VSE: adding scene stripe resets animation of colour stripe with mask #98287

New Issue

forceengine · 2022-05-20T13:39:20+02:00

forceengine commented

2022-05-20 13:39:20 +02:00

System Information
Operating system: Linux-5.13.0-41-generic-x86_64-with-glibc2.33 64 Bits
Graphics card: AMD RENOIR (DRM 3.41.0, 5.13.0-41-generic, LLVM 12.0.1) AMD 4.6 (Core Profile) Mesa 21.3.8 (git-813ee839be)

Blender Version
Broken: version: 3.1.2, branch: master, commit date: 2022-03-31 17:40, hash: cc66d1020c
Worked: none tested

Short description of error
I have added a colour channel and applied a mask to it. I have created and animated a clock as UI element.
When I add the clock as scene stripe to the scene with the masked colour stripe, the mask of the colour strip is not applied correctly.

Exact steps for others to reproduce the error
I have no idea what really causes the issue or how I could reproduce it. See the attached blend file and video files.

Notes

When the clock strip is muted in the VSE overview, the problem does no longer exist. So I guess it's some weird interaction between the mask and the scene strip
The bug can be solved when you create a new scene and add the colour strip as scene strip and on top of that the clock strip. It works, but performance is incredibly bad.

Files

The actual output: bug 01.mp4
How it should look like: bug 02.mp4
blend file: bug.blend

**System Information** Operating system: Linux-5.13.0-41-generic-x86_64-with-glibc2.33 64 Bits Graphics card: AMD RENOIR (DRM 3.41.0, 5.13.0-41-generic, LLVM 12.0.1) AMD 4.6 (Core Profile) Mesa 21.3.8 (git-813ee839be) **Blender Version** Broken: version: 3.1.2, branch: master, commit date: 2022-03-31 17:40, hash: `cc66d1020c` Worked: none tested **Short description of error** I have added a colour channel and applied a mask to it. I have created and animated a clock as UI element. When I add the clock as scene stripe to the scene with the masked colour stripe, the mask of the colour strip is not applied correctly. **Exact steps for others to reproduce the error** I have no idea what really causes the issue or how I could reproduce it. See the attached blend file and video files. **Notes** - When the clock strip is muted in the VSE overview, the problem does no longer exist. So I guess it's some weird interaction between the mask and the scene strip - The bug can be solved when you create a new scene and add the colour strip as scene strip and on top of that the clock strip. It works, but performance is incredibly bad. **Files** - The actual output: [bug 01.mp4](https://archive.blender.org/developer/F13094245/bug_01.mp4) - How it should look like: [bug 02.mp4](https://archive.blender.org/developer/F13094246/bug_02.mp4) - blend file: [bug.blend](https://archive.blender.org/developer/F13094247/bug.blend)

forceengine commented

2022-05-20 13:39:20 +02:00

Added subscriber: @forceengine

Omar Emara commented

2022-05-23 10:59:12 +02:00

Added subscriber: @OmarEmaraDev

Omar Emara commented

2022-05-23 10:59:12 +02:00

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

Omar Emara commented

2022-05-23 10:59:12 +02:00

It seems the file does not include the images necessary to make the clock, so I can't exactly replicate the sequence. Can you pack into the file or attach the necessary images?

It seems there is a depsgraph error, not sure if it related. Will check when we get the file.

add_relation(Sequence Scene Audio -> Sequencer) - Could not find op_from (ComponentKey(SCclock, AUDIO))
add_relation(Sequence Scene Audio -> Sequencer) - Failed, but op_to (ComponentKey(SCbug, SEQUENCER)) was ok

It seems the file does not include the images necessary to make the clock, so I can't exactly replicate the sequence. Can you pack into the file or attach the necessary images? It seems there is a depsgraph error, not sure if it related. Will check when we get the file. ``` add_relation(Sequence Scene Audio -> Sequencer) - Could not find op_from (ComponentKey(SCclock, AUDIO)) add_relation(Sequence Scene Audio -> Sequencer) - Failed, but op_to (ComponentKey(SCbug, SEQUENCER)) was ok ```

Richard Antalik commented

2022-05-24 00:50:50 +02:00

Added subscriber: @iss

Richard Antalik commented

2022-05-24 00:50:50 +02:00

I often see errors mentioned above, but these do not affect image output.

@forceengine does this issue persist when you reopen .blend file? I can't reproduce this issue here.

I often see errors mentioned above, but these do not affect image output. @forceengine does this issue persist when you reopen .blend file? I can't reproduce this issue here.

forceengine commented

2022-05-24 18:20:18 +02:00

In #98287#1362823, @OmarEmaraDev wrote:
It seems the file does not include the images necessary to make the clock, so I can't exactly replicate the sequence. Can you pack into the file or attach the necessary images?

@OmarEmaraDev
Sorry, tired me forgot that the files are linked in blender. This zip file should include every file needed for the animation.

bug archive.zip

Regarding the animation: The first part of the animation of the red and masked colour layer has finished before the clock starts to expand.
When the clock starts to expand, the mask flickers. When the clock has expanded all the way and does no longer change it's size, the mask looks as it should.
The error also is visible with single image rendering.
If you go through the frames one after the other with Alt + mouse wheel it looks like as if the animation of the colour layer was reset.

@iss Yes, the bug also occurs when I reopen the file. It can be reproduced every time.
I did a quick scene render (extract files -> render animation) after I created the archive in this post and I got the same bug again.

EDIT: I just tested this bug on Windows. I get the same bug in the preview window on Windows, but the rendered Video looks ok.
However, a rendered single frame does show the same bug like in the viewport (I always thought that Eevee VSE gives you the same result for movies and single frames.)

A single frame direct from VSE (F12 + Alt+S). The clock is almost at it's full size. What you see is the 'reset' effect that seems to be caused in the viewpor and single rednered frames.

> In #98287#1362823, @OmarEmaraDev wrote: > It seems the file does not include the images necessary to make the clock, so I can't exactly replicate the sequence. Can you pack into the file or attach the necessary images? @OmarEmaraDev Sorry, tired me forgot that the files are linked in blender. This zip file should include every file needed for the animation. [bug archive.zip](https://archive.blender.org/developer/F13103140/bug_archive.zip) Regarding the animation: The first part of the animation of the red and masked colour layer has finished *before* the clock starts to expand. When the clock starts to expand, the mask flickers. When the clock has expanded all the way and does no longer change it's size, the mask looks as it should. The error also is visible with single image rendering. If you go through the frames one after the other with Alt + mouse wheel it looks like as if the animation of the colour layer was reset. @iss Yes, the bug also occurs when I reopen the file. It can be reproduced every time. I did a quick scene render (extract files -> render animation) after I created the archive in this post and I got the same bug again. EDIT: I just tested this bug on Windows. I get the same bug in the preview window on Windows, but the rendered Video looks ok. However, a rendered single frame does show the same bug like in the viewport (I always thought that Eevee VSE gives you the same result for movies and single frames.) A single frame direct from VSE (F12 + Alt+S). The clock is almost at it's full size. What you see is the 'reset' effect that seems to be caused in the viewpor and single rednered frames. ![Windows.png](https://archive.blender.org/developer/F13103485/Windows.png)

Richard Antalik commented

2022-05-24 23:51:52 +02:00

I am checking on Windows, so this could be the reason for me not being able to reproduce. From videos in description it looked to me that issue is flickering of animated color strip. I don't understand how can I see the issue from provided still frame in comment above - what frame number is that? Looks like frame 61, which is what I see in preview. Is that incorrect?

If I can reproduce on windows, it would help a lot, because I don't have linux machine with dev environment set up.

I am checking on Windows, so this could be the reason for me not being able to reproduce. From videos in description it looked to me that issue is flickering of animated color strip. I don't understand how can I see the issue from provided still frame in comment above - what frame number is that? Looks like frame 61, which is what I see in preview. Is that incorrect? If I can reproduce on windows, it would help a lot, because I don't have linux machine with dev environment set up.

Omar Emara commented

2022-05-25 09:48:47 +02:00

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

Omar Emara commented

2022-05-25 09:48:47 +02:00

I can reproduce the issue. What seems to be happening is that the red color strips have animations that sometimes gets evaluated at the active scene frame and sometimes gets evaluated at the frame of the scene of the scene strip of the clock.
While they should always be evaluated at the frame of the active scene as far as I can tell.

I can reproduce the issue. What seems to be happening is that the red color strips have animations that sometimes gets evaluated at the active scene frame and sometimes gets evaluated at the frame of the scene of the scene strip of the clock. While they should always be evaluated at the frame of the active scene as far as I can tell.

forceengine commented

2022-05-26 20:54:45 +02:00

In #98287#1363806, @iss wrote:
I don't understand how can I see the issue from provided still frame in comment above - what frame number is that? Looks like frame 61, which is what I see in preview. Is that incorrect?

If I can reproduce on windows, it would help a lot, because I don't have linux machine with dev environment set up.

Yes, it looks like frame 61. Since the bug does not show up on Windows when you render a video, but does show up when you render a single frame and in the viewport, the bug can sorta be reproduced on windows. The still frame shows that it's sorta possible since when the clock does expand, the colour strip is not animated and should keep it's size.

Regarding a Linux machine: I don't know how much time and/or effort you want to put into this, but you could use a live USB Linux distro, download blender from flatpak/snap or as AppImage and test it there. It for sure is not a good setup at all if you want to test more than this bug. This is at least how I tested the few bugs I reported before I got my Linux machine.

In #98287#1363953, @OmarEmaraDev wrote:
What seems to be happening is that the red color strips have animations that sometimes gets evaluated at the active scene frame and sometimes gets evaluated at the frame of the scene of the scene strip of the clock.
While they should always be evaluated at the frame of the active scene as far as I can tell.

This sounds about right. If you remove the mask from the colour stripe you can see better what happens:
010001-0170.mp4

> In #98287#1363806, @iss wrote: > I don't understand how can I see the issue from provided still frame in comment above - what frame number is that? Looks like frame 61, which is what I see in preview. Is that incorrect? > > If I can reproduce on windows, it would help a lot, because I don't have linux machine with dev environment set up. Yes, it looks like frame 61. Since the bug does not show up on Windows when you *render a video*, but does show up when you render a single frame and in the viewport, the bug can sorta be reproduced on windows. The still frame shows that it's sorta possible since when the clock does expand, the colour strip is not animated and should keep it's size. Regarding a Linux machine: I don't know how much time and/or effort you want to put into this, but you could use a live USB Linux distro, download blender from flatpak/snap or as AppImage and test it there. It for sure is not a good setup at all if you want to test more than this bug. This is at least how I tested the few bugs I reported before I got my Linux machine. > In #98287#1363953, @OmarEmaraDev wrote: > What seems to be happening is that the red color strips have animations that sometimes gets evaluated at the active scene frame and sometimes gets evaluated at the frame of the scene of the scene strip of the clock. > While they should always be evaluated at the frame of the active scene as far as I can tell. This sounds about right. If you remove the mask from the colour stripe you can see better what happens: [010001-0170.mp4](https://archive.blender.org/developer/F13110468/010001-0170.mp4)

Richard Antalik commented

2022-05-26 23:12:15 +02:00

@forceengine Yes I can check on Linux, but not really debug at least easily. But since you report different OS to be working correctly which implies different settings, can you check if you have disk cache enabled on Linux installation and if issue will happen when you disable it?

Richard Antalik commented

2022-05-26 23:15:55 +02:00

But I can reproduce difference in color strip animation when scene strip is enabled/disabled, so I guess this would be enough to investigate.

Philipp Oeser removed the

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

VSE: adding scene stripe resets animation of colour stripe with mask #98287