Power to the Smartphone

There's been a fair amount of interest in an offhand comment I made recently about saving a lot of standby time by making an LED blink less frequently.  So let's spend some time talking about what burns power on your Smartphone.  Hopefully this will give you an understanding of where that battery power went and maybe help you figure out how to squeeze a little more life out of your devices.

Milli who?

If you look at your phone's battery, you'll probably find a label that says something like "1000 mAh".   Think of this number as the size of your battery's gas tank.  A bigger number means that the battery can hold more power.  A lower number means that it holds less.  The "mAh" is an abbreviation for "milliamp hours." 

I won't go into the differences between amps, volts, and watts here (unless you folks really want me to).  You could correctly say that mAh doesn't represent power, but, for various reasons, it's reasonable to think of it as power.  And, when you do, the math becomes really easy.

If your phone had a 1000 mAh battery and only burned 1 mA, then it would last for 1000 hours.  If it burned 2 mA, it would last for 500 hours.  If it burned 1000 mA, it would last for one hour.  Milliamp hours divided by milliamps equals hours.   It's that simple.

Who you callin' "typical"?

Every phone is different.  Any numbers I give for a "typical" phone might be wildly different on your phone.  Hardware changes all the time, and so do these numbers.  So don't hold me to them.  But, a "typical" WM Smartphone has around a 1000 mAh battery.  And, when in standby (backlight off, screen off, cpu doing very little, phone only listening for incoming calls), the typical device burns around 5 mA.  Doing the math, you'll see that the typical device should run for 200 hours on standby.

200 hours???  I'd kill for 200 hours!

Are your batteries not lasting that long?  Here are the biggest power users on your phone.  Maybe one of them is hurting your standby time.

Backlight: The backlight eats power like a whale eats plankton.  "Typical" backlights burn 45 mA.  Your phone with the backlight on is probably burning ten times as much power as it does during standby.  In other words, every minute you spend with the backlight on is ten minutes less standby time you'll have.  If you keep your backlight on continuosly, your 200 hour phone will become a 20 hour phone.

CPU: This one is really complicated, but the simple story is that faster CPUs burn more power than slower ones, and CPUs burn a lot less power when they're not being used than when they're busy. 

The first factor is why you still see some devices with a 200 MHz CPU even though there are much faster ones available today.  All else being equal, a 200 MHz CPU uses half as much power as a 400 MHz one.

The bigger deal, though, is CPU usage.  When the CPU is idle, it goes into a lower power mode that burns much less power than when it's active.  When you're not doing anything with it (and not syncing), the Smartphone CPU is generally idle 99.9% of the time.  If you were to load an app that kept the CPU busy just 1% of the time, that app would be using ten times more CPU than normal.  Depending on the CPU and other factors, this could turn your 200 hour phone into a 20 hour one.

This is why I did those "Power to the Developers" blog entries and begged people to not use the CPU when the phone was idle.  It's fine to do animations and things when the user is interacting with the device (especially when the backlight is on).  But the app really should stop when the screen turns off.  If the user can't see the screen, why burn the power trying to show him something? 

As a user, you can pay attention to the apps you add to your phone.  Did your battery life plunge after adding something?  Do you have a great new stock ticker on your home screen that's still scrolling text at 3 AM when you're asleep and the phone is on your bedside table?  Do you have ActiveSync set to sync every 5 minutes, but you only look at the device every 10?  Maybe in off peak times you should set it to something very long, or even manual.  Or, you might ask, "Is always up to date (AUTD) email better or worse for batteries than 5 minute sync?"  That depends.  Do you typically get email more often then every 5 minutes or less often than every 5 minutes?  Any time spent syncing is CPU power being spent.  The less you do it, the better your battery life will be.

Cell Radio:  The cell radio burns a lot more power when actively sending data than when idle.  This factors into those active sync decisions above.  It also means that an app which sends or receives data will burn much more power than one that doesn't. 

Even when idle, the cell radio will burn around 2 mA of the device's 5 mA standby current.  If you need to stretch your battery life out, you can get quite a bit of mileage out of putting it into flight mode.  Of course you can't send or receive phone calls then, but maybe you don't want phone calls at 3 AM when you're asleep.  You might benefit from putting the phone into flight mode before going to bed.  (Personally, I go one step further and turn mine off at night.  Other people don't want to be disconnected and leave theirs on.  Do what works for you.)

Signal strength is a huge factor as well.  The closer a cell radio is to a cell tower, the less power it takes for the two of them to communicate.  When you move into a poor signal strength area, the radio needs to turn up the power to stay connected.  I frequently hear people at odds over whether a given phone has good battery life.  One will say the phone is great, and the other will say that it's terrible.  More often than not, the one who thinks battery life is great lives in an area with great signal strength and the one who thinks it's bad lives in an area with poor signal strength.  Personally, the only place in my home where I can get any reception at all is standing next to the window in my son's room upstairs.  I might as well turn on flight mode as soon as I get home…

Bluetooth:  BTh uses another radio.  Having it on at all burns more power than having it off.  Having it on and sending data over it burns more power than having it off and not sending data over it.  If you're using BTh, you need to have it on.  If you're not, you should turn it off to conserve power.

Vibration:  Here's a prime example of why the battery capacity is measured in milliamp hours.  Time is just as important as current draw in figuring out who is burning all your power.  If the backlight is on for a second, it'll burn 1/60th as much power as if it's on for a minute.  Case in point, the vibration motor is probably the single biggest instantaneous power draw in the device.  It's got to physically move the device around, and that takes a lot of power.  But the vibration motor is never on for very long.  It shakes the phone for a few seconds every so often, but does nothing most of the time.  So the vibration motor isn't a substantial drain on the battery.  That said, if something were to happen to cause your vibration motor to turn on and not turn off, it would suck your batteries dry in no time.

LEDs:  These are the little blinking lights on the outside of your phone.  Typically you've got a green one that blinks whenever the cell radio is on.  You usually have a blue one that blinks whenever Bluetooth is connected.  Sometimes you have a yellow one for other sorts of notifications, like upcoming calendar appointments or unread email. 

LEDs annoy me.  In my opinion, they burn more power than they're worth.  But there's usually no way to turn them off.  (They're completely OEM controlled.)  Everyone assumes that users are best off with a LED telling them that the radio is connected.  I think we should be weighing that against how much power the LEDs draw.

A typical LED, while on, burns 5 mA of current.  Remember that the entire phone in standby is burning 5 mA.  Turn an LED on and leave it on, and you'll cut your battery life in half.  Fortunately, we don't usually leave LEDs on.  We blink them.  Typically, the blink rate for a LED is something like 0.2 sec on, 1.8 sec off.  Since that results in the LED only being on 10% of the time, overall it just burns 10% of its 5 mA, or 0.5mA. 

When I say that a typical phone on standby uses 5 mA, it's actually using 4.5 mA, with the other 0.5 going toward blinking that green LED.  I recently worked on an extremely power efficient device whose standby current was 2.5 mA with the LED off.  It just killed me that we had to blink the LED and bring the standby current up to 3.0 mA.  So I convinced people to let me change the blink from the standard 0.2 seconds out of every 2 to 0.05 seconds out of every 2.  Now, rather than being on 10% of the time, the LED was on 2.5% of the time.  And the power went from 0.5 mA to 0.125 mA.  On a 1000 mAh battery, that change alone would buy almost 50 hours of standby time. 

I'm not ready to start carrying "Down with LEDs" placards.  But if you want to put pressure on your Mobile Operators to tell their OEMs to get rid of the darned things, you'll get no objection from me.

What about the PocketPC?

That's the majority of the story on Smartphones.  There are definitely similarities in the PocketPC, but there are some differences too.  I'm personally more interested in Smartphones, but if you folks are interested, I could do a follow up entry on PocketPCs too.  I'll be out next week, but when I get back, I'll gauge interest and write one if necessary.

In the mean time, I hope this has helped you understand where your battery life is going.

Mike Calligaro

Comments (59)

  1. frank says:

    i’ve said death to leds on handheld devices for 10 years.

    on my windows mobile i’m lucky to get 50 hours of standby time TOTAL, much less extra.

    i’m always amazed by how much power is wasted by people who think unnecessary notifications are valuable.

  2. Solnyshok says:

    Well, I will speak for Pocket PC PDAs… ipaq wifi and bt led actually do not blink. it is are ON continuously when bt or wifi is on. even before you explained the power drain, I thought that it will burn through my eyes in no time.

    Getting rid of ipaq BT/WiFi led is very easy


  3. Andy says:

    Speaking of the backlight sucking up copious amounts of battery power – it would be *GREAT* if the MS Mobiles/Smartphone team would develop and release a "backlight powertoy" that would allow us users who don’t want or need the backlight to turn it off completely.  I am perfectly able to see my homescreen and menus in normal light without the backlight, and I honestly would prefer if it never came on.  Can you guys please make something like this?  Powertoys for WM5 in general are non-existent at this point so maybe if you guys have something in the works you could add this incredibly helpful little option for those of us dying for longer battery life!  Thanks!

  4. Corentin says:

    Would love to read a similar article about pocket PC!!!

  5. LazerFX says:

    I’d love to see a PocketPC article.  I’m rather interested in the differences, as a potential .NET developer for PocketPC devices, and as a consumer (With a Dell X51v).  I was shocked that a LED consumed that much power – I’ve always thought of them as low-drain devices.

    If you could tell us how to get rid of flashing LED’s on PocketPC’s, that’d be great 😉  If not, details of the power drains would be much appreciated.

    Incidentally, how much does memory drain?

  6. MobileRead says:

    Everyone seems to have their own ideas about what drains the battery power in their smartphone. We’ve previously seen an application that helps you figure out what drains the battery. And we’ve seen a very technical explanation  of it all.


  7. Levo says:

    GREAT article.  Please continue to write on power. Stuff I’d like to see:

    * Why do color screens take more power?  How much more?

    * Do different screen technologies take different amounts of power?

  8. Jason says:


  9. Celularis says:

    En el blog del equipo de Windows Mobile hay una excelente nota sobre como se usan las baterías en los Smartphones. En la nota se explica el concepto de miliamperes/hora con la que se están midiendo las baterías hoy en…

  10. Riki says:

    >4.5 mA

    @ 3.4V -> 15mW

    very interesting – i was working on a device (commercial – no radios, Freescale CPU) that did 1.0mA @ 12V -> 12mW in idle, interesting to see we weren’t too far off cell phones.

  11. Marc says:

    I’ve always hoped to see "power Profiles" for mobile devices, similar to laptops.

    When the battery reaches 50% then cpu = x, backlight brightness = x, etc. When battery reaches 25% then… You get the idea.

    Basically you could have profiles for: Best performance, Audio & Video and Power Saving etc.

    Either that or OEM’s need to supply higher capacity batteries 😉

  12. Matt Lacey says:

    Can I make another request for a PocketPC equivilent article also, please.

  13. Very interesting, the same about Pocket PC would be great to compare.

    I must admit the Qtek 8310 (HTC QVGA Smartphone) has an impressive battery life, while the 9000 (HTC VGA Pocket PC) is a bit dissapointing.

  14. Matthew says:

    LEDs shouldn’t be banished, but they should be configurable. Blink to say I’m on the network? No, that’s normal. Blink to say I have a message waiting? Yes, that’s useful, and consumes far less power than waking the whole phone, turning on the screen and backlight, and letting all the idle processes wake up and start running just to see if maybe I have a message waiting.

    Care to speculate why the Motorola Q gets 20 hours battery life in standby if I’m really lucky? My PocketPCs can go for several days in standby without the obnoxious Q default of turning off the screen several second after the last button push, making it impossible to read a message without constant scrolling about to keep it awake.

  15. Alberto Silva says:

    Some people believe that using ‘light’ colors as background instead of dark colors can increase power consumption. Is there any true on this?

  16. CpDbob says:

    And what about Ti OMAP 850 in HTC WizardProphet.

    Overall, I can’t understand why power consumption is so different usind the same hardware. Take for example Asus A632636N639

  17. Hi,

    I am sorry to add a completely off the topic comment – but I just wanted to let everyone on the Windows Mobile team know that we just launched the Mobile Monday Chicago chapter and we’ve had one event already with another planned for the 21st of this month (Neal Ford will be talking this month on Mobile Web 2.0). I was hoping to get one of you on the WM team to come out here and speak at one of the events… We’re also having a demo of the latest Neuros device on the 21st.

    Any interest?

    Please email me @ kiran.bellubbi [aT] gmaiL . com

  18. DougT says:

    I’m wondering if the author – by use of the word Smartphone – was referring to both varieties?  Most of what I see here, I would expect to be the same in both platforms – no?  The LED on my K-Jam probably takes up the same power as the LED on another phone.  Same for BT radio, backlight on a 320×240 screen (ok, the digitizer takes up a bit extra), etc.   Seems to me that most of this is probably on target for PPC’s….



  19. The One Eyed Man says:

    Not sure if WM devices still do this or not, but under CE 3.0 and Pocket PC (CE 4.0) they DID:  WHY DO YOU HAVE TO HAVE A BLINKING RED LED TO TELL ME THAT MY BATTERY IS ABOUT TO DIE???  JUST LET IT DIE…. DON’T CONSUME IT FASTER!!!!

  20. Rob F says:


    Thanks for another interesting and informative article.  A follow-up on PPC power consumption would be most appreciated.  

    I was surprised at the power consumption of LEDs.  In light of this, I would prefer to have information such as network connection and Bluetooth radio state indicated by Today icons, rather than LEDs, at least as a user-configurable option.

    You mentioned that the LEDs are "completely OEM-controlled".  Does this mean that their function is set in the ROM image, and could not be changed by a software "powertoy", as suggested above?

    Also, I am not surprised at the high power consumption of the Bluetooth radio and agree with you that disabling it some of the time would yield better battery life.  I think the main reason I don’t do this is because of the UI design.  I typically carry my SMT 5600 in a belt holster with the keypad locked, to prevent unintended audio recordings (easy to do with this model) and speed-dial calls.  To turn off the BT radio, I have to do the following: Start (to wake up), Unlock, # (to confirm unlock), Start, Settings, Bluetooth, Bluetooth, Left (to select "Off"), Done, and press-and-hold End (to go back to locked state); ten keypresses.  Then when I am walking to my car, where I may want to use my BT headset, another ten keypresses to re-enable it (trying not to walk into a pole in the garage as I walk while staring at my phone 🙂 ).  

    This isn’t directed at you, if UI is not your area, but it is something to consider for the next iteration of the OS.  A little thought here on surfacing valuable features could help you to help all of us more easily maximize our battery life.



  21. I too have always thought that the ‘blink’ rate on the phone was way too much.  I would really love to have this under the user control, much like the "Accessibility" options under Settings on a SmartPhone.

    For me, on my SmartPhone I have a very short backlight time out, followed by a slightly longer display timeout.

    Do we know how much power the "Light Sensor" takes?  On devices such as the SPV C550???  I am sure that this too is going to consume about the same power as the LED!

  22. dahua007 says:

    Yes, the backlight eats power like a whale eats plankton :). My method is to have backlight not to openned when environment light is power enough. There is a software named Nine Way Keypad that can replace left softkey with a numeric key. For some device, if you unlock keypad or phone by this numeric key but not real left softkey, the backlight would not be openned unless you press real left softkey.

  23. MikeCal says:

    Thanks for all the replies folks.  Sorry I haven’t been responding.  I was on vacation last week.

    I’ll definitely do an equivalent PPC entry.  Hopefully I’ll be able to get to it this week.  

    Here are a few answers to some of the questions asked here:

    One of the reasons you see so few PowerToys for WM is that most of the stuff you’d want PowerToys for are under OEM control and not standard.  For instance, the backlight driver is written by OEMs and there is no interface into the driver to say "turn the backlight off."  Some OEMs have ways to do this, but that way is going to be specific to that OEM.  On many PocketPCs there’s a control panel setting that lets you disable the backlight.  On some if you press and hold the power button, it will turn off the backlight and keep it off.  But there’s no standard method (that I’m aware of, at least) that works on all devices.

    Similarly, there’s no generic way to get rid of the blinking LEDs.  Again, OEMs own the drivers that control those LEDs.  Sometimes it’s something even deeper than the drivers (we’ve had devices where the Cell radio hardware blinks the green LED).  

    I’m currently working on a registry controlled LED service that would be designed to let the OEMs not need to worry about blinking LEDs.  It so happens, because I’m a typically devious sort of guy, that it would allow me (and other power users) to disable blinking these darned things.  However, it’s unlikely that we’ll force our OEMs to use this service.  We’ll make it available to them and encourage them to use it, but we generally let them do what they want when it comes to hardware and drivers.

    Regarding the Q, I’ve seen pretty conflicting reports on battery life.  For some people, it’s great.  For others, it’s not.  My best speculation is that the people with poor battery life live in poor signal strength areas.

    Regarding older versions that blinked a red LED when the battery got low.  Hey, we used to do even better than that!  We used to wake the device up, turn the backlight on, play a sound, and pop up a notification that the battery was critically low.  Repeatedly.  We’re a bit smarter than that now.

    Regarding how hard it is to turn on and off BTh: yeah, that annoys me too.  I wrote a quick little program to make it easier.  Check out my ToggleBTh (linked from here) http://blogs.msdn.com/windowsmobile/archive/2006/01/11/511660.aspx

    Re: Light Sensors.  Assuming that’s done right, it wouldn’t take much power at all.  For one thing, it takes a lot less power to passively sense light than to actively generate it.  For another, the sensor would probably only be on while the backlight is on, and the backlight’s power drain would dwarf the sensor’s.  Finally, the code most likely would only check the sensor every once in a while, not continuously.  One of the guys who works for me wrote the code for a light sensor recently and we had it only sense once when the backlight turns on and then not again until the backlight had been left on for over a minute.  The effect would be miniscule.

    Regarding the various LCD (not backlight) questions, I’m pretty sure I know the answers, but I want to cross check them with some of my hardware friends to be sure.  I’ll get back to you on them.


  24. There’s been a lot of concern over MSFP and how it affects battery life on Windows Mobile devices. …

  25. In my recent “Power to the Smartphone” entry, I talked about the biggest drains on Smartphone batteries. …

  26. Nino.Mobile says:

    Software / Hardware SOTI Pocket Controller Professional v5.07 is out Pocket PC Thoughts is reporting

  27. Here are a set of interesting URL’s from my news feeds from August and September 2006 (in no particular order!): Listen to Radio on Your Pocket PC (Phone Edition) with Mundu Radio Screencasts on the subject of Windows Mobile development…

  28. Ron says:

    I have my phone check for email every 15 minutes.  Does that process use a lot of power?

  29. Anon says:

    Very nice article. Could you please cite the source from where you are getting these energy measurements? Did you actually run the experiments or these are published as part of the device manual? For instance you said that a device in running at 1% CPU throughout will roughly have a 10 times shorter life as that which is in complete standby. Intuitively, I see how your numbers make sense but if you could quote a source it would be more legitimate.

  30. MikeCal says:

    We have these cool power monitor boxes here.  The power monitor boxes provide power and measure exactly how much they provide.  We can remove a device’s battery, clip on some leads from the power monitor box, and watch exactly what is happening.  The box stores data over time and we can then go back and analyze that data.  

    Some of the data I gave above came from those power monitors and some of it comes from comparing specifications for parts to parts we actively measured.  Because these are generalizations, though, you’ll always be able to find devices with parts that are more or less power hungry than others.  You can really only use what I said above as a rule of thumb.  


  31. KASHIF says:

    hi there i was wondering if someone could please explain to me how if a portable power device was producing a certain power output per hour it can be related to the mAh rate? i do not understand how to determine the voltage from this portable power device.



  32. MikeCal says:

    KASHIF, electrical power is usually measured in Watts, which is current times voltage (Amps x Volts).  You might think you would need to look at both the current and the voltage to figure out where the power is going.  But on a mobile device the voltage doesn’t change.  So you can pretty much ignore it and just look at the current.  If the voltage is fixed, the only thing that can affect power is the current.  

    The tools we use all measure amps.  Many of them do it by sticking a small resistor in line with the battery and measuring the voltage drop across it.  Because the resistance is known, we can use ohms law (V=IR) to get the current.  

    Once you know how much current is flowing out of the battery (how many milli-Amps), you just need to multiply by hours to get the mAh rate.  For instance, if 1 mA flows out of the battery for an hour, you’ve just used 1 mAh.  

    Does that help?


  33. KASHIF says:

    Hi Mike,

    Thanks for the help, that explains it alot better now.

    If I wanted to caluclate how much power is used by the mobile phone in 1 hour of call-time how would I do it?

    Would it just be P=IV, "POWER = 1mA x 3.5" "POWER = 3.5 mW"? so divide by 360 for power per second?

    Thanks for the help


  34. MikeCal says:

    KASHIF, Power is an instaneous value. It doesn’t have time in it.  If you look at the bill from your power company, you’ll find that your usage isn’t measured in Watts (or KiloWatts).  It’s measured in Watt Hours (probably KiloWatt Hours, or kWh).  

    Think in terms of a 60 Watt light bulb.  Every instant that the light bulb is on, it burns 60 Watts of power.  If you leave the light bulb on for an hour, you’ve burned 60 Watt Hours of power.

    In the example you gave, if the phone runs at 3.5 volts, and if the extra current when on a phone call is 1mA, then the power from a phone call is 3.5mW, as you said.  If you talk for an hour, it’s 3.5 milliwatt hours, or 3.5mWh.


  35. Neotialbe says:

    Yes, LED can last for 10 years but do you know after you use it for 2-3 years, the color gets a lot dimmer. This is something that need to develop more in LED but normally people don’t know and never relize about it. I got this information about this from talking with somebody via http://en.mysnglife.com/index.php?p=3265#comment-687

  36. Neotialbe says:

    Oh, sorry, I just gave the wrong link. I meant this link  www.lunaraccents.com/technology-programmable-LED-controller.html

  37. Lim says:

    HI Mike, thanks for the informative article.You mentioned a lot about current usage in each component that drains off the battery life.My question is if you use a CPU that is of lower operating voltage,will it helps to save some power since P=VI? If i try to reduce I but increase V,although the P may be the same,however now I is reduced, so based on all the examples you mentioned that is based on I consumption, does this mean we can have longer battery time?

  38. MikeCal says:

    Lim, the reason I can ignore V and just talk about I is that, on a given device, V is fixed.  Components have a required voltage and software doesn’t have the option of changing that voltage to use less power.

    It IS true that a phone designed to use lower power components can burn less power than one designed to use higher power components.  For instance, 5 volt DRAM parts usually burn more power than 3.3V ones.  As phones started being designed with the lower voltage RAM, those phones were able to use less power than the previous phones with the higher voltage RAM.

    However, once the phone hardware has been designed and built, it’s almost impossible to change these components.  So, at that point, we need to focus on current.


  39. JJ says:

    Hello Mike,

    Your article is great!Thanks for the great sharing.

    One quick question about power consumption in Smartphone or mobile devices.Normally, concern for power is mainly for more battery life? What are the other major reasons to reduce the power consumption? Heat and component reliability or life span?

  40. MikeCal says:

    Hi JJ,

    Battery life is the primary concern.  Because phones are small, their battery capacities are also small.  There’s not enough power in there to really worry about heat dissipation.  If a component was burning enough power to create unwanted heat, it would just destroy the battery life.  So worrying about the battery life covers the other bases.


  41. luker says:

    pleeeeeease discuss PocketPC (WM5/6) as well.

    One main issue I  am discussing with users is that; the email and UpToDateNotifications eats battery when using Push.

    them-For – "Direct Push" keeps the data connection open, therefore its using less energy using true "Push" technology via Exchange server.  

    me-Against – "Pull" increases my battery life (and other users’) by 25% or more so i can go 1 or 2 days before charging – setting the phone to get email say, every 10 or 15 minutes.

    This is a query for unread mail and does not use push at all (?)

    Once this change is enacted battery life improves no matter what you are using the device for.


  42. isaiah says:

    would be good if you had a post on Pocket PCs

  43. Trevor Morris says:

    Is there a utility that will tell me how much power (or CPU cycles) my application is using?  I think I’ve effectively eradicated all power wasting threads, but would like some definitive proof.

  44. MikeCal says:

    Trevor, I’ve been meaning to write one, but haven’t had a chance.  The information is available, but buried deep in the kernel and needs a trusted app to call into it.  That makes it hard to write for smartphone, since the app needs to be signed.  

    I’m not aware of any third party apps that extract this data and give it to you.  



  45. po188 says:

    The Po188 is a photoelectrical integrated optical sensor designed at λp=520nm, with double sensitive receiver. It is highly sensitive to visible light, and varies linearly with illumination changing. It is used in saving energy, automatic sensitivity to light, and self-adaptive control in TV, LCD back light, Digital code product, instrument, industry device etc.

    ■ Electric Characteristic

     Small dark current, low illumination response, high sensibility, output current in proportion to illumination;

     Built in double sensitive receiver and Optical filter-less, attenuate near infrared automatically. Spectral response close to human eye sensitivity;

     Embedded micro signal CMOS amplifier, high precision voltage source, and correct circuit; Output current is high;

     Two Optics material packages to choose, to permeate visible light, cut off ultraviolet, relative damp infrared ray, and improve optical-filter effect;

     Accord with RoHS standard. Pb free, CdS free.

  46. Alz says:

    Hi guys! Is it possible to write an app to wake up the device when external power is connected?

  47. MikeCal says:

    Alz, does the device you want to write this app for have a touch screen?  There are two flavors of Windows Mobile, "Professional" (aka "PocketPC") and "Standard" (aka "Smartphone").  The easiest way to tell which you have is that PocketPC’s have a touch screen and Smartphones don’t.  

    When a PocketPC goes to sleep, the CPU turns off completely.  So there’s no way to write an app that runs while the PocketPC is asleep.  If this is the type of device you have, then, no, you can’t write an app to wake it up on external power.  Some OEMs have designed their hardware to enable waking up on external power, though.  You might find a setting for this under Settings.

    When a Smartphone goes to "sleep" it turns off the screen and puts the CPU into a low power state, but keeps the CPU able to run applications.  Because the CPU is still able to run, it’s possible to write apps that do things in this state. "Wake up" in this case, though, really amounts to turning on the screen.


  48. vijayPalvai says:

    Hi Mike,

    your Article on power consumption is really helpfull! could you please help me to get the following info:

    1.Is there any tool or microsoft powertoy that could help,monitor the amount of power an executable on the windows mobile smartphone is consuming ?



  49. Craig Hockenberry is a smart guy. He’s been around for a while, and has an impressive track record in the Mac software world. I don’t, but that’s not going to stop me from arguing with him about background apps on…

  50. Over the years, I’ve delivered several “Top 10” sessions and called them different things. Top Snafus,

  51. amidan says:

    Hi All,

    I have to determine the power consumption of the CPU (a part of the SOC)for a series of smart phones (e.g. Nokia, BlackBerry, Motorola and HTC). My problem is that I don’t have the perfect solution to do this but and old fashion one using an amp meter to measure the current on stand-by and after that during the execution of a test software program. More specifically I must measure the power consumption for each phone while running a series of Java benchmarks (for example GrinderBench).

    To obtain the measurements I searched for software that helps you measure your phone’s power consumption. For example I found: ”Nokia Energy Profiler” , which works on Nokia sets only. My problem is that for the other phones I did not find similar software applications. If anyone knows where I can find similar software for other models,  please send me an email(adrian.midan@vivaja.com) or post a reply.

    I also want your input and comments if I am on the right track. Is this the correct way to measure the power consumption, or you would like to suggest other methods?

    Best Regards,


  52. Mike,

    Thanks for another interesting and informative article.  A follow-up on PPC power consumption would be most appreciated.  



  53. Karthi says:

    thanks for the info.. but i assume a counter act will be there in future to keep the battery life not consumed too much while maintaining the same speed.



  54. montaser says:

    hey plzzzzzzzzzzzzzzzz

    i want to convert my htc p3300 to wm6

    how? plz tell me step by step

    and from where

  55. Alex Atkin UK says:

    That’s interesting about PocketPCs basically turn off the main CPU.

    I have my PPC set to not turn on the screen when it receives a new message.  This greatly saves on the battery, but am I to believe then that once I have received a message the PPC will still be awake but with the screen off?

    Theoretically it could drop back into standby as soon as the new SMS sound effect finished playing, but I cant really check that.

  56. Narendra Choudhary says:

    Hi Friends,

    I want to start or develop small windows mobile application,

    i have gone through MSDN library http://msdn.microsoft.com/en-us/library/bb158509.aspx and followed all sets to set development environment,  

    1. Installation of Windows XP SP2,

    2. Microsoft Visual Studio 2006,

    3. Microsoft .NET Compact Framework

    4. ActiveSync 4.5.

    5. Windows Mobile 6 Professional SDK,

    but now i am unable to know that what should i do next to start develop a application for windows mobile……………

    I am confused with following points…

    1. In Visual Studio which type of project should i select …

    (e.g. Win console application for desktop development)

    2. Should and how i link mobile SDK 6.0 and Visual studio 6.0….

    Please help me in this topic …

    Please send a solution if u have on folowing mail id ….


  57. Joshuapa says:

    Hi Narendra,

    Here is a quick walkthrough for creating a Hello World application on WM6 with Visual Studio 2005. There are probably simpler Hello Worlds, but it should get you started. http://msdn.microsoft.com/en-us/library/bb158524.aspx

    If you have Visual Studio 2008, the process is similiar, but there are a few differences. Take a look at http://msdn.microsoft.com/en-us/windowsmobile/bb250560.aspx for more details.

    Does that help?


Skip to main content