Mantaflow simulation is not simulating. #77135

New Issue

Scott VanKirk · 2020-05-28T17:18:31+02:00

Scott VanKirk commented

2020-05-28 17:18:31 +02:00

System Information
Operating system: Windows-10-10.0.18362-SP0 64 Bits
Graphics card: GeForce RTX 2070 SUPER/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 442.92

Blender Version
Broken: version: 2.90 (sub 4), branch: master, commit date: 2020-05-27 10:07, hash: efa4ae17f0
Worked: (newest version of Blender that worked as expected)

Short description of error
I have set up a simple scene with mantaflow. There is 1 domain, 1 body of fluid, and one flow source. It was working for a while, but the effector (volcano) stopped effecting the fluid at around frame 100. I added a physics domain to try and address this and tweaked a few settings and now when I try to run the simulation, nothing happens. The first frame shows the particles, but nothing changes after that. I removed the physics domain, still doesn't work.

Exact steps for others to reproduce the error
volcano restart5 bug.blend

Open file
Play animation

**System Information** Operating system: Windows-10-10.0.18362-SP0 64 Bits Graphics card: GeForce RTX 2070 SUPER/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 442.92 **Blender Version** Broken: version: 2.90 (sub 4), branch: master, commit date: 2020-05-27 10:07, hash: `efa4ae17f0` Worked: (newest version of Blender that worked as expected) **Short description of error** I have set up a simple scene with mantaflow. There is 1 domain, 1 body of fluid, and one flow source. It was working for a while, but the effector (volcano) stopped effecting the fluid at around frame 100. I added a physics domain to try and address this and tweaked a few settings and now when I try to run the simulation, nothing happens. The first frame shows the particles, but nothing changes after that. I removed the physics domain, still doesn't work. **Exact steps for others to reproduce the error** [volcano restart5 bug.blend](https://archive.blender.org/developer/F8560332/volcano_restart5_bug.blend) - Open file - Play animation

Scott VanKirk commented

2020-05-28 17:18:31 +02:00

Added subscriber: @ScottVanKirk

blender-admin commented

2020-05-28 17:18:31 +02:00

#77134 was marked as duplicate of this issue

Scott VanKirk commented

2020-05-29 08:27:30 +02:00

Here is another example of mantaflow not running the sim. Here I was playing with the size of the particles, trying to see what different settings would look like. The simulation stopped running and even returning the numbers didn't get it to restart.

This seems to be part of a larger problem with Mantaflow. It doesn't reset everything in the simulation as it should. For example, when I'm using "replay" mode, stop the simulation, then restart from frame one, My effectors often stop working as expected and after the previous stopping point the fluid just falls through the effectors with no interaction.

Again, I have been unable to reduce this any further since it occurs randomly as far as I can tell.
volcano restart6 bug.blend

Here is another example of mantaflow not running the sim. Here I was playing with the size of the particles, trying to see what different settings would look like. The simulation stopped running and even returning the numbers didn't get it to restart. This seems to be part of a larger problem with Mantaflow. It doesn't reset everything in the simulation as it should. For example, when I'm using "replay" mode, stop the simulation, then restart from frame one, My effectors often stop working as expected and after the previous stopping point the fluid just falls through the effectors with no interaction. Again, I have been unable to reduce this any further since it occurs randomly as far as I can tell. [volcano restart6 bug.blend](https://archive.blender.org/developer/F8562538/volcano_restart6_bug.blend)

Ankit Meel commented

2020-05-30 10:37:11 +02:00

Added subscriber: @ankitm

Ankit Meel commented

2020-05-30 10:37:11 +02:00

The thing with effectors sounds like the following task. Did you move the effector ?
#76095 (Liquid cache of type "Replay" doesn't get invalidated when moving participating objects)

The thing with effectors sounds like the following task. Did you move the effector ? #76095 (Liquid cache of type "Replay" doesn't get invalidated when moving participating objects)

Scott VanKirk commented

2020-06-01 03:28:15 +02:00

No, nothing was moved, though I have seen that error as well.

Richard Antalik commented

2020-06-01 23:01:37 +02:00

Added subscriber: @iss

Richard Antalik commented

2020-06-01 23:01:37 +02:00

Changed status from 'Needs Triage' to: 'Needs User Info'

Richard Antalik commented

2020-06-01 23:01:37 +02:00

In #77135#942058, @ScottVanKirk wrote:
No, nothing was moved, though I have seen that error as well.

I can confirm, that liquid is not simulated. In this case it may be good to know, how to recreate this file from scratch, because files provided are fairly complex.
Can you give us steps to create example file?

> In #77135#942058, @ScottVanKirk wrote: > No, nothing was moved, though I have seen that error as well. I can confirm, that liquid is not simulated. In this case it may be good to know, how to recreate this file from scratch, because files provided are fairly complex. Can you give us steps to create example file?

Scott VanKirk commented

2020-06-02 20:56:33 +02:00

I have spent hours trying to understand all the various failure modes of Mantaflow in replay mode with little success. If I can find the time, I'll try and track this down some more.

Scott VanKirk commented

2020-06-03 00:43:35 +02:00

Okay, this might have something to do with changing the cache format to openvdb

Richard Antalik commented

2020-06-04 10:12:20 +02:00

I am not sure if this is useful as a bug report.

This files may got to state that looks OK for engine, and to fix the issue, actual workflow needs to be sanitized. So I could confirm this, but developer may ask you what exactly you did to get to this state.

I am not sure if this is useful as a bug report. This files may got to state that looks OK for engine, and to fix the issue, actual workflow needs to be sanitized. So I could confirm this, but developer may ask you what exactly you did to get to this state.

Richard Antalik commented

2020-06-04 10:14:14 +02:00

Changed status from 'Needs User Info' to: 'Confirmed'

Scott VanKirk commented

2020-06-07 16:36:51 +02:00

You can definitely see this bug if you run a simulation, then change the cache format to openVDB. After that the simulation will not simulate.

Philipp Oeser commented

2020-06-08 10:02:22 +02:00

Added subscriber: @lichtwerk

Philipp Oeser commented

2020-06-08 10:02:22 +02:00

So this has been reported again, see #77432 (Mantaflow Liquid not work when "Data File Format" is changed to "OpenVDB") but this time straight to the point from the beginning [which will make the report a bit to understand].
Will merge these reports...

So this has been reported again, see #77432 (Mantaflow Liquid not work when "Data File Format" is changed to "OpenVDB") but this time straight to the point from the beginning [which will make the report a bit to understand]. Will merge these reports...

Philipp Oeser commented

2020-06-08 10:02:31 +02:00

Closed as duplicate of #77432

Philipp Oeser closed this issue

2020-06-08 10:02:31 +02:00

Sign in to join this conversation.

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

Mantaflow simulation is not simulating. #77135