Artifacts on screen in F44

Ask Fedora
, ,
1

This thread is a continuation of other threads here on the forum:
watching videos in browser show artifacts
artifacts when playing videos

This morning I read this thread on the KDE forum: graphical -glitches and artifacts since 6.7.4 inwhich @barryascott wrote that people have mentioned since kernel 7.1.8 this matter is resolved. I updated my Kinoite and saw I was now on kernel 7.1.10:

System:
  Kernel: 7.1.10-200.fc44.x86_64 arch: x86_64 bits: 64 compiler: gcc v: 16.2.1
  Desktop: KDE Plasma v: 6 tk: Qt v: N/A wm: kwin_wayland dm: N/A
    Distro: Fedora Linux 44 (Toolbx Container Image)

Unfortunately the artifacts still show up, so, at least for me, it is not over yet. Are there any others who, with kernel 7.1.8 and/or higher, still have these problems?

$ inxi -Fzxx
System:
  Kernel: 7.1.10-200.fc44.x86_64 arch: x86_64 bits: 64 compiler: gcc v: 16.2.1
  Desktop: KDE Plasma v: 6 tk: Qt v: N/A wm: kwin_wayland dm: N/A
    Distro: Fedora Linux 44 (Toolbx Container Image)
Machine:
  Type: Laptop System: LENOVO product: 82RG v: Legion 5 Pro 16ARH7H
    serial: <superuser required> Chassis: type: 10 v: Legion 5 Pro 16ARH7H
    serial: <superuser required>
  Mobo: LENOVO model: LNVNB161216 v: SDK0T76461 WIN
    serial: <superuser required> part-nu: LENOVO_MT_82RG_BU_idea_FM_Legion 5
    Pro 16ARH7H Firmware: UEFI vendor: LENOVO v: JUCN68WW date: 12/05/2024
Battery:
  ID-1: BAT0 charge: 69.3 Wh (97%) condition: 71.5/80 Wh (89.3%) volts: 15.37
    min: 15.44 model: Celxpert L21C4PC1 serial: <filter> charging:
    status: not charging cycles: 187
  Device-1: hidpp_battery_0 model: Logitech Wireless Keyboard
    serial: <filter> charge: normal status: discharging
CPU:
  Info: 8-core model: AMD Ryzen 7 6800H with Radeon Graphics bits: 64
    type: MT MCP arch: Zen 3+ rev: 1 cache: L1: 512 KiB L2: 4 MiB L3: 16 MiB
  Speed (MHz): avg: 1096 min/max: 404/4787 boost: enabled cores: 1: 1096
    2: 1096 3: 1096 4: 1096 5: 1096 6: 1096 7: 1096 8: 1096 9: 1096 10: 1096
    11: 1096 12: 1096 13: 1096 14: 1096 15: 1096 16: 1096 bogomips: 102208
  Flags-basic: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3
Graphics:
  Device-1: NVIDIA GA106M [GeForce RTX 3060 Mobile / Max-Q] vendor: Lenovo
    driver: nvidia v: 610.57.04 arch: Ampere pcie: speed: 5 GT/s lanes: 8 ports:
    active: HDMI-A-1 empty: DP-7,DP-8,eDP-2 bus-ID: 01:00.0 chip-ID: 10de:2560
  Device-2: Advanced Micro Devices [AMD/ATI] Rembrandt [Radeon 680M]
    vendor: Lenovo driver: amdgpu v: kernel arch: RDNA-2 pcie: speed: 16 GT/s
    lanes: 16 ports: active: eDP-1 empty: DP-1, DP-2, DP-3, DP-4, DP-5, DP-6,
    Writeback-1 bus-ID: 34:00.0 chip-ID: 1002:1681 temp: 43.0 C
  Display: wayland server: N/A compositor: kwin_wayland driver:
    gpu: amdgpu,nv_platform,nvidia,nvidia-nvswitch display-ID: 0
  Monitor-1: HDMI-A-1 model: Philips 27M1C5500V res: 2560x1440 dpi: 109
    diag: 685mm (27")
  Monitor-2: eDP-1 model: BOE Display 0x0a9b res: 2560x1600 dpi: 189
    diag: 406mm (16")
  API: OpenGL v: 4.6 vendor: amd mesa v: 26.1.5 glx-v: 1.4 es-v: 3.2
    direct-render: yes renderer: AMD Radeon 680M (radeonsi rembrandt ACO DRM
    3.64 7.1.10-200.fc44.x86_64) device-ID: 1002:1681 display-ID: :0.0
  API: EGL Message: EGL data requires eglinfo. Check --recommends.
  Info: Tools: api: glxinfo gpu: nvidia-smi x11: xdriinfo, xdpyinfo, xprop,
    xrandr
Audio:
  Device-1: NVIDIA GA106 High Definition Audio driver: snd_hda_intel v: kernel
    pcie: speed: 5 GT/s lanes: 8 bus-ID: 01:00.1 chip-ID: 10de:228e
  Device-2: Advanced Micro Devices [AMD/ATI] Radeon High Definition Audio
    vendor: Lenovo driver: snd_hda_intel v: kernel pcie: speed: 16 GT/s
    lanes: 16 bus-ID: 34:00.1 chip-ID: 1002:1640
  Device-3: Advanced Micro Devices [AMD] Audio Coprocessor vendor: Lenovo
    driver: snd_pci_acp6x v: kernel pcie: speed: 16 GT/s lanes: 16
    bus-ID: 34:00.5 chip-ID: 1022:15e2
  Device-4: Advanced Micro Devices [AMD] Ryzen HD Audio vendor: Lenovo
    driver: snd_hda_intel v: kernel pcie: speed: 16 GT/s lanes: 16
    bus-ID: 34:00.6 chip-ID: 1022:15e3
  API: ALSA v: k7.1.10-200.fc44.x86_64 status: kernel-api
  Server-1: PipeWire v: 1.6.8 status: active (process) with:
    1: pipewire-pulse status: active 2: wireplumber status: active
    3: pipewire-alsa type: plugin 4: pw-jack type: plugin
Network:
  Device-1: MEDIATEK MT7922 802.11ax PCI Express Wireless Network Adapter
    vendor: Lenovo driver: mt7921e v: kernel pcie: speed: 5 GT/s lanes: 1
    bus-ID: 02:00.0 chip-ID: 14c3:0616
  IF: wlp2s0 state: up mac: <filter>
  Device-2: Realtek RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet
    vendor: Lenovo driver: r8169 v: kernel pcie: speed: 2.5 GT/s lanes: 1
    port: 4000 bus-ID: 03:00.0 chip-ID: 10ec:8168
  IF: eno1 state: up speed: 1000 Mbps duplex: full mac: <filter>
Bluetooth:
  Device-1: Foxconn / Hon Hai Bluetooth 5.2 Adapter [MediaTek MT7922]
    driver: btusb v: 0.8 type: USB rev: 2.1 speed: 480 Mb/s lanes: 1
    bus-ID: 3-3:3 chip-ID: 0489:e0d8
  Report: rfkill ID: hci0 rfk-id: 2 state: up address: see --recommends
Drives:
  Local Storage: total: 953.87 GiB used: 1.18 TiB (127.0%)
  ID-1: /dev/nvme0n1 vendor: Samsung model: MZVL21T0HCLR-00BL2
    size: 953.87 GiB speed: 63.2 Gb/s lanes: 4 serial: <filter> temp: 42.9 C
Partition:
  Message: No partition data found.
Swap:
  ID-1: swap-1 type: zram size: 8 GiB used: 0 KiB (0.0%) priority: 100
    dev: /dev/zram0
Sensors:
  System Temperatures: cpu: 46.1 C mobo: 44.8 C gpu: amdgpu temp: 44.0 C
  Fan Speeds (rpm): fan-1: 0 fan-2: 0
Info:
  Memory: total: 16 GiB note: est. available: 14.8 GiB used: 5.05 GiB (34.1%)
  Processes: 419 Power: uptime: 47m wakeups: 0 Init: systemd v: 259
    default: graphical
  Packages: pm: rpm pkgs: N/A note: see --rpm Compilers: gcc: 16.1.1
    Shell: conmon running-in: systemd inxi: 3.3.41

This is just an example of what I am still seeing:

image
image380×859 156 KB

In the middle of the picture you see a vertical line with different colors on either side although they should be the same.

2

I upgraded my workstation to Fedora 44 this week and I’m having the exact same artifacts watching YouTube. I’m on an AMD GPU and they look exactly the same. Your logs show two different GPU vendors so I don’t know if you were having these issues on the NVIDIA or the AMD one.

3

Hello,
Is there in the meantime a solution for the artifacts issue?
I am momentarily watching one Youtube video on 2 laptops, both with AMD CPU and GPU, one older laptop with openSUSE Tumbleweed fully updated and on my present laptop Fedora Kinoite fully updated.
The older laptop shows the movie just fine, the newer (with Kinoite) is a mess if I can be so honest.
Which packages need to be updated, and how can I see which version of those packages I am using so I can compare those between Tumbleweed and Kinoite?

Thanks.

4

As far as I know this was a Mesa issue and is fixed in Mesa 2.1.8. If updating to 2.1.8 is not possible on your system at this time then perhaps you can disable GPU media decoding in your browser as a temporary workaround to decode videos with CPU instead.

In Chrome/Chromium navigate to chrome://flags and search for Hardware-accelerated video decode, set it to disabled from the dropdown and restart your browser.

In Firefox-based browsers navigate to about:config and look for media.hardware-video-decoding.enabled, double click it to set it to “false” which will bold out the option, indicating it has been changed. Then restart your browser.

5

Thank you.
Do you also know how I can check the version number and the name of the package involved to see if it is updated already or not? Guess not because the screen still looks awful.
Thanks

6

The paste in your original post it has the info: API: OpenGL v: 4.6 vendor: amd mesa v: 26.1.5
So that’s still a few versions behind the fixed version. I have not personally used Kinoite but Mesa should update to a newer version in the near future just by keeping the system up to date I would assume. Hard to say how long that will take.

7

Current stable should be on 26.1.8. Would try updating the packages and see if that helps.

8

I’m running 26.1.8 currently and as you can see in these screenshots from this video, it’s not fixed yet, admittedly I’m on F44 KDE rather than Kinoite but data is data. I’m waiting now to see if this fix listed in the 26.2.1 notes: “RadeonSI VA-API: vertical color discontinuity / split on YouTube (AV1/VP9) with RX 7900 XTX” fixes it.

image
image2560×1440 751 KB

image
image1920×1080 265 KB

9

Interesting. I am also on F44 KDE and for me the issue was fixed on 2.1.8. I have an RX 9070 XT. I checked the exact part of the same video you posted a screenshot of and confirmed no artifacting (right clicking on video and checking the stats reveals av01 as codec). I guess it might still not be fixed on all AMD GPUs then.

10

I don’t think I can do that since it is probably part of the immutable part of the OS. I guess there is nothing else to do then waiting.
I now have my old computer as backup should I want to see a movie.

Thanks all.

11

Are those the versions in the toolbx container that inxi was run from, or the versions on the host?

12

No idea which data inxi gives out, it did run inside a toolbox container. Do you think it might give the data from the container instead of the host? If that is so then inxi is useless on an atomic system, right?

Graphics:
  Device-1: NVIDIA GA106M [GeForce RTX 3060 Mobile / Max-Q] vendor: Lenovo
    driver: nvidia v: 610.57.04 arch: Ampere pcie: speed: 5 GT/s lanes: 8 ports:
    active: HDMI-A-1 empty: DP-7,DP-8,eDP-2 bus-ID: 01:00.0 chip-ID: 10de:2560
  Device-2: Advanced Micro Devices [AMD/ATI] Rembrandt [Radeon 680M]
    vendor: Lenovo driver: amdgpu v: kernel arch: RDNA-2 pcie: speed: 16 GT/s
    lanes: 16 ports: active: eDP-1 empty: DP-1, DP-2, DP-3, DP-4, DP-5, DP-6,
    Writeback-1 bus-ID: 34:00.0 chip-ID: 1002:1681 temp: 43.0 C

The Nvidia driver is not installed in the toolbox, still that data is part of the inxi output. Doesn’t that mean it must be the host data?

13

I ran glxinfo -B which gave me all the info amount my GPU and mesa version, I’ve snipped the relevant lines below.

Extended renderer info (GLX_MESA_query_renderer):
    Device: AMD Radeon RX 7800 XT (radeonsi, navi32, ACO, DRM 3.64, 7.1.13-200.fc44.x86_64) (0x747e)
    Version: 26.1.8
14

This is my output:

glxinfo -B
name of display: :0
display: :0  screen: 0
direct rendering: Yes
Extended renderer info (GLX_MESA_query_renderer):
    Vendor: AMD (0x1002)
    Device: AMD Radeon 680M (radeonsi, rembrandt, ACO, DRM 3.64, 7.1.13-200.fc44.x86_64) (0x1681)
    Version: 26.1.8

I don’t have a dedicated AMD GPU, only the integrated one. At the moment I don’t use the Nvidia driver, only Nouveau is doing something now and then. Would it get better to install Nvidia as well? I must say I like the present situation because the laptop fan don’t spin as much and so the laptop is much quieter.

15

I don’t know. I put an Nvidia card in my system as an experiment earlier this year, maybe last year, to see if I could use it just as a Physx/Cuda accelerator but I couldn’t make it behave that way and it caused performance issues in GPU demanding loads.

16

As far as mesa is concerned, it seems to be giving you the data from the container (because glxinfo on the host shows you that you have Mesa 26.1.8).

It’s not exactly useless, but running it in a container has limitations. Ideally it would be run on the host. It might be interesting for the System Extensions SIG to consider ‘packaging’ it as a sysext.

That said, inxi is a single perl script and it could also be run on the host just by putting it in ~/.local/bin.

Kernel drivers are a slightly different case, because the container doesn’t have its own kernel and uses the host’s.

17

Im also running a 9070XT and i am still seeing the vertical bar issue on some videos from time to time on the current 26.1.8 mesa version. Its weird because i only noticed this happening after an update a couple of weeks ago.

18

Same with me. I’m also using an RX 9070 XT. No artifacts in that video. Strange … :thinking: