How to use integrated graphics for Plex transcoding

Plex is my platform of choice for streaming local media. It’s great for streaming my Blu-ray collection to any platform I use that have a Plex app on it. But not all platforms support direct playback of Blu-ray rips (mkv-files). I can still play it but the video needs to be transcoded by the Plex media server first to a format the player supports. That process can be quite intensive for an uncompressed Blu-ray.

Plex supports hardware acceleration to make this process more efficient and so that it won’t eat up your entire CPU. You can activate hardware acceleration in the settings for the Plex media server under Transcoder.

When you do this Plex will use your graphics card instead for transcoding. If you only have one graphics card in your computer then there is not much more you can do. But if you have a desktop PC chances are you have an integrated GPU inside your CPU and a discrete GPU that you maybe use for gaming.

This is very likely if you are on Intels mainstream platform. For instance I have a 7700K which have an integrated GPU. If you open up Task Manger in Windows 10 and lookup the model number for your CPU you can search for it on Google and find out if you also have an integrated GPU. If you have one it should show up in Task Manager as GPU0 or 1. This is what I see.

If your CPU is supposed to have an integrated GPU but you can’t see it then it’s probably because it has been deactivated in the EUFI (BIOS) of your PC. For instructions on how to activate it you can check the manual for your motherboard.

Windows 10 makes it possible to choose which GPU a specific program should use. This makes it possible for us to assign the Plex transcoder to run on the integrated GPU instead of the default, in my case, Nvidia GPU. That means I can max out a AAA game while I’m also transcoding Blu-rays without taking almost any performance hit.

To set this up you need to open the Settings app in Windows 10. Then go to System -> Display and scroll down to the bottom. There click on the link called Graphics settings. Choose Classic app and then click on Browse.

You need to do this twice to add the files Plex Media Server.exe and PlexTranscoder.exe. In my case these files are located in: C:\Program Files (x86)\Plex\Plex Media Server

When you have done that click on each of the entries and choose Options. Your integrated GPU should show up as Power saving GPU. Choose the Power saving alternative for each of the programs and then save.

When you are done it should look like this.

It used to work to only add the PlexTranscoder.exe but for some reason I now also have to add the Plex Media Server.exe for it to work.

To be on the safe side I would recommend you reboot your computer now. When you have done that you can try to stream to a device where you know transcoding will have to be used and look in Task Manager as the GPU usage for your integrated GPU rises.

3 Replies to “How to use integrated graphics for Plex transcoding”

  1. Thanks. I did that. I think there is an issue with Plex. When I watch plextrasncoder.exe in Task Manager, GPU Engine shows that for 1080p and below hw decoding, it uses the igpu. For 4k it goes to the NVIDIA. Not sure why!

  2. This is great, thanks. I just enabled on my plex server system with an i3 10105 and a NVIDIA GT1030.

    When I transcode a video via plex, I notice in task manager both the Intel gpu and 1030 “video decode” monitor share the exact same graph, as in they are both being used? Meanwhile, the 3D graph shows a big spike in usage with the intel gpu, but not the NVIDIA one. How can I make sure it’s only using the Intel GPU?

    1. I’m not using Plex on Windows anymore and things might have changed since I wrote this. But you can while transcoding something look in in task manager and see which processes are using the GPU, it might be that you need to and more exes to the list of exes that use the iGPU.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.