Ran Bluetooth mode for months because I assumed simpler equals less drain. Sleep detection sounded cool but I figured all the sensor processing running in the background would cancel out whatever it saved by killing ANC early.
Switched my Somnipods to auto mode with 1h ANC limit a few weeks ago mostly out of curiosity. Expected it to be a wash at best.
Every morning I'm waking up with 15 to 25 percent more charge than I used to. Not a little more. Noticeably more. The thing that's supposed to do more is somehow doing less to the battery, which is kind of wild when you think about it.
The only explanation that makes sense to me is that the ANC is a bigger drain than anything the sleep detection is running, so once the algorithm decides I'm out and cuts it, the savings stack up over however many hours I'm actually asleep. My podcast window is usually 20 to 40 minutes, so there's a solid 5 or 6 hours where it's just sitting there in low power mode.
What I'm lowkey curious about is how it's reading me specifically. I'm a still sleeper, basically don't move once I'm horizontal, so I'd think the buds would flag me as asleep pretty fast. But the timing feels right, it's not cutting ANC while I'm still listening. For people who move around a lot during sleep the detection is probably tracking something completely different, and I wonder if that changes how efficiently the whole thing works for them.