Change object to rigid body far away modifies existing physics behaviour #77814

New Issue

Josef Schandl · 2020-06-12T22:12:30+02:00

Josef Schandl commented

2020-06-12 22:12:30 +02:00

System Information
Operating system: Windows 10
Graphics card: NVIDIA GeForce GTX 1060 3GB

Blender Version
2.83.0

Short description of error
Change object to rigid body far away modifies existing physics behaviour

Exact steps for others to reproduce the error
See attached blender file cubes18d1g.blend cubes18d1g.blend

First run the animation by starting on frame 0. That's the original behaviour of the cubes.
Then set object "Plane_hold" to "Rigid Body": just click on button "Rigid Body" in tab Physics properties.
The run the animation again at frame 0. The behaviour of the pushed cubes is quite different!

Why should a change of an object far away setting to "Rigid body" influence rigid body behaviour of the original animation?

**System Information** Operating system: Windows 10 Graphics card: NVIDIA GeForce GTX 1060 3GB **Blender Version** 2.83.0 **Short description of error** Change object to rigid body far away modifies existing physics behaviour **Exact steps for others to reproduce the error** See attached blender file cubes18d1g.blend [cubes18d1g.blend](https://archive.blender.org/developer/F8612818/cubes18d1g.blend) 1. First run the animation by starting on frame 0. That's the original behaviour of the cubes. 2. Then set object "Plane_hold" to "Rigid Body": just click on button "Rigid Body" in tab Physics properties. 3. The run the animation again at frame 0. The behaviour of the pushed cubes is quite different! Why should a change of an object far away setting to "Rigid body" influence rigid body behaviour of the original animation?

Josef Schandl commented

2020-06-12 22:12:30 +02:00

Added subscriber: @Josef_Schandl

Richard Antalik commented

2020-06-19 04:37:04 +02:00

Added subscribers: @sreich, @iss

Richard Antalik commented

2020-06-19 04:37:04 +02:00

I could imagine this being a known limitation but I have no real idea. @sreich Is this supposed to happen?

In case this is limitation, I would recommend asking on one of the community websites: https://www.blender.org/community/ if there are known workarounds.

I could imagine this being a known limitation but I have no real idea. @sreich Is this supposed to happen? In case this is limitation, I would recommend asking on one of the community websites: https://www.blender.org/community/ if there are known workarounds.

Sergej Reich commented

2020-06-19 14:26:44 +02:00

Unfortunately I can't say for sure without investigating exactly how our rigid body library (Bullet) behaves, which would take quite a lot of time.
Most likely this is how it's supposed to work. The reasoning is that the sort of numerical solvers we use are quite sensitive to things like how many objects there are, the order in which objects are added to the simulation, etc.
Determinism is only really guaranteed if you run the exact same simulation multiple times, adding objects changes the simulation.
It's possible to improve this behaviour, but depending on how exactly things are implemented it could be a lot of work.

Unfortunately I can't say for sure without investigating exactly how our rigid body library (Bullet) behaves, which would take quite a lot of time. Most likely this is how it's supposed to work. The reasoning is that the sort of numerical solvers we use are quite sensitive to things like how many objects there are, the order in which objects are added to the simulation, etc. Determinism is only really guaranteed if you run the exact same simulation multiple times, adding objects changes the simulation. It's possible to improve this behaviour, but depending on how exactly things are implemented it could be a lot of work.

Richard Antalik commented

2020-06-20 04:05:57 +02:00

Changed status from 'Needs Triage' to: 'Archived'

Richard Antalik closed this issue

2020-06-20 04:05:57 +02:00

Richard Antalik self-assigned this 2020-06-20 04:05:57 +02:00

Richard Antalik commented

2020-06-20 04:05:57 +02:00

In #77814#958412, @sreich wrote:
Unfortunately I can't say for sure without investigating exactly how our rigid body library (Bullet) behaves, which would take quite a lot of time.
Most likely this is how it's supposed to work. The reasoning is that the sort of numerical solvers we use are quite sensitive to things like how many objects there are, the order in which objects are added to the simulation, etc.
Determinism is only really guaranteed if you run the exact same simulation multiple times, adding objects changes the simulation.
It's possible to improve this behaviour, but depending on how exactly things are implemented it could be a lot of work.

Thanks for providing details. I will close this report since it is a limitation.

> In #77814#958412, @sreich wrote: > Unfortunately I can't say for sure without investigating exactly how our rigid body library (Bullet) behaves, which would take quite a lot of time. > Most likely this is how it's supposed to work. The reasoning is that the sort of numerical solvers we use are quite sensitive to things like how many objects there are, the order in which objects are added to the simulation, etc. > Determinism is only really guaranteed if you run the exact same simulation multiple times, adding objects changes the simulation. > It's possible to improve this behaviour, but depending on how exactly things are implemented it could be a lot of work. Thanks for providing details. I will close this report since it is a limitation.

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

Change object to rigid body far away modifies existing physics behaviour #77814