Database Users

[TUT] SetCPU and Advanced Settings

I've been using setcpu for a while now, but have never bothered to mess with the advanced settings. Searching around I have only found out what most of this stuff means, but I'm missing some still. I thought I would share my findings. I have included SetCPU's descriptions (in italics) supplemented with my findings.
Governor choices (I'm using king's bfs kernel #1 on fresh 3.1.0.2) -
Ondemand - Uses the highest frequency when tasks are started, decreases step by step
Conservative - Increases frequency step by step, decreases instantly
Interactive - I couldn't figure this one out... any help?
Powersave - Uses the lowest possible clock speed to complete its tasks
Userspace - Manual controll of the frequency
Performance - Always uses the highest clock speed
Advanced Settings -
Sampling Rate - An interval (in microseconds) at which the governor will poll for updates. When this happens, the governor will decide whether to scale the CPU up or down. It uses such little power that it is better at lower values when using profiles such as screen off.
Up Threshold (1%-100%) - Defines a percentage from 1% to 100%. When the CPU load reaches this point, the governor will scale the CPU up. When using low min values (245), this happens instantly, using higher values (768) it overclocks less often. With Conservative lower values are better because it slowly increases your clock speed to what you need, with Ondemand, higher is better, as it overclocks less often.
Down Threshold (1%-100%) (conservative only) - Defines a percentage from 1% to 100%. When the CPU load reaches this point, the governor will scale the CPU down. Higher values will offer more aggressive battery saving, lowering the clock speed quicker.
Ignore Nice Load (0/1) - If this value is "1," the system will ignore "Nice" processes when deciding to scale up or down. I need a little more info for this one, what exactly is a nice process? DO NOT GOOGLE 'NICE LOAD' ESPECIALLY AT WORK OR AROUND CHILDREN/WIFE
Freq Step (0%-100%) (conservative only) - Defines how much (as a percentage of the maximum CPU speed) the conservative governor will increase the CPU speed by each time the CPU load reaches the Up Threshold. Increased the value slightly to be able to overclock quicker, but not to high to avoid unnecessary overclocking.
Powersave Bias (0-1000) (ondemand only) - Setting this value higher will "bias" the governor toward lower frequencies. This is a percentage, where 1000 is 100%, 100 is 10%, and 0 is 0%. The ondemand governor will scale the CPU to a frequency lower than its "target" speed according to this value. Gives ondemand some more battery saving potential. High values give worse performance than conservative with equal or worse battery saving. If you want the performance of ondemand with some more battery use values under 200.
I hope this info was helpful to someone, and here are my setcpu settings. I have attempted to target 150-175ms for short and 350-400ms for long benchmarks to match my performance governor and save battery at the same time.
With ondemand I get about 170ms short and 380ms long. I use 90 for up and 50 for powersave. The performance is slightly better than the default settings, and the battery is about equal. I might play with this more, as it should hit the same values as performance with better battery life.
In conservative long benchmarks in setcpu are actually faster than short ones because it takes setcpu time to adjust the speed. Run a short one immediately after a long one to see its actual value. Up changed 75 and down to 25, not much of a change, but drastic performance increase with no battery change. I also increased freq step to 10% to obtain higher speeds faster. Getting the same 170ms short and 370ms long.
My Settings
Conservative 245-1190
Temp > 50C - 768 conservative
Screen Off - 499 ondemand (allows for the screen to be unlocked faster, especially useful with incoming calls)
Charging/Full - 1190 performance
Battery < 15% - 652 conservative
Sampling - 200000
Up Thresh - 75
Down Thresh -25
Ignore Nice - 0
Freq - 10
More DFS Info
SetCPU Info

davebu said:
DO NOT GOOGLE 'NICE LOAD' ESPECIALLY AT WORK OR AROUND CHILDREN/WIFE
Click to expand...
Click to collapse
LMAO 10chars

HondaCop said:
LMAO 10chars
Click to expand...
Click to collapse
Yeah. I almost spit out my Vanilla Coke on that one. LOL

Anytime have any info about nice load or anything to add?
Sent from my HTC EVO 4G.

HondaCop said:
LMAO 10chars
Click to expand...
Click to collapse
I missed this yesterday... Post of the day in my opinion

Thanks dave...good write up

This is what I found about the interactive governor in github:
cpufreq: interactive: New 'interactive' governor
New interactive governor.
This governor is designed for latency sensitive workloads, UI interaction for
example.
Advantages:
+ significantly more responsive to ramp cpu up when required (UI interaction)
+ more consistent ramping, existing governors do their cpu load sampling in a
workqueue context, the 'interactive' governor does this in a timer context, which
gives more consistent cpu load sampling.
+ higher priority for cpu frequency increase, rt_workqueue is used for scaling
up, giving the remaining tasks the cpu performance benefit, unlike existing
governors which schedule rampup work to occur after your performance starved
tasks have completed.
Existing governors sample cpu load at a particular rate, typically
every X ms. Which can lead to under powering UI threads when the user has
interacted with an idle system until the next sample period happns.
The 'interactive' governor has a different approach. Instead of sampling the cpu
at a specified rate, the governor will scale the cpu frequency up when coming
out of idle. When the cpu comes out of idle, a timer is configured to fire
within 1-2 ticks. If the cpu is 100% busy from exiting idle to when the timer
fires then we assume the cpu is underpowered and ramp to MAX speed.
If the cpu was not 100% busy, then the governor evaluates the cpu load over the
last 'min_sample_rate' (default 50000 uS) to determine the cpu speed to ramp down
to.
There is only one tuneable for this governor:
/sys/devices/system/cpu/cpufreq/interactive/min_sample_rate:
The minimum ammount of time to spend at the current frequency before
ramping down. This is to ensure that the governor has seen enough
historic cpu load data to determine the appropriate workload.
Default is 5000 uS.
Also, in the original application thread as explained by the dev, "nice" processes are:
2. Nice processes are used by the IO scheduler to designate a low-priority process. Ignore nice load basically tells ondemand to disregard processes with higher nice values.

Good topic. You covered the bases pretty well. Glad someone finally put this together as it is useful to know. Now prepare for 1000 threads in the next month asking for the information you just posted.

hey question. i went and purchased SetCPU and attempted to follow your instruction. problem is, whenever SetCPU tries to gain super user permission, it says "no root access granted. Are applications allowed root access?" i dunno what to do. can someone advise me?

Umm, is your phone rooted?
Sent from the void...

Yessir. Since day 2 ^_^ (plus its in my sig)
Sent from my Evo using Tapatalk

SilverStone641 said:
Yessir. Since day 2 ^_^ (plus its in my sig)
Sent from my Evo using Tapatalk
Click to expand...
Click to collapse
Try uninstalling and reinstalling it.
Then double check all of your superuser settings.

SilverStone641 said:
Yessir. Since day 2 ^_^ (plus its in my sig)
Sent from my Evo using Tapatalk
Click to expand...
Click to collapse
Sorry, using the xda app which doesn't display the sig.
Sent from the void...

So, as far as speed/responsiveness of governors goes:
Fastest ------------------------------------------------------> Slowest
Performance ------> ondemand ------> Interactive? ------> Conservative
Poor battery consumption --------------------> Best battery consumption

This thread is exactly what i was looking for, thanks for the detailed explanation of the what and why.
Will try it out this week with Fresh 3.1 and KK#8.

this thread helped a lot, i was just in setCPU messing around with things, now i can use this thread to help get what i want. i bookmark'd the hell out of this thread

Thanks...OP...hopefully people will read it first...try things..then ask questions...
I am still working to see how to get the best battery life from cm6 and snap..

Thanks for the helpful post!

I experienced a "nice load" when I unboxed my EVO. Anyway the only setting I use is:
Screen off: 245/128 on demand.
Works for me. And thanks for this helpful post to help us understand all that technical mumbo jumbo.

So I got a rooted Vanilla install of the latest Sprint OTA Froyo build on my EVO. (the 3.29.651.5 build).
I purchased the latest version of SetCPU (2.03) last night and used the autodetect method for the CPU governor.
I notice on my EVO that I only have these 3 options:
Scaling:
ondemand, userspace and performance....
Is this normal to not have the conservative setting since I have the defacto kernel with a vanilla rom?
Thanks
Sheldon
Okay, so I figured it out, my default kernel does not have these other options, oh well......
Nice app though, so far its working really well.

[GUIDE] Governors for NOOBS

Interactive - Instead of sampling the cpu at a specified rate, the governor will scale the cpu frequency up when coming out of idle. When the cpu comes out of idle, a timer is configured to fire within 1-2 ticks. If the cpu is 100% busy from exiting idle to when the timer fires then we assume the cpu is underpowered and ramp to MAX speed.
Smartass- Is an improved version of interactive governor
Ondemand – Available in most kernels, and the default governor in most kernels. When the CPU load reaches a certain point (see “up threshold” in Advanced Settings), ondemand will rapidly scale the CPU up to meet demand, then gradually scale the CPU down when it isn't needed.
Conservative– Available in some kernels. It is similar to the ondemand governor, but will scale the CPU up more gradually to better fit demand. Conservative provides a less responsive experience than ondemand, but can save battery.
Performance – Available in most kernels. It will keep the CPU running at the “max” set value at all times. This is a bit more efficient than simply setting “max” and “min” to the same value and using ondemand because the system will not waste resources scanning for CPU load.
Powersave – Available in some kernels. It will keep the CPU running at the “min” set value at all times.
Userspace– A method for controlling the CPU speed that isn't currently used by SetCPU. For best results, do not use the userspace governor.
Hope those will able to help newbies to SetCPU or No Frills CPU.
Credit to LeeDroid for the infos

I believe Interactive is the most responsive kernel out there. Faster than ondemand. So it'll technically use more power when you are using the phone.
HOWEVER, isn't the best feature of Smartass the ability to cap frequencies when the screen is OFF? This essentially negates the need for using SetCPU to limit frequenceis when screen is off. Yes it's an improved version of interactive, but it should be just as responsive, but with these set limits in. I'm just not sure what the specific rules are. I used it for my Moto Milestone, but I'm unsure of what the rules are here.

window7 said:
Interactive - Instead of sampling the cpu at a specified rate, the governor will scale the cpu frequency up when coming out of idle. When the cpu comes out of idle, a timer is configured to fire within 1-2 ticks. If the cpu is 100% busy from exiting idle to when the timer fires then we assume the cpu is underpowered and ramp to MAX speed.
Smartass - Is an improved version of interactive governor
Ondemand – Available in most kernels, and the default governor in most kernels. When the CPU load reaches a certain point (see “up threshold” in Advanced Settings), ondemand will rapidly scale the CPU up to meet demand, then gradually scale the CPU down when it isn't needed.
Conservative – Available in some kernels. It is similar to the ondemand governor, but will scale the CPU up more gradually to better fit demand. Conservative provides a less responsive experience than ondemand, but can save battery.
Performance – Available in most kernels. It will keep the CPU running at the “max” set value at all times. This is a bit more efficient than simply setting “max” and “min” to the same value and using ondemand because the system will not waste resources scanning for CPU load.
Powersave – Available in some kernels. It will keep the CPU running at the “min” set value at all times.
Userspace – A method for controlling the CPU speed that isn't currently used by SetCPU. For best results, do not use the userspace governor.
Hope those will able to help newbies to SetCPU or No Frills CPU.
And do correct me if I am wrong. Thank you
Click to expand...
Click to collapse
You should credit the source you copy and pasted that from. Or even better post the link.
Sent from my Nexus S using XDA App

dmo580 said:
I believe Interactive is the most responsive kernel out there. Faster than ondemand. So it'll technically use more power when you are using the phone.
HOWEVER, isn't the best feature of Smartass the ability to cap frequencies when the screen is OFF? This essentially negates the need for using SetCPU to limit frequenceis when screen is off. Yes it's an improved version of interactive, but it should be just as responsive, but with these set limits in. I'm just not sure what the specific rules are. I used it for my Moto Milestone, but I'm unsure of what the rules are here.
Click to expand...
Click to collapse
SMARTASS is a rewrite of the INTERACTIVE governor, and it moves the CPU up/down depending on whether the phone is idle/locked/ ... etc.
But some problems that I experience with SMARTASS is that my music playback sometimes stutters(?). So I'm just sticking with INTERACTIVE.

http://setcpu.com
edit:/#7 didnt seem to work right?

matt2053 said:
You should credit the source you copy and pasted that from. Or even better post the link.
Sent from my Nexus S using XDA App
Click to expand...
Click to collapse
Thanks for telling me that. I had changed it
However, I cant post the link as I cannot find the source I find it.
I only know that LeeDroid provided me the details

window7 said:
Thanks for telling me that. I had changed it
However, I cant post the link as I cannot find the source I find it.
I only know that LeeDroid provided me the details
Click to expand...
Click to collapse
I believe it comes from SetCPU's website.
Http://www.setcpu.com
Sent from my Nexus S using XDA App

olorin86 said:
SMARTASS is a rewrite of the INTERACTIVE governor, and it moves the CPU up/down depending on whether the phone is idle/locked/ ... etc.
But some problems that I experience with SMARTASS is that my music playback sometimes stutters(?). So I'm just sticking with INTERACTIVE.
Click to expand...
Click to collapse
Is this because your screen is off? Yeah, smartass has some issues when the screen is off. Main issue is wakeup and whatever else your phone has to do. I know the Netarchy kernel with smartass just got a revision lately to deal with more wake issues.
But in terms of when your screen is on smartass should function exactly like interactive. So that's why I said smartass is essentially interactive, but with those preset rules regarding screen off that you would otherwise have to create using SetCPU.
The reason you might have worse battery might be because during screen off, your phone is struggling to do work at a slower pace (ideally you shouldn't have to do much, but maybe some people's phones have a lot of stuff running in the background), and so the more time you spent with the CPU active ends up eating more power.
This is one of the arguments about Intel Atom vs. Intel i3. Both idle at the same wattage, but the i3 uses massively more power in load. Probably 2x-3x more. However, given that the i3 is like so FAST, it gets say an encoding job done in like 1/4 the time. Your overall power (Watts * time) used is actually less with the faster CPU. This might apply with the Smartass governor. If your CPU is struggling and maxing out at its cap for a long time because its not fast enough when the screen is off, then perhaps this can be an issue. Just a thought. This is why I think the max_freq for sleep should probably be set around 500-600mhz instead of like 200mhz.
But once again I'm not sure what the governor is set at right now. If someone knows, please do tel

Dynamic Overclocking of CPU?

Is this possible on Androids (mostly, this phone?) Before this I had a windows phone and I had a software where I could set a minimum and maximum cpu clock rate.. and then it will overclock dynamically as needed! If you are running a game and it needs more power, it will overclock it while the game is running.. dynamically without you having to change the clock-rate manually.
This will solve battery life problems when overclocking to 1.8ghz. There is no need for it to be draining the battery at 1.8ghz if you are only texting and probably even a 200mhz cpu can do that lol
Hope what I said makes sense.

it's easily done by installing Setcpu,provided that your phone has been rooted and kernel modded to allow overclocking.
Sent from my IDEOS X5

Actually it is already done in stock ROM without root, setCPU provides a way to change the default way of handling the CPU frequencies and governors.

The stock ROM comes right out of the factory can do "cpu stepping" itself, just like the intel "Speedstep" thing. But the factory powersaving profile (or if you wish to call the scaling) might not match our taste. Then there it comes the apps called Setcpu which enhances the cpu stepping behavior according to our preference. The enhancement considers the frequency of sampling cpu load, thresholds modifications, and other considerations.
In Setcpu, we can accord one of the following scaling setting that suits our taste. From my experience, each scaling setting behaves as follows:
ondemand - runs at Min or at Max (eg. 200 MHz or 1200 MHz)
interactive - adjusts frequencies according to the real load
conservative - behaves like "interactive" but incline to power saving
performance - always runs at Max
powersave - always runs at Min
userspace - disables Setcpu scaling and uses stock scaling
smartass - behaves like "interactive" but always runs at Min when LCD screen is off
And again, the phone must be rooted and kernel modded for Setcpu to work properly.

I've been using setcpu with conservative. I like conservative because it gives you a lot of control over the throttling of the cpu. It seems to work pretty good.
Are you getting any benefit from overclocking to 1.8GHz? I also overclocked to 1.8GHz, but I didn't see any noticeable improvement past 1.0-1.2GHz.

I like to use interactive Cox I enjoy the fast sampling rate and hence the high responsiveness.
Sent from my IDEOS X5

Personally I have not tried overclocking yet.. still need to update the kernel, but I'm afraid it might get unstable.
Didn't know that CPU could do it dynamically using different profiles.
Thanks all for the help

Tcm9669 said:
Is this possible on Androids (mostly, this phone?) Before this I had a windows phone and I had a software where I could set a minimum and maximum cpu clock rate.. and then it will overclock dynamically as needed! If you are running a game and it needs more power, it will overclock it while the game is running.. dynamically without you having to change the clock-rate manually.
This will solve battery life problems when overclocking to 1.8ghz. There is no need for it to be draining the battery at 1.8ghz if you are only texting and probably even a 200mhz cpu can do that lol
Hope what I said makes sense.
Click to expand...
Click to collapse
I have setcpu but the x5 is not listed....??

Setcpu Governers ?

okay im sorry if this is one of the most noobish questions, but what exactly are governors , and how do they help ?
Sent from my Mytouch 4G Running Capychimps Sense 3.5 rom (v.4.0)

Cpu governors "govern" how the cpu works. There is a list somewhere that explains what a few of them do. Basically they dictate how your cpu behaves and therefore the performance you see and power consumption you experience. I'll see if I can find that list.
Edit:
- The ondemand governor is the default option used by Android. It scales the CPU speed between the minimum and maximum speeds depending on CPU load. If the system needs more speed, the kernel will rapidly scale up the CPU speed. - The conservative sets the CPU speed in a similar way to the ondemand governor, but scales the CPU up much less rapidly. This would theoretically save battery power, but may lead to less responsiveness. - The userspace governor is currently useless. It's another way for applications to set the CPU speed that SetCPU does not use. - The powersave governor always keeps the CPU at the minimum set frequency. - The performance governor always keeps the CPU at the maximum set frequency.
Except for "userspace," no matter which governor you set, the CPU will always stay within the bounds of the maximum and minimum speeds you set in SetCPU.
smartass governor -is based on the concept of the interactive governor. I have always agreed that in theory the way interactive works -by taking over the idle loop -is very attractive. I have never managed to tweak it so it would behave decently in real life. Smartass is a complete rewrite of the code plus more. I think its a success. Performance is on par with the "old" minmax and I think smartass is a bit more responsive. Battery life is hard to quantify precisely but it does spend much more time at the lower frequencies. Smartass will also cap the max frequency when sleeping to 352Mhz (or if your min frequency is higher than 352 -why?! -it will cap it to your min frequency). Lets take for example the 528/176 kernel, it will sleep at 352/176. No need for sleep profiles any more!
Sent from my HTC Glacier using XDA App

estallings15 said:
Cpu governors "govern" how the cpu works. There is a list somewhere that explains what a few of them do. Basically they dictate how your cpu behaves and therefore the performance you see and power consumption you experience. I'll see if I can find that list.
Edit:
- The ondemand governor is the default option used by Android. It scales the CPU speed between the minimum and maximum speeds depending on CPU load. If the system needs more speed, the kernel will rapidly scale up the CPU speed. - The conservative sets the CPU speed in a similar way to the ondemand governor, but scales the CPU up much less rapidly. This would theoretically save battery power, but may lead to less responsiveness. - The userspace governor is currently useless. It's another way for applications to set the CPU speed that SetCPU does not use. - The powersave governor always keeps the CPU at the minimum set frequency. - The performance governor always keeps the CPU at the maximum set frequency.
Except for "userspace," no matter which governor you set, the CPU will always stay within the bounds of the maximum and minimum speeds you set in SetCPU.
smartass governor -is based on the concept of the interactive governor. I have always agreed that in theory the way interactive works -by taking over the idle loop -is very attractive. I have never managed to tweak it so it would behave decently in real life. Smartass is a complete rewrite of the code plus more. I think its a success. Performance is on par with the "old" minmax and I think smartass is a bit more responsive. Battery life is hard to quantify precisely but it does spend much more time at the lower frequencies. Smartass will also cap the max frequency when sleeping to 352Mhz (or if your min frequency is higher than 352 -why?! -it will cap it to your min frequency). Lets take for example the 528/176 kernel, it will sleep at 352/176. No need for sleep profiles any more!
Sent from my HTC Glacier using XDA App
Click to expand...
Click to collapse
wow ! , thanks i guess im gonna get rid of my sleeping profile and start using smartass, thanks for the help
Sent from my Mytouch 4G Running Capychimps Sense 3.5 rom (v.4.0)

do you happen to know any info about smartassv2 ? , i just checked my list and i have both smartass and smartassv2
Sent from my Mytouch 4G Running Capychimps Sense 3.5 rom (v.4.0)

You're welcome! By the way, I'd appreciate it if you'd hit the Thanks button. I help people all the time and am rarely thanked. I'm starting to get a complex. Ha.
Sent from my HTC Glacier using XDA App

jjbat87 said:
do you happen to know any info about smartassv2 ? , i just checked my list and i have both smartass and smartassv2
Sent from my Mytouch 4G Running Capychimps Sense 3.5 rom (v.4.0)
Click to expand...
Click to collapse
Smart ass v2 is just an update of smart ass v1
Sent from my HTC Glacier using XDA App

Frequency and voltage settings.

Caveats: Every CPU and GPU does not come from the same bin, fabricated on the same date and possibly not manufactured in the same facility. They may each display different physical properties and a wider range of stability than others. What works for me may not work for you.
That being said, I've been stress testing my device with different settings for the past couple weeks trying to find a sweet spot of stability, speed, battery life and heat output.
I'm going to share two setups: my current one that I've stress tested for less than 24hrs but has proved stable through all conditions encountered thus far and my tried and true setup I've used for over a week with no trouble.
Tried and true setup:
Governor - ondemand
Range - 100MHz through 1.6GHz
100MHz - 800mV
200MHz - 825mV
300MHz - 850mV
400MHz - 900mV
500MHz - 900mV
600MHz - 900mV
700MHz - 925mV
800MHz - 950mV
900MHz - 1000mV
1000MHz - 1025mV
1100MHz - 1100mV
1200MHz - 1125mV
1300MHz - 1150mV
1400MHz - 1175mV
1500MHz - 1250mV
1600MHz - 1350mV
Experimental but stable battery saver:
Governor - ondemand
Range - 100MHz through 1.6GHz
100MHz - 775mV
200MHz - 775mV
300MHz - 800mV
400MHz - 800mV
500MHz - 825mV
600MHz - 850mV
700MHz - 875mV
800MHz - 900mV
900MHz - 950mV
1000MHz - 1000mV
1100MHz - 1100mV
1200MHz - 1125mV
1300MHz - 1150mV
1400MHz - 1175mV
1500MHz - 1225mV
1600MHz - 1350mV
GPU setup:
Low power state - 100MHz @ 800mV
High performance state - 400MHz @ 1050mV
Notes:
Custom governors were not stable for me AT ALL! I've found ondemand to be the best one for me and my needs, personally.
100MHz @ 750mV was so, SO close to being stable for me but my phone would routinely reboot in the screen off state. I'm assuming the stress of apps updating in the background, notifications etc was just too much.
As much as I love WidgetLocker (and I really do!), I found it to consume valuable resources, have more pronounced wake up lag and generally contribute to instability.
I use Chainfire3D to run my games etc. at x4 MSAA. As previously stated by Chainfire, the Mali can run at x4 with almost no extra overhead. I imagine that if one doesn't use x4 MSAA, one *might* be able to get away with 400MHz @ the stock 1000mV setting. That being said, I consider an extra 50mV to run at 133MHz faster to be a bargain.
Many games can be run with x16 MSAA with minimal overhead but I've found that for some resource intensive ones, especially multiplayer, they'll slow down unless the GPU is fed at 1200mV but this in turn causes a lot of heat generated so I would advise to avoid turning on x16 MSAA for those that you do find slowing down.
I use and recommend Voltage Control (donate version for extra features!) for setting up clock range and voltage for both the CPU and GPU. It also allows one to set boot settings (at setup or init.d script) and create multiple profiles. I do not recommend init.d script for untested settings as it could cause you issues.
Edit: Not everyone's kernels may support GPU OC/UV or the CPU ranges listed here. I am not responsible if you bork your device.
Here's someone else's method for testing settings:
Here's how I test UV settings.
Turn on everything. Wifi, bluetooth, max brightness, the whole works. This ensures the system is at maximum strain.
Start at maximum CPU clock
Lock the CPU clock (set the minimum and maximum allowed clock to the clock you are currently undervolting)
Lower the voltage by one step
Start a benchmark for a few minutes to see if undervolted clock is stable
If it passes, lower it again go back to step 4
When it freezes up your phone, reboot it and increase the voltage at that clock by two steps and consider it safe
Move to next frequency and go back to step 3.
You reached your lowest clock? Congrats, you should have a well undervolted CPU
Your voltages should always be lowering when your go from the highest clock to the lowest. If it happens that you have to increase the voltage at a lower clock, then also increase the higher clock frequency. I had a few hard locks because of this.
Example.
1000mAh (1GHz) > 900 mAh (900MHz) *< 950 mAh (800MHz) * > 700mAh (600mAh)
The 800MHz voltage is now higher than the 900MHz voltage. Also increase the 900MHz voltage to the same or higher voltage of the lower one.
1000mAh (1GHz) > 950 mAh (900MHz) > 950 mAh (800MHz) > 700mAh (600mAh)
Now that you have it undervolted, you may find that it could hardlock/reboot on you. When it happens do this:
Increase the voltage on all undervolted clocks by one step.
Continue using the device for a day
If the device locks up again, go back to back step 1
If its ok for a day, then every day lower the voltage back to what you had of only one clock (I suggest you go from highest to lowest)
You should be able to find which undervolt caused the reboot fairly quickly and still be able to normally use the phone and keep the rest of the "optimal" undervolts.
Click to expand...
Click to collapse
Sent from my GT-N7000 using xda premium

I don't think UV saves battery. It is display that sucks most of the juice.
You save less than 2% with extreme UV and after a single reboot caused by instability - you lose even more battery.
There's an excellent thread in Nexus S forums - "battery drain benchmarks" (please search it).
I had similar UV settings and my phone never crashed during benchmarks or stress tests.
But it always crashed while installing 100+ apps with app backup restore, restoring backups with TB or MBR, gaming.
After removing UV, it never crashed.
I haven't tested UV with ICS... would see and report if it really saves battery.

Boy124 said:
I don't think UV saves battery. It is display that sucks most of the juice.
You save less than 2% with extreme UV and after a single reboot caused by instability - you lose even more battery.
There's an excellent thread in Nexus S forums - "battery drain benchmarks" (please search it).
I had similar UV settings and my phone never crashed during benchmarks or stress tests.
But it always crashed while installing 100+ apps with app backup restore, restoring backups with TB or MBR, gaming.
After removing UV, it never crashed.
I haven't tested UV with ICS... would see and report if it really saves battery.
Click to expand...
Click to collapse
I'm not sure if you've read everything through carefully or you would have seen that I've covered several of your points.
You also would have seen the method I use for stress testing and would have noted that I aim for four things: speed/performance, stability, power management AND thermal regulation.
While I agree that the display, barring a wonky or misbehaving app, will almost always be the #1 battery drainer - power management will certainly help to conserve battery life.
You also would have seen I mention profiles. There may not be a one size fits all setting for everyone but one can most certainly set up profiles for different scenarios.. Such as TiB backups/restores.
Sent from my GT-N7000 using xda premium

Did you do some benchmarks at the highest speed several times to make sure you are getting extra performance? With this phone I noticed that while the phone wont crash.. .some times performance will drop when running at settings now fully correct.
Sent from my GT-N7000 using XDA

You covered a lot of points but UV is total waste of time.
You get nothing out of it.
http://forum.xda-developers.com/showthread.php?t=1478406
Could you please post your data, how much battery do you save after UV?

Disagree boy, cause with wakelock screen is off, there is significant battery drain, I went to 10 hours life on single charge, due to wakelock.
Normally with deepsleep about 2 days. That's a reduction of 87.5% with screen off. Cpu running @200mhz.
Do the same with undervolting will dramatically increase battery life in that situation. So overal it will be a fraction compared to using the device with screen on, but still significant.
Edit: guess I was wrong here
Sent from my GT-N7000 using Tapatalk 2

baz77 said:
Disagree boy, cause with wakelock screen is off, there is significant battery drain, I went to 10 hours life on single charge, due to wakelock.
Normally with deepsleep about 2 days. That's a reduction of 87.5% with screen off. Cpu running @200mhz.
Do the same with undervolting will dramatically increase battery life in that situation. So overal it will be a fraction compared to using the device with screen on, but still significant.
Sent from my GT-N7000 using Tapatalk 2
Click to expand...
Click to collapse
I actually did the test on Gingerbread.
I set min and max to 200 MHz, activated flight mode and had stock music player running for 3 hours - with undervolt and without undervolt.
To my surprise battery consumption was the same.
May be experts who know about our processor architecture can shed some light here.

Boy124 said:
You covered a lot of points but UV is total waste of time.
You get nothing out of it.
http://forum.xda-developers.com/showthread.php?t=1478406
Could you please post your data, how much battery do you save after UV?
Click to expand...
Click to collapse
I understand where you're coming from, boy.
I don't have data at the moment though I wish I did. But to be honest, it'd be scrambled anyway since whenever I'm not working or mission critical when I need proven stability, I'm testing out all different sorts of settings leading to lots and LOTS of reboots and such!
That being said, anecdotally, I have seen improved battery life for myself but maybe it's a placebo and I could be wrong about it - I have been before in the past. I do feel though that under my normal usage scenarios, I am experiencing less battery drain. It's difficult to quantify though exactly what this is due to since I experiment with kernels, voltages and frequencies.
But if all I'm getting is a 2% boost, man - I'll take it! Like any modder, whether it's min/maxing in a game, working on a car or whatever else, every little bit of a parameter squeezed out is something.
I also feel that you're too caught up on a single aspect, the battery life thing, to the detriment of my overarching holistic goal - efficiency.
Originally I started undervolting and experimenting with frequencies because of thermal output. I had wanted to experiment with x16 MSAA settings, which led to my GPU needing 400MHz and 1200mV which led to lots of heating up which led to me experimenting with everything I could.
Efficiency is what I want. The best performance at the best speeds at the best battery life at the best thermal regulation I can manage.
Now I'm looking at energy efficiency. I'm seeing suggestions that 100MHz may not be as efficient as 200MHz on our Exynos because the tradeoff in frequency power usage isn't worth the longer time spent completing tasks. I'm also seeing that in some situations, a performance best governor targeting max freq may be efficient because less time is spent completing a task and a quicker return to sleep.
I'm just sharing what I'm doing and hopefully others can benefit.
http://forum.xda-developers.com/showthread.php?t=1369817
Sent from my GT-N7000 using xda premium

Wow, thats illogical makes me wonder the math behind it.
Sent from my GT-N7000 using Tapatalk 2

While I appreciate the effort thrown into this, I humbly acknowledge the conclusion is incorrect.
When you lower Voltage slightly, without affecting stability, you pretty much put a toll on the processor for extra "wear and tear" and reduce its lifespan. However, this comes at the reward of reduced current.
So, it should be saving you battery. Underclocking it (safely) is also going to save you battery. And the same thing with different governors, like interactivX compared to regular ondemand, by finishing off processes quicker and reducing the frequency and voltage quicker, and going into Deep Sleep quicker.
I don't have the means to run a Scientific Experiment to prove these claims, nor the time to conduct them. But the majority of "hackers" synonymously agree it saves a noticeable power. These include themers, kernel developers and the casual user. I don't think an educated MAJORITY can be incorrect to the scale of this test's claims.