All videos has white dots/speckles while VLC doesnt have

[Leopard1907]

mpv Information

mpv v0.41.0 Copyright © 2000-2025 mpv/MPlayer/mplayer2 projects
 built on Dec 22 2025 14:58:51
libplacebo version: v7.351.0
FFmpeg version: n8.0.1
FFmpeg library versions:
   libavcodec      62.11.100
   libavdevice     62.1.100
   libavfilter     11.4.100
   libavformat     62.3.100
   libavutil       60.8.100
   libswresample   6.1.100
   libswscale      9.1.100

Other Information

- Linux version:NAME="EndeavourOS"
PRETTY_NAME="EndeavourOS"
- Kernel Version:6.18.2-arch2-1
- GPU Model:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Navi 31 [Radeon RX 7900 XT/7900 XTX/7900 GRE/7900M] [1002:744c] (rev c8)
12:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Raphael [1002:164e] (rev cb)
Display hooked to 7900 XTX, single display.
- Mesa/GPU Driver Version:25.3.2
- Window Manager and Version:Kde Wayland,Plasma version 6.5.4
- Source of mpv:Arch Linux repo, not AUR
- Latest known working version:Unknown
- Issue started after the following happened:Unknown

Reproduction Steps

Video renders as expected on Firefox, also renders as expected on VLC media player at local playback. Same file produces white dots/speckles on MPV, also happens with no config.

Expected Behavior

Actual Behavior

Log File

output.txt

Reproducing log file leads to this, not sure if this is expected. Otherwise normally video plays but with those speckles.

Sample Files

https://www.youtube.com/watch?v=1Wyb6vFVy-M

I used yt-dlp to download it, downloaded it a while ago. Highest quality version ( basically what yt-dlp does by default)

Still, in case it got reencoded to something else. Here is what mpv says ( with my config )

I carefully read all instruction and confirm that I did the following:

• I tested with the latest mpv version to validate that the issue is not already fixed.
• 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.

[llyyr]

Could you try with libplacebo from master, instead of the latest release?

[avih]

For reference, the image seems to have dithering.

[RipleyTom]

Not sure that's any help but I have no dithering issue on that particular video with the exact same mpv version and same version of the libraries(my version is Gentoo/self built).

Notable difference is a different video decoder is selected(software vs yours is hardware?):

[kasper93]

Transfer: linear

This output is currently broken and should be avoided. If you see it, add target-trc to override it.

[Leopard1907]

Transfer: linear

This output is currently broken and should be avoided. If you see it, add target-trc to override it.

Need to add that to where exactly? As i dont force it anywhere on my own, important to note issue reproduces without a config as well.

But sure, i can still try adding that to config but how? As what do i need to pass to change it something else?

profile=high-quality
scale=ewa_lanczossharp
cscale=ewa_lanczossharp
vo=gpu-next
gpu-api=vulkan
gpu-context=waylandvk
target-colorspace-hint
hwdec=vaapi

Things that i have in my conf file, do note issue still repros without any conf and with no conf it still defaults to whatever that linear thing is.

[kasper93]

I think @llyyr had custom profile somwhere.

[Leopard1907]

@llyyr @avih I found a workaround thanks to another Linux user somewhere else. I'm using Asus's official ICM profile from product page. Turning it off solves the problem. Now problem is:

Do note setting KDE to use "built-in" profile has the same issue.

1-) Why other apps are not affected by it?

2-) MPV was not affected by it before either because i use that ICM for well over a year now.

Said ICM file.

ASUS XG32VQR.tar.gz

https://www.asus.com/displays-desktops/monitors/tuf-gaming/tuf-gaming-vg32vqr/helpdesk_download?model2Name=TUF-Gaming-VG32VQR

To sum it up:

Color profile on KDE set to nothing: MPV works ok

Color profile on KDE set to either built in or said ICM above: MPV becomes problematic

[Leopard1907]

Also difference is:

Not using a color profile makes mpv use bt.1886 instead of linear, using color profiles makes it use linear which is broken.

@kasper93

[llyyr]

I'm using Asus's official ICM profile

Do not use ICC profiles you find on manufacturer websites, they're extremely unlikely to be accurate.

Do note setting KDE to use "built-in" profile has the same issue.

I'm not sure what built-in means here.

But yes as noted, linear is a bit broken. Try with --target-trc=bt.1886 instead. Linear shouldn't be picked already on newer libplacebo, though it needs a release still

[Leopard1907]

@llyyr

I'm not sure what built-in means here.

https://planet.kde.org/xavers-blog-2024-05-10-hdr-and-color-management-in-kwin-part-3/

There is a fourth option though: Use the color information from the display’s EDID. In Plasma 6.1, you can simply select this in the display settings.

Doing mpv --target-trc=bt.1886 '/home/beast/RDR2： But You Said You Knew Spanish - BUT IN LEGO (4K) [1Wyb6vFVy-M].webm'

when built-in option from KDE or icc file is selected still defaults to linear

[Headcrabed]

I'm using Asus's official ICM profile

Do not use ICC profiles you find on manufacturer websites, they're extremely unlikely to be accurate.

Do note setting KDE to use "built-in" profile has the same issue.

I'm not sure what built-in means here.

But yes as noted, linear is a bit broken. Try with --target-trc=bt.1886 instead. Linear shouldn't be picked already on newer libplacebo, though it needs a release still

Built-in seems to be using monitor EDID.

[llyyr]

Could you provide logs without --gpu-debug for both cases where mpv uses linear and mpv uses bt.1886? The version if libplacebo used fails validation and crashes, so the logs are completely useless when made with --gpu-debug

[Leopard1907]

@llyyr Sure. Here they are.

linear.txt
bt1886.txt