• Login
  • Register
  • Login Register
    Login
    Username/Email:
    Password:
    Or login with a social network below
  • Forum
  • Website
  • GitHub
  • Status
  • Translation
  • Features
  • Team
  • Rules
  • Help
  • Feeds
User Links
  • Login
  • Register
  • Login Register
    Login
    Username/Email:
    Password:
    Or login with a social network below

    Useful Links Forum Website GitHub Status Translation Features Team Rules Help Feeds
    Jellyfin Forum Support Troubleshooting [PSA] - Nvidia Drivers HEVC Transcoding

     
    • 0 Vote(s) - 0 Average

    [PSA] - Nvidia Drivers HEVC Transcoding

    Fatal Error on HEVC transcoding
    DTM450
    Offline

    Junior Member

    Posts: 2
    Threads: 1
    Joined: 2024 May
    Reputation: 0
    #1
    2024-05-28, 02:49 PM
    Hi all,

    Just a heads up I had to roll back my NVIDIA drivers to 552.44 and apply the Keylase-Nvenc Patch to restore HEVC transcoding and fix the fatal error message popup with my setup

    System Setup:
    Windows 10 WSL2
    Docker-Desktop v4.30.0
    Docker version 26.1.1
    Jellyfin version (LinuxServer Docker Image):
    Server version 10.9.3
    Web version 10.9.3
    Build version 10.9.3
    TheDreadPirate
    Offline

    Community Moderator

    Posts: 15,375
    Threads: 10
    Joined: 2023 Jun
    Reputation: 460
    Country:United States
    #2
    2024-05-28, 03:03 PM
    Which driver version was the source of your issue? On which GPU? Do you happen to have an ffmpeg log with the error you're describing.
    Jellyfin 10.10.7 (Docker)
    Ubuntu 24.04.2 LTS w/HWE
    Intel i3 12100
    Intel Arc A380
    OS drive - SK Hynix P41 1TB
    Storage
        4x WD Red Pro 6TB CMR in RAIDZ1
    [Image: GitHub%20Sponsors-grey?logo=github]
    DTM450
    Offline

    Junior Member

    Posts: 2
    Threads: 1
    Joined: 2024 May
    Reputation: 0
    #3
    2024-05-29, 03:55 AM
    Here is the additional info

    Driver Version 555.85 (GRD)
    GPU RTX 3080ti
    FFMPEG LOG: https://pastebin.com/raw/v4uK8gTR

    After doing some digging on the Keylase Git Repo apparently Nvidia has updated something in the new driver that makes it so you no longer require the Keylase Patch but I don't know if I cleaned up my installation properly. Despite running an nvidia/cuda container and it reporting the correct nvidia-smi details on the new driver additional testing was erroring when using the nbody container to test for cuda functionality.
    nyanmisaka
    Offline

    Team Member

    Posts: 232
    Threads: 0
    Joined: 2023 Jun
    Reputation: 8
    #4
    2024-05-29, 06:14 AM
    There's nothing we can do since it's an NVIDIA/CUDA driver regression.

    https://github.com/tensorflow/tensorflow/issues/68711
    « Next Oldest | Next Newest »

    Users browsing this thread: 1 Guest(s)


    • View a Printable Version
    • Subscribe to this thread
    Forum Jump:

    Home · Team · Help · Contact
    © Designed by D&D - Powered by MyBB
    L


    Jellyfin

    The Free Software Media System

    Linear Mode
    Threaded Mode