Symmetry not working after duplicate-with-symmetry
Diomede
Posts: 15,173
I am trying to model a head in the vertex modeler in C8.5. Following a C5 tutorial, I created the basic forms of the face on half, then used duplicate-with-symmetry to generate the other half (including weld). However, after I use duplicate-with-symmetry (including weld), the symmetry feature does not work when I try to continue editing. Am I missing a step?
Symmetry_2.JPG
714 x 651 - 85K
Symmetry_1.JPG
609 x 663 - 51K
Comments
You need to use the "Center Symmetry Plane on Selection" command after performing an operation like that. Switch to polygon selection and do control-A to select everything, then choose the command from the "Model" menu.
When symmetry is active, you should see the green symmetry plane rectangle - and it is not there in your screenshot.
By the way,
Nice mug, Diomede! :)
I often have the same problem as Diomède, even when the center symmetry plane is visible.
It is also similar problem to select zones of growths of the hairs...
Thanks for the replies. Fenric, I did what you suggested but it still didn't work. Not sure why. Here are my steps (in Windows)
- create mesh half
- ctrl A (selects entire mesh)
- make sure plane selection is zy for symmetry
- ctrl click on vertex on edge (future middle) of mesh, moves working box plane to edge of mesh
- Duplicate with symmetry (and either let C8.5 try to weld, or select/weld after duplicate if I have too many vertexes within default tolerance)
- View, reset working box
[so far, this is what I had done before Fenric's suggestion]
- per Fenric, Ctrl A to select entire mesh
- under model menu, select Center Symmetry Plane on Selection
Then I chose a polygon and made sure symmetry was enabled, yet the symmetry feature still isn't working. Seems odd. If the mesh isn't symmetrical, seems like it would be Carrara's fault.
And to confirm Dudu_00001, I just tried the exact same steps per Fenric, but with a simpler mesh, but this time it worked. Argh.
Yes, sometime that work, sometime not...
Is that could depend of shader domain selected ?
Good suggestion. So I just tried deleting all of the shader domains except texture 0. Still didn't work. I have a few other ideas to experiment with, such as exporting the mesh as an obj and re-importing it. I don't hold out much hope, but if I identify what causes the inconsistent treatment, or a good work-around, I'll update the thread.
Has anyone tried Project Gemini from DCG? This is how it is described.
"Project Gemini is a collection of tools to assist symmetrical modeling. No more fumbling around with duplicate with symmetry and weld. The Gemini plug-in mirrors your modeling, welds the two halves together, and even cleans up the stray vertices in the middle saving you time and trouble."
Obviously, I'd rather C8.5's symmetry tool work correctly, but I don't mind throwing a little love to Fenric, DCG, and the other plugin producers when they solve problems.
Thanks. The tutorial I am following is from a handbook for C5 Pro that I found used on Amazon.
I couldn't get that first mesh in symmetry mode after duplicate-with-symmetry, even by exporting and re-importing as an obj. I then started over, and was able to create a headlike mesh that re-achieve symmetry, and could be attached to the body.
Very mysterious.
After some tests on a simple sphere, I saw that it is only symmetry on the axis X which does not function, Y and Z are OK…
I duplicated (without symmetry) the sphere then connected both by a bridge: X does not function, Y is OK but on the first sphere only and Z is OK… (See screen shots)
Then, I made a rotation with 90 degrees of all the object: Only Z functions…
I made the same test with "duplicate with symmetry", everything work except that the Z axis stay on the first sphere.
I do not understand the logic of this option…
Would it be possible to move the axis of separation (I don't thing so)?
Thanks for doing the tests, Dudu_00001. I've been doing some tests also. I haven't been able to reproduce the problem on a mesh that I begin in the same file session that I use the duplicate-with-symmetry. I wonder if it has something to do with how something is saved and reopened? My third try (screenshot below) worked without a hitch. I can only think of two differences between this third try and the my first try above (1) my first try had the eyes made separately then combined with the head mesh by exported as an obj and then re-importing as a single obj, and (2) the first try was conducted over 3-4 sessions while the one below was completed in a single modeling session. Other than one of those, I can't think of any difference between try 1, which lost the symmetry function, and try 3 (symmetry function continues to work after duplicate with symmetry). The mesh statistics should be visible in the screenshot.
Aha! It may have been the eyes. Here, within the same model, I've used the construct command to add identical spheres as eyes. The head still can use symmetry, but the eyes can't. I've selected a polygon on the forehead and the right eye to illustrate.
Did you weld the whole?
You are correct. Whatever is causing the problem in the first post remains a mystery.
as casual observer my guess is that after you duplicate the mesh you no longer are getting a symmetrical mesh, courtesy of a glitch in the vertex welding , assuming the modelling with symmetry needs a symmetrical mesh to work with