Yes, completely on device. I happen to personally know one of the maintainers and got to discuss some of the code with her.
Also completely open source, so if you don’t trust the packaged apps, you could even build it yourself from source
I am running a snapcast server (on a very small VM, could be a RasPi) and have a few RasPi Zero 2W with micro-USB to USB-C adapter -> Apple USB-C to 3.5mm adapter and then feeding into various speakers/amps distributed through the house, running snapcast clients. Audio sync is absolutely flawless, devices can be grouped on demand and there are plugins for all kinds of inputs on the server.
Inital setup was a bit fiddly, bit still done in one afternoon (although I admit it helped being a software engineer to compile some code myself).
IIRC there is no dedicated „Air Force One“. It’s just whatever plane the POTUS happens to be on. Of course usually that boils down to very specific few aircrafts, but if he boards the bribe-jet, this would be „Air Force One“ at that point.
Bei mir/uns lädt automatisch alles zu einem immich zu Hause auf den Server.
Zeitstempel und GPS-Location versteht das und Lokal macht das ein Machine-Learning für Personen- und Objekterkennung.
Die erkannten Personen werden einmal mit Namen versehen und dann kann man da sehr gut drin suchen und finden nach Ort, Zeit und Inhalt der Bilder/Videos
I’ve read somewhere some time ago that they generally prefer the songs/files that are already cached somewhere close to the user to reduce costs for traffic whenever the user leaves the choice to Spotify
Yes, completely on device. I happen to personally know one of the maintainers and got to discuss some of the code with her. Also completely open source, so if you don’t trust the packaged apps, you could even build it yourself from source
https://gitlab.com/bloodyhealth/drip
I am running a snapcast server (on a very small VM, could be a RasPi) and have a few RasPi Zero 2W with micro-USB to USB-C adapter -> Apple USB-C to 3.5mm adapter and then feeding into various speakers/amps distributed through the house, running snapcast clients. Audio sync is absolutely flawless, devices can be grouped on demand and there are plugins for all kinds of inputs on the server. Inital setup was a bit fiddly, bit still done in one afternoon (although I admit it helped being a software engineer to compile some code myself).
https://github.com/snapcast/snapcast
IIRC there is no dedicated „Air Force One“. It’s just whatever plane the POTUS happens to be on. Of course usually that boils down to very specific few aircrafts, but if he boards the bribe-jet, this would be „Air Force One“ at that point.
Edit: I just checked. https://en.wikipedia.org/wiki/Air_Force_One
Bei mir/uns lädt automatisch alles zu einem immich zu Hause auf den Server. Zeitstempel und GPS-Location versteht das und Lokal macht das ein Machine-Learning für Personen- und Objekterkennung. Die erkannten Personen werden einmal mit Namen versehen und dann kann man da sehr gut drin suchen und finden nach Ort, Zeit und Inhalt der Bilder/Videos
Pretty sure that’s Carrot Weather
I’ve read somewhere some time ago that they generally prefer the songs/files that are already cached somewhere close to the user to reduce costs for traffic whenever the user leaves the choice to Spotify