Adjustment Clip turns black when using Size, Position & Rotate

Bug
1

Bug Report: Adjustment Clip turns black when using Size, Position & Rotate

Shotcut version: 26.9
Operating system: Loc-OS 23 (Linux)
Video mode: OpenGL

Description

When I add an Adjustment Clip above my video clips and apply the Size, Position & Rotate (SP&R) filter to the Adjustment Clip, the entire result becomes black.

The problem occurs both in the preview and in the exported video.

Steps to reproduce

  1. Create a new project.

  2. Add a video clip to the timeline.

  3. Add an Adjustment Clip on a track above the video.

  4. Select the Adjustment Clip.

  5. Add the Size, Position & Rotate filter.

  6. Change any of the SP&R parameters, such as scale, position, or rotation.

Expected behavior

The Adjustment Clip should apply the SP&R transformation to the underlying video while keeping the video visible.

Actual behavior

The video becomes completely black as soon as SP&R is applied to the Adjustment Clip. The black output is also present after exporting the project.

Additional information

  • The issue is reproducible in both the preview and the final exported video.

  • Shotcut is using OpenGL.

  • The problem occurs specifically when SP&R is applied to an Adjustment Clip.

Could you please confirm whether this is a known limitation or a bug in the Adjustment Clip implementation?

Thank you.

2

That is a limitation and will not be addressed. Don’t use that filter on adjustment clip.

3

Thanks for clarifying.

I understand that this is a limitation and that SP&R will not be supported on Adjustment Clips.

Could you please explain why SP&R cannot be used with Adjustment Clips? Is there a recommended workflow for applying a position, scale, or rotation transformation to multiple clips at once?

For example, if I have several clips on different tracks and I want to zoom, rotate, or reposition the entire composition as a single unit, what would be the recommended approach in Shotcut?

Thanks for the clarification.

4

It is a technical limitation due to how it gets the image from the source to take advantage of its maximum resolution. Think about slow zooms on digital photo. You can study the source code. You can use --experimental and the UI generator. There is a Composite and transform available that can do it, but the UI for this is crude and unusable. If you can do high level (QML script) programming, you can grab the generated code and modify it. There is a good chance this will be added officially with a good UI because there is a PR to add motion blur to it, which is exciting. There is also a TransformOFX OpenFX plugin that kinda works with the generated UI.

Another approach is to use a nested project file and attach filters to it.