3 ms·
> But this seems rather unlikely to me. Are they saying that an application running continuously in the background uses the same amount of battery as an applica
by izacus 7y ago
> But this seems rather unlikely to me. Are they saying that an application running continuously in the background uses the same amount of battery as an application that's not running at all? Or am I just misreading that statement?
I don't see how that could be true - Android has seen massive battery life improvements after introduction of Doze which forces apps to sleep and stops the from constantly waking up radio and main ARM core. FCM is the only service that can wake up the device remotely and includes throttling code.
Before these changes Android had severe idle battery life issues (which were happily mocked by owners of locked down and proprietary iOS devices on this very website) since multiple apps waking up radio for notifications/synces caused radio/cpu to run pretty much constantly. It's worth noting that just doing a radio ping can increase power consumption by as much as 100x for next few minutes.
Maybe the Telegram report ignores the fact that they aren't allowed to wake up the device anymore. If you're measuring in isolation it's easy to make a case for "not making a difference".
- AstralStorm 7y agoThere is no way to wake up a device remotely ever. Firebase is doing the same thing anyone can, running on a schedule. It just batches the comms in one time slot. Android scheduler mechanism was supposed to offer the same benefit.