Hey, Lutz from UC here. Things got a bit heated in here, so just a quick update to cool it down and clear things up: yes, we’re on the battery topics and continue to improve.
Two things are getting mixed up: battery percentage reporting vs. real standby drain.
The “0% / sudden drops” behavior is a known battery gauge issue with coulomb counting. It’s not a real battery level drop, but the level is reported incorrectly. We’re testing fixes internally.
Some good news: we’ve already got standby consumption improvements internally. The bigger gains involve an MCU firmware update for a microcontroller, so we’re being extra careful and testing a lot before rolling it out. With the current internal work we reached a significant improvement in sleep current while keeping Wi-Fi connected in standby.
Current best settings (today):
Wi-Fi on in standby / Bluetooth off, the Bluetooth part of our radio module can be a bit erratic right now and can push standby up to roughly ~20 mA.
About Prevent Sleep option in activies: if that’s enabled, the remote is basically fully on, so ~6 hours of battery life can be totally normal. Awake power is on average around ~1.5 W, but it depends a lot on the use case, especially how many integrations are running and what they are doing. For example, integrations trying to constantly connecting to unavailable devices or producing lots of log information.
One more important point: we can’t fully control what third-party integrations do. We’ve already seen cases where some keep the CPU busy or trying to constantly connect even when it looks like “nothing is happening”, and that will hit battery life hard. So if you’re seeing heavy drain, please try disabling extra/third-party integrations temporarily to see if the behavior changes, it’s a very quick way to narrow things down.
Please be nice - We’ll share more as soon as the next improvements are ready to ship broadly.