Server Version#: Latest 1.14, not installed yet
Player Version#: not installed yet
Hello, I am getting ready to install my first Plex server and I wanted to bounce it off the community here to see if see if I have missed anything or should be doing something different.
I am almost finished building my new dedicated home theater. To begin with I will watch movies just with a 4K BD, Lumagen video processor, and a 4K/5K/Cinemascope Barco projector that supports HDR.
I want to start the work of setting up a system that can get as close as I can to being able to playback ripped 4K disks through Plex, with quality as close as I can get it to when watching those discs directly from a BD player. And of course all lesser formats as well. It’s especially important that can I can do all the sound formats like Atmos/DTS/Auro-3D.
The client I am planning to use will be an NVidia shield pro, which I haven’t bought yet.
I have a file server running Windows Server 2022 (soon to be 2025). It runs on an older SuperMicro server with a Xeon E5 2698v3, 128GB RAM, and an LSI hardware RAID controller. I currently have 6x22TB enterprise SATA drives (with a total of 16 slots available for expansion), as well as 400GB of cache for that volume using enterprise SAS3 write-intensive SSDs. The OS drive and a separate volume for Hyper-V VMs also run on the same type of SSDs. The server has 10Gig networking as well. This yields about 80TB of storage with R6. The only VMs the server runs right now are some Minecraft servers for my kids.
So, my questions are:
- Should I just install the Windows version of the Plex server and call it a day? Or is there a compelling reason to do something different?
- I don’t intend to stream/transcode to lesser devices, although I am aware subtitles might require that. The CPU in the server is powerful but dated, so I don’t know if that will be an issue. I will eventually upgrade the server and could even toss a GPU in there if I wanted to.
- IF there is an issue with this, I could build a smaller physical server with whatever hardware would be best, enough local SSD storage for the database, and then the server would only be hosting the media itself, but basically a dedicated Plex server.
- I could see also streaming to the living room TV, that would also be 4K and maybe another Shield, unless I could just install the Plex client on the TV directly (Google).
- With the BD player and Lumagen, I will do HDR tone mapping and so forth in hardware, and I am assuming the Shield could do the same with the Lumagen, but I am new to HDR.
- Any reason to use a different platform for the Plex server rather than Windows directly?
- I considered Docker, and if I did that I would not use WSL, more likely Ubuntu or even Photon running as a VM on the Hyper-V server, or in the case of the separate dedicated server, just installed directly without docker (again doesn’t need to be Windows). My current docker containers run on Photon on a separate ESXi host on a Photon VM with docker installed, right now just Pi-Hole and Portainer but I plan to do others. That server is also really old so I have been considering something newer to be my docker host.
- But, so far I don’t know if there is any particular advantage to running it on the current Windows server, vs. docker, vs. a separate Linux machine either as a VM or physical machine. If I need hardware acceleration I can pass-thru a GPU to the guest and/or have a host with Quicksync support (which the Xeon in the current server does not).
So, overall I think I am pretty close based on my research, but trying to match the BD player in quality, or perhaps soon kaleidescape are big shoes to fill, so I could be missing something important.
My expectations are that I might be able to get within 90-95% of the quality of the BD player if I am lucky. But the rest of the content I plan to put in the Plex server is all my standard BDs that I don’t plan to upgrade to 4K, DVDs, Laserdiscs, etc. so that I don’t have to go digging in the closet for the program I want to watch, or dust off a 30-year-old LD player ![]()
Thanks very much in advance.
-Jeff