Page MenuHome

Multires not smoothing right
Closed, ArchivedPublic

Description

System Information
Operating system: Windows-10-10.0.19041-SP0 64 Bits
Graphics card: GeForce RTX 2080 SUPER/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 460.89

Blender Version
Broken: version: 2.93.0 Alpha, branch: master, commit date: 2021-02-08 23:42, hash: rBeab9165c2512
Worked: Not sure, same behavior in 2.92 that I have.

Short description of error
Multires modifier is distorting the mesh in the attached scene. Disabling it and adding another Multires has the same behavior, however deleting the misbehaving modifier and adding it again makes it right. This is a retopo mesh and I had a Shrinkwrap modifier when I noticed the issue, though the distortion remains with just the mesh and the Multires.

Exact steps for others to reproduce the error
You can see the behavior on opening the attached file.

Event Timeline

Germano Cavalcante (mano-wii) changed the task status from Needs Triage to Needs Information from User.EditedFeb 10 2021, 9:07 PM

Disabling it and adding another Multires has the same behavior

This seems to be the expected behavior of multires.
Perhaps it was not supposed to be possible to add two multires modifiers (@Pablo Dobarro (pablodp606) is this normal?).

In any case, we cannot know what happened in that mesh without being able to reproduce the problem from scratch.
Even because that result is not impossible.

There are some parts of the code that take into account multiple multires modifiers in the stack, but as far as I understand the idea is to always consider the first multires modifier as the "active". I'm not sure if allowing this in the UI has another purpose.

As long as you keep adding multires modifiers, they will always get the same data from the same CD_MDISP datalayer. If you remove all of them, then a new multires modifier will create a clean CD_MDISP with no displacement info on it.

The multires modifier in that file is working as expected, it contains valid displacement data and the limit surface is also right (you can test this using the erase displacement mesh filter). Those artifacts are correctly stored in the CD_MDISP datalayer of the mesh, but there is no information on how they were produced.