines, which I want to straighten. See 2nd screenshot. Thank you in advance!

enter image description here

$\endgroup$
4
  • $\begingroup$ You can connect the end vertices with a straight line (j) and then dissolve the unneeded geometry... Why do you need this? I mean what do you care if the lines are straight? Wouldn't it be more logical to get rid of unnecessary geometry instead? $\endgroup$ Mar 17 at 13:07
  • $\begingroup$ You mean using the Knife tool; otherwise I would not get new vertices in-between. Dissolving the original vertices also removes edges. Why I need this? I am still learning, I might be wrong. But I think non-straight lines give different (unpredictable) results in subsurface modelling. Removing unnecessary geometry would lead to lots of ngons which again is not ideal for subsurface modelling. $\endgroup$ Mar 17 at 13:30
  • $\begingroup$ Geometry you have has lots of n-gons... It might be better to use bevels for something that has bevels and subsurf for organic forms. Subsurf creates loads of unneeded geometry in cases like this. I mean... There are no rules... you can do whatever works. If you select 2 vertices and hit j, the vertices will be connected in a shortest path on surface between them. Knife tool also works. $\endgroup$ Mar 17 at 13:42
  • $\begingroup$ Thank you. That's a very valid argument to limit subsurf to really applicable cases and that it's not a tool for everything. $\endgroup$ Mar 17 at 14:52

1 Answer 1

0
$\begingroup$

I do not know if this would work exactly the same in all cases you might need it, but for your example there is a relatively simple way to do this:

  1. Select the left and right vertex of the crossing line.

    select outer vertices

  2. Press J to join those vertices. The edges inbetween will be split automatically.

    join vertices

  3. In Edge Select mode, select the obsolete zigzag edges you do not need anymore and press X > Dissolve Edges.

    dissolve edges

  4. And that's it:

    finished result

$\endgroup$
2
  • $\begingroup$ Thank you Gordon! I think that solves my use-case. (It is similar to Martynas' proposal, to be fair). I also found out, that I can ring-select the obselete geometry by alt-clicking on one of the obsolete edges. So it can be an automated process. $\endgroup$ Mar 18 at 14:35
  • $\begingroup$ Yes, it is basically his proposal, but it is always good to have an answer that can be accepted, to indicate other users this question is solved (if they have the same problem) or for people wanting to help not to have to read through the comments to see it actually had been solved already. Also an illustrated answer sometimes helps more with understanding than pure text. But there are more ways to do this, I just decided to show the one I would use. $\endgroup$ Mar 18 at 14:51

You must log in to answer this question.

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