I use a seedbox to run qBittorrent with Jackett, and have a bunch of sources on there. I know there are other methods, but that works for me and I’m comfortable with it.

Rarbg has been gone for about five months and it’s harder and harder to find 4k TV rips. For current shows, sometimes you can get episodes as they air, but it’s rare to see full seasons.

The tipping point was when I searched for a Netflix show from two weeks ago just now, with zero 4K results.

I’ve always hated private trackers since they’re a pain in the ass, but is that the only option now? Do I need to look at a different method entirely? I’ve been torrenting since that existed, but if I need to do something else, let me know.

I don’t expect to get torrent sites in any responses, but you can do that if you have any 😂

  • Wes_Dev@lemmy.mlEnglish
    3·
    1 year ago

    For years I’ve assumed there was some kind of invisible image manipulation that effectively encrypts the Netflix user ID in the video feed itself, so they can use leaked video to pinpoint users that share recordings.

    Same type of thing paper printers do, and AI giants are trying to push now.

    • Auli@lemmy.caEnglish
      2·
      1 year ago

      How would that work? And if it was being done I’m sure it would have been discovered already. I mean it is just video.

      • Wes_Dev@lemmy.mlEnglish
        1·
        1 year ago

        Basically small deliberate changes to the brightness or hue of chunks or even individual pixels scattered across the frame. Netflix would have the original, and would know where to look for the changes that embed the user ID data, or whatever.

        They just get a copy of the leaked video, locate the ID, and take the user to court or ban them.

        I’m not great at explaining it, so look up image stenography to get an idea of what I mean.

        Anyway, I don’t have any proof they do that, but I guess I always assumed they did, because… why not?

    • Krauerking@lemy.lolEnglish
      2·
      1 year ago

      LoL I remember in the early 2010s a porn comic that actually did this and people were shocked because it seemed like a lot of work for a small niche thing.

      And then we realized that converting to a png and then back to jpg killed the identifier. But the system was in place and I have seen so many versions of and workarounds for it. Someone even had a custom clipper software that just looked for the excess identifier data on the end of the file and would trim it.

      I think it’s wise to always think they might be doing this kind of stuff, especially big companies, but breaking it is pretty easy usually even by accident.