1. Created new "Path" variable named RAM Transcode Path: /RAM
2. Container path is /RAM
3. Host path is /dev/shm
4. Access Mode is Read/Write
5. Modified Plex in Settings > Transcoder to reflect the transcoding directory change of /RAM
6. Restarted Plex server to apply changes

Configure a path if not one exists:

Name: Transcode
Container Path: /transcode
Host Path: /dev/shm/

That's all that's needed for the container setup

Then, inside Plex settings
Settings > Transcoder > Transcoder temporary directory
/transcode

Side note: the host path can be /dev/shm/ or /tmp

They are both RAM drives /dev/shm/ using no more than 1/2 your RAM while /tmp will use up to all of it., neither of which will use more than 1/2 of your available RAM.

There's no wrong answer which you use.

I am using Intel 10100 w/ Quick Sync RAM transcoding.. This is all i had to do to get it running...

Step 1:

In the container, I created a path from the RAM to /transcode, so the location transcodes occur in are in the system ram shared with the unRaid OS.

in the container (binhex-plexpass) in my case... *you need plexpass active for quick sync transcoding...

Name: Transcode

Container Path: /transcode

Host Path: /dev/shm

Then you will need to go into the docker container and change transcoding inside plex to /transcode under settings-transcoder-transcoder temporary directory

Step 2:

This was required for my system to use Quicksync properly as the out-of-box docker container wouldn't work naturally. I followed this guide to give permissions and change access settings for the docker to properly use the iGPU driver for transcoding.

Guide i followed: https://forums.unraid.net/topic/77943-guide-plex-hardware-acceleration-using-intel-quick-sync/

Those two steps is all i needed to get mine working on the Intel 10100 for video transcoding, Audio still uses the CPU unfortunately, but that has minimal impact on the CPU overhead as i can run 20+ streams at the same time transcoding with no issue.. The only way i have found to get the iGPU to perform hardware transcoding on both video and audio is to use a Linux VM, but when i tried it out the system resources required to run the linux VM to gain this was not worth it. In linux gaining audio transcode my system would only do 17 transcodes at the same time, but in the docker container, I could get 20 before maxing out the CPU.

ZFS Essentials: Array Disk Conversion to ZFS or Other Filesystems - No Data Loss, No Parity Break!
ZFS Essentials: Auto-Converting Folders to Datasets on Unraid
Mastering ZFS: Easily Auto Snapshot & Replicate a ZFS Dataset - Part A
Mastering ZFS: Snapshot to Rsync Replication to Non-ZFS Destinations - PART B