Database Users

SetCPU...Beneficial for Battery Life?

I've seen a few different posts in some of the kernel threads debating whether SetCPU is helping or hurting battery life. SO, I'm just kind of curious to see what results are on a larger scale? Based on your own experiences, do you have SetCPU installed and if so, does it help or hurt battery life generally? Also, if you do have it installed, do you use profiles? What are the most beneficial settings to use?

1. Not in right section
2. SetCPU not intended for battery life
3. It only adjusts CPU clockspeed
4. This thread is mostly meaningless

5. It's been discussed ad nauseam.

charnsingh_online said:
1. Not in right section
2. SetCPU not intended for battery life
3. It only adjusts CPU clockspeed
4. This thread is mostly meaningless
Click to expand...
Click to collapse
SetCPU is not intended for battery life? Go to the Market and look at the description. If I posted this in the wrong section I apoligize. But, I think you are mistaken with your comment about SetCPU not being intended to increase battery life or increase performance...

THATTON said:
SetCPU is not intended for battery life? Go to the Market and look at the description. If I posted this in the wrong section I apoligize. But, I think you are mistaken with your comment about SetCPU not being intended to increase battery life or increase performance...
Click to expand...
Click to collapse
SetCPU only sets clock speeds and governors already in the kernel. If you just install SetCPU and adjust no settings your battery life will not change. Thus, "does SetCPU help battery life?" is utterly and completely meaningless.
Discussion of different governors and clock speeds has occurred (and is still occurring) ad nauseum and is really more suited for the General forum.

Thread moved as it does not pertain to N1 development.

I see very little gains from setcpu but I use it because I purchased it from the market and why not use it if you bought it right?
This method does not apply to drug addiction LOL
-Charlie

bri3d said:
SetCPU only sets clock speeds and governors already in the kernel. If you just install SetCPU and adjust no settings your battery life will not change. Thus, "does SetCPU help battery life?" is utterly and completely meaningless.
Discussion of different governors and clock speeds has occurred (and is still occurring) ad nauseum and is really more suited for the General forum.
Click to expand...
Click to collapse
lol Why would you download an application, not use it, and expect results?
If you throttle your CPU down you WILL get better battery life. My phone is set to never go over 600mhz and I get bettter life with it than if I turn off setcpu altogether.

charnsingh_online said:
1. Not in right section
2. SetCPU not intended for battery life
3. It only adjusts CPU clockspeed
4. This thread is mostly meaningless
Click to expand...
Click to collapse
You have a lot of knowledge, this is obvious but you're unnecessarily harsh (mean).
It's boring but's it's a legitimate question because people still make inaccurate conclusions about CPU and battery life. Those of us with some knowledge can really help those that are trying to understand.
#2 above is correct. But the question remains, does a forced lower clock speed ceiling have an effect on battery life? It could do, of course it could, but without a baseline and a control environment it's impossible to prove either way. I suspect the OP is simply looking for subjective opinions.
And on this basis I offer:
The CPU only has a material effect on battery drain when it's being utilised.
When the Nexus CPU is not required to work it idles using the lowest power possible
The radio (network) interface is the second most demanding element of on your battery over time (next to the display). Although the CPU peak demand is higher than the radio.
SetCPU does not impact radio battery use.
SetCPU can not have a positive effect on battery usage if it's using more power to run it's clock cycles.
SetCPU can force the processor to use less power (wind down speed).
Slowing the processor means some tasks will take longer to perform.
If those tasks require a high-drain elements (display, radio, WiFi or BT for example) then it's counter-productive (battery wise) to slow them down.
However, because CPU power consumption does not have a liner relationship to clock speed, then some tasks that don't use high-drain elements will consume less power to complete.
So, whilst it's unlikely that your battery life will benefit from the use of SetCPU alone there is a chance that it will.
SetCPU is a fantastic app designed for overclocking, the profiles are niche facilities that may offer battery benefit to a narrow range of users.

Hey djmcnz thanks for the indept look at this app but more importantly thanks for showing respect to those of us who are just learning. We all have to learn information at some point and there are people that forget that at one point some one had to tell them.

Thank you for the clarification on that! Djmcnz-that was exactly what I was looking for in terms of an answer. I really appreciate you taking the time out to explain everything for me and anyone else that may have been curious.

charnsingh_online said:
1. Not in right section
2. SetCPU not intended for battery life
3. It only adjusts CPU clockspeed
4. This thread is mostly meaningless
Click to expand...
Click to collapse
Don't know why you're so pissed off by a thread...
1. Not a very big issue. We have mods here to take care of this.
2. I don't know if SetCPU affects battery life or not but similar thing on a PSP device does increase the battery life. I have tried it on my PSP and setting the clock speed to the lowest acceptable level (depending upon what you're doing) does help maximizing the battery life.
3. You're absolutely right here.
4. Don't know what to say man.. but being a little humble doesn't hurt....

I never meant to be rude. I always get pissed off when people post in wrong sections Seriously. If people post in right section it just frees up moderator time. And about CPU nexus CPU has same voltage for many frequencies like 998,960 have same voltage. Going so down doesn't mostly benefit. So setcpu is only good for overclocking IMO. Display uses most of the power along with radio n CPU is one of those in middle of usage maybe 3rd or 4th. So underclocking will give a big battery boost is just a placebo. Atmost 10 minutes more is what underclocking can provide. N its not worth sacrificing the performance. Go for something underpowered if u want to underclock IMO. So setcpu serves more purpose of power than battery

I use it for the cool widget and standby/idle profile. B-)

you know what?youre allright.i follow your threads and you explain things well for someone like me learning all this ****.i got no time for keyboard commandos.thanks for the explanation.
djmcnz said:
You have a lot of knowledge, this is obvious but you're unnecessarily harsh (mean).
It's boring but's it's a legitimate question because people still make inaccurate conclusions about CPU and battery life. Those of us with some knowledge can really help those that are trying to understand.
#2 above is correct. But the question remains, does a forced lower clock speed ceiling have an effect on battery life? It could do, of course it could, but without a baseline and a control environment it's impossible to prove either way. I suspect the OP is simply looking for subjective opinions.
And on this basis I offer:
The CPU only has a material effect on battery drain when it's being utilised.
When the Nexus CPU is not required to work it idles using the lowest power possible
The radio (network) interface is the second most demanding element of on your battery over time (next to the display). Although the CPU peak demand is higher than the radio.
SetCPU does not impact radio battery use.
SetCPU can not have a positive effect on battery usage if it's using more power to run it's clock cycles.
SetCPU can force the processor to use less power (wind down speed).
Slowing the processor means some tasks will take longer to perform.
If those tasks require a high-drain elements (display, radio, WiFi or BT for example) then it's counter-productive (battery wise) to slow them down.
However, because CPU power consumption does not have a liner relationship to clock speed, then some tasks that don't use high-drain elements will consume less power to complete.
So, whilst it's unlikely that your battery life will benefit from the use of SetCPU alone there is a chance that it will.
SetCPU is a fantastic app designed for overclocking, the profiles are niche facilities that may offer battery benefit to a narrow range of users.
Click to expand...
Click to collapse

djmcnz said:
You have a lot of knowledge, this is obvious but you're unnecessarily harsh (mean).
It's boring but's it's a legitimate question because people still make inaccurate conclusions about CPU and battery life. Those of us with some knowledge can really help those that are trying to understand.
#2 above is correct. But the question remains, does a forced lower clock speed ceiling have an effect on battery life? It could do, of course it could, but without a baseline and a control environment it's impossible to prove either way. I suspect the OP is simply looking for subjective opinions.
And on this basis I offer:
The CPU only has a material effect on battery drain when it's being utilised.
When the Nexus CPU is not required to work it idles using the lowest power possible
The radio (network) interface is the second most demanding element of on your battery over time (next to the display). Although the CPU peak demand is higher than the radio.
SetCPU does not impact radio battery use.
SetCPU can not have a positive effect on battery usage if it's using more power to run it's clock cycles.
SetCPU can force the processor to use less power (wind down speed).
Slowing the processor means some tasks will take longer to perform.
If those tasks require a high-drain elements (display, radio, WiFi or BT for example) then it's counter-productive (battery wise) to slow them down.
However, because CPU power consumption does not have a liner relationship to clock speed, then some tasks that don't use high-drain elements will consume less power to complete.
So, whilst it's unlikely that your battery life will benefit from the use of SetCPU alone there is a chance that it will.
SetCPU is a fantastic app designed for overclocking, the profiles are niche facilities that may offer battery benefit to a narrow range of users.
Click to expand...
Click to collapse
HUH English Please
Kidding

mikey1022 said:
huh english please :d
kidding
Click to expand...
Click to collapse
34567890

Personaly done many tests and the result was:
Test config: WiFi tethering all the way, screen 100% Playing video all the time 2G only
4:10 @ 245Mhz hard
3:30 @ 998Mhz hard
No use actually - using N1 on 245Mhz is impossible - too sluggish.
SetCpu is ussefull:
1) If u have OC kernel to set OC mode for games like Asphalt
2)For letting android vary frequence ondemand instead of 998 all the time
3)For downclocking while in sleep mode (why use full power when u dont use it?)
4)For using Failsafe profile, to prevent battery and hardware damage.
That's all.
No use trying saving battery setting profiles like 100% - 998, 50% - 576, 20% - 499. This is useless.
On UV kernels the same thing +\-30 minutes battery life. And UV kernels themselfs dont give segnificant battery life increase, only lags and unstability ti system.
Dont believe me try yourself - Create yourself some actions fo testing and repeat them 2 time (Min cpu and Max cpu) on any kernel. Results will be very close.

SeriousX said:
3)For downclocking while in sleep mode (why use full power when u dont use it?)
Click to expand...
Click to collapse
The CPU steps down to it's minimum speed by itself. It never uses more juice than it needs to.

As far as i know, it is always at maximum, but maybe im wrong and you are right - then theres even less sence in this app.

Overclocking N1

You can overclock n1 only to 1.190ghz, while desire hd 1.9ghz and the htc desire Z (G2) 2.0ghz. Does N1 has to old cpu?
-------------------------------------
Sent via the XDA Tapatalk App with my Sexy Nexy

Yes. 1st Gen snapdragon
Sent from my Nexus One using XDA App

if you want to OC you N1 go and OC you Desktop is the best choice

Why would you wanna over clock your phone? I have my N1 clocked @ 691 and works really fast with the MIUI rom and battery performance is better than stock. I'm not a fan of custom rom & rooting but I been pretty pleased so far. overclocking the nexus one will drain your battery like crazy plus the 1st Gen of snapdragons weren't as good with graphics as the A4chips and humming birds.

i have mine underclocked too and it works fine. try going a step further and underclocking it to like 422 when it's sleeping/standby. it'll help your battery

josemedina1983 said:
Why would you wanna over clock your phone? I have my N1 clocked @ 691 and works really fast with the MIUI rom and battery performance is better than stock. I'm not a fan of custom rom & rooting but I been pretty pleased so far. overclocking the nexus one will drain your battery like crazy plus the 1st Gen of snapdragons weren't as good with graphics as the A4chips and humming birds.
Click to expand...
Click to collapse
The connection between clockspeed and power consumption is not as strong as you think. But without a doubt it has an influence. Much more important is the voltage. If you "undervolt" the Nexus One CPU you can even get better battery live with higher clockspeed.
And if you use a tool to change the clockspeed depending on the situation (display on/off, battery % left, workload) and undervolt the cpu you can safe A LOT of juice.
With Wildmonks kernel, MIUI and SetCPU I get a much better lifetime than ever before even though my Nexus runs at 1152MHz.

Actually, the frequency makes a BIG difference in power consumption. Think of it this way - each clock causes changes propagating in transistors, which are the actual power draw. More clocks = more changes = more power drawn. As easy as that.
So, having 10% higher frequency and 10% lower voltage compensates each other.
Nexus has examples that overclock to 1.5GHz when overvolted, like Desire Z and Desire HD (both of those have to be overvolted to go up stable from 1.2GHz). Most of Nexus Ones fail when overclocking and don't reach higher than 1.2GHz, but it might be not because of the CPU, but because of other devices on system board.

Generally, it is only when you change the voltage (which is required to stabilize the higher frequency) that you see noticeable differences in battery life.
Jack_R1 said:
Actually, the frequency makes a BIG difference in power consumption. Think of it this way - each clock causes changes propagating in transistors, which are the actual power draw. More clocks = more changes = more power drawn. As easy as that.
So, having 10% higher frequency and 10% lower voltage compensates each other.
Nexus has examples that overclock to 1.5GHz when overvolted, like Desire Z and Desire HD (both of those have to be overvolted to go up stable from 1.2GHz). Most of Nexus Ones fail when overclocking and don't reach higher than 1.2GHz, but it might be not because of the CPU, but because of other devices on system board.
Click to expand...
Click to collapse

willverduzco said:
Generally, it is only when you change the voltage (which is required to stabilize the higher frequency) that you see noticeable differences in battery life.
Click to expand...
Click to collapse
Ok, some additions required.
Leakage is also dependent on power, and the dependency graph isn't linear - and starts breaking upwards at some point, usually being a tad above the max designed voltage.
Going down in voltage makes leakage change approximately linear, and doesn't affect nearly as much as going up.
Overclocking will draw power just as I noted above - exactly with the same percentage difference - only when the clock is reaching the overclocked area, which happens only when you're playing games or doing CPU-intensive tasks.
Undervolting will affect leakage, which happens 100% of the time.
So yes, when running in dynamically scaled environment, undervolting has more effect than overclocking. On desktop PC, running the same clock frequency constantly, the effect is the same.

Very True. And I wasn't saying that overclocking, while at the same voltage, didn't draw ANY more power... I just am trying to say that (for example in this graph) overclocking only has a small effect on power draw until you actually change the voltage. In that same example, going from 3.4 to 3.8 GHz only adds about 6% current draw while at the same vCore, while going up a similar amount in clock speed.
I'd even wager to say that if you're slightly under-volted and as heavily overclocked as you can go at that given voltage, you'll save some trivial amount of power versus stock because of the fact that voltage affects power draw significantly more than clock speed. I would also wager that if you are at an overclocked speed and are at stock voltage, the amount of current and power draw will be almost indistinguishable to the end user, since things like display will almost always use much more power if the display is on for any appreciable amount of time.
Jack_R1 said:
Ok, some additions required.
Leakage is also dependent on power, and the dependency graph isn't linear - and starts breaking upwards at some point, usually being a tad above the max designed voltage.
Going down in voltage makes leakage change approximately linear, and doesn't affect nearly as much as going up.
Overclocking will draw power just as I noted above - exactly with the same percentage difference - only when the clock is reaching the overclocked area, which happens only when you're playing games or doing CPU-intensive tasks.
Undervolting will affect leakage, which happens 100% of the time.
So yes, when running in dynamically scaled environment, undervolting has more effect than overclocking. On desktop PC, running the same clock frequency constantly, the effect is the same.
Click to expand...
Click to collapse

Jack_R1 said:
Actually, the frequency makes a BIG difference in power consumption. Think of it this way - each clock causes changes propagating in transistors, which are the actual power draw. More clocks = more changes = more power drawn. As easy as that.
So, having 10% higher frequency and 10% lower voltage compensates each other
Click to expand...
Click to collapse
I wouldn't call 10% more peak power consumption big if you take in account that the cpu is only running at the max clock speed a very small amount of time. 90% of the time the device is sleeping anyway and even if it's not you barely need the max clock speed. But if you do you will recognize the difference.
On the other side the reduced voltaged can safe you power all the time.
willverduzco said:
I'd even wager to say that if you're slightly under-volted and as heavily overclocked as you can go at that given voltage, you'll save some trivial amount of power versus stock because of the fact that voltage affects power draw significantly more than clock speed. I would also wager that if you are at an overclocked speed and are at stock voltage, the amount of current and power draw will be almost indistinguishable to the end user, since things like display will almost always use much more power if the display is on for any appreciable amount of time.
Click to expand...
Click to collapse
That's exactly what I experienced.

Pommes_Schranke said:
I wouldn't call 10% more peak power consumption big if you take in account that the cpu is only running at the max clock speed a very small amount of time. 90% of the time the device is sleeping anyway and even if it's not you barely need the max clock speed. But if you do you will recognize the difference.
On the other side the reduced voltaged can safe you power all the time.
Click to expand...
Click to collapse
Yes, you're right, and that's why I corrected myself in my second post. I totally forgot about the frequency scaling.

Off topic, but this is why I love XDA. Rational debate over a subject by intelligent people, where there usually isn't flaming. Thanks added to the two of your posts.

Safe Undervolting

Hi, I've lowered all the figures by 100mv, but I notice no difference in my battery life. Is it safe to lower the figures more?
Sent from my Nexus 7 using xda premium

Define what you mean by "safe".
Higher core voltages are used at higher clock rates in order to preserve logic margin in the face of more power supply noise and tighter timing. So, undervolting by definition means you are giving up margin.
Whether your SoC happens to have a worst-case timing path which is "fast" or "slow" relative to other devices cannot be deduced by anyone here on XDA or even at Nvidia. About the best you can hope for is some "feel good reports" about what others do with undervolting and completely undisciplined "testing". But even that provides no information about your individual chip.
good luck

Thanks for your reply. By "safe" I mean without damaging my device. And yes, it would be interesting to hear some undervolting stories. I'd quite like to extend my battery life. Cheers!
Sent from my Nexus 7 using xda premium

I've been using my nexus 7 runing a AOKP pub build by mrRobinson and franco's r47 kernel. i've undervolted the entire board by 100mv it is runing pretty solid without any noticible lag or instability. Yet every chip is diferent and can or cannot this margin you'll have to test for your self. Just leave the option 'set on boot' unticked untill your sure that your device is capable of using those kind of voltages. And don't use very large steps. Just like overclocking in fact! Try it - it runs - get a little bit lower. If it crashes you shoukd be able to reboot the nexus and the settings you've changed reseted
Sent from my Nexus 7 using Tapatalk 2

antmasi said:
I've been using my nexus 7 runing a AOKP pub build by mrRobinson and franco's r47 kernel. i've undervolted the entire board by 100mv it is runing pretty solid without any noticible lag or instability. Yet every chip is diferent and can or cannot this margin you'll have to test for your self. Just leave the option 'set on boot' unticked untill your sure that your device is capable of using those kind of voltages. And don't use very large steps. Just like overclocking in fact! Try it - it runs - get a little bit lower. If it crashes you shoukd be able to reboot the nexus and the settings you've changed reseted
Sent from my Nexus 7 using Tapatalk 2
Click to expand...
Click to collapse
Thanks! Have you noticed any improvement in battery life?
Sent from my Nexus 7 using xda premium

Been testing for 2 days now! Battery seems a little bit better yet maybe placebo! But i feel it does not get has hot as runing stock voltages! It was the main reason i did the undervolt!
Sent from my Nexus 7 using Tapatalk 2

antmasi said:
Been testing for 2 days now! Battery seems a little bit better yet maybe placebo! But i feel it does not get has hot as runing stock voltages! It was the main reason i did the undervolt!
Sent from my Nexus 7 using Tapatalk 2
Click to expand...
Click to collapse
Placebo
Sent from my Nexus 7 using XDA Premium HD app

For what i've read that's the truth but if the frequencies set for the soc are getting less power shoudn't it draw less power from battery? 100mv its almost 10% of the stock voltages!
Sent from my Nexus 7 using Tapatalk 2

antmasi said:
For what i've read that's the truth but if the frequencies set for the soc are getting less power shoudn't it draw less power from battery? 100mv its almost 10% of the stock voltages!
Click to expand...
Click to collapse
For the SoC, whether it's leakage (V^2/R) or dynamic power dissipation (f*C*V^2), yes, you might expect let's say (0.9^2) = 81% battery use at the same operating frequencies.
OTOH, if 75% of the power drain normally is used by the LCD backlight (for instance in a "reading web pages" use case), then reducing the supply voltage will get you only 1/4 of that 20% savings - about a 5% improvement - because the power is being dissipated elsewhere. (Display, DRAM, 3G radio, WiFi radio, etc)

I agree with what you've pointed! In fact i've just taken out the undervolt because Inhad a reboot under heavy multitasking (torrent download, xbmc opened, chrome also downloading a smal file) i'm not certain that it was caused by the uv but it's possible. The Nexus started to lag and then froze completly! From time to tima i've the need of heavy multitask and it wasn't up to the task!
Sent from my Nexus 7 using Tapatalk 2

Kernels, number of cores, deep sleep

Hi, I am trying a lot of different modded kernels for my brand new Nexus 4. However, due probably to the fact that I am not competent enough to look at the code of the patches they apply I have difficulties in understanding some technical differences between them. In particular:
1) It seems to me that some kernels use only some of the four cores of the processor. Namely, I have noticed that Franco kernel uses only two cores, while Trinity hotlplugs all of them when needed. Is there some evidence in support of one or the other choice? How the other kernels behave under this point of view?
2) There seems to be a huge difference for what concerns the time spent by my device in so-called "deep sleep". E.g. when I use Trinity the device, looking at the stats, seems to stay in deep sleep for almost all the time in which the screen is off, while in Franco this happens for a shorter time and in Matrix kernel almost never. However, this seems to have a negligible impact on battery life: e.g. with Matrix the processor is most of the time at the lowest frequency instead of deep sleep, but I get excellent battery life nonetheless. So I wonder if I should care about deep sleep at all.
Thanks in advance for any clarification.

patroclo7 said:
Hi, I am trying a lot of different modded kernels for my brand new Nexus 4. However, due probably to the fact that I am not competent enough to look at the code of the patches they apply I have difficulties in understanding some technical differences between them. In particular:
1) It seems to me that some kernels use only some of the four cores of the processor. Namely, I have noticed that Franco kernel uses only two cores, while Trinity hotlplugs all of them when needed. Is there some evidence in support of one or the other choice? How the other kernels behave under this point of view?
2) There seems to be a huge difference for what concerns the time spent by my device in so-called "deep sleep". E.g. when I use Trinity the device, looking at the stats, seems to stay in deep sleep for almost all the time in which the screen is off, while in Franco this happens for a shorter time and in Matrix kernel almost never. However, this seems to have a negligible impact on battery life: e.g. with Matrix the processor is most of the time at the lowest frequency instead of deep sleep, but I get excellent battery life nonetheless. So I wonder if I should care about deep sleep at all.
Thanks in advance for any clarification.
Click to expand...
Click to collapse
1) Franco kernel doesn't only use two cores. You can select 2 or 4 cores and hot plug via his app or script.
2) yes you should care about deep sleep, of anything it will provide batter idle battery life by a small margin
Sent from my Nexus 4 using Tapatalk 2

Vanhoud said:
1) Franco kernel doesn't only use two cores. You can select 2 or 4 cores and hot plug via his app or script.
2) yes you should care about deep sleep, of anything it will provide batter idle battery life by a small margin
Click to expand...
Click to collapse
Thanks for your reply. About 1), I meant to say that it uses two cores by default. What is the advantage in being restrained to use only half the power at disposal instead of using time by time what is needed?Am I right in thinking that in this way always the same two cores are used while the two others stay virgin? Or is there some kind of rotation, so that in each moment at most two are used but not always the same two?

patroclo7 said:
Thanks for your reply. About 1), I meant to say that it uses two cores by default. What is the advantage in being restrained to use only half the power at disposal instead of using time by time what is needed?Am I right in thinking that in this way always the same two cores are used while the two others stay virgin? Or is there some kind of rotation, so that in each moment at most two are used but not always the same two?
Click to expand...
Click to collapse
In franco kernel, hotplug is usually the default.
In hotplug mode, the default is set so that when there is minimal activity only the first two cores are active. This is to save battery and reduce heat i am assuming. When the cpu load increases, the 3rd and 4th core are activated to assist.this is also helpful in reducing the amount of times each cpu core turns on/off (i.e. hot plugging) because this action has an impact on performance. The first two cores stay on constantly until the load increased, then cpu 3 and 4 are activated . Once the cpu load falls down, 3/4 are turned off
when in dual core mode, yes only two cores are used period. You really cant even tell a difference when only using two cres, except for heavy gaming.
Quad core mode makes all the cores active all the time, best for gaming

Quad isn't always needed...Many android apps aren't even multi threaded so having every core on is foolish.
Sent from my Nexus 4 using xda app-developers app

by default, the device hotplugs. itll use the number of cores it needs when it needs them. you can set yourself how many cores the device can use as a minimum(with root). i use trinity, and i usually set that all four cores are used all the time. i see slightly better battery life that way.
you will see better idle battery drain the better the device deep sleeps. btw battery life is all dependent on how you use the device. if youre one of those people that barely touch your device, and let it lay around for log periods if time, then a deep sleeping device would be beneficial to you. if youre a power user thats constantly using the device nonstop then it wouldnt make any difference to you. most of us fall in between these two types of users

(Q) Can undervolting too much cause more heat?

As the title says, can undervolting too much cause more heat than just a little undervolting? I'm at - 175mv on every frequency and my phone usually gets quite warm even with normal use, music, etc. Maybe about 45°c...thanks! (I don't over clock, or play games)
Sent from my Carbon powered AKFAUX fueled monster NEXUS 4

45C is normal for this device, when its in the 60+ range then it's time for concern. Undervolting is in millivolts not volts so the change in temperature and battery is not very significant.

However back in the Nexus S days some genius dev measured that undervolting too much forced the cpu to redo some calculations, which made it use more power than the uv would save. Whether that's what's happening to you I wouldn't know, but it could be.
In the process the dev also found out that undervolting didn't save any noticeable amount of battery. Of course, whether the same goes for our cpu isn't known for sure, but I would guess the same concept applies.
Sent from my Nexus 4 using Tapatalk 4 Beta

One of the main reasons to UV is to prevent overheating/reduce overall temperature increase. So in other words, no. Just don't UV too much (-125 mV MAX)!
Sent by carrier pigeon

your cpu temp is completely normal, a bulls-eye type of normal too. and yes, undervolting too much can cause the phone to work harder, which in turn can cause more heat.