Page MenuHome

Smoke sim glitch in Cycles
Confirmed, NormalPublicBUG


System Information
Operating system: Windows-10-10.0.17763 64 Bits
Graphics card: GeForce RTX 2080/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 441.12

Blender Version
Broken: version: 2.81 (sub 16), branch: master, commit date: 2019-12-04 11:32, hash: rBf1aa4d18d49d
Worked: (optional)

Short description of error
While rendering in cycles my smoke is missing pieces and is very blocky.

Exact steps for others to reproduce the error
Make a quick smoke sim above 60 res or up with adaptive domain (not sure if has to be with a smoke collision object) then add a principled volume to world output and change density to .03 or up. Once a light source is added like a sun and you give it a second to render you can see the weird distortions.

From the file attached just click render and you can see the weird issue

If you want to know what this looks like in a rendered video here is one

Ok so I did some more testing and it looks a principled volume smoke sim cant be in the same space as an existing principled volume area (world or just a cube)

Event Timeline

Jacques Lucke (JacquesLucke) changed the task status from Needs Triage to Needs Information from User.Jan 14 2020, 1:17 PM

This file has been created using the old smoke modifier. Are you able to reproduce the issue with the new one in the latest builds?

I upgraded to 2.81a and did a quick smoke on the default cube and it was the same

Heres what I got in 2.82, might be a bit better but still off

Please provide the updated file, so that we can reproduce it. It should be as simple as possible.

Just bake the sim and look threw the camera in cycles

Jacques Lucke (JacquesLucke) changed the task status from Needs Information from User to Confirmed.Jan 15 2020, 3:56 PM

Thanks. I can see the issue now. Seems to be related to the fact that you have a volumetric world material. I'm not sure if this is a bug, limitation or just a wrong combination of settings for this kind of scene in Blender.

@Stefan Werner (swerner), do you know the category of the issue in the above file?

Brecht Van Lommel (brecht) changed the subtype of this task from "Report" to "Bug".Feb 3 2020, 12:46 PM

I noticed that the same issue exists when using the new volume object type in master branch with a vdb file.

Sorry for the late reply.

Two observations:

  1. It does not happen with branched path tracing.
  2. It may look like noise at low sample counts, but does not disappear with higher sample counts.