[03/25/13] A Governor a Day - HTC Rezound

Today I'm launching this not-so-little experiment of mine.
Purpeltendire said:
Now that I've familiarized myself with much of the Rezound forum, I was wondering how I could begin contributing in my own way. I don't have much skill in the way of development, and with school I doubt I'd have the time. But then another thought came to mind... guinea pig!
A Governor a Day
So basically what I'm planning on doing is testing the effects of the various CPU governors. I was planning on doing this for myself anyways, as I'm very curious what the practical differences are between governors.
I'll be setting several constants (Min/Max frequencies, voltages, ROM/Kernel, etc) and the only variable (at first) will be the governor. I'll probably start off with a one day time period, just because of the simplicity. Overall routine will be to set the governor when I plug my phone in, either at night before I go to bed or just before I run out of juice -- say 10% mark. And then just use it as I always do, because I have the benefit of a fairly regular class schedule there shouldn't be any significant changes to my usage pattern.
I'll be using several apps to help document all this data; if you want to suggest another to add to the accuracy and range of the stats, feel free. So far the list is:
Statistics
GSam Battery Monitor.
SystemPanel [Paid].
CPU Spy.
BetterBatteryStats (Added 02/29/13)
Settings
Kernel Tuner
Android Tuner Free
I'm guessing that for the first week or so I'll be taking community feedback from you all and incorporating it into the OP as constants and such, so the final experiment probably won't start for another few days and I'll have to redo my first few days of statistics. But that really doesn't bother me.
This is sort of my introductory post, I'm almost done with my day on DanceDance. As I told some people in the Infection2.1 thread I'd be making a very detailed post of my battery life and settings, I'll also be copying that post into a new thread and reserving the first few posts for myself. You're all welcome to make suggestions in both this thread and the dedicated thread, but let's try to keep Neo's ROM thread clean for him.
Click to expand...
Click to collapse
I'm thinking of starting with the more popular governors first. It seems as if Wheatley and DanceDance are both very popular and also very similar, so that might be an ideal starting point for my first two days. Thoughts, and are there any apps you would like to see me use to record data?

Using this post to display the most current governor post.
Purpeltendire said:
Intellidemand redo!
First off, pay less attention to the BetterBatteryStats stuff. I had a weird usage pattern today... used it normally until about 9 PST, then had a mild crisis to deal with over GTalk until like now. So non-stop typing for a solid hour and a half kind of ruined the battery usage percentage... it was way lower earlier.
I'm also glad to be redoing this now that Intellidemand has been getting so much attention in Infection 2.3 and Hiro as a very good governor.
Intellidemand Redo 1
Intellidemand Redo 2
Intellidemand Redo 3
Great governor. I think this can give Wheatley a run for its money. From memory, my usage up until my typing storm was around 2 hours screen on time, and 3.9% battery usage. That's truly fantastic. I ended with a pretty high hourly percentage, BUT! I also ended with nearly 3 hours of screen-on time. For many people that's not only high, it can't be done. So all the more amazing because of it.
Performance wise, this is a very close runner to OnDemand. Little lag unless you go out of your way to look for it, and good scaling too.
Final conclusion is that this is a capable all-rounder, and it supports tweaking so you can go even further to customize it to your needs. I highly recommend you do so.
Click to expand...
Click to collapse

Links to individual governor writeups, as the thread is a bit of a mess.
Wheatley
DanceDance
Intellidemand
Ondemand
Lionheart
SmartassV2
Lulzactive
Scary
Interactive
Lazy
Badass
SavageZen
Skywalker
Conservative
Performance
Comparison: Wheatley v. DanceDance v. SmartassV2

Re: A Governor a Day
Sounds legit man good luck can't wait to see results.
What is your usage like? How much screen on time are you looking to get?
I'd like to see android tuner used to set governors it also has a nice battery stats portion of the app.
Sent from my ADR6425LVW using xda app-developers app

Re: A Governor a Day
Squirrel1620 said:
Sounds legit man good luck can't wait to see results.
What is your usage like? How much screen on time are you looking to get?
I'd like to see android tuner used to set governors it also has a nice battery stats portion of the app.
Sent from my ADR6425LVW using xda app-developers app
Click to expand...
Click to collapse
I usually have between 1.5 to 2 hours of screen-on time. If I do a lot of document editing away from home it might be closer to 2.5, but that's on the long side for me.
When I used Android Tuner within Ecliptic, I wasn't really fond of its interface. But I'll take a look again, at the least I could use the battery stats from it.
Edit: Spent about 15 minutes digging through the settings in the free version of Android Tuner. Unfortunately it looks like a lot of the battery and statistic recording is only on the paid version of the app, or else takes up almost double the resources of GSam and SystemPanel together. Plus I really don't feel like paying $10 for an app I don't like. But take a look at everything when I post in the OP later, hopefully most if not all of the data you're looking for will be in there from either GSam or SystemPanel.
Sent from my Infected Rezound.

I ran the Wheatley governor today, CPU Min at 192 MHz and max at 1350. I undervolted another -50mV from standard Infected, so it's 100mV below stock. The second CPU core was disabled unless needed, and was running the OnDemand governor.
The phone spent as much time at 1350Mhz as it spent with the screen on, at about 2 hours 15 minutes. Total time before I plugged the phone in to upload screenshots was at 16 hours 20 minutes, with about 27% battery left (That Battery Stats screenshot was taken after I had it plugged in for a couple minutes, sorry). According to GSam, the screen took about 49% power, apps took about 45%, and Wifi/Radios took about 5.6% together.
The biggest battery hogging app was Apex Launcher by a long shot, at 17%.
Most of the stats you can see just by looking through the screenshots, I had to upload them to dropbox because they're too large to include in the post and XDA attachments downsizes them far too much.
If you have any other questions or want to see additional stats that I didn't include, feel free to ask in this thread (link in my signature). I'll be doing this sort of thing daily with different governors as a sort of real-world experiment, so if you have any suggestions please let me know. Suggestions on everything from how I should set the kernel or governor up, to applications I should use to track usage, everything is welcome.
Screens! (Click to zoom if you need to)
Wheatley End 1
Wheatley End 2
Wheatley End 3

Yesterday when I was running Wheatley, I took some screenshots when the battery hit 50%, but I didn't end up using them in the final post. Now that I'm on DanceDance today, I did the same thing to see if there were differences. I didn't expect many, because lots of people say that these two governors are very similar.
DanceDance at 50%
Wheatley at 50%
I must say, I was surprised. So far I'm getting about an hour less battery life than yesterday. The differences in screen on time are minimal, about seven minutes, but the screen is using about 2 percent more power on DanceDance. Applications are also using more power, by about 3%. Somehow this is all adding up to about an hour less battery... which might be considerable to some.

Man.. I'm talking to myself here. Oh well. At least I'm good company! :highfive:
Alright, DanceDance. I'll link the screenshots, and then do my writeup.
DanceDance 1
DanceDance 2
DanceDance 3
Right off the bat, according to CPU Spy the phone spent a little more time at various intermediate clocks than in Wheatley. However, this may have been due to the fact that when on Wheatley I had CPU1 set to OnDemand and not Wheatley like CPU0 was. Thanks for the catch, brenuga;
brenuga said:
I think the reason the cpu spent so much time at 1350 is because your cpu1 governor is set to ondemand. I would change it to wheatley or Intellidemand.
Click to expand...
Click to collapse
I'll have to come back to Wheatley on both CPUs sometime soon.
Moving on to GSam. Phone Radio, Screen on, Wifi, and Wake time all look pretty similar if you compare them to the Wheatley screenshots; everything was within about 30 minutes of each other, which makes sense because I took the DanceDance screenshots 30 minutes earlier in the evening than the Wheatley ones. However, in Wheatley, the screen used more power than the apps, while in DanceDance the apps were using slightly more power than the screen. I'm not sure if this was due to my usage or not...
App usage were largely the same as well.
SystemPanel offers a bit more insight. The Battery and Device Usage plots on Wheatley and DanceDance were very close to each other, sharing the same shape and general line. However the lowest graph (which has its label cut off, sorry. It's CPU Usage!) tells a different story. Wheatley appears to have a significantly lower CPU Usage average, at times appearing to be by as much as ~20%.
These two governors are very very similar, as you'll read time and time again in these forums. But sometimes it's not about stats... sometimes it's about how your device feels. That's where DanceDance really wins in this comparison. To me, it feels a fair bit smoother and more fluid -- more Buttery, I suppose. This does come at a cost though; while your screen is on, you can practically watch the battery meter drop. It's a lot more exciting than watching paint dry... fair bit of action here haha. Wheatley didn't seem to have the same kind of battery drop while using applications as DanceDance did. However it seemed to make up for it, with CPU Usage appearing to fall to lower thresholds than in Wheatley at one or two points.
TL;DR: DanceDance felt a little better from a smoothness standpoint, while Wheatley seemed to be a bit better at conserving battery with screen on. I think that they're so close though that DanceDance is the better choice. Who doesn't like smooth?

Thank you for the review! I like.

Pretty cool idea. Keep it up.

Re: A Governor a Day
I'll make a thread in the general (non device specific) to try and get more attention here. I think a lot of people would like to see these results
Here is the thread in general discussion
http://forum.xda-developers.com/showthread.php?t=2171844
Oh and may I request intellidemand? I've always used it. Thank you
Sent from my ADR6425LVW using xda app-developers app

Awesome! Great idea.
Now do Lionheart!

Squirrel1620 said:
I'll make a thread in the general (non device specific) to try and get more attention here. I think a lot of people would like to see these results
Here is the thread in general discussion
http://forum.xda-developers.com/showthread.php?t=2171844
Oh and may I request intellidemand? I've always used it. Thank you
Sent from my ADR6425LVW using xda app-developers app
Click to expand...
Click to collapse
Appreciate that man, thanks.
You're in luck! I'm actually doing Intellidemand today, and then OnDemand tomorrow.
chupajr said:
Awesome! Great idea.
Now do Lionheart!
Click to expand...
Click to collapse
Recommend me an AOSP kernel with Lionheart enabled, and I'd be more than happy to. Snuzzo's Funky doesn't have it.

Halfway through Intellidemand day, and I'll compare it with DanceDance because that's the most recent one.
Intellidemand at 50%
DanceDance at 50%
Through the day I've noticed battery life is considerably worse than either DanceDance or Wheatley. DanceDance took almost 1.5 hours longer to drain to 50%. The screen is taking about 11% more power than in DanceDance, even though it's been on for 15 minutes less. I'm fairly sure my brightness settings have been about the same (Auto in daylight, set level around 20% indoors).
Full write up will come tonight.

Re: A Governor a Day
Purpeltendire said:
Recommend me an AOSP kernel with Lionheart enabled, and I'd be more than happy to. Snuzzo's Funky doesn't have it.
Click to expand...
Click to collapse
Mine has it (Ermahgerd), and at least one (maybe both) of iHateWebOS's has it.
Sent from my SCH-I605 using Tapatalk 2

shrike1978 said:
Mine has it (Ermahgerd), and at least one (maybe both) of iHateWebOS's has it.
Sent from my SCH-I605 using Tapatalk 2
Click to expand...
Click to collapse
Alright, thanks. I'll have to give your kernel and iHate's Deluxe a try when I finish up with the governors in Snuzzo's,
Without further ado, here's Intellidemand as requested. [Now with BetterBatteryStats!]
Intellidemand End 1
Intellidemand End 2
Intellidemand End 3
The governor seemed to be decent at scaling, keeping the CPU spread out more than either DanceDance or Wheatley. Still between one and five minutes spent at the more popular clocks, with the majority of the time logged at 1350, 192, and Deep Sleep as usual. Nothing major to report here, pretty standard for what I've seen so far.
GSam has something different to say. For one, battery life seemed strained. There was considerably more usage at idle or deep sleep than either DanceDance or Wheatley. I watched a movie at one point this evening, and when I checked my phone again it had dropped 6% over 2.5 hours, without touching it once. It also ramped up quickly while browsing, I think in the time it took me to make a 2 sentence post the battery dropped 6 or 7%. That didn't happen with the others. The battery discharge graph is pretty steep, which makes sense given the increased discharge.
Also, app usage seemed abnormally high. I checked on that and found that the Android OS/Kernel was using a LOT of power: 65%. Yesterday on DanceDance it was using 6%... there's something weird here.
If someone running Intellidemand (Squirrel1620?) could run GSam for a day and see if this happens to you, that would be great. I'd like to see if this is a random event or something consistent with the governor.
SystemPanel reports similar findings under "System Processes" with 43% of my total CPU usage. But quite frankly, other than that there's nothing really weird going on. No wakelocks, no apps freezing... it just seems like the OS wasn't idle for some reason.
My verdict: Not a governor for idle users, partly because it's based off the GPU. If you do gaming or use apps that require a bit more graphical power, then by all means go for it. But for users like me, who text, browse the web a bit, and do some forum work, this governor has very little to offer except minimal noticed lag.
Tomorrow [Saturday] I'll do OnDemand, and Sunday I'll pick another kernel to flash so I can do Lionheart on Monday.

Re: A Governor a Day
Purpeltendire said:
Alright, thanks. I'll have to give your kernel and iHate's Deluxe a try when I finish up with the governors in Snuzzo's,
Without further ado, here's Intellidemand as requested. [Now with BetterBatteryStats!]
Intellidemand End 1
Intellidemand End 2
Intellidemand End 3
The governor seemed to be decent at scaling, keeping the CPU spread out more than either DanceDance or Wheatley. Still between one and five minutes spent at the more popular clocks, with the majority of the time logged at 1350, 192, and Deep Sleep as usual. Nothing major to report here, pretty standard for what I've seen so far.
GSam has something different to say. For one, battery life seemed strained. There was considerably more usage at idle or deep sleep than either DanceDance or Wheatley. I watched a movie at one point this evening, and when I checked my phone again it had dropped 6% over 2.5 hours, without touching it once. It also ramped up quickly while browsing, I think in the time it took me to make a 2 sentence post the battery dropped 6 or 7%. That didn't happen with the others. The battery discharge graph is pretty steep, which makes sense given the increased discharge.
Also, app usage seemed abnormally high. I checked on that and found that the Android OS/Kernel was using a LOT of power: 65%. Yesterday on DanceDance it was using 6%... there's something weird here.
If someone running Intellidemand (Squirrel1620?) could run GSam for a day and see if this happens to you, that would be great. I'd like to see if this is a random event or something consistent with the governor.
SystemPanel reports similar findings under "System Processes" with 43% of my total CPU usage. But quite frankly, other than that there's nothing really weird going on. No wakelocks, no apps freezing... it just seems like the OS wasn't idle for some reason.
My verdict: Not a governor for idle users, partly because it's based off the GPU. If you do gaming or use apps that require a bit more graphical power, then by all means go for it. But for users like me, who text, browse the web a bit, and do some forum work, this governor has very little to offer except minimal noticed lag.
Tomorrow [Saturday] I'll do OnDemand, and Sunday I'll pick another kernel to flash so I can do Lionheart on Monday.
Click to expand...
Click to collapse
I'll run intellidemand on lunar kernel with EclipticRez. Starting off at 100% and ill post gsam when it gets to about 14
Sent from my ADR6425LVW using xda app-developers app

Re: A Governor a Day
Thanks for your testing info. Battery and performance are important to me which I'm sure is much like others. Great reviews so far
Sent from my ADR6425LVW using xda app-developers app

Squirrel1620 said:
I'll run intellidemand on lunar kernel with EclipticRez. Starting off at 100% and ill post gsam when it gets to about 14
Sent from my ADR6425LVW using xda app-developers app
Click to expand...
Click to collapse
Thanks, I appreciate it.
Jwezesa said:
Thanks for your testing info. Battery and performance are important to me which I'm sure is much like others. Great reviews so far
Sent from my ADR6425LVW using xda app-developers app
Click to expand...
Click to collapse
Thank you. Anything in particular you'd like to see?
EDIT: Phone decided to shut itself off while I wasn't paying attention, so the stats were all reset. No results today unfortunately.

Re: A Governor a Day
Ok so this was intellidemand underclocked 1.35 ghz normal usage mostly 3g/wifi
Sent from my ADR6425LVW using xda app-developers app

Related

battery temp

root rookie here.. First device I ever rooted, really I'm new to cell phones in general.
I flashed this rom adryn bamf 4.1 (non remix) and its working good and all
But I noticed on the smartass setting, which underclocks, that after about an hour or so of playing a game/tv whatever my batter will be around 42C, it seems very hot compared to around 32C before I put this rom and kernal on
Two questions
Am I doing something wrong or is this normal
And, How hot can a battery get before it becomes something to worry about?
Thanks =P
Yea in the 40's is way high. Remember, heat is inefficiency (wasted energy). After playing a game for an hour you will probably be in the mid 30's C. I would recommend trying a different kernel and wiping devlik cache. Also, see what apps and services are running in the background and increase the length of time between syncs. Also, smartass scaling did great on my incredible but I have found that it is not perfected yet for the tb. I would suggest ondemand unless the dev or OP specifically says that smartass is the way to go. As far as heat and battery I find that adrenylyn's kernels do the best. As far as performance, drod and ziggy's seem to fly. Isoman or whatever seems to also be a favorite but I personally didn't have great results. It is important to not supremely OC or UC. I would simply stray away from UC in general. For a daily driver, I would stay under 1.5 Ghz, and honestly 1.2 seems to work best for me. Hope this helps.
thanks for taking the time to respond
So why no under clocking? I assumed it would help battery life but not so much?
Just set to 1200 for both min/max? Or on demand scaling up to 1200?
I prob had a bunch of apps running in the backround, i'm new to droid.. came from an old feature phone
42C is definately too hot! As the previous poster suggested, try a different kernel. There also could be a remote chance your battery is defective.
I really feel like it is more likely that something I am doing is impacting it
I had the phone stock for a week and no problems (same usage)
Rooted but stock rom for a week and no problems
Put this kernal on (which everyone says is really good) and this rom and it was getting hot.. So far it is good around 38-39 (still too high?) with 1200mhz on demand min 256mhz
I flashed 2 different kernals and both times after less than 45 mins of browsing forums and playing home run baseball my battery gets to 42C. Could this be caused by the rom? I wonder if I damaged my battery or phone at this point
Mine got up to 113 degrees over the weekend. It felt like I was cooking my phone off and getting ready to throw it like I do in black ops.
I realized my phone was trying to search for a 4g signal when there was none. I entered #*#*#4636#*#*# on my dial pad to turn it off. It reduced my battery heat by 5 degrees. I should point out I was out in the field. Aka in the middle of nowhere where.
Sent from my rooted Thunderbolt with VirusROM AirborneTB. Xda premium
I'm @ 53.4C right now and it doesn't seem to be charging, lol.
raider3bravo said:
Mine got up to 113 degrees over the weekend. It felt like I was cooking my phone off and getting ready to throw it like I do in black ops.
I realized my phone was trying to search for a 4g signal when there was none. I entered #*#*#4636#*#*# on my dial pad to turn it off. It reduced my battery heat by 5 degrees. I should point out I was out in the field. Aka in the middle of nowhere where.
Sent from my rooted Thunderbolt with VirusROM AirborneTB. Xda premium
Click to expand...
Click to collapse
Sorry but to fly off topic what do you do? I'm generally out in the boondocks when I'm on wellsite...
Sent from my Thunderbolt running CM7...
42c isn't anything out of the ordinary, or outside operational parameters for a battery of that type. Not by a long shot, actually. That battery can safely operate at 59c, but the phone won't charge it north of about 47c due to the fact charging will increase the temperature even more.
As for the governors:
The smartass governor operates similarly to the interactive governor, but isn't as aggressive and allows for wake up lag reduction. The governor SHOULD be your main line of controlling clock speed, not your min and max settings. Some might like to argue that point, but that's mainly because they haven't messed around with governor parameters. With and ondemand or interactive governors, and even some smartass governors (Not Ziggy's modified), it's entirely possible to set min as low as 61MHz, and max at 1.65GHz and have the CPU rarely, if ever, reach those speeds. It depends entirely on what you've set the governor parameters to be. I don't know of any app that allows you to modify those because the locations of the parameters aren't always the same name or in the same places so it's best done through a script. The script I uses does routinely reach up to my freq_max, but it does it fairly aggressively, then ramps down just as aggressively. Part of that is the nature of the lagfree governor, part of it is where I've set the thresholds. In any event, governor control via a script is a much better solution than castrating your device.

Need SetCPU or SGS2 varies CPU already?

I use SetCPU to help battery life but someone told me it's not needed on the SGS2 as it already scales CPU for demand. True?
leedavis said:
I use SetCPU to help battery life but someone told me it's not needed on the SGS2 as it already scales CPU for demand. True?
Click to expand...
Click to collapse
Interesting point you raise actually.
I've just installed Setcpu and used the on demand governor. I left the values as default (200mhz for minimum and 1.2 ghz for maximum) - with no overclock.
I've immediately noticed swiping through the screens is a bit smoother and the biggest improvement is the gallery. All my photos appear much smoother. Before the gallery app was a bit lagy.
I haven't set any profiles yet such as screen off.
Every Android phone I've owned scaled the cpu, I think they all do. I've found that with setCPU my battery gets drained much faster en no real benefit in smoothness.
jzuijlek said:
Every Android phone I've owned scaled the cpu, I think they all do. I've found that with setCPU my battery gets drained much faster en no real benefit in smoothness.
Click to expand...
Click to collapse
Have you tried this yet though on the Galaxy S2?
There is definitely less lag than before - as stated, specifically in the gallery app. Just generally swiping feels more responsive as well. Battery is still pretty awesome, especially when using Lightening Rom 1.1 and the Android battery calibration app.
Hmm. I'll try SetCPU on the SGS2 and post back the findings (Performance+Battery).
I don't know how can it get any more smoother, I mean its already SO smooth!
there are many points to use setcpu on gs2:
-for some reason I dunno, gs2 can't manage it's 1.2ghz without gettin too warm. downclock and get rid of the burn effect.
-gs2 sports a good management of gpu (it does most of the work and setcpu doesnt down\overclock that). downclockin doesnt affect UI or video o browsing experience at all. can even downclock at 500 max speed without any sides.
-the only side u ll see it's benchmark (quadrant downgrading to 2000) but I hope u won't pay attention to such an unseful thing. benchmark doenst mean nothing, daily usage it's the only point to look at.
my settings: conservative, 200min 800max.
battery draining doesnt belong to setcpu this time, look to other settings.
alexleon said:
there are many points to use setcpu on gs2:
-for some reason I dunno, gs2 can't manage it's 1.2ghz without gettin too warm. downclock and get rid of the burn effect.
-gs2 sports a good management of gpu (it does most of the work and setcpu doesnt down\overclock that). downclockin doesnt affect UI or video o browsing experience at all. can even downclock at 500 max speed without any sides.
-the only side u ll see it's benchmark (quadrant downgrading to 2000) but I hope u won't pay attention to such an unseful thing. benchmark doenst mean nothing, daily usage it's the only point to look at.
my settings: conservative, 200min 800max.
battery draining doesnt belong to setcpu this time, look to other settings.
Click to expand...
Click to collapse
IT IS TRUE. I agree with every single line you wrote, it is just my expericence.
I have too setcpu conservative and undercloked 800 Mhz.
There no slow down or lag at all... But I am wondering if it gives a real boost to battery life. I am not sure of this.
I'll keep you guys posted... But I think that an undevolted Kernel it is really a need as for the solution of the damn dual core ginger bug that is sucking 20% of my battery every day
Well,from my experience with my Desire and Desire HD(won't even bother with the Hero,I had no real knowledge then),governors can make a huge difference.I for one like smartass or interactive governors(mostly the same).I wouldn't suggest conservative,interactive does the job much better.Tasks get done in less time and the CPU throttles down more quickly.Other than that,you can underclock or overclock all you like,it never made any big difference in battery life for me(Unless Sammy's CPUs are different in that aspect-Snapdragons are really "overclock-friendly").That's personal preference after all!
Anyway,the best solution IMO would be a vdd_levels interface.For those who don't know what it is,it is a mod made by -snq(Meet him at the Desire forums-That guy's a true LEGEND!He can patch/modify anything!),which practically allows you to change the voltage levels of the CPU on the fly rather than having to stick with the values hardcoded into the kernel.Using this and a simple script in GScript to change values that won't survive reboot or in init.d to be applied on boot,you can find the optimal voltages for your CPU(Don't forget,every CPU is unique and different),thus reducing heat and maximizing battery life.
If a dev brings that to the SGS2 it will be a big step in the right direction as far as I'm concerned.
I use SetCPU without issue, but only to run profiles (i limit the device to 500mhz when the screen is off). The rest of the time it scales itself up to 1.4GHz without fuss and using stock voltage. Battery life is fine, best ive had for an android device.
Wow, I've taken SetCPU off but left JuiceDefender on and my battery life is fantastic. At 70% after slightly heavier than normal use (used for listening to music for a couple of hours this morning) and been off charge for 8.5 hours.
SetCPU seems counterproductive on SGS2
leedavis said:
I use SetCPU to help battery life but someone told me it's not needed on the SGS2 as it already scales CPU for demand. True?
Click to expand...
Click to collapse
I did set ondemand which is a little more reactive and slightly smoother.
Though System Tuner is less cpu-consuming and much more useful on the SGS2. No need for all those complicated settings from setCPU. Only changing governor and changing frequencies on standby are useful.
leedavis said:
Wow, I've taken SetCPU off but left JuiceDefender on and my battery life is fantastic. At 70% after slightly heavier than normal use (used for listening to music for a couple of hours this morning) and been off charge for 8.5 hours.
SetCPU seems counterproductive on SGS2
Click to expand...
Click to collapse
Juice Defender uses as much battery as it saves this is fact, i have SepCPU set to 200 Min - 800 Max - On demand and have Juice Defender Ultimate and i thought it was great but it was recommended to me that i could save more battery by not using this, initially i was skeptical but tried it and i was astonished at the results, my battery life improved by 9 hours (i carried out a test with JD and without)
Anyone who says SetCPU uses up loads of battery is talking nonsense,it actually saves battery if configured correctly.
I am using Check Rom with set CPU I have it 1.2ghz max and 200. Using conservative governer. I been off charge for 15hrs, however I am using light usage I am on 72% screen on has been 5h 25m at time of writing. Not yet calibrated the battery.
jonny68 said:
Juice Defender uses as much battery as it saves this is fact, i have SepCPU set to 200 Min - 800 Max - On demand and have Juice Defender Ultimate and i thought it was great but it was recommended to me that i could save more battery by not using this, initially i was skeptical but tried it and i was astonished at the results, my battery life improved by 9 hours (i carried out a test with JD and without)
Anyone who says SetCPU uses up loads of battery is talking nonsense,it actually saves battery if configured correctly.
Click to expand...
Click to collapse
What are the settings are you using for the setcpu program in your S2?? Did you remove the Juice defender application?

setcpu with ics roms

has anyone else tried using setcpu with ics roms? does it actually do anything?? I'm trying now, and i set my lowest frequency to 192 and like max at 894...thing is I don't notice the phone being any slower... is it actually doing anything? it says it is.. system panel shows it maxing out at 894..but I'm still dropping battery like crazy lol clearly it doesn't help enough if it even works.
Sent from my ADR6425LVW using XDA App
Known bug with ics roms, will not work at all until source is released and we get someone to to build a kernel without HTC perflock built in
Sent from my aokpcb_tenderloin using xda premium
don't use any apps that control the CPU. that goes for frequency,governors,etc. it messes with perflock which causes the CPU to run at only the min/max frequency and no proper scaling. stick with this advice until custom ics kernels come
It works for me. Tell me I'm stupid if you want but my battery doesn't drain as fast with it. If I set CPU to 500mhz it runs slower so it does something. It does bounce around due to perflock. But it helps me
Sent from my ADR6425LVW using Tapatalk 2
Evocm7 said:
It works for me. Tell me I'm stupid if you want but my battery doesn't drain as fast with it. If I set CPU to 500mhz it runs slower so it does something. It does bounce around due to perflock. But it helps me
Sent from my ADR6425LVW using Tapatalk 2
Click to expand...
Click to collapse
Ehhh far to many times I'd have the phone go 1.5Ghz full blast non stop even with the screen off. Be at work and wonder why my leg was on fire.
Sent from my ADR6425LVW using XDA
It works great for me, and was just updated recently and fixed all the little minor problems and occasional crashes that I used to have.
None of the stock kernels should ever be govern with a CPU app. I have no idea why people say it only applies to ICS. In fact this applies to BOTH GB and ICS kernels. GB and ICS kernels have the same battery life on stock roms.
If the kernel does not allow you to disable perflock, it won't scale right, stock kernels cannot have it disabled. I say again, BOTH ICS and GB because people like to give out FALSE INFORMATION.
Sent from my ADR6425LVW using XDA
nosympathy said:
Ehhh far to many times I'd have the phone go 1.5Ghz full blast non stop even with the screen off. Be at work and wonder why my leg was on fire.
Sent from my ADR6425LVW using XDA
Click to expand...
Click to collapse
That has happened to me once. But not since last update. It does bounce around sometimes but never goes above the max I specify so I'm happy with it. Keeps temps down under heavy use as well even though temps aren't much of an issue with latest leak
Sent from my ADR6425LVW using Tapatalk 2
Works fine after the latest update on conservative governor, scales throughout the range given. Cleanrom 4.4 stock kernel.
richii0207 said:
I say again, BOTH ICS and GB because people like to give out FALSE INFORMATION.
Click to expand...
Click to collapse
exactly...
lol so people are saying it works, and people are saying it doesn't.
doesn't really answer my question lol
Sent from my ADR6425LVW using XDA App
Quadrant only shows my CPUas being single core . I'm on CleanRom Standard Edition 4.5. Wasn't sure if that had been addressed in a thread yet or not . So I wasn't sure where to address this, but this seemed like a good place . I took a screenshot of it .
Sent from my ADR6425LVW using xda premium
jayochs said:
lol so people are saying it works, and people are saying it doesn't.
doesn't really answer my question lol
Sent from my ADR6425LVW using XDA App
Click to expand...
Click to collapse
If it works its doing something, it works. It works on cleanrom ics pro 4.4 on conservative governor since the last update before that it wasn't working right for me it would jump to Max all the time now it scales right.
Kane5581 said:
If it works its doing something, it works. It works on cleanrom ics pro 4.4 on conservative governor since the last update before that it wasn't working right for me it would jump to Max all the time now it scales right.
Click to expand...
Click to collapse
i'm running cleanrom 4.5 right now. when i set to conservative, it jumps all over, but it still maxes ouat at 1512 often.
i just stopped touching it because i didn't feel like breaking anything lol but again, some people are saying due to perflock it won't work, while others say they have experience of it working. so confusing.
Perflock being the pain it is, I'm staying away from it right now.
When I had it installed, my battery life was complete fail.
Sent from my ADR6425LVW using Tapatalk 2
jayochs said:
lol so people are saying it works, and people are saying it doesn't.
doesn't really answer my question lol
Sent from my ADR6425LVW using XDA App
Click to expand...
Click to collapse
All you can do is try it. Worst case scenario your battery drains too fast. Then uninstall it. Why it helps some and not others, I don't know. It helped me get 16 hrs of battery life with 4hr screen time with 4g on stock battery (ics) yesterday. No way in hell I could get that without setcpu
Sent from my ADR6425LVW using Tapatalk 2
Well at the risk of posting terrible "FALSE INFORMATION!", I'll let you know my experience with setCPU so far and you can make your own decisions from there.
First, here's my current config:
Android 4.0.1, SebastianFM's 1.20 OC Kernel, CFQ IO scheduler, interactive or on-demand governor depending on situation (cpu profiles enabled).
Profiles enabled, listed lowest to highest priority:
bat <=100% -- min=384, max=1566, interactive, CFQ
bat < 90% -- min=384, max=1350, interactive, CFQ
bat < 75% -- min=192, max=1188, interactive, CFQ
bat < 50% -- min=192, max=1026, interactive, CFQ
temp > 105*F -- min=192, max864, interactive, CFQ
bat < 34% -- min=192, max=864, conservative, CFQ
bat < 25% -- min=192, max=648, conservative, CFQ
SCREEN OFF -- min=192, max=540, interactive, CFQ
temp > 120*F -- min=192, max=540, conservative, CFQ
bat < 15% -- min=192, max=540, conservative, CFQ
So, I realize Perflock is enabled, so frequencies do tend to jump around. However, when you set and enable profiles, setCPU does seem to control the frequencies still, even though you can watch the min/max frequencies move around.
I use the setCPU widget on my homescreen too. So lets assume for example my battery is between 90-100% charge, in which case my min=384 and my max=1566. With profiles enabled, I can watch my minimum frequency occasionally go as high as my max, or 1566. Generally, this happens when my phone is actually using the 1566 frequency at the present moment, which is indicated by the current frequency display on the widget. However, if my current CPU speed happens to be anywhere in between the range i selected, lets just say its at 864mhz currently. In this situation, my current frequency will show 864mhz, but my min frequency might also show 864mhz and my max might show 1188mhz, at that particular time.
OK so now my battery has depleted to 49%, and so my profile kicks in and sets the minimum frequency at 192 and the maximum frequency at 1026. In this case, I'll NEVER see a current CPU speed over 1026mhz. I might see both the minimum and maximum at 1026mhz for a period of time, or I may see the current frequency say 648mhz while the minimum is at 192mhz and the maximum reports 648mhz also.
The point is, YES, I can watch perflock doing its thing and changing my min/max frequencies around constantly. However, by simple observation I can confirm it does actually scale within my specified parameters.
There was a bug with setCPU up to about a month ago where it would occasionally crash and your CPU would just go to its normal default (384-1512mhz, deadline scheduler, on-demand governor), however the recent updates have fixed the crashes! SetCPU now continually runs without crashing, and when you use profiles, it DOES seem to scale exactly how you tell it to.
a.mcdear said:
Well at the risk of posting terrible "FALSE INFORMATION!", I'll let you know my experience with setCPU so far and you can make your own decisions from there.
First, here's my current config:
Android 4.0.1, SebastianFM's 1.20 OC Kernel, CFQ IO scheduler, interactive or on-demand governor depending on situation (cpu profiles enabled).
Profiles enabled, listed lowest to highest priority:
bat <=100% -- min=384, max=1566, interactive, CFQ
bat < 90% -- min=384, max=1350, interactive, CFQ
bat < 75% -- min=192, max=1188, interactive, CFQ
bat < 50% -- min=192, max=1026, interactive, CFQ
temp > 105*F -- min=192, max864, interactive, CFQ
bat < 34% -- min=192, max=864, conservative, CFQ
bat < 25% -- min=192, max=648, conservative, CFQ
SCREEN OFF -- min=192, max=540, interactive, CFQ
temp > 120*F -- min=192, max=540, conservative, CFQ
bat < 15% -- min=192, max=540, conservative, CFQ
So, I realize Perflock is enabled, so frequencies do tend to jump around. However, when you set and enable profiles, setCPU does seem to control the frequencies still, even though you can watch the min/max frequencies move around.
I use the setCPU widget on my homescreen too. So lets assume for example my battery is between 90-100% charge, in which case my min=384 and my max=1566. With profiles enabled, I can watch my minimum frequency occasionally go as high as my max, or 1566. Generally, this happens when my phone is actually using the 1566 frequency at the present moment, which is indicated by the current frequency display on the widget. However, if my current CPU speed happens to be anywhere in between the range i selected, lets just say its at 864mhz currently. In this situation, my current frequency will show 864mhz, but my min frequency might also show 864mhz and my max might show 1188mhz, at that particular time.
OK so now my battery has depleted to 49%, and so my profile kicks in and sets the minimum frequency at 192 and the maximum frequency at 1026. In this case, I'll NEVER see a current CPU speed over 1026mhz. I might see both the minimum and maximum at 1026mhz for a period of time, or I may see the current frequency say 648mhz while the minimum is at 192mhz and the maximum reports 648mhz also.
The point is, YES, I can watch perflock doing its thing and changing my min/max frequencies around constantly. However, by simple observation I can confirm it does actually scale within my specified parameters.
There was a bug with setCPU up to about a month ago where it would occasionally crash and your CPU would just go to its normal default (384-1512mhz, deadline scheduler, on-demand governor), however the recent updates have fixed the crashes! SetCPU now continually runs without crashing, and when you use profiles, it DOES seem to scale exactly how you tell it to.
Click to expand...
Click to collapse
awesome analysis. ...so what's your battery life like?
I've gottent it to work a couple of times, other times it seems to do nothing at all, and i have also had it almost burn a hole in my pocket due to heat. I would say dont try to OC for now and wait until we have something more stable as they have previously stated. To be honest its a hit or miss type of mod for now.
jayochs said:
awesome analysis. ...so what's your battery life like?
Click to expand...
Click to collapse
My battery life is pretty good. I'm very surprised by how little i actually notice the difference in responsiveness and performance as my CPU throttles lower as my battery drains throughout the day.. However, I will admit that the only reason I have it scaling so low like that was to match the same battery life that I was getting on totally de-sensed GB builds...

[Q] Overclocking

Im taking a poll on the average rezound user cpu frequency for development
bunchies said:
Im taking a poll on the average rezound user cpu frequency for development
Click to expand...
Click to collapse
lol... I would always underclock to the 1200Mhz range to save battery, or use the stock 1512 with a more conservative governor like Intellidemand, rarely ever overclocked and when I did it was just to test something or run a quick benchmark, never really saw any real gain to it at all, always thought it was better to lower the bottom end as low as you could get it to 96-192Mhz so that even if something kept it awake, it would barely be running.
Whenever I'm running "Intensive items" I sometimes overclock. Most of the time I try to underclock to save battery as ace said. Although Doesn't really save that much battery. I stay at stock 1.5GHz a lot though...
pball52998 said:
Whenever I'm running "Intensive items" I sometimes overclock. Most of the time I try to underclock to save battery as ace said. Although Doesn't really save that much battery. I stay at stock 1.5GHz a lot though...
Click to expand...
Click to collapse
True... Underclocking the top end doesn't do much, although if you can lower the bottom end, and undervolt some, it can make a big difference when sleeping.
Sent from my Nexus 4 using Tapatalk
acejavelin said:
True... Underclocking the top end doesn't do much, although if you can lower the bottom end, and undervolt some, it can make a big difference when sleeping.
Sent from my Nexus 4 using Tapatalk
Click to expand...
Click to collapse
O you're saying when sleeping. Holy crap YES!! It makes a HUGE difference. I run sense 5 a lot (running that right now) and the battery life on "sleep" is already great, but when I underclock like I did the other day, I forgot to plug my phone in that night, it was at 5% I woke up and it was at 4% 8 hours later only 1% lost. Of course thats on Snuzzo's funky kernel and underclocked, but DANG!!! NEVER seen my battery life that good. The 4% also lasted another 5 hours without use! Don't know how but it did. :laugh::good:
Depends what I'm doing, usually under clocked to save battery throughout the day. I used to record YouTube videos on my rez, I would over clock for recording and benchmarks, finding the perfect i/o scheduler, governor and clock speed to run stable was fun.
Sent from my ADR6425LVW
1728. Fastest mine goes without being unstable.
bunchies said:
Im taking a poll on the average rezound user cpu frequency for development
Click to expand...
Click to collapse
1728 all the time.
Sent from my ADR6425LVW using Tapatalk

Whats the most power saving rom

around?
You post in the wrong spot and also u need to find that out on your own because all ROMs have deferent stuff inside them also all phones not all work the same
Sent from my LG-E970 using XDA Free mobile app
Posting in the right forum is my favorite way to save battery.
this is the rom sub-forum isnt it? if isnt can someone please move it to the right place
Don't listen to what anybody says: battery life isn't too ROM-dependent for our LG OG. I get the same battery life no matter what ROM I use. It all depends on your kernel settings. I get at least 4 hours of screen on time with the following settings in the Official CM11 rom with the LGE Gproj Kernel:
162 min, 1026 max, Wheatley with a powersave bias of 1, MPDecision ON (from what I've seen, it's better at turning off cores than the in-kernel implementation), max GPU at 128 if you don't plan on playing too many games with the interactive governor, Multicore Power-Saving set to 1, and -150 UV across the board (voltage for 162MHz starts at 650; I have a fast-binned E970). Also, I use auto-brightness but I've edited the brightness table heavily so it's JUST bright enough for most situations. I also keep GPS, Bluetooth, NFC, LTE, notification light, and cap keys off. My data is off whenever possible too. And every time before I turn off my device, I press Greenify's "hibernate now" widget to prevent any wakelocks. Lastly, I use ART instead of Dalvik but I don't think that it makes that much of a difference in battery life (that should change by the time Android 4.5 comes out though).
Hope all this helps lol. But seriously, you posted in the wrong forum noob
i apologize for the mistake but in other forums i havent really been meet with such issues then again i suppose this phone is more popular then my previous ones. In any case i got around 4:30 hours right now, i was just hoping to get more out of it. I would expect a phone with such a huge battery to do 8 or 9 hours but maybe thats just too much
topdawg7793 said:
Don't listen to what anybody says: battery life isn't too ROM-dependent for our LG OG. I get the same battery life no matter what ROM I use. It all depends on your kernel settings. I get at least 4 hours of screen on time with the following settings in the Official CM11 rom with the LGE Gproj Kernel:
162 min, 1026 max, Wheatley with a powersave bias of 1, MPDecision ON (from what I've seen, it's better at turning off cores than the in-kernel implementation), max GPU at 128 if you don't plan on playing too many games with the interactive governor, Multicore Power-Saving set to 1, and -150 UV across the board (voltage for 162MHz starts at 650; I have a fast-binned E970). Also, I use auto-brightness but I've edited the brightness table heavily so it's JUST bright enough for most situations. I also keep GPS, Bluetooth, NFC, LTE, notification light, and cap keys off. My data is off whenever possible too. And every time before I turn off my device, I press Greenify's "hibernate now" widget to prevent any wakelocks. Lastly, I use ART instead of Dalvik but I don't think that it makes that much of a difference in battery life (that should change by the time Android 4.5 comes out though).
Hope all this helps lol. But seriously, you posted in the wrong forum noob
Click to expand...
Click to collapse
I consistantly get ~4 hours screen on with 384 min/1512max ONDEMAND. Everything else (undervolting, alternate governors, disabling radios) just hurts it for my OG.
Like honestly, Mine has gotten a lot better since I quite trying to do so much to it and just let it do its thing.
droidkeyboardfan12 said:
i apologize for the mistake but in other forums i havent really been meet with such issues then again i suppose this phone is more popular then my previous ones. In any case i got around 4:30 hours right now, i was just hoping to get more out of it. I would expect a phone with such a huge battery to do 8 or 9 hours but maybe thats just too much
Click to expand...
Click to collapse
You're never gonna see 8 to 9 hours on any phone for on screen time. I had a huge extended battery on my SGS3 and still got only 7 at best (4200mah battery)
Sent from my LG-E970 using XDA Premium 4 mobile app
well is there at least a rom that is more stable then others for this phone?
droidkeyboardfan12 said:
well is there at least a rom that is more stable then others for this phone?
Click to expand...
Click to collapse
From my personal experience. CM nightlys are very stable. I've read good things about mokee but haven't tried it out yet. I think CM with xposed framework and AC display from the play store is perfect.
I get about 3 to 3.5 hours screen time with about 12 hours off charger (this is with location turned off... Until Google fixes all the nlp wakelocks it's not worth having it turned on).
Sent from my LG-E970 using XDA Premium 4 mobile app
I still use the 4.4 beanstalk rom with the adreno kernel.. from my personal experience it's the best.. power saving wise..

Categories

Resources