I have an extensive movie/tv show library. Been collecting for years and is a mixed bag of media types. I would like to convert all to h265 or something similar to try and conserve space. Currently my media is sitting at around 30TB’s I know from a previous post that is not as significant as some others but converting one by one would be a nightmarish undertaking. I am hoping someone has a cli script to do the conversion.
I am currently using radarr, sonarr, lidarr and readarr for my media I know lidarr and readarr are dead and am looking for an alernative either for those or the full stack. I would also like to perform the conversion of media from the downloaded format to what is suggested to the best format.
You must log in or register to comment.
Lidarr ain’t dead.
They juat don’t have too much time to fix the metadata issue.
Jesus folks. Have some patience with devs donating their free time after work to fix your little media addiction.
If they arent fast enough for you, do it manually or use another program…I never stopped using it and just got an update today or yesterday for it. I was going on what was posted here. I never meant to start any trouble.
It’s just what was sad that irked me.
No offense taken (and hopefully non given). I hope you continue enjoying this hobby :)
The question is why do you want to convert them?
If you use something like JellyFin as your media server and client, it will transcode them as you watch them. If you’re on your phone using crappy WiFi, it will adapt the bitrate down automatically. If you’re at home using a projector connected by Ethernet, you’ll get the original file.
If you think you’re running out of space, it’s often easier and cheaper to buy more HDDs. H265 and modern audio codec will save you a maximum of 50%. For about £200 you can get a 15TB HDD.
Hell, I’ve found ripping a DVD to MKV results in a 3-5gb file.
Then converting that MKV using handbrake, I can bring the size down as much as 75%. When you’re talking about a thousand videos, that adds up.
TV series (especially older stuff) I can consistently reduce 80%+. This makes a real difference for shows that were on for 10 years.
And these all look fine on a 65" TV from 6’ away. Why store more if I don’t have to?
Ripping straight from physical to media is just the remuxing process.
What you are doing is called an encode.
It can be good. But not as original as the original.It’s like saying an MP3 is as good as a FLAC…
It is not necessarily about saving space so much as it is about uniformity. And yes my server is beefy but you get 3 or 4 people all transcoding at the sane time and that beefy server will choke. As I have said I have been collecting for many years with a very large mixed bag of codecs I am just trying to clean up the mess that is my media.
OK, but why do you care about uniformity? If it’s just some OCD-adjacent urge - that’s fine; do what makes you happy.
But from a technical point of view, VLC will play back no matter what the codec and compression level.
Not exactly what you asked:
Is your plan to also reduce resolution and bitrate? If not:
- It’s going to reduce visual quality anyhow.
- Are you sure “changing codecs” would save you as much space as you think it would?
It depends where your material comes from. If it’s ripped from streaming services it’s (generally) already as compressed as it can be regardless of codec. E.g. Amazon mostly provides their streams in h264 (a 20 yo codec) and they’re very high quality at very small filesize.
If you still want to do that: I don’t believe there is one automated solution that will save quality and space across the board. (If there is, I’d like to hear about it)
A scenario I’d envision for myself:
- tinkering with ffmpeg for test results
- grouping material according to these results
- set up a separate ffmpeg-script for each group
- wrap the whole thing into a systemd service and make sure it survives reboots, i.e. picks up where it left.
I am not looking to adjust quality of the video or audio just change the codec. I am not necessarily converting ONLY from h264 my media consumption goes back many years and as such is a huge mixed bag of codecs. My videos are not coming exclusively from streaming services.
Some of it is likely still quite findable and assuming quite a few titles are many seasons of 1 show: use your known channels and redownload in more recent repacks would be the easiest, least hassle least risk of quality loss. Use Sonarr and/or jellyfin exports to identify shows with high GB per minute of runtime…
Don’t, unless your library is entirely in UHD/4K you’re going to seriously hurt quality by compressing with h265. And it won’t be that big if a difference in size. And if you ignore that advice at least don’t recompress it from already compressed sources. There are lots of h265 files made from high quality sources out there you should download to replace them instead.
There is a *darr program (tdarr maybe) that you can use to on-the-fly convert media, but unless you know what you’re doing you’ll just waste power/time over looking for torrents that are already HEVC. Also, depending on the age of the content, HEVC really wont save that much space.
Lidarr and Readarr DO work, but it relies on community support for it to be incredibly accurate (IIRC Lidarr uses Musicbrainz for its backend, and unless someone had added an album there you wont ever pull metadata for it) and takes a lot more time to correctly set up and configure in my experience.
There are a few guides as far as “the best” settings for media, trash guides comes to mind but his guides are aimed at having everything in “the best format” vs saving space
I’ll just leave this here https://locatarr.github.io/
I have no experience using it https://home.tdarr.io/
Internet people say that if you reencode the movies, you will lose information and sacrifice quality. I do not know if it is noticable! It is probably visible on a large screen and definitely not on a small one. Like always: shit in, shit out.
You should always use the best source possible. If you don’t have it, maybe someone who has it should do it and not you. I would wait for this person to do it and redownload it once it is available.
To me it just makes more sense that the person creating the 264/265 video does it instead of all the guys who just downloaded it. It’s also better for the environment if only one person does it.
265 is not the future gold standard anymore but av1 is. If you are to do it, go for av1 but make sure all your devices can play it natively.
I made a Handbrake profile with the settings i wanted, and used the “bulk add” function to add entire shows to the queue.
The media lives on the network, my desktop does the heavy lifting, and saves the resulting video to the hard drive. Then I copy everything back to the network.
It’s not terribly efficient, but I couldn’t get tdarr to work, so this works for me. Going forward, i just have to make sure my sonarr settings are on point
That’s what I did back in the day. When blockbuster closed down I ended up with over 1000 movies and a few bad sets of TV shows. I had to encode low back then as I only had 4tb of space.
Tdarr is what I use to unify my media. It does a good job of converting my files to h265. Runs in the background with my tesla p4 transcoding 3 streams at a time. Ripped through my 30ish tb library in a few days. It can change bitrate and resolution as well, but I haven’t had the time to play around with that yet. Can watch you library for change and transcode automatically, or you can run it when you like.
+1 for Tdarr. 10tb almost 11 saved since I set it up.
I am investigating TDARR I know a few people have suggested it. I like the nodes feature so I can use my desktop running a heavy GPU as my 2U server cant really run a graphics card (no power connectors for it). I am using a multi node server with E5 XEON chips.
No graphics card? Why do you even think about transcoding? Serious question
My server currently doesn’t have a video card (just the crappy on-board), it transcodes fine for one user (which is all I ever have). I don’t even notice an uptick in cpu when it’s transcoding (I’m sure it does, it just doesn’t seem to impact performance).
This is a 5 year old Dell SFF, running Windows, with 3 Windows VMs in VMware. It has no trouble transcoding, while converting videos using Handbrake. It’s maxed, but it does it.
I do plan to get a video card, it’s just not urgent.
Edit: Just did a test and with 2 simultaneous transcodes, Jellyfin will jump the cpu about 5% on video start, but settle back down to less than 1%.
Disk usage skyrockets with the second transcode, bouncing between 50% and 90%, and the network connection has some hard spikes at video start (with a gigabit connection).
The server that runs the bulk of my homelab does not have a graphics card. My TrueNas Scale server does as well as my desktop. The video’s are stored on the NAS and the remote shares are available on any system. So no graphics card in the server(s) is not ideal but They are beefy servers and still plenty capable of running ffmpeg but the conversion does not have to happen there.
Got it 👌🏼
