Physics simulation cache appears to share start & end frame drivers (but not if numeric start/end values) #119710

New Issue

James Tomkinson · 2024-03-20T17:57:29+01:00

James Tomkinson commented

2024-03-20 17:57:29 +01:00

System Information
Operating system: Windows-10-10.0.22631-SP0 64 Bits
Graphics card: NVIDIA GeForce RTX 2060/PCIe/SSE2 NVIDIA Corporation 4.6.0 NVIDIA 546.33

Blender Version
Broken: version: 4.0.2, branch: blender-v4.0-release, commit date: 2023-12-05 07:41, hash: 9be62e85b727
Worked: (newest version of Blender that worked as expected)

Short description of error
In a multi-cache simulation, I can set separate start/end frames, but a driver is shared by both cache settings.

Exact steps for others to reproduce the error
I'm not sure if the following is a bug or current product limitation.

I need to cease a physics simulation for a series of frames, and possibly re-bake the simulation when I adjust the animation by scaling areas that will include certain timeline markers. This will also allow for correct re-bakes from command line, driven by timeline markers.

To achieve this I create two caches with appropriate stop and stop times:

These start/end times appear to be set per-cache. Now I want to drive the start/end frames from timeline marker names and problems begin. If I add drivers to start/end of both caches end edit in Drivers Editor, I can see that the cache names are distinguished in the editor, but changes to the first or second caches driver variables:

are then forced upon the other caches values:

example blend file: separate simulations.blend

**System Information** Operating system: Windows-10-10.0.22631-SP0 64 Bits Graphics card: NVIDIA GeForce RTX 2060/PCIe/SSE2 NVIDIA Corporation 4.6.0 NVIDIA 546.33 **Blender Version** Broken: version: 4.0.2, branch: blender-v4.0-release, commit date: 2023-12-05 07:41, hash: `9be62e85b727` Worked: (newest version of Blender that worked as expected) **Short description of error** In a multi-cache simulation, I can set separate start/end frames, but a driver is shared by both cache settings. **Exact steps for others to reproduce the error** I'm not sure if the following is a bug or current product limitation. I need to cease a physics simulation for a series of frames, and possibly re-bake the simulation when I adjust the animation by scaling areas that will include certain timeline markers. This will also allow for correct re-bakes from command line, driven by timeline markers. To achieve this I create two caches with appropriate stop and stop times: ![0.png](/attachments/d8b02b65-e4c8-4dc2-93d5-f8c9438b095f) ![1.png](/attachments/d5ba90a9-fb01-4935-a76f-bb79a50cd9e5) These start/end times appear to be set per-cache. Now I want to drive the start/end frames from timeline marker names and problems begin. If I add drivers to start/end of both caches end edit in Drivers Editor, I can see that the cache names are distinguished in the editor, but changes to the first or second caches driver variables: ![2.png](/attachments/dc3d69a7-d3b2-4854-977e-e1357d71680b) are then forced upon the other caches values: ![3.png](/attachments/a0b75d9d-d6ca-4e35-8dd7-cd8b705d848a) example blend file: [separate simulations.blend](/attachments/18dbbb83-3e5c-49a9-adf8-59f695756f34)

0.png

13 KiB

3.png

48 KiB

2.png

180 KiB

separate simulations.blend

170 KiB

1.png

13 KiB

James Tomkinson added the

labels 2024-03-20 17:57:29 +01:00

Germano Cavalcante commented

2024-03-21 14:40:21 +01:00

Hi @jamestomk, thanks for the report.

What is happening is that the driver is updating the point_cache that is active.
It does not differentiate which point_cache is being used.
I'm not sure if it's possible to specify which point cache is the driver's focus, but the behavior described is expected as far as I can see.

One way I've found to specify a point_cache is by accessing the list of point caches within a point_cache.
C.object.modifiers['Softbody'].point_cache.point_caches['second']

But I'm not sure if it's possible to add a driver for it.

Not a bug AFAICS, but I think it deserves input from a developer... maybe @JacquesLucke (Nodes & Physics) or @ideasman42 (Python API).

Hi @jamestomk, thanks for the report. What is happening is that the driver is updating the `point_cache` that is active. It does not differentiate which `point_cache` is being used. I'm not sure if it's possible to specify which point cache is the driver's focus, but the behavior described is expected as far as I can see. One way I've found to specify a point_cache is by accessing the list of point caches within a `point_cache`. `C.object.modifiers['Softbody'].point_cache.point_caches['second']` But I'm not sure if it's possible to add a driver for it. Not a bug AFAICS, but I think it deserves input from a developer... maybe @JacquesLucke (`Nodes & Physics`) or @ideasman42 (`Python API`).

Germano Cavalcante added

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

Physics simulation cache appears to share start & end frame drivers (but not if numeric start/end values) #119710