Tdarr
Tdarr will batch-transcode your media, for efficient storage and maximum-compatibility streaming. Tdarr utilizes unused CPU / GPU capacity across our platform to perform batch transcodes, keeping your library lean and mean!
How do I access Tdarr?
Tdarr will be available from your dashboard (https://<username>.elfhosted.com>), or directly, at https://<username>-tdarr.elfhosted.com
How to use Tdarr
Tdarr is powerful and flexible, and can be intimidating to configure. Our instance is pre-configured to work as best as possible with our hardware / resources, but this section will detail some of the important configurations necessary.
Worker setup
The following is required for the worker component of Tdarr:
Configure node
From the main Tdarr window, select the node, set 1 GPU transcode and 1 CPU healthcheck, and then click Options to add node-specific config:
Node options
Set the hardware encoding type to vaapi, allow the GPU worker to do CPU tasks, and set a low priority for FFMpeg/handbreak tasks:
Staging section
Back on the main Tdarr page, under Staging Section, ensure that Auto accept successful transcodes is checked (without this, the transcoded file will sit in /transcodes waiting manual approval, eventually filling up the /transcodes volume and crashing the pod):
Library setup
Each library which you want Tdarr to process needs to be setup as follows:
Source
For each source folder, ensure that the following options are checked for optimum performance:
Transcode cache
The Transcode cache section must be set to /transcode:
Transcode options
Your may require a custom combination of plugins, but the transcoding plugin which works best with our QSV hardware is the one highlighted below:
How do I get support for Tdarr?
- For general use of Tdarr, refer to the official site or to one of the links below.
- For specific support re your ElfHosted configuration / account, either post a topic in the #elf-forum channel, or call for 1:1 #elf-help.