Thanks for watching guys! 👾 EDIT: I should note that this video assumes you are wanting to use a CPU which has iGPU capabilities for the transcoding. Apologies for any confusion. Don't forget to like and subscribe - it really helps! 🙌 Can't wait to share more soon - including our follow-up one month on from installing Ubiquiti UDM.✔️
Thanks for another vid mate, making us Aussies look good! Managed to combine this RAM technique with GPU transcoding enabled and has significantly sped things up. For anyone else wondering, to enable GPU transcoding (assuming you have an Nvidia card installed) you must first add the extra parameter "--runtime=nvidia" (without quotations) and then separated by a space from the RAM parameters/any other paramaters. You'll of course also need to have set up the appropriate drivers which you can find in Community Applications. A working config will look a little something like this "--runtime=nvidia --device=/dev/dri --mount type=tmpfs,destination=/tmp,tmpfs-size=20000000000 --no-healthcheck". This will still use your GPU to transcode but will instead offload that data to RAM instead of smashing your CPU and Cache resources. Happy homelab-ing! Edit: I should add that I have this tried and tested on linuxserver's repository but YMMV.
"--device=/dev/dri" is only for CPU transcoding. Since a recent update, the container will fail to spin up if you use this with GPU transcoding, so just remove it to get "--runtime=nvidia --mount type=tmpfs,destination=/tmp,tmpfs-size=20000000000 --no-healthcheck". And of course, adjust "tmpfs-size=xxxxxxxxxxx" to be appropriate for for how many bytes of RAM you're able to allocate in your server or you might exceed your total RAM which may lead to containers/vm's etc. starting to crash.
Warning to new guys here- I learned this the hard way. Setting the AppData config path to the cache (mnt/cache/appdata/PlexMediaServer) without a designated cache drive on your server will cause you to lose the settings on every reboot. For me this caused *every* container of mine to lose the appData every reboot, and so I had to set things up multiple times before figuring this out!
Another great video! I was really baffled why I didn't have some settings that you had thinking I had an old version. Turns out those were Plex Premium options! I need to get that.
Love the videos, impressive tutorials! - I have one question, seems they have changed the container a bit. Now there is a "Host Path 2" with a container path: "/transcode" Should we just leave it empty or delete it after implementing this method?
Hello, the video is very well explained, I have a question. Within the Unraid Plex configuration you must create a patch configuration and enter the following data. Container Path: /transcode Host Path: /tmp/plex_transcode ? Do I have to create this folder in the plex configuration directory? Thanks in advance Regards.
@@phillipmckeough8406 If it disappeared, worry not! Go to your /Docker page on unraid. At the bottom, clikc on "Add container". There should be in the template list your docker with all its configuration. Select the template and correct what broke it!
Giving your suggestion for RAM a shot over the dev/shm option I went with. Really keen to see your video on ports relating to remote access as this has always been one of my biggest pain points in plex.
Once I changed this, my libraries all disappeared from the network and I had to reassign the library folders and re-grant access to my users. Is there a reason for this? It is now also going through detecting intros and the likes as well. Does this mean its duplicating intros and other things, as its stating it is vengeraring video chapter thumbnails...when they have already been generated before? I am new to Unraid so I am not sure why it is doing this. Sorry for the question. Is there also a way to verify it is transcoding in the RAM? I am not sure where to see that at. Thank you in advance.
@ibracorp, hey folks, I followed the guide (thx) can anyone tell me how to confirm that the transcodes are in RAM in a dir called tmp instead of on the array in a dir called tmp? Previously I could use explorer to see the actual transcodes on my cache.... any insights would be appreciated.
Isn't /tmp a standard mapping in unraid for ram? I added a custom path (/transcode) to /tmp in the template and all seems to be transcoding through the ram 😊 transcode itself of course happens with a quadro M4000 card 😜
@@brandonakey6616 If you use the method given in this video, then you can control the maximum amount of RAM that plex will use for transcodes. I used the other method for awhile, and over time Plex actually filled my RAM up with a bunch of transcodes. This way you can limit the space, and Plex will actually remove the old transcodes to make room. Personally I double what they did here, and reserve 4GB of space since my homelab has 32GB of RAM anyway.
I... should've watched this video last night before I did my mass change on my media library. Now at least I know why I couldn't do anything for hours. Lesson well learnt I'd say!
Very good content. The only negative I would point out is that it's pretty unwatchable on a phone. Need to be zooming in on the settings as you talk about them so text is readable. I'll have to rewatch on my computer later.
Using unraid 6.9.2 I got a docker error due to /dev/dri not being available - bit of research, i just mapped /transcode to /dev/shm, I assume this does the same thing? However, no idea how to check that the transcoding is actually occurring within this ramdisk space. Video still helped though, so thanks for this although you uploaded it some months ago.
That's very interesting, how much ram does it really need? I don't have that much ram available and wonder how much ram can I set for transcoding while being useful.
I would say it really depends on how many transcodes you typically get but I always have to recommended getting a decent amount of RAM first - in helps in so many ways. But - if that's not a viable option, I recommend perhaps 8GB and see if that works well for you.
@@IBRACORP Yes, it uses Intel CPU's with integrated graphics so it offloads the CPU intensive task to the GPU. That's why you use a RAM disk if you don't have any GPU, let's say most AMD processors or some non GPU Intel versions and if you don't have plex pass. If you have any form of GPU I'd use that instead of a RAM disk. See: support.plex.tv/articles/115002178853-using-hardware-accelerated-streaming/
I should have clarified it a little more so thanks for correcting me on that one. I can see how it may be confusing for those unsure of their CPU capabilities.
@@IBRACORP While appreciating it's a non-trivial task to preface every video with an explanation of who the video is aimed at: explaining what kind of hardware set up they should be running for doing what follows to make sense, I really do think it's worth that preamble. Thankfully, I watched the video through and read the comments before I made any changes, but as I have an old GTX1050ti set up to do my transcoding, using RAM would be pretty pointless. I'd recommend pinning a comment to clarify that this is for people with an iGPU as currently it comes across as being something you'd recommend for everyone running Plex on unraid.
for who doesn't have Flex pass consequently doesn't have GPU transcoding --mount type=tmpfs,destination=/tmp,tmpfs-size=2000 --no-healthcheck 16GB = 17,179,869,184 bytes, thank me later
lifesaver mate - thought I'd just lost my entire frickin media library and was in the process of restoring from an old backup, thought id scroll comments first - cheers heaps man!
Use hardware acceleration when available is for GPU, NOT your CPU. If you don’t have that selected your CPU will perform the transcoding…. Edit: still not using your “CPU” it’s using the GPU integrated in the cpu. A faster cpu doesn’t change the GPU specs so it’s still GPU
Thanks for watching guys! 👾
EDIT: I should note that this video assumes you are wanting to use a CPU which has iGPU capabilities for the transcoding. Apologies for any confusion.
Don't forget to like and subscribe - it really helps! 🙌
Can't wait to share more soon - including our follow-up one month on from installing Ubiquiti UDM.✔️
Was wondering, mine doesn't have APU. Looking for cheap gtx that can do 265.
Thanks for another vid mate, making us Aussies look good!
Managed to combine this RAM technique with GPU transcoding enabled and has significantly sped things up. For anyone else wondering, to enable GPU transcoding (assuming you have an Nvidia card installed) you must first add the extra parameter "--runtime=nvidia" (without quotations) and then separated by a space from the RAM parameters/any other paramaters. You'll of course also need to have set up the appropriate drivers which you can find in Community Applications. A working config will look a little something like this "--runtime=nvidia --device=/dev/dri --mount type=tmpfs,destination=/tmp,tmpfs-size=20000000000 --no-healthcheck". This will still use your GPU to transcode but will instead offload that data to RAM instead of smashing your CPU and Cache resources. Happy homelab-ing!
Edit: I should add that I have this tried and tested on linuxserver's repository but YMMV.
Thanks Andrew appreciate the support! And love the additional info for those using dedicated cards
what is --device=/dev/dri use for, i have nvidia runtime but not this,
Dev/dri should be there even with NVIDiA?
"--device=/dev/dri" is only for CPU transcoding. Since a recent update, the container will fail to spin up if you use this with GPU transcoding, so just remove it to get "--runtime=nvidia --mount type=tmpfs,destination=/tmp,tmpfs-size=20000000000 --no-healthcheck". And of course, adjust "tmpfs-size=xxxxxxxxxxx" to be appropriate for for how many bytes of RAM you're able to allocate in your server or you might exceed your total RAM which may lead to containers/vm's etc. starting to crash.
@@DestinyKey13 I get an error that says the directory already exists. Any suggestions on how to fix this?
Warning to new guys here- I learned this the hard way. Setting the AppData config path to the cache (mnt/cache/appdata/PlexMediaServer) without a designated cache drive on your server will cause you to lose the settings on every reboot. For me this caused *every* container of mine to lose the appData every reboot, and so I had to set things up multiple times before figuring this out!
Welp, this is the answer I have been looking for. Thanks! I guess I need to set up a Cache drive.
Another great video! I was really baffled why I didn't have some settings that you had thinking I had an old version. Turns out those were Plex Premium options! I need to get that.
Love the videos, impressive tutorials! - I have one question, seems they have changed the container a bit.
Now there is a "Host Path 2" with a container path: "/transcode"
Should we just leave it empty or delete it after implementing this method?
Same question as @skalikid. Any suggestions would be great!
Still need an answer..
/tmp
that pre-roll video is really cool
Hello, the video is very well explained, I have a question. Within the Unraid Plex configuration you must create a patch configuration and enter the following data.
Container Path: /transcode
Host Path: /tmp/plex_transcode ? Do I have to create this folder in the plex configuration directory?
Thanks in advance Regards.
I followed your steps right up the the 4:59 mark and after I saved, the container vanished. What happened?
did you figure this out. happened to me too
@@phillipmckeough8406 If it disappeared, worry not! Go to your /Docker page on unraid. At the bottom, clikc on "Add container". There should be in the template list your docker with all its configuration. Select the template and correct what broke it!
what can really help is the scripts to copy and paste in the description, trying to type the huge strings is pointless
Giving your suggestion for RAM a shot over the dev/shm option I went with. Really keen to see your video on ports relating to remote access as this has always been one of my biggest pain points in plex.
Best thing about Sunday’s are your videos. Keep up the good work
Thank you mate that means a lot! Enjoy your Sunday!
Once I changed this, my libraries all disappeared from the network and I had to reassign the library folders and re-grant access to my users. Is there a reason for this? It is now also going through detecting intros and the likes as well. Does this mean its duplicating intros and other things, as its stating it is vengeraring video chapter thumbnails...when they have already been generated before? I am new to Unraid so I am not sure why it is doing this. Sorry for the question.
Is there also a way to verify it is transcoding in the RAM? I am not sure where to see that at. Thank you in advance.
Thanks for the video!! really interesting.
Is there a way to check that the config is properly done, and it is using ram?
Could you make one of these videos for Emby with the ram transcoder dir.
@ibracorp, hey folks, I followed the guide (thx) can anyone tell me how to confirm that the transcodes are in RAM in a dir called tmp instead of on the array in a dir called tmp? Previously I could use explorer to see the actual transcodes on my cache.... any insights would be appreciated.
Isn't /tmp a standard mapping in unraid for ram? I added a custom path (/transcode) to /tmp in the template and all seems to be transcoding through the ram 😊 transcode itself of course happens with a quadro M4000 card 😜
This is also true! More than one way to skin a cat 😸
I was wondering if either method has an advantage?
@@brandonakey6616 If you use the method given in this video, then you can control the maximum amount of RAM that plex will use for transcodes. I used the other method for awhile, and over time Plex actually filled my RAM up with a bunch of transcodes. This way you can limit the space, and Plex will actually remove the old transcodes to make room. Personally I double what they did here, and reserve 4GB of space since my homelab has 32GB of RAM anyway.
I... should've watched this video last night before I did my mass change on my media library. Now at least I know why I couldn't do anything for hours. Lesson well learnt I'd say!
That's how we learn!
thank you for the video! is there a difference if you add a path to the plex container called /transcode and path /tmp? if you have plex pass.
Nope as long as you use the same name inside Plex to make sure it's mapped to the same place
@@IBRACORP thank you!
thanks for the tips! i hope it helps my Synology.
Very good content.
The only negative I would point out is that it's pretty unwatchable on a phone. Need to be zooming in on the settings as you talk about them so text is readable.
I'll have to rewatch on my computer later.
Thanks Barrie yeah I'll be trying to change that you very soon
hey how can I check if it is really transcoding using RAM. I know for nvidia cards its nvidia-smi
Came to try to find the same answer. Been looking everywhere. And does it matter if I already have things set to transcode to my i5's iGPU?
is 8GB enough? and what happens if the allocated space gets filled up?
thank you!!
Using unraid 6.9.2 I got a docker error due to /dev/dri not being available - bit of research, i just mapped /transcode to /dev/shm, I assume this does the same thing? However, no idea how to check that the transcoding is actually occurring within this ramdisk space. Video still helped though, so thanks for this although you uploaded it some months ago.
As always good stuff man
Thank you John, was bit rocky getting back into it this week
¿Se puede transcodificar con Intel 12600, dev/dri? UnRAID 6.10. RC4
In the "Remote access settings" the values for "Internet upload speed" and "Limit remote stream bitrate" are in megabit (Mb), not megabyte (MB).
You are correct! Thanks for the correction
That's very interesting, how much ram does it really need? I don't have that much ram available and wonder how much ram can I set for transcoding while being useful.
I would say it really depends on how many transcodes you typically get but I always have to recommended getting a decent amount of RAM first - in helps in so many ways. But - if that's not a viable option, I recommend perhaps 8GB and see if that works well for you.
Is this applicable for Jellyfin?
no
Great video, just a comment though. Hardware acceleration doesn't use your CPU, it uses your GPU.
Thanks for watching mate. I believe you might be incorrect? Intel Quick Sync will be utilised by hardware acceleration
@@IBRACORP Quick Sync Video
Clear Video - video decoding using a general purpose Intel GPU
@@IBRACORP Yes, it uses Intel CPU's with integrated graphics so it offloads the CPU intensive task to the GPU. That's why you use a RAM disk if you don't have any GPU, let's say most AMD processors or some non GPU Intel versions and if you don't have plex pass. If you have any form of GPU I'd use that instead of a RAM disk. See: support.plex.tv/articles/115002178853-using-hardware-accelerated-streaming/
I should have clarified it a little more so thanks for correcting me on that one.
I can see how it may be confusing for those unsure of their CPU capabilities.
@@IBRACORP While appreciating it's a non-trivial task to preface every video with an explanation of who the video is aimed at: explaining what kind of hardware set up they should be running for doing what follows to make sense, I really do think it's worth that preamble. Thankfully, I watched the video through and read the comments before I made any changes, but as I have an old GTX1050ti set up to do my transcoding, using RAM would be pretty pointless. I'd recommend pinning a comment to clarify that this is for people with an iGPU as currently it comes across as being something you'd recommend for everyone running Plex on unraid.
Chào bạn bạn chia sẻ rất hay và thú vị, ngày mới chúc bạn luôn vui khỏe bình an HP và thành công trong mọi lĩnh vực nhé ❤️❤️❤️
for who doesn't have Flex pass consequently doesn't have GPU transcoding
--mount type=tmpfs,destination=/tmp,tmpfs-size=2000 --no-healthcheck
16GB = 17,179,869,184 bytes, thank me later
My container never starts if I add the following --device=/dev/dri--mount type=tmpfs,destination=/tmp,tmpfs-size=200000000000 --no-healthcheck
You're missing a space where --mount starts
not to mention copying that completely means 20GB of RAM will be used, you have to change the 200000000000 to something that coincides with your build
Your connection is 40 mega bits, not bytes. divede by 8 and you've got it in bytes.
Cheers!
setting appdata to cache assumes you haven't renamed your cache pool. if you have, use the name of the cache pool
lifesaver mate - thought I'd just lost my entire frickin media library and was in the process of restoring from an old backup, thought id scroll comments first - cheers heaps man!
For Ubuntu and ram transcoding this is all you need: /dev/shm:/transcode:rw
Nice one thanks for sharing!
thx works for unraid as well
Do you add this on the Plex transcoder directory?
Thanks
Use hardware acceleration when available is for GPU, NOT your CPU.
If you don’t have that selected your CPU will perform the transcoding….
Edit: still not using your “CPU” it’s using the GPU integrated in the cpu. A faster cpu doesn’t change the GPU specs so it’s still GPU
CPUs with integrated iGPU do the job, just like ours!
FIRST