01-12-2014 07:39 PM
85 ... 234
tools
  1. falconrap's Avatar
    Power is mainly impacted by 1) the display, 2) the CPU, 3) the cellular radio, and 4) the OS. Newer displays tend to be more efficient per inch than older displays, so I would guess the displays on both phones would use similar power.The second is the CPU. The lower the voltage a CPU can run at, the longer it can stay at low clock speed and still get the work needed done, the deeper the standby state, and the lower the overall leakage current on the gates of the particular design/process the less power will be used. The 800 is likely more power efficient in terms of wattage/unit of work done. The Snapdragon 800 is on the newer 28nm HPm process at TSMC (the Snapdragon 600 is on the older 28nm LP process), so the amount of power both CPU's is probably around the same under most conditions, despite the names of the processes involved. The LTE radio on the 800 is a newer, more efficient one than the 600, and you can bet that this impacts your battery life when using any data. The advances in these LTE radios have brought about significant power improvements the past 2 years.

    Finally, the OS is another key difference. Android, under real world use, tends to use more power than WP, mostly because there tend to be a lot more background processes running, especially when a user starts loading up on widgets. Overall, if I had to guess, I would say the 929 would probably get a lot more standby time, and probably somewhere between a half hour to an hour more talk time and similar for data. This phone should easily last through a day of moderate use. Proof will be in the pudding.
    DustinRyan and puttlikezach like this.
    12-09-2013 08:09 PM
  2. DustinRyan's Avatar
    @falconrap As a current non-WP user, do you think live tiles drain battery like Android widgets? Or possibly less/more depending how many are running? I haven't been able to test. I've been told live tiles use more data on my data plan, but never heard about how it affects battery. Just something I thought of while reading these battery posts.
    12-09-2013 08:35 PM
  3. manicottiK's Avatar
    do you think live tiles drain battery like Android widgets?
    Tiles can be updated in two ways: 1) new data pushed by a server or 2) a "periodic task" that updates the tile locally.

    The first method uses virtually no power on the phone as the OS just has to respond to a small data push from the server. Although a server could send such updates at a really high frequency, using more battery power, most don't. (The push server for our university portal app sends updates when new grades are posted at the end of an academic term and on a few other conditions -- that works out to about 16 push notifications a year.)

    The second method is the more common way of updating tiles because everything can be self-contained in the app doing the update (i.e., no server needed). These "periodic tasks" are run by the OS approximately every 30 minutes. The app cannot request more (or less) frequent updates -- 48 times a day is the only option. When the task is run, it must finish within 25 seconds and must not use more than about 20MB of memory.

    As such, Live Tiles can't use much power because they run for less than 2% of the time and use less than 2% of memory. Of course, a skilled programmer can write an Android widget that uses less than that, but it's also possible to write one that uses substantially more.
    12-09-2013 09:36 PM
  4. DustinRyan's Avatar
    Thanks for the clarification! That answers my question.
    12-09-2013 11:05 PM
  5. falconrap's Avatar
    He beat me to it. Most Android widgets update more often, and generally can tend to use more power in general.
    DustinRyan likes this.
    12-10-2013 04:50 AM
  6. manicottiK's Avatar
    I noted that our app does, on average, 16 pushes (technically 16 tiles plus 16 corresponding toasts) each year, but forgot to add info on our periodic task.

    In our app, if the user enables it, the task runs every half hour to download fresh data used to update between zero and a lot of tiles. The sample below shows the main app tile, the tile for a specific class, the schedule tile (we're in week 11 of the term), and the all courses tile (this student has only one course with recent activity, which happens to be the one above it in the picture). Other tiles provide the balance on our "DragonCard", counts of upcoming job interviews, etc. There's lots to update, but not much never every half hour.

    We had to restructure things twice and switch our data serialization library twice to get the task to run in 11MB. (I mentioned 20MB above, but that's for 1 GB phones. We stuck with the 11MB limit used for 512MB phones to support more of our students.) The restructuring reduced both memory usage and cellular data used. By only sending the app data when there's a change (i.e., if a student adds or drops a class, the new class list is sent to the phone, otherwise, the server sends back a null list to tell the app to keep using the one already on the phone) the background task generally has less updating work to do, which helps us stay within the memory limits.

    We also built in logic so that it a task dies or is killed, the next time that it runs, it finishes updating any remaining tiles using the data retrieved the last time. In such cases, the tiles may be an hour out of date, but at least it keeps the task running. If the app does die twice in a row such that the OS drops the periodic task from the list of enabled apps, we let the user know that and provide why the task was unscheduled (memory, speed, blocked by user via Settings, etc).
    Attached Thumbnails 480-8-live-tiles.png  
    Last edited by manicottiK; 12-10-2013 at 06:36 AM.
    12-10-2013 06:18 AM
  7. puttlikezach's Avatar
    Thanks to everyone who chimed in. Extremely insightful information. Not surprisingly, it sounds like there is a lot that effects battery usage and that WP may have some inherent advantages over Android in that regard...but as it was said, the proof will be in the pudding. Out with this phone already!!
    Quantify likes this.
    12-10-2013 01:53 PM
  8. kbilly70's Avatar
    Rumors of a possible international version of the 929:

    Flagship Nokia Lumia phone with 5.2" 1080p display shows up - TechShout
    Loco5150 likes this.
    12-11-2013 07:05 AM
  9. Maanasee Deshmukh's Avatar
    He beat me to it. Most Android widgets update more often, and generally can tend to use more power in general.
    See this is why windows is so much better!! Thanks for all the insightful info people ;)

    Sent from my device using Tapatalk
    01-12-2014 05:27 PM
  10. undulose's Avatar
    If it's true, that would probably be a good news. The Lumia 1020 has 2000 mAh only but it can last a day of heavy use. Then consider the fact that it is also rumored to have Snapdragon 800 which is more energy-efficient that its predecessors.

    On a second note, WP is more energy-conservative than Android, in general.
    01-12-2014 07:39 PM
85 ... 234

Similar Threads

  1. Replies: 25
    Last Post: 08-29-2016, 01:43 AM
  2. So, who plans to get a Nokia Lumia 525?
    By Muessig in forum Nokia Lumia 525
    Replies: 10
    Last Post: 10-15-2014, 05:04 AM
  3. how to change default text size?
    By TheKinkajou in forum Dell Venue 8 Pro
    Replies: 5
    Last Post: 12-02-2013, 03:13 PM
  4. Can Microsoft work with Nokia now?
    By jedpatrickdatu in forum Microsoft News & Rumors
    Replies: 11
    Last Post: 11-29-2013, 05:42 PM
  5. Do live tiles drain your battery?
    By Oneofthesedays in forum Windows 8
    Replies: 1
    Last Post: 11-27-2013, 04:55 PM
LINK TO POST COPIED TO CLIPBOARD