4 ms·

I sort of went the opposite way. I had a giant homelab already and paid for Plex lifetime (just because I thought it was good software after years of use, not b

I sort of went the opposite way. I had a giant homelab already and paid for Plex lifetime (just because I thought it was good software after years of use, not because I really needed the features or anything) but then I ended up consolidating all of my media to just being a bare metal Linux standard PC case running a plain NFS share (I guess that's still a NAS, but perhaps more spartan than the usual connotation) which clients like Infuse or a local media player app can just load directly.

bombcar 10mo ago
I used a SMB share with Infuse for awhile and it worked well, but adding Jellyfin in the middle made it much easier for the kids/wife to understand how to find things. (For awhile I was VLC off a ram SMB share but that was confusing even to me sometimes.)
- zamadatix 10mo ago
  Interesting, I'm curious how it helps. Maybe with non-video content or something? So far Infuse seems to do everything for shows/movies with the sorting, categorizing, posters, descriptions, actor info, search, release dates, etc already so long as you point it to where the video files are and name them based on the TVDB naming. I even set it up so my parents TV has access to it and they use it more than Netflix now. I don't have any non-video content though, so I'm not sure whether Infuse covers that part well (or at all).
  - bombcar 10mo ago
    I seem to recall the main issue was it finding new/moved/renamed content without forcing a full rescan (which would take forever).
    - zamadatix 10mo ago
      Ah, that makes sense. I only have just the media files (no metadata) stored flat in a single SSD based mount and the background autoscan seems to pick up changes/additions consistently for that snappy (especially since I only add new content but once a week or so) but if I were constantly adding things I wanted to watch on demand, used a more complicated directory structure, stored any metadata locally, had a bit more latency in the storage, or had multiple sources I'm sure that would turn into a bit of a sync mess real quick. I also found Infuse's scanning performance to be absolute dog shit over even clean 5 GHz Wi-Fi. I solved that with a wire, but it sounds like middleware would have skipped that problem as well.

bombcar 10mo ago

I used a SMB share with Infuse for awhile and it worked well, but adding Jellyfin in the middle made it much easier for the kids/wife to understand how to find things. (For awhile I was VLC off a ram SMB share but that was confusing even to me sometimes.)

zamadatix 10mo ago
Interesting, I'm curious how it helps. Maybe with non-video content or something? So far Infuse seems to do everything for shows/movies with the sorting, categorizing, posters, descriptions, actor info, search, release dates, etc already so long as you point it to where the video files are and name them based on the TVDB naming. I even set it up so my parents TV has access to it and they use it more than Netflix now. I don't have any non-video content though, so I'm not sure whether Infuse covers that part well (or at all).
- bombcar 10mo ago
  I seem to recall the main issue was it finding new/moved/renamed content without forcing a full rescan (which would take forever).
  - zamadatix 10mo ago
    Ah, that makes sense. I only have just the media files (no metadata) stored flat in a single SSD based mount and the background autoscan seems to pick up changes/additions consistently for that snappy (especially since I only add new content but once a week or so) but if I were constantly adding things I wanted to watch on demand, used a more complicated directory structure, stored any metadata locally, had a bit more latency in the storage, or had multiple sources I'm sure that would turn into a bit of a sync mess real quick. I also found Infuse's scanning performance to be absolute dog shit over even clean 5 GHz Wi-Fi. I solved that with a wire, but it sounds like middleware would have skipped that problem as well.