Regression: `--background=tiles` shows solid black instead of checkerboard on `vo=gpu-next` (works fine on `vo=gpu`)
Nobody has claimed this yet.
- Dominant language
- C
- Stars
- 37k
- Forks
- 3.5k
- Avg merge
- 1d 10h
- Merged PRs (30d)
- 22
Description
mpv Information
mpv v0.41.0-1012-ge8673660a Copyright © 2000-2026 mpv/MPlayer/mplayer2 projects
built on Aug 29 2026 00:29:12
libplacebo version: v7.371.0 (v7.360.0-118-g41ac298-dirty)
FFmpeg version: N-126314-g3386acd2f
FFmpeg library versions:
libavcodec 63.8.101
libavdevice 63.2.100
libavfilter 12.3.101
libavformat 63.6.100
libavutil 61.6.100
libswresample 7.2.100
libswscale 10.2.100
Other Information
- Windows version: Windows 11
- GPU model, driver and version: NVIDIA
- Source of mpv: Windows build
- Latest known working version: mpv v0.41.0-dev-g41f6a6450 (Built on Dec 21 2025)
- Issue started after the following happened: Upgrading to recent 2026 builds.
Reproduction Steps
Expected behavior
When playing a video with an alpha channel (e.g. MOV Animation or WebM yuva420p), using --background=tiles should render a checkerboard pattern behind the transparent video.
Actual behavior
With --vo=gpu-next, the background renders as solid black instead of tiles.
However, switching to --vo=gpu renders the checkerboard perfectly on the exact same build.
This seems to be a regression in gpu-next / libplacebo handling of alpha/tiles, or possibly related to the recent FFmpeg 9.0/libavcodec 63 update which seems to drop A=straight metadata for some files.
Reproduction steps
- Open a video with transparency.
- Run:
mpv --no-config --background=tiles --vo=gpu-next video.mov-> (Result: Solid Black) - Run:
mpv --no-config --background=tiles --vo=gpu video.mov-> (Result: Checkerboard appears correctly)
Expected Behavior
When playing a video with an alpha channel (like MOV Animation or WebM yuva420p), using the --background=tiles option should render a checkerboard pattern behind the transparent video to indicate transparency.
Actual Behavior
With --vo=gpu-next, the background renders as solid black instead of the expected tiles.
However, switching the renderer to --vo=gpu renders the checkerboard perfectly on the exact same build.
This seems to be a regression in how gpu-next or libplacebo handles alpha/tiles, or possibly related to the recent FFmpeg 9.0/libavcodec 63 update which seems to drop A=straight metadata for some files during decoding.
Reproduction steps:
- Open a video with transparency (e.g. MOV).
- Run:
mpv --no-config --background=tiles --vo=gpu-next video.mov(Result: Solid Black) - Run:
mpv --no-config --background=tiles --vo=gpu video.mov(Result: Checkerboard appears correctly)
Log File
Sample Files
No response
I carefully read all instruction and confirm that I did the following:
- I tested and confirmed that the issue exists with the latest release version or newer.
- I provided all required information including system and mpv version.
- I produced the log file with the exact same set of files, parameters, and conditions used in "Reproduction Steps", with the addition of
--log-file=output.txt. - I produced the log file while the behaviors described in "Actual Behavior" were actively observed.
- I attached the full, untruncated log file.
- I attached the backtrace in the case of a crash.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Reproduce the issue with the attached input.zip using --background=tiles and compare --vo=gpu-next against --vo=gpu, then read mpv_log.txt for the rendering and alpha-handling differences. Done means the transparent video shows the checkerboard with --vo=gpu-next as it does with --vo=gpu, without regressing the existing renderer.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- c
- Domain
- audio-video-rtc, cli, computer-graphics
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100