Battery draining
-
@giampy
Mobile data is known to drain the battery faster than WIFI (especially 4G) but that doesn't explain how OTA16 may have changed something.
The power consumption for the modem should not have changed... -
@applee said in Battery draining:
Mobile data is known to drain the battery faster than WIFI (especially 4G)
Is that specific to this particular phone or does it apply to all of them?
-
@cliffcoggin
This is the same on my phone, that's why i run most of time on 3G and only switch to 4G when i need more bandwich (more bandwich = more energy).
Wifi consume less because you are near source (most of time at home), so power needed is lower for same bandwich.This probably also has to do with frequencies used in each case.
-
@cliffcoggin
I would say the same as @Keneda
Of course depending on your hardware it could be mitigated, but in overall yes 4G uses more power than wifi. -
@kugiigi my Xperia x i don't use it as a daily driver, but it does get a sim and always on wifi. Prior to otp 16, I easily lasted for more than 48 hours, but now it's like 30-34 hours.
-
@applee said in Battery draining:
@cliffcoggin
I would say the same as @Keneda
Of course depending on your hardware it could be mitigated, but in overall yes 4G uses more power than wifi.Thank you both. I was not aware of that.
-
I also estimate the higher loss at about 30%.
It's not a big deal for me either, but in the blog <the release of Ubuntu Touch OTA-16> an improvement in battery life on the Xperia X was announced.Translated with www.DeepL.com/Translator (free version)
-
Maybe it's related to the latest kernel changes (Jan and Feb):
https://github.com/fredldotme/device-kernel-loire/commits/ubports/LA.UM.5.7.r1-ng -
@c4pp4 This sounds like a clue, because I changed directly in the <stable channel> from OTA-15 to OTA-16. So without fluent update in the <developer channel>.
-
@mario-ch Look how many changes was made in a couple of days so I think fluent update wouldn't help.
-
@c4pp4 I meant that I noticed the difference, between 2 Dec 20 to 15 Mar 2021 better than if I had done the small updates week by week.
-
@mario-ch I know what you mean but e.g.:
Jan 21 - 7 commits
Jan 24 - 7 commits
Jan 27 - 13 commits
Jan 29 - 13 commits
Jan 31 - 17 commitsNo chance to recognize when the change was made.
-
After discussing here, my Xperia X just went down to 20% in just 13 hours
Do you have apps that is always running on your phone? Mine is usually TELEports and Pesbuk. -
Recent changes of CPU governors related to suzu:
Jan 22: https://github.com/fredldotme/device-kernel-loire/commit/2feb357a3e4a605b687cef3d3d6c572ae11cb9fc
CPU_QUIET_GOVERNOR_RQBALANCE:
This governor will scale the number of CPUs online depending on both CPU load and the number of runnable threads with careful but simple and precise calculations.
CPU_FREQ_GOV_SCHEDUTIL:
This governor makes decisions based on the utilization data provided by the scheduler. It sets the CPU frequency to be proportional to the utilization/capacity ratio coming from the scheduler. If the utilization is frequency-invariant, the new frequency is also proportional to the maximum available frequency. If that is not the case, it is proportional to the current frequency of the CPU. The frequency tipping point is at utilization/capacity equal to 80% in both cases.
Maybe @fredldotme could bring back availability of previous default governor, I'm not sure if it was "interactive" or what. Then we can try to test it via command:
echo "interactive" | sudo tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor -
I got mixed responses from the recent Xperia X work. Some say battery life has improved, now some claim it got worse. Generally those who got worse battery life had suspension prevention enabled in some apps.
@c4pp4 I'm not sure that schedutil is at fault here. Also we need it as it sets CPU frequencies in a kthread context, which is required for some mutexes to work correctly and not drag down the system into a panic.
You could try reverting this commit locally and see for yourself if it's due to the way the CPU is actually idling: https://github.com/fredldotme/device-kernel-loire/commit/84d9466cc10cc869b3a19086208cf4c6c59ef9b0
IMPORTANT: You will lose the ability to get notifications on time.
I'd rather suggest not using suspension prevention in UTTT as that's not a supported configuration for exactly this reason.
-
@fredldotme
Can you describe how they are using the phone so the battery life has improved? I'm not using UTTT at all. Wi-Fi OFF. Data 3G ON. I have Dekko2 and TELEports installed. I'm using Morph and listening to the music. Anbox is off. That's all and I'm "happy" with 12 hours. Unfortunately I have no hardware to build the test at the moment.
Idle test from full charge: Wi-Fi OFF, Data OFF, two active SIMS, display turned OFF all the time, after 12 hours it's on 80%. Is it OK?
I'll take a notes about phone using (time and task) and try to compare it with battery usage graph and let you know.
-
@fredldotme Stable channel OTA-16 and I've reinstalled all my apps for sure. It looks like Morph browser (and maybe some other apps?) is eating more battery power (javascript sites?). On the other hand I've noticed better system response. I think it's the result of the governor changes. Without using Morph too frequently it lasts quite a long in my case (cca 20h). I think it's a great work after all. I will be monitoring usage again and when I find out something new I'll place a note.
-
@c4pp4 cool
-
As I mentioned before, the more consumption of the battery, is not a serious problem for me. Because I think that if the changes made contribute to a more stable and secure system, that is more important to me. Thank you for your work and the 'commitment'. I wish you a happy Easter.
Translated with www.DeepL.com/Translator (free version)
-
Lately, the more battery drain is more apparent on my Xperia X. The drain happens on idle. I get around 20% drain overnight with wifi turned off. I have Dekko and Ubsync though, so perhaps those have impact too.
