Cloth cache bug #91227

Closed
opened 2021-09-06 21:25:59 +02:00 by Elfak · 10 comments

System Information
Operating system: macOS-10.15.7-x86_64-i386-64bit 64 Bits
Graphics card: AMD Radeon Pro 575 OpenGL Engine ATI Technologies Inc. 4.1 ATI-3.10.22

Blender Version
Broken: version: 2.93.2, branch: master, commit date: 2021-08-03 05:58, hash: 1eb06de260
Worked: (newest version of Blender that worked as expected)

Short description of error

  1. Cloth cache with negative range (for example -50 to 200) has some glitch. It being baked, but if you close file and open it again there will be no simulation, but cache appeared as baked. Cache can be deleted and baked again, but after reopening file – same error.
  2. Memory cache has a bug with negative range too – with -1 frame. All negative frames calculated, but -1st frame stands still. And if we switch to disk cache and bake it – all frames will be OK.
  3. When using memory cache, sometimes (not often) cloth simulation with wind goes wrong and cloth moves very slow, unnatural. But if we select object and switch to edit mode and back to object mode the simulation will be OK, but only for current object. So you have to do this trick with all objects. Or recreate FF Wind object. Or try to reopen file. Some magic. I've got this bug a few times, but I was in the middle of deadline, so I didn't catch it. If it appears again I will add it here.
  4. When using memory cache if you delete FF Wind the simulation stays the same, even after reopening file. And you have to change something in Cloth settings to force simulation to recalculates (or press Update All to Frame), but you have to do it with all separate objects with cloth, one by one. You can add new FF Wind (not any other Force Field, won't work) and all caches will be calculated again. But I think this is still an issue.
  5. When using different objects with cloth simulation, as I understand it, you using one cache. But all the object has Cloth Cache settings, and if you, for example, forget to change frame range in one of them to match the others simulation will be calculated with maximum range. Also there is a check box for Disk Cache and you have to check it once on any objects to use disk cache and you have to remember on which object you did it, or it might cause a little panic. :D I don't think it's a bug, but it is really hard to operating like this.

Exact steps for others to reproduce the error
wind-cloth-bug-01.zip – baked not baked cache

**System Information** Operating system: macOS-10.15.7-x86_64-i386-64bit 64 Bits Graphics card: AMD Radeon Pro 575 OpenGL Engine ATI Technologies Inc. 4.1 ATI-3.10.22 **Blender Version** Broken: version: 2.93.2, branch: master, commit date: 2021-08-03 05:58, hash: `1eb06de260` Worked: (newest version of Blender that worked as expected) **Short description of error** 1. Cloth cache with negative range (for example -50 to 200) has some glitch. It being baked, but if you close file and open it again there will be no simulation, but cache appeared as baked. Cache can be deleted and baked again, but after reopening file – same error. 2. Memory cache has a bug with negative range too – with -1 frame. All negative frames calculated, but -1st frame stands still. And if we switch to disk cache and bake it – all frames will be OK. 3. When using memory cache, sometimes (not often) cloth simulation with wind goes wrong and cloth moves very slow, unnatural. But if we select object and switch to edit mode and back to object mode the simulation will be OK, but only for current object. So you have to do this trick with all objects. Or recreate FF Wind object. Or try to reopen file. Some magic. I've got this bug a few times, but I was in the middle of deadline, so I didn't catch it. If it appears again I will add it here. 4. When using memory cache if you delete FF Wind the simulation stays the same, even after reopening file. And you have to change something in Cloth settings to force simulation to recalculates (or press Update All to Frame), but you have to do it with all separate objects with cloth, one by one. You can add new FF Wind (not any other Force Field, won't work) and all caches will be calculated again. But I think this is still an issue. 5. When using different objects with cloth simulation, as I understand it, you using one cache. But all the object has Cloth Cache settings, and if you, for example, forget to change frame range in one of them to match the others simulation will be calculated with maximum range. Also there is a check box for Disk Cache and you have to check it once on any objects to use disk cache and you have to remember on which object you did it, or it might cause a little panic. :D I don't think it's a bug, but it is really hard to operating like this. **Exact steps for others to reproduce the error** [wind-cloth-bug-01.zip](https://archive.blender.org/developer/F10381354/wind-cloth-bug-01.zip) – baked not baked cache
Author

Added subscriber: @AlexMelkin

Added subscriber: @AlexMelkin
Author

UPD. Also there is a problem with alternative caches. When you add a new cache to a cloth object only (!) this object will be allowed for modification. All other cloth objects remain the same with baked state. But, when you hit "Bake all Dynamics" it bakes everything, right? So, alternative cache will have all other caches? What will happen if I add another cache to different cloth object? Will it bake all other caches as well? And which one it would be if other cloth objects have a few caches? And how this will affect time of simulation? Or I should do everything one by one?
It's not common to use an alternative cache, but this emphasising the problem.
Also, for now the only way to bake cloth simulation for sure is export to Alembic or, probably, to another format. Without closing file.

UPD. Also there is a problem with alternative caches. When you add a new cache to a cloth object only (!) this object will be allowed for modification. All other cloth objects remain the same with baked state. But, when you hit "Bake all Dynamics" it bakes everything, right? So, alternative cache will have all other caches? What will happen if I add another cache to different cloth object? Will it bake all other caches as well? And which one it would be if other cloth objects have a few caches? And how this will affect time of simulation? Or I should do everything one by one? It's not common to use an alternative cache, but this emphasising the problem. Also, for now the only way to bake cloth simulation for sure is export to Alembic or, probably, to another format. Without closing file.

Added subscriber: @dave9828

Added subscriber: @dave9828

For the cache, it won't work because the starting bake frame is -50. Change it to more than 0. And as for the cache I've managed to fix it. Go to the physics properties. Then scroll down and find the cache tab. Click "Delete All Bakes" and "Update All to frame". And then click Bake All Dynamics. And for the blender that I have fixed is in the attachments.wind-cloth-fixed-01.zip

For the cache, it won't work because the starting bake frame is -50. Change it to more than 0. And as for the cache I've managed to fix it. Go to the physics properties. Then scroll down and find the cache tab. Click "Delete All Bakes" and "Update All to frame". And then click Bake All Dynamics. And for the blender that I have fixed is in the attachments.[wind-cloth-fixed-01.zip](https://archive.blender.org/developer/F10384790/wind-cloth-fixed-01.zip)
Author

In #91227#1217417, @dave9828 wrote:
For the cache, it won't work because the starting bake frame is -50. Change it to more than 0. And as for the cache I've managed to fix it. Go to the physics properties. Then scroll down and find the cache tab. Click "Delete All Bakes" and "Update All to frame". And then click Bake All Dynamics. And for the blender that I have fixed is in the attachments.wind-cloth-fixed-01.zip

It works with negative range, and it is very (VERY!) important that it works. Because motion in shots don't start with 0 frame. For example, wind already blows and cloth already affected by it when shot starts. And it is hell to have pipeline based on constant shift of timeline. But there is more. Commonly you have to work with tracking data, and commonly tracking data starts from 0 (or 1st) frame. So negative range for ANY simulation is necessary.
Simulation goes well and cache been baked well. So the problem only that cache isn't working after reopening file. It is a bug.

> In #91227#1217417, @dave9828 wrote: > For the cache, it won't work because the starting bake frame is -50. Change it to more than 0. And as for the cache I've managed to fix it. Go to the physics properties. Then scroll down and find the cache tab. Click "Delete All Bakes" and "Update All to frame". And then click Bake All Dynamics. And for the blender that I have fixed is in the attachments.[wind-cloth-fixed-01.zip](https://archive.blender.org/developer/F10384790/wind-cloth-fixed-01.zip) It works with negative range, and it is very (VERY!) important that it works. Because motion in shots don't start with 0 frame. For example, wind already blows and cloth already affected by it when shot starts. And it is hell to have pipeline based on constant shift of timeline. But there is more. Commonly you have to work with tracking data, and commonly tracking data starts from 0 (or 1st) frame. So negative range for ANY simulation is necessary. Simulation goes well and cache been baked well. So the problem only that cache isn't working after reopening file. It is a bug.

It's not possible with a negative range as the player minimal is 0. So it's not a bug. Just set your start frames to whatever you want and make the end frames longer. Then as always delete the bakes and bake it.

It's not possible with a negative range as the player minimal is 0. So it's not a bug. Just set your start frames to whatever you want and make the end frames longer. Then as always delete the bakes and bake it.
Author

In #91227#1217704, @dave9828 wrote:
It's not possible with a negative range as the player minimal is 0. So it's not a bug. Just set your start frames to whatever you want and make the end frames longer. Then as always delete the bakes and bake it.

Have you opened Blender? There is practically unlimited negative range! Player starts with 0 frame for rendering, and this has a point – you don't need negative frames for render. Render frame range should not be (however it might) the same as simulation or animation frame range. Blender has negative frames because of the reason – they needed for a lot of things.
How is it impossible if it works? Everything goes well except a few things. I did a job with negative range and everything went well. But I had to use some tricks to override the bugs, like exporting my simulation to Alembic to get it baked.
Again – Blender already have this ability to make simulations with negative frame range! It works! And this is awesome that we have this ability, it is very useful. So, this is a bug, and from my point of view a major one.

> In #91227#1217704, @dave9828 wrote: > It's not possible with a negative range as the player minimal is 0. So it's not a bug. Just set your start frames to whatever you want and make the end frames longer. Then as always delete the bakes and bake it. Have you opened Blender? There is practically unlimited negative range! Player starts with 0 frame for rendering, and this has a point – you don't need negative frames for render. Render frame range should not be (however it might) the same as simulation or animation frame range. Blender has negative frames because of the reason – they needed for a lot of things. How is it impossible if it works? Everything goes well except a few things. I did a job with negative range and everything went well. But I had to use some tricks to override the bugs, like exporting my simulation to Alembic to get it baked. Again – Blender already have this ability to make simulations with negative frame range! It works! And this is awesome that we have this ability, it is very useful. So, this is a bug, and from my point of view a major one.

Added subscriber: @iss

Added subscriber: @iss

There are more than 1 issues reported here all but point 5 are reported already (which seems to be more a feature request). I will merge to existing report.

There are more than 1 issues reported here all but point 5 are reported already (which seems to be more a feature request). I will merge to existing report.

Closed as duplicate of #70631

Closed as duplicate of #70631
Sign in to join this conversation.
No Label
Interest
Alembic
Interest
Animation & Rigging
Interest
Asset Browser
Interest
Asset Browser Project Overview
Interest
Audio
Interest
Automated Testing
Interest
Blender Asset Bundle
Interest
BlendFile
Interest
Collada
Interest
Compatibility
Interest
Compositing
Interest
Core
Interest
Cycles
Interest
Dependency Graph
Interest
Development Management
Interest
EEVEE
Interest
EEVEE & Viewport
Interest
Freestyle
Interest
Geometry Nodes
Interest
Grease Pencil
Interest
ID Management
Interest
Images & Movies
Interest
Import Export
Interest
Line Art
Interest
Masking
Interest
Metal
Interest
Modeling
Interest
Modifiers
Interest
Motion Tracking
Interest
Nodes & Physics
Interest
OpenGL
Interest
Overlay
Interest
Overrides
Interest
Performance
Interest
Physics
Interest
Pipeline, Assets & IO
Interest
Platforms, Builds & Tests
Interest
Python API
Interest
Render & Cycles
Interest
Render Pipeline
Interest
Sculpt, Paint & Texture
Interest
Text Editor
Interest
Translations
Interest
Triaging
Interest
Undo
Interest
USD
Interest
User Interface
Interest
UV Editing
Interest
VFX & Video
Interest
Video Sequencer
Interest
Virtual Reality
Interest
Vulkan
Interest
Wayland
Interest
Workbench
Interest: X11
Legacy
Blender 2.8 Project
Legacy
Milestone 1: Basic, Local Asset Browser
Legacy
OpenGL Error
Meta
Good First Issue
Meta
Papercut
Meta
Retrospective
Meta
Security
Module
Animation & Rigging
Module
Core
Module
Development Management
Module
EEVEE & Viewport
Module
Grease Pencil
Module
Modeling
Module
Nodes & Physics
Module
Pipeline, Assets & IO
Module
Platforms, Builds & Tests
Module
Python API
Module
Render & Cycles
Module
Sculpt, Paint & Texture
Module
Triaging
Module
User Interface
Module
VFX & Video
Platform
FreeBSD
Platform
Linux
Platform
macOS
Platform
Windows
Priority
High
Priority
Low
Priority
Normal
Priority
Unbreak Now!
Status
Archived
Status
Confirmed
Status
Duplicate
Status
Needs Info from Developers
Status
Needs Information from User
Status
Needs Triage
Status
Resolved
Type
Bug
Type
Design
Type
Known Issue
Type
Patch
Type
Report
Type
To Do
No Milestone
No project
No Assignees
3 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: blender/blender#91227
No description provided.