Related
Hey guys.
I have a clean Mahdi ROM (the latest from like 1 week ago) and have the feeling that the battery drains slightly faster than in stock. Apart from other things I started trying different kernel tweaks and I discovered the following:
The governor, although showing explanations for a whole 7 different modes, only allows me to either pick "interactive" or "performance".
Since I never really need the CPU settings from interactive mode, I actually want to use conservative.
How can I realise that now?
Please feel free to post your battery life stats, I'm really curious.
Sidenote: I have "transparent weather clock" on my homescreen and it seems to need a lot of battery. Does anyone have a recommendation for a lightweight & goodlooking homescreen widget?
The kernel for that ROM only contains two governors, please do not use performance because it only uses your max frequency, which is bad for temp & battery life.
Ace42 said:
The kernel for that ROM only contains two governors, please do not use performance because it only uses your max frequency, which is bad for temp & battery life.
Click to expand...
Click to collapse
Ok, thanks for the info. I wondern why it only contains those two though.
I'd never use performance, interactive already provides more power than I need.
I'm just looking for a way to actually use the conservative mode.
Something weird just happened to me: my phone was at about 38% and when I looked 1h later, it was pretty warm and powered off 1min later because battery had dropped to 1%.
I think this is pretty weird. I did not use the phone in that hour and had no apps running.
As illogical it might sound, my first suspect was the energy saving mode I enabled. I dont know why but I think it messed something up.
Does anyone have similar issues or any useful advice for me?
w00ZzL said:
Ok, thanks for the info. I wondern why it only contains those two though.
I'd never use performance, interactive already provides more power than I need.
I'm just looking for a way to actually use the conservative mode.
Click to expand...
Click to collapse
The developer believes those to be the best choices, I have no clue why. Interactive mostly stays near the max freq & makes my phone uber hot like an oven. Feel free to use Render kernel, its in the general section.
w00ZzL said:
Something weird just happened to me: my phone was at about 38% and when I looked 1h later, it was pretty warm and powered off 1min later because battery had dropped to 1%.
I think this is pretty weird. I did not use the phone in that hour and had no apps running.
As illogical it might sound, my first suspect was the energy saving mode I enabled. I dont know why but I think it messed something up.
Does anyone have similar issues or any useful advice for me?
Click to expand...
Click to collapse
I've never experienced that, are you on the latest nightly?
Transmitted via Geass
Ace42 said:
The developer believes those to be the best choices, I have no clue why. Interactive mostly stays near the max freq & makes my phone uber hot like an oven. Feel free to use Render kernel, its in the general section.
Click to expand...
Click to collapse
Do you have Render kernel? If yes which governor mode are you using?
Ace42 said:
I've never experienced that, are you on the latest nightly?
Click to expand...
Click to collapse
I'm on the version from 6-28 and patch V6 which should be the latest?!
w00ZzL said:
Do you have Render kernel? If yes which governor mode are you using?
I'm on the version from 6-28 and patch V6 which should be the latest?!
Click to expand...
Click to collapse
That is the latest, I'm currently using Render kernelon 6-28 with ondemand.
http://forum.xda-developers.com/lg-g2/development/kernel-render-kernel-aosp-lg-roms-t2803929/page5
Ace42 said:
That is the latest, I'm currently using Render kernelon 6-28 with ondemand.
http://forum.xda-developers.com/lg-g2/development/kernel-render-kernel-aosp-lg-roms-t2803929/page5
Click to expand...
Click to collapse
Thanks for bringing that kernel up, I've flashed it yesterday and am delighted. Will try the release from today now.
BTW my low battery life is gone now so thanks again.
Following the advice of @shadowstep and @Davey126 I have created a thread to separate this from the whole Interactive guide thread by @soniCron. This is due to (1) one, the thread is no longer updated for almost a year and (2) two; to explain the tweak, establish baselines/guides for testers and to gather more interested individuals to try it out.
-------------------------------------------
Introducing GlassCannon!
Description: A sound modification to the famous interactive parameters. Provides the smoothest interface, great performance while bestowing the lowest frequencies available. Ramping up quickly to maximize "inputs" from I/O overheads then immediately ramping down once tasks are done. The perfect balance between lowering down your frequency, and finishing up tasks quickly.
Why Glasscannon? Why not?
Who am I? I came from the hammerhead scene. I have been modding interactive parameters for more than 2 years and owns a community called: Android Battery Community whom has its own fair share of followers but has been quite stagnant for almost a year ever since my hammerhead broke. Though, I have been tweaking things with other devices; LG G4, Samsung Note 6, OPO and other smartphones that I got my hands on. Now I think it's time for me to get into the Nexus 6P scene. And after just literally 2 days after I posted the tweak, there are messages and posts tempting me to port it to N5X and LG G4, and with enough encouragement as well as a promise of testers; I edited the values matching my GlassCannon tweak with the former LG G4 .txt file I have in my desktop and here it is!
The difference? After years of being a paranoid about my battery (literally looking at dashboards, cpu cycles etc. you know, that guy who just tends to not be satisfied about everything) I finally settled down and read a lot of things and made it as my basis. Most tweaks in here uses target load as an optimal way to force cores to stick into lower frequencies but we won't be doing that with GlassCannon. We will be using two underrated tunables: above_hispeed_delay and input_boost. This two underrated tunables are being neglected for years, though some used it quite efficiently; I have yet to see a tweak that maximizes the two tunable's potential. We would be using above_hispeed_delay as a substitute to the unpredictable target_load. Instead of assigning too much within a tunable that we can't even lay our finger on how it works properly, why not let the SoC handle it and assign a delay along with timer_rate so it can run efficiently? And let input_boost jump up here and there to provide quick surge whenever there are tasks running under the hood.
Look under (Version: N5X l G4):
Code:
[COLOR="Gray"]/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor interactive
/sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq 384000
/sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq 1440000
/sys/devices/system/cpu/cpu0/cpufreq/interactive/go_hispeed_load 93
/sys/devices/system/cpu/cpu0/cpufreq/interactive/above_hispeed_delay 0 600000:19000 787200:20000 960000:24000 1248000:38000
/sys/devices/system/cpu/cpu0/cpufreq/interactive/timer_rate 50000
/sys/devices/system/cpu/cpu0/cpufreq/interactive/hispeed_freq 600000
/sys/devices/system/cpu/cpu0/cpufreq/interactive/timer_slack 380000
/sys/devices/system/cpu/cpu0/cpufreq/interactive/target_loads 29 384000:88 600000:90 787200:92 960000:93 1248000:98
/sys/devices/system/cpu/cpu0/cpufreq/interactive/min_sample_time 60000
/sys/devices/system/cpu/cpu0/cpufreq/interactive/boost 0
/sys/devices/system/cpu/cpu0/cpufreq/interactive/align_windows 1
/sys/devices/system/cpu/cpu0/cpufreq/interactive/use_migration_notif 1
/sys/devices/system/cpu/cpu0/cpufreq/interactive/use_sched_load 0
/sys/devices/system/cpu/cpu0/cpufreq/interactive/max_freq_hysteresis 0
/sys/devices/system/cpu/cpu0/cpufreq/interactive/boostpulse_duration 0
/sys/devices/system/cpu/cpu4/cpufreq/scaling_governor interactive
/sys/devices/system/cpu/cpu4/cpufreq/scaling_min_freq 384000
/sys/devices/system/cpu/cpu4/cpufreq/scaling_max_freq 1824000
/sys/devices/system/cpu/cpu4/cpufreq/interactive/go_hispeed_load 150
/sys/devices/system/cpu/cpu4/cpufreq/interactive/above_hispeed_delay 20000 960000:60000 1248000:30000
/sys/devices/system/cpu/cpu4/cpufreq/interactive/timer_rate 60000
/sys/devices/system/cpu/cpu4/cpufreq/interactive/hispeed_freq 960000
/sys/devices/system/cpu/cpu4/cpufreq/interactive/timer_slack 380000
/sys/devices/system/cpu/cpu4/cpufreq/interactive/target_loads 98
/sys/devices/system/cpu/cpu4/cpufreq/interactive/min_sample_time 60000
/sys/devices/system/cpu/cpu4/cpufreq/interactive/boost 0
/sys/devices/system/cpu/cpu4/cpufreq/interactive/align_windows 1
/sys/devices/system/cpu/cpu4/cpufreq/interactive/use_migration_notif 1
/sys/devices/system/cpu/cpu4/cpufreq/interactive/use_sched_load 0
/sys/devices/system/cpu/cpu4/cpufreq/interactive/max_freq_hysteresis 0
/sys/devices/system/cpu/cpu4/cpufreq/interactive/boostpulse_duration 0
/sys/module/cpu_boost/parameters/input_boost_freq 0:600000 1:600000 2:600000 3:600000 4:960000 5:960000
/sys/module/cpu_boost/parameters/sync_threshold 1248000
/sys/module/cpu_boost/parameters/boost_ms 40
/sys/module/cpu_boost/parameters/migration_load_threshold 15
/sys/module/cpu_boost/parameters/load_based_syncs Y
/sys/module/cpu_boost/parameters/shed_boost_on_input N
/sys/module/cpu_boost/parameters/input_boost_ms 300
/sys/module/cpu_boost/parameters/input_boost_enabled 1
/sys/module/msm_performance/parameters/touchboost 0
/sys/module/msm_thermal/core_control/enabled 0
/sys/module/msm_thermal/parameters/enabled Y[/COLOR]
Explanation:
go_hispeed_load: Anything between 85-94 is average by my tests, it tends to stick to hispeed_load 40-60 which is what I think would be "optimal" for our cat.
above_hispeed_delay: The most important part of our tweak along with input_boost. This should set up the perfect "delay" so cpu cores could adjust and decide before ramping up to higher frequencies. Our frequency set as hispeed from little cluster is 600000.
target_loads: We won't be dwelling much in this. We put a small amount so that our frequency could ramp up if needed and take a pause on frequencies we deemed to be really significant; 600mhz, 672mhz, 960mhz and 12348mhz. Along with hispeed_delay, it should provide low consumption and tend to stick on 384mhz and 600mhz 89% of the time.
timer_slack: Put it to 380000, just trust me.
max_freq_hysteresis: Asking why theres jitters and a bit of stuttering on your screen? this is the culprit. Turn it to 0 and you will be fine as long as you correct the other tunables. This is because hysteresis actually uses "former" frequencies to calculate which frequency would be best to ramp up to next. If you tend to stick to lower frequencies more, with this on then you will be sticking at low frequencies almost forever, which obviously isn't something good.
input_boost: Now, to explain why this is important; there are two things I would like to emphasize. (1) One, this value was made so that under the hood tasks as well as simple bumps to your frequency if there is a notification, sync etc. would be possible. (2) this value is extremely useful to provide that quick boost to complete tasks with ease. The big clusters would have an input_boost of 633000 which was supposed to be the stock mhz. Why? I deemed that the big cluster isn't that necessary unless you are running an extremely graphic heavy game. So tone it down to 384mhz underclocking it and provide input boost to the mhz that was supposed to be the stock frequency, increasing battery life without sacrificing anything.
Will this be updated? NO, NEVER. I am asking myself why does this other "tweaks" update from time to time if there are no kernel parameters that are changing? Aren't all those tweaks supposed to be tested hundreds of times before releasing? Then why change the parameters again? Am I asking too much? Am I fat? NO. Just NO. EXCEPT if there will be changes within the interactive parameters then sure.
Maximal, Minimal and Nominal Frequencies? Not to hurt anyone's heart but I personally think that this is a retardified(if that is in fact a word) version of "I shall say something smart". I don't see the point of determining what the nominal, maximal and minimal frequencies are as we can't even put a finger what they literally mean. This is actually "subjective" and I really find this so called nominal frequencies irrelevant. Please don't hate.
Do note that most of the explanations above were not edited and is still pretty much the same as N6P description because I'm lazy af.
Credits: @soniCron, @xSilas43, @shadowstep for pointing out the missing timer_slack and input_boost, google for many things to read, this website for music while typing, that website for providing me more things to read, a bucket of fries and two chickens to make me company.
Go ahead and feel free to download the .txt file below and happy tweaking!
======================A SINCERE THANKS TO EVERYONE WHO TESTED AND PARTICIPATED
Tips Hat
After many things happening and after more than a month, I am really glad that we have succesfully ported GlassCannon to N5X and LG G4 (or any other device with same cpu structure). I am proud to present to you the fruits of the small labors of both me and the team as well as others behind the scene testing GlassCannon for your devices.
What has been the verdict? Overall, via landslide; the GlassCannon's Beta2.0 has come out on top. I figured and of course realized after the alphatest that people in N5X community complained (though in small dose) about stutters and a small lag in some apps. Most were debunked especially if you are using an online app but I had to wrap around everything and investigate further. Even though I promised that the Beta would be more battery efficient, I adjusted it so that it should provide a much better performance and smoothness without compromising the battery life of our cats. And thus, the verdict is Beta 2.0 *autistic screeching*
Special thanks to: @Curlyfry2121 @shadowstep @crian @spartan268 @IcyGlacial @Lazerlord
Just a headsup, I will be leaving the Alpha below for other people to download if they still want some!
Grab a copy below! Enjoy!
Official Testers:
@shadowstep N6P
@Lazerlord
@Curlyfry2121
@mesmurized
@@spartan268
@deani77 N6P
@igoorsoouza
@CazeW
Credits to:
Everyone whom are secretly using the GlassCannon profile. To my Kentucky Fried Chicken for keeping me company. @Davey126 for encouragement and other stuffs, @The Flash because you are the fastest man alive, @shadowstep for being a MTG fan, @flar2 for being an awesome developer and supporting this kind of tweaks, @soniCron for starting this and many more. Huge thanks! and I hope this will go on for the better!
Thanks
Sent from my Pixel using Tapatalk
Firefox, Google plus, and Snapchat, with a little Hangouts during school where the data connection is little to none half the day.
Curlyfry2121 said:
Firefox, Google plus, and Snapchat, with a little Hangouts during school where the data connection is little to none half the day.
Click to expand...
Click to collapse
Thanks for the second full cycle *takes notes*
Well thats a huge steep slope we have. I can see where thise battery drains come from and you really have to do something about those. Have you ever considered airplane mode while you are at an area where there's little to no signals? Those red marks should have drained your battery like crazy! At first I was skeptical when you said you have a bad signal lol.
Now about the cpu times etc.:
(1) One, that is pretty much a great cpu stats for both little and big. I can see that it sticks to 600mhz the way I tuned it too. Those little percentage above 600mhz are pretty nice, that means thst they never got stuck in that number for more than 30-40ms (above_hispeed_delay and timer_rate) doing the trick. (2) Two, I can even see the big cluster's cpu5 off, what more can I ask for? (Purely intentional to set a really high ts
Arget_load and go_hispeed_load at lowest frequency to make this happen) this means everythung works fine as intended (the way I tweaked it to be.)
Anyway this is to be expected for light usage. Btw could you please install accubattery for me and turn on "cpu core" on the overlay settings within the app? There should be 6 overlay batteries at your bottom right continuously changing to see cpu frequencies. Now with thst try a full cycle again and note this:
*At what apps does most of the cores get to red (above 3/4)
*In your observation, how long does it stick to yellow (1/2 above) and red frequencies? (Above 3/4)? Please tell in miliseconds e.g. red at 20ms most of the time etc.
You will see what I'm talking about when you install the app
Anyways continue again with a normal full cycle bro and I'll be waiting for the next update!
phantom146 said:
Thanks for the second full cycle *takes notes*
Well thats a huge steep slope we have. I can see where thise battery drains come from and you really have to do something about those. Have you ever considered airplane mode while you are at an area where there's little to no signals? Those red marks should have drained your battery like crazy! At first I was skeptical when you said you have a bad signal lol.
Now about the cpu times etc.:
(1) One, that is pretty much a great cpu stats for both little and big. I can see that it sticks to 600mhz the way I tuned it too. Those little percentage above 600mhz are pretty nice, that means thst they never got stuck in that number for more than 30-40ms (above_hispeed_delay and timer_rate) doing the trick. (2) Two, I can even see the big cluster's cpu5 off, what more can I ask for? (Purely intentional to set a really high ts
Arget_load and go_hispeed_load at lowest frequency to make this happen) this means everythung works fine as intended (the way I tweaked it to be.)
Anyway this is to be expected for light usage. Btw could you please install accubattery for me and turn on "cpu core" on the overlay settings within the app? There should be 6 overlay batteries at your bottom right continuously changing to see cpu frequencies. Now with thst try a full cycle again and note this:
*At what apps does most of the cores get to red (above 3/4)
*In your observation, how long does it stick to yellow (1/2 above) and red frequencies? (Above 3/4)? Please tell in miliseconds e.g. red at 20ms most of the time etc.
You will see what I'm talking about when you install the app
Anyways continue again with a normal full cycle bro and I'll be waiting for the next update!
Click to expand...
Click to collapse
Gotcha! And yeah I did use airplane mode one time but I forgot to today lol. I will install it and I'll follow your instructions for tomorrow's cycle, and yeah now you get what I meant ?
phantom146 said:
Thanks for the second full cycle *takes notes*
Well thats a huge steep slope we have. I can see where thise battery drains come from and you really have to do something about those. Have you ever considered airplane mode while you are at an area where there's little to no signals? Those red marks should have drained your battery like crazy! At first I was skeptical when you said you have a bad signal lol.
Now about the cpu times etc.:
(1) One, that is pretty much a great cpu stats for both little and big. I can see that it sticks to 600mhz the way I tuned it too. Those little percentage above 600mhz are pretty nice, that means thst they never got stuck in that number for more than 30-40ms (above_hispeed_delay and timer_rate) doing the trick. (2) Two, I can even see the big cluster's cpu5 off, what more can I ask for? (Purely intentional to set a really high ts
Arget_load and go_hispeed_load at lowest frequency to make this happen) this means everythung works fine as intended (the way I tweaked it to be.)
Anyway this is to be expected for light usage. Btw could you please install accubattery for me and turn on "cpu core" on the overlay settings within the app? There should be 6 overlay batteries at your bottom right continuously changing to see cpu frequencies. Now with thst try a full cycle again and note this:
*At what apps does most of the cores get to red (above 3/4)
*In your observation, how long does it stick to yellow (1/2 above) and red frequencies? (Above 3/4)? Please tell in miliseconds e.g. red at 20ms most of the time etc.
You will see what I'm talking about when you install the app
Anyways continue again with a normal full cycle bro and I'll be waiting for the next update!
Click to expand...
Click to collapse
Accubattery requires me to pay to use the overlays, and I have no money what so ever on my play account :crying: is there anything else I could use instead?
Curlyfry2121 said:
Accubattery requires me to pay to use the overlays, and I have no money what so ever on my play account :crying: is there anything else I could use instead?
Click to expand...
Click to collapse
I forgot about all of that ? well I dont really recommend the other alternatives since most are outdated. Though cpu float should do the trick
I'm going to start all over with my testing again in case my previous one was skewered for some strange reason. Downloaded gboard instead of bb keyboard and rebooted my phone. I'll reset the CPU stats as well after my app updates are done since that'll likely tax things a little too much. I'll have proper results probably 24 hours from now.
My main culprit right now is that the g4 skin is more heavyweight than thought to be so a nudge more CPU power is needed for general tasks to account for it if that makes any sense. Otherwise i run my phone fairly lean as it's debloated and i make use of greenify, amplify, and powernap so there's minimal services and apps in the background at most times.
spartan268 said:
I'm going to start all over with my testing again in case my previous one was skewered for some strange reason. Downloaded gboard instead of bb keyboard and rebooted my phone. I'll reset the CPU stats as well after my app updates are done since that'll likely tax things a little too much. I'll have proper results probably 24 hours from now.
My main culprit right now is that the g4 skin is more heavyweight than thought to be so a nudge more CPU power is needed for general tasks to account for it if that makes any sense. Otherwise i run my phone fairly lean as it's debloated and i make use of greenify, amplify, and powernap so there's minimal services and apps in the background at most times.
Click to expand...
Click to collapse
Not to jump on some hate-wagon here but I am really against powernap, thought that's just me.
I have gboard as well and has chosen a substratum dark overlay for it (if you don't have a substratum support just get the material black theme) also, gboard consumes battery a little tinsy bit than your normal keyboard, don't forget to disable the snippets and share to google statistics in the preference tab to avoid continuous syncing. I'll be waiting for the results and if everything still seems to lag, I might cook something different especially for you
Reasons for new thread:
3) This is a different method to the once described before.
Android O, Naptime FORCE DOZE, My previous stats are as follows;
(This was using my custom governor tweaks and the skipped/unknown lost battery in the middle is a bootloop)
Forgot this xD
After 2-3 days on the latest version, I continue to see overnight battery usage (sleep+doze+battery) trending a bit lower then with my prior profile HawkFlyv1.2. This along with a noticeable increase in responsiveness, puts GlassCannon at the top of my current list.
Good work @phantom146. I'll try to work on providing details for this alpha test in the next day or so
LazerL0rd said:
Reasons for new thread:
3) This is a different method to the once described before.
Click to expand...
Click to collapse
LazerL0rd said:
Android O, Naptime FORCE DOZE, My previous stats are as follows;
(This was using my custom governor tweaks and the skipped/unknown lost battery in the middle is a bootloop)
Click to expand...
Click to collapse
That instagram... :laugh:
LazerL0rd said:
Forgot this xD
Click to expand...
Click to collapse
Any stutters or jitters? phone freezes etc.?
About the battery life, any comments on that? and the one we talked about the pm (regarding the ramping down and input_boost) any thoughts after this cycle?
Will wait for your answer
Mesmurized said:
After 2-3 days on the latest version, I continue to see overnight battery usage (sleep+doze+battery) trending a bit lower then with my prior profile HawkFlyv1.2. This along with a noticeable increase in responsiveness, puts GlassCannon at the top of my current list.
Good work @phantom146. I'll try to work on providing details for this alpha test in the next day or so
Click to expand...
Click to collapse
I am glad you are liking this profile and will wait for more results, also; it would help me if you can try GlassCannon with heavy usage using heavy graphical games such as need for speed asphalt, nba etc. I really want to see if we have any kind of lags or stutters using high performance games. Thanks for the continuous support!
phantom146 said:
That instagram... :laugh:
Any stutters or jitters? phone freezes etc.?
About the battery life, any comments on that? and the one we talked about the pm (regarding the ramping down and input_boost) any thoughts after this cycle?
Will wait for your answer
I am glad you are liking this profile and will wait for more results, also; it would help me if you can try GlassCannon with heavy usage using heavy graphical games such as need for speed asphalt, nba etc. I really want to see if we have any kind of lags or stutters using high performance games. Thanks for the continuous support!
Click to expand...
Click to collapse
Was getting stuttering with Geo Dash ?? but I think it's just the settings I've enabled and or it's bad optimisations. Battery is equal to or slightly less than my previous governor. After a while I don't see much stuttering (btw i had stuttering before on Geo Dash too xD). There's one thing I enabled that you didn't though. And it's a value called fast_ramp_down. I'm gonna try without it now and see if there's still stuttering.
Just checked and stuttering is MASSIVELY reduced by disabling that option. Well, I thought it would be good. Anyway here are today's stats;
PS - I've decided to keep it, with a few of my own changes xD and sorry about the recharging, I was watching some TV shows ?
LazerL0rd said:
Was getting stuttering with Geo Dash but I think it's just the settings I've enabled and or it's bad optimisations. Battery is equal to or slightly less than my previous governor. After a while I don't see much stuttering (btw i had stuttering before on Geo Dash too xD). There's one thing I enabled that you didn't though. And it's a value called fast_ramp_down. I'm gonna try without it now and see if there's still stuttering.
Click to expand...
Click to collapse
LazerL0rd said:
Just checked and stuttering is MASSIVELY reduced by disabling that option. Well, I thought it would be good. Anyway here are today's stats;
PS - I've decided to keep it, with a few of my own changes xD and sorry about the recharging, I was watching some TV shows
Click to expand...
Click to collapse
well thats good to hear I was really wondering why you were getting stutters as 4 of my testers never got them. Its cool it is your phone to begin with so just tweak away but thanks for keeping the tweak it means a lot to me that somebody will use them as a daily profile :good:
Btw, those stats are crap hahahaha I can't even see where to begin with those charging cycles haha :silly:
phantom146 said:
well thats good to hear I was really wondering why you were getting stutters as 4 of my testers never got them. Its cool it is your phone to begin with so just tweak away but thanks for keeping the tweak it means a lot to me that somebody will use them as a daily profile :good:
Btw, those stats are crap hahahaha I can't even see where to begin with those charging cycles haha :silly:
Click to expand...
Click to collapse
Tomorrow I'm trying a mix of yours and a few I've created before. I done a little research and it seems my idea should work. But it's more experimental than Android O :laugh:
---------- Post added at 07:41 PM ---------- Previous post was at 07:40 PM ----------
The day after I'll test yours again, and I'll stay with whatever is better ?
Good morning Le Pro3 forum -
I've been switching between Paranoid Android and Lineage OS for the past few weeks. One thing I've noticed very specifically is that PA 7.3.0 (and 7.2.0...) does not seem to effectively go into 'doze' as well as Lineage OS. What I mean is that battery drain overnight is pretty high on PA (10-20%) while Lineage is very low (less than 5%) using the same applications.
So that I can better understand and troubleshoot, would anyone be willing to point me to other forum posts either here or on other forums that dive into troubleshooting this kind of thing? I'm not finding anything that useful that is *recent*.
Thanks in advance!
mdkathon said:
Good morning Le Pro3 forum -
I've been switching between Paranoid Android and Lineage OS for the past few weeks. One thing I've noticed very specifically is that PA 7.3.0 (and 7.2.0...) does not seem to effectively go into 'doze' as well as Lineage OS. What I mean is that battery drain overnight is pretty high on PA (10-20%) while Lineage is very low (less than 5%) using the same applications.
So that I can better understand and troubleshoot, would anyone be willing to point me to other forum posts either here or on other forums that dive into troubleshooting this kind of thing? I'm not finding anything that useful that is *recent*.
Thanks in advance!
Click to expand...
Click to collapse
Slightly tangential, but how is the battery life overall compared to LineageOS? I've been running LineageOS and while it performs well, the battery life isn't nearly as good as stock EUI. As for over night battery drain, I get about 10% in 8-9 hours in stand by.
Hey. I tried every single rom out there from MoKee to Lineage and i can asure u one thing. Best for battery is Omni ROM (latest) + BlackScreen kernel version 3.6
Now i am a medium - heavy user. Lots of time on snapchat and playin games for about half an hour to a hour. I got 6 hours of screen on time wich is great really compared to other roms.
Now i also tried various combinations of using different governors and so on so i tested an option for about three days and then tried another one. I was seeking for the best possible battery life while don't have any stuttering or lags. Cuz if u put the cpu governor for example on impulse and then underclock it too... It lags like hell.
Soo what do u need to do to get also good standby performance?? I loose less than 3% over night.
Try following:
IN KERNEL AUDITOR
CPU: Both big and little cores governor set on alucard
Underclock the big cores to 1900 and little cores to 1440
GPU: governor set to msm-adreno-tz; underclock max frequency to 560 MHz; enable Adreno idler
I/O sheduler: set governor to noop; read-ahead 128 kB
Low Memory Killer: set profile to Agressive
Misc: Vibration strenght set on about 15%, disable Fsync, TCP must be westwood.
_________________________________________
Developer options:
Set logger buffer size to off
Set all three animation scales on 0.5x
_________________________________________
That is it. If u have any questions or something doesn't work as written -> reply back
P.S. turn off wi-fi during night. Helps a bunch.
insomnixXx said:
Hey. I tried every single rom out there from MoKee to Lineage and i can asure u one thing. Best for battery is Omni ROM (latest) + BlackScreen kernel version 3.6
Now i am a medium - heavy user. Lots of time on snapchat and playin games for about half an hour to a hour. I got 6 hours of screen on time wich is great really compared to other roms.
Now i also tried various combinations of using different governors and so on so i tested an option for about three days and then tried another one. I was seeking for the best possible battery life while don't have any stuttering or lags. Cuz if u put the cpu governor for example on impulse and then underclock it too... It lags like hell.
Soo what do u need to do to get also good standby performance?? I loose less than 3% over night.
Try following:
IN KERNEL AUDITOR
CPU: Both big and little cores governor set on alucard
Underclock the big cores to 1900 and little cores to 1440
GPU: governor set to msm-adreno-tz; underclock max frequency to 560 MHz; enable Adreno idler
I/O sheduler: set governor to noop; read-ahead 128 kB
Low Memory Killer: set profile to Agressive
Misc: Vibration strenght set on about 15%, disable Fsync, TCP must be westwood.
_________________________________________
Developer options:
Set logger buffer size to off
Set all three animation scales on 0.5x
_________________________________________
That is it. If u have any questions or something doesn't work as written -> reply back
P.S. turn off wi-fi during night. Helps a bunch.
Click to expand...
Click to collapse
Hi mate, thank you for your detailed review. It helped me especially for kernel auditor settings. But i have an objection in my honest oppinion. I have tried both roms, AICP 12.1 and OmniROM and AICP is much better at battery department. But both rom is great in the rest.
SUMMERBREEZE said:
Hi mate, thank you for your detailed review. It helped me especially for kernel auditor settings. But i have an objection in my honest oppinion. I have tried both roms, AICP 12.1 and OmniROM and AICP is much better at battery department. But both rom is great in the rest.
Click to expand...
Click to collapse
Really, well u know it is mostly how u use it. Cuz for my type of usage Omni works perfect. I was using AICP for a month before i tried other roms as described and it was really good but idk... When i was using it i got 4h and 30min at best with my kind of usage but with omni i got easily 6h. It depends from user to user but well... I will give AICP another shot anyway. I am glad that my review helped u.
insomnixXx said:
Really, well u know it is mostly how u use it. Cuz for my type of usage Omni works perfect. I was using AICP for a month before i tried other roms as described and it was really good but idk... When i was using it i got 4h and 30min at best with my kind of usage but with omni i got easily 6h. It depends from user to user but well... I will give AICP another shot anyway. I am glad that my review helped u.
Click to expand...
Click to collapse
yes it helped a lot. especially kernel tweaking thing. both rom is great imho and 1-1.5 hr screen on time difference may marry -as you said- on which programs/apps we use. we must be glad that our le pro 3 has a great nougat support from these beautiful devs and i dont know how to thank to them. i just used 2-3 days this AICP rom and i will give also a shot to OmniROM to compare them again. I love both roms and i am sure i will keep switching between the two
Hey there,
Just to let you know that with bootloader unlocked + root (magisk) there is a nice set of tunable option with kernel adiutor. With a bit of entropy and governor tuning, you get even better performance from your a2. That's pretty cool ??
Can you share your settings? Hours of screen, performance... And kernel adiutor setup
RawKnighy said:
Can you share your settings? Hours of screen, performance... And kernel adiutor setup
Click to expand...
Click to collapse
disabled cpu boost to avoid cpu stuck at 2,2
Change governor to on demand
Entropy increase to get 40% free
Mostly it.
SOT depend on my use and vary between 4-5 hours to 6+ hours. This is mainly due to two factors:
-network coverage (this can suck your battery like nothing)
- use if BT + viperfx for hours ?
The best way to test it.. is to test it you may have the same increase as me (4 hours sit at most before mod)
Grüß,
H.
*stucked cpu ppl, please don't hate me, I had to wait to get something to be shown ??
Look at that sleep deepness ?
dodjob said:
*stucked cpu ppl, please don't hate me, I had to wait to get something to be shown ??
Look at that sleep deepness ?
Click to expand...
Click to collapse
You rooted your mi a2 right.so are you facing any issue after root?
Lol No, except that I installed a gcamera which wrecked my original camera.. after countless system flash it went but all ok. The ondemand thing is a real treat mate, and fast as hell ?
dodjob said:
*stucked cpu ppl, please don't hate me, I had to wait to get something to be shown ??
Look at that sleep deepness ?
Click to expand...
Click to collapse
Does the clock alarm work with such a deep sleep?
ILA said:
Does the clock alarm work with such a deep sleep?
Click to expand...
Click to collapse
Of course it does ?
Can I get the process of installation
I don't see any improvements with on demand actually. Interactive gives me 0.3%/h drain in deep sleep and I doubt it can get much better
Benjamin_L said:
I don't see any improvements with on demand actually. Interactive gives me 0.3%/h drain in deep sleep and I doubt it can get much better
Click to expand...
Click to collapse
On demand does not improvement battery life as soon as you keep cpu boost in cpu 1. Did you disabled it? For me it was night and day
Chaitanya Kumar said:
Can I get the process of installation
Click to expand...
Click to collapse
Unlock bootloader + root the phone and then just install an app called "Kernel auditor" from the Play store and tune the kernel from within the app.
dodjob said:
Hey there,
Just to let you know that with bootloader unlocked + root (magisk) there is a nice set of tunable option with kernel adiutor. With a bit of entropy and governor tuning, you get even better performance from your a2. That's pretty cool ??
Click to expand...
Click to collapse
Care share settings with screenshot?
dodjob said:
disabled cpu boost to avoid cpu stuck at 2,2
Change governor to on demand
Entropy increase to get 40% free
Mostly it.
SOT depend on my use and vary between 4-5 hours to 6+ hours. This is mainly due to two factors:
-network coverage (this can suck your battery like nothing)
- use if BT + viperfx for hours ?
The best way to test it.. is to test it you may have the same increase as me (4 hours sit at most before mod)
Grüß,
H.
Click to expand...
Click to collapse
Danke! Can you share your entropy settings, please!
e34v8 said:
Danke! Can you share your entropy settings, please!
Click to expand...
Click to collapse
?
wow, this deep sleep was awesome
Just install hex kernel... It's awesome
syncstar32 said:
Just install hex kernel... It's awesome
Click to expand...
Click to collapse
whats the advantages ?
Hey guys. Just wanted to share some tweaks with which you guys can get amazing battery life on your redmi note 3. I used this setup for 3 full charge cycles, and averaged about 10Hrs SOT with a 1 year old battery, when I was getting about 5-6 hrs of SOT only on normal pie/10 roms. Do give this setup a try if you want the absolute best battery, and don't mind about performance degradation a bit... The device now behaves like it's got a 5000Mah battery now. If you guys manage a better setup than this, then do share your setup down below. And please hit the thanks button if this helped you. Cheers:fingers-crossed::fingers-crossed:.
SETUP:
Havoc OS latest 8.1
Latest agni kernel 13.0
Spectrum profile to battery
In any kernel manager (I use fk kernel manager):
.Big cores disabled
.Little cores max clock at 1382Mhz. Min clock to 400Mhz.
.Undervolt to lowest stable value (It's different for everyone, so try out a stable value for you).
.GPU max freq to 432MHz. Min to 133Mhz.
.Adreno Idler to on.
**Guys who have the 2gb kenzo, try the mcaosp 8.0 ROM instead of havoc. It has better
RAM management for you guys.**
shivaay1234 said:
Hey guys. Just wanted to share some tweaks with which you guys can get amazing battery life on your redmi note 3. I used this setup for 3 full charge cycles, and averaged about 10Hrs SOT with a 1 year old battery, when I was getting about 5-6 hrs of SOT only on normal pie/10 roms. Do give this setup a try if you want the absolute best battery, and don't mind about performance degradation a bit... The device now behaves like it's got a 5000Mah battery now. If you guys manage a better setup than this, then do share your setup down below. And please hit the thanks button if this helped you. Cheers:fingers-crossed::fingers-crossed:.
SETUP:
Havoc OS latest 8.1
Latest agni kernel 13.0
Spectrum profile to battery
In any kernel manager (I use fk kernel manager):
.Big cores disabled
.Little cores max clock at 1382Mhz. Min clock to 400Mhz.
.Undervolt to lowest stable value (It's different for everyone, so try out a stable value for you).
.GPU max freq to 432MHz. Min to 133Mhz.
.Adreno Idler to on.
Click to expand...
Click to collapse
Lol why u disabling cores. If u fantasize about 5000mah battery then buy redmi 8. Most of the kenzo users gone to android 10.
Black_Stark said:
Lol why u disabling cores. If u fantasize about 5000mah battery then buy redmi 8. Most of the kenzo users gone to android 10.
Click to expand...
Click to collapse
You don't like it don't use it. Don't criticize him for giving his opinion. You're talking as if he's forced you to disable your phones cores.
By the way, I'll try out the setup. Thanks for posting it for all kenzo users.
delete
Delete
@shivaay1234, Good tutorial - thank you! Could you please share your undervolting preferences?
wxtester said:
@shivaay1234, Good tutorial - thank you! Could you please share your undervolting preferences?
Click to expand...
Click to collapse
Glad you liked it mate. I'll tell you my undervolting values, but it's different for everyone, so, do experiment.
400 and 695 mhz with 710mv
806mhz 745mv
1017mhz 825mv
1190mhz 880mv
1305mhz 915mv
1382mhz 935mv
shivaay1234 said:
Glad you liked it mate. I'll tell you my undervolting values, but it's different for everyone, so, do experiment.
400 and 695 mhz with 710mv
806mhz 745mv
1017mhz 825mv
1190mhz 880mv
1305mhz 915mv
1382mhz 935mv
Click to expand...
Click to collapse
Thanks a lot mate. You're setup is giving me close to 12hrs of sot, which is insane. I still can't believe it. Thanks man.
i will try, i dont care about android 10
rickymanx said:
i will try, i dont care about android 10
Click to expand...
Click to collapse
Lol. Same here mate.
shivaay1234 said:
Hey guys. Just wanted to share some tweaks with which you guys can get amazing battery life on your redmi note 3. I used this setup for 3 full charge cycles, and averaged about 10Hrs SOT with a 1 year old battery, when I was getting about 5-6 hrs of SOT only on normal pie/10 roms. Do give this setup a try if you want the absolute best battery, and don't mind about performance degradation a bit... The device now behaves like it's got a 5000Mah battery now. If you guys manage a better setup than this, then do share your setup down below. And please hit the thanks button if this helped you. Cheers:fingers-crossed::fingers-crossed:.
SETUP:
Havoc OS latest 8.1
Latest agni kernel 13.0
Spectrum profile to battery
In any kernel manager (I use fk kernel manager):
.Big cores disabled
.Little cores max clock at 1382Mhz. Min clock to 400Mhz.
.Undervolt to lowest stable value (It's different for everyone, so try out a stable value for you).
.GPU max freq to 432MHz. Min to 133Mhz.
.Adreno Idler to on.
Click to expand...
Click to collapse
Where I should get that spectrum profile?
Divyeshj66 said:
Where I should get that spectrum profile?
Click to expand...
Click to collapse
It's an app in the play store. Link:
https://play.google.com/store/apps/details?id=org.frap129.spectrum&hl=en
If anyone wondering, you can use Kernel Aduitor to disable cores and undervolting, it's free and it is quite comprehensive.
@shivaay1234 are you the guy who made that rapid os rom?
Ehsnf said:
@shivaay1234 are you the guy who made that rapid os rom?
Click to expand...
Click to collapse
Yup.
shivaay1234 said:
Yup.
Click to expand...
Click to collapse
Oh cool. Bro, but why did you stop development of rapid os? I liked that ROM so much...
Divyeshj66 said:
Oh cool. Bro, but why did you stop development of rapid os? I liked that ROM so much...
Click to expand...
Click to collapse
When Q sources become stable, I will release the Q version of the ROM.
I used to disable the big cores as well on a per-app basis. For light usage like listening to music or reading ebooks, downloading stuff in the background etc it used to yield tangible improvements in battery life. Also while charging to keep battery temp. in check. Now I have Tasker automate it all.
Funnily enough in Redmi Note 3 with only the little cores active device didn't feel much slower but in my Poco F1 whose little cores are much faster (higher IPC, 25% higher clock speeds) it would feel a little laggy with big cores disabled.
Its better to undervolt only the higher frequencies and leave alone/reduce only a little the lower ones as they are much closer to the threshold voltage and more likely to cause a forced reboot if undervolted too much.
@amn1987 Thanks for bumping this thread dude, i really needed to improve my SOT. Hopefully this will help.
PRemBh0i said:
@amn1987 Thanks for bumping this thread dude, i really needed to improve my SOT. Hopefully this will help.
Click to expand...
Click to collapse
Bear in mind the performance degradation can be quite noticeable. Instead of completely disabling the big cores maybe experiment with clocking them lower at 1-1.2GHz where they are most efficient in terms of perf./W IIRC. At these low clocks they will still completely outperform the little cores due to nearly 80% higher performance per clock.
SHADOW Kernel by @energyspear17
Already have QUAD core and battery mode.
You can also UV while flashing
Which would result in better battery backup.
Anyways different peoples different choices.
Good works?