Yep, I only noticed because I got prompted to update when I ran it today!

It took about a week to generate for my library without hardware-accelerated MJPEG generation, at the default resolution in the Trickplay configuration. I let it use 8 threads but CPU use was close to nothing the whole time, even with priority bumped up to above normal.

It wound up consuming about 10GB of storage by the time it was done, for a library of 2.6TB. My library is mostly 1080p stuff, a mix of h.264 and x265.

Mine’s been running for about 5-6 days now, also not a huge library. I’m running Jellyfin in an LXC container on a host with 16 CPU cores. Started with 4 cores, but have bumped it up to 8. I have noticed that when it is generating the Trickplay images for h.264 content it only uses about 8% of the available CPU resources. When generating images for x265 it uses about 60-70%.It doesn’t seem to matter what the priority for the trickplay job is set to.

I assume I should probably wait for my multi-day running Trickplay task to finish before attempting an update, right? :)

My library isn’t huge (in my opinion anyway, a few hundred episodes of TV and maybe 100 movies). My Trickplay job is about 16% complete after 3 days, lol. My AMD iGPU doesn’t appear to be supported for the MJPEG stuff so I don’t get GPU acceleration, but I have Jellyfin set to allow 4 threads for generating Trickplay images, and am running on a 4-core VM that sits on physical hardware that isn’t slow at all. Looks like even though it is using 4 threads it is still only using one of the cores, as CPU utilization for the ffmpeg process doing it is always at about 25%.

At the rate my Jellyfin server is generating trickplay images right now the Android client might have support for them by the time it finishes.

I use one of those tiny mini PCs, with an AMD mobile CPU in it. It sips power but has enough oomph for transcoding when necessary. I’m sure the NAS that my library actually sits on uses way more power with its mechanical HDDs.

I run mine in an LXC container. I just snapshotted it in case of disaster and then ran apt update && apt upgrade.

That was a fascinating read, thanks for sharing!

Jellyfin is excellent. You can always just download it and run it on whatever you’re running your existing setup on and give it a try. The server’s available for a bunch of platforms.

I have a regular-ass Amazon Fire TV stick 4k max that I use to play my Jellyfin content. It has native hardware support for h.265. I can’t remember the last time it needed a stream transcoded. Of course, it is encumbered with the Amazon ecosystem, but it was cheap.

I also have one of those $20 Walmart Android TV boxes. The UI is a little slow in it but it plays the same Jellyfin content just fine, and you can replace the stock launcher on it with whatever you want.

Currently playing: Yakuza Kiwami, released 2016. Yep.

Free software that essentially lets you roll your own Netflix.

Malicious compliance. I hope the EU stomps on them hard.

Update: I wound up getting one of the Chinese mini PCs from Amazon. $300+tax got me a Ryzen 7 5700U, 32GB of RAM, 1TB NVMe storage, and a single 2.5Gb ethernet port. I can add a second interface via USB-C if necessary. Really not bad at all. I have Proxmox up and running on it already, with PiHole and Jellyfin already running in LXC containers. Jellyfin took a bit of screwing around to get the CIFS shares from my NAS and hardware-accelerated transcoding going, but everything works now!

Yes, and in fairness to them that’s how it has been. The ads are only for stuff sold on the kindle store too, so not like you’re seeing ads for hot singles in your area or some shit on your kindle. It’s only now that the store is flooded with this AI garbage that the ads have become annoying.

I do almost all of my reading on a kindle - one of the ones with ads on the lock screen, and for months now all of the ads have been for this type of no-effort, low-quality, AI-generated garbage. Amazon clearly doesn’t give a fuck as long as they make money.

0.16.4 has now been released hot on the heels of 0.16.3.

It was one of the earlier podcasts to blow up huge. It used to be great.

That’s a pretty reasonable ad load. Have you listened to 99% Invisible lately? Each episode is about 10 minutes long, and it seems like at least 50% of it is now ads and sponsorship stuff.