It's not only the price hikes and the atomization/fragmentation of content, what made me cut the wire was that I couldn't get high quality in my computer unless I used X browser for Y service.
It's not only the price hikes and the atomization/fragmentation of content, what made me cut the wire was that I couldn't get high quality in my computer unless I used X browser for Y service.
I have a friend who calculated up their hardware costs (before ram/ssd/hdd-ageddon) and paying for all the streaming services costs came in lower over the span of something like 10+ years. That said, they have the best UI/UX experience I've seen and don't see ads so I'd still count it as a win.
Oh sure, you can use an old laptop (or seedbox-type thing), an external drive, and get pretty far but if you want to replicate streaming services for all the shows you care about, it's not cheap.
You shouldnt be spinning up cloud resources to host your media stack. You should be buying used hardware off cycle and cobbling it together. All of which costs time to operationalize it, ie tweaking gpu drivers on linux to get that old hardware working.
It only really makes sense if youre a real enthusiast, or amortize it across your family, friends, etc. But then youre breaking a bunch more laws and as a working professional that just isnt worth the liability for the value provided.
The real answer I think is simply to consume less content.
It's something that a good weekend with a local abliterated Qwen 3.8 local model could cook.
1. How much content are we talking about?
2. How many people use your server?
Yes, QuickSync is amazing but hard drives alone (for a larger collection) are incredibly expensive (today). Even before prices shot up I know people that sunk $8K+ into hard drives alone.
A single 20TB drive can hold quite a bit of media.
Not quite. Before the insane prices I was able to get 4x 20TB for $1,200, so $300/20TB, so $8000/300 = 26.6, call it 27 drives, so 27x20 = 540TB, a little over half a petabyte, not "petabytes" (before mirror/raid/etc).
If you were buying that over time and keeping your price-per-harddrive about the same, you'd come up with much lower (since you would have been buying smaller drives in the past. Also smaller/older drives have to be rotated out due to age/wear/too-small so it's not all additive.
But you are correct, there is a difference between "data hoarders" and people that pirate and it's an important distinction (eww, that sounds a little AI-y, but I wrote it).
The same old 6th-gen i7 desktop rig that I do everything else with also does this job quite well. It's not beefy by any modern measure, and I'd have built it and kept it going (and powered on) even if I never had any intention of running these background processes on it.
The only added hardware expense is one external hard drive (redundancy? hah, no -- not for this). This cost a couple of hundred bucks a several years ago, which is the equivalent of surprisingly few trips to Taco Bell.
One certainly can spend a quite lot on nice kit to accomplish any goal, but that's not an indication of any necessity to do so.
If a film of acceptable quality averages around 2GB (yay h.265), then a 10TB hard drive can hold 5,000 average films.
I have access to a well-seeded 20+ year old private tracker but usually use the above method to stream content for free these days.
(Couple with Plex/Jellyfin).
Once you set it up with Sonarr and Radarr and Plex/Jellyfin you will never go back to HBO, Netflix, blah blah blah blah.
My non-techie partner honestly uses it way more than I do because it's just so damn easy. You punch a request into radarr and after some time it magically appears in Jellyfin.
It was a bear to set up the first time, but smooth sailing after that
FWIW, I didn't want to spend hours or days on the setup, but Claude set the whole *arr stack for me in an hour on an older NUC, under Proxmox.
* The Free Software Media System | Jellyfin // https://jellyfin.org/