Much has been written in blogs these past 18 months about how Feedback and Phone Update (can find these in Settings) transfer boatloads of data and drain your battery down. While I have never believed either of these myths to be true I thought it would be good to put them to the test. While I figured this would be a simple slam dunk kind of test, things got a little complicated so bear with me (or scroll to the bottom for the results).
I am not privy to the internal code that controls these functions, but can apply a little logic here. Feedback most likely consists of a log (txt) file somewhere on your device that either uploads at some timed interval or after it has reached some kind of size threshold. While I know that log files can grow quickly if left unchecked I don’t imagine Microsoft would want to move that much data. My guess is that uploading a photo to SkyDrive or Facebook, or attaching to an email probably consumes as much (or much more) data and power as a 24 hour cycle of Feedback. Phone Update most likely uses even less energy. Anyone who has waited for their phone to announce , “an update is available”, and instead plugged into their PC/Zune to get said update immediately, knows that Phone Update does not check very often. Maybe once or twice every 24 hours. While I can’t quantify data transfer rates for these functions, more data transfer should equate to more power drain. Sounds logical enough.
For review, all testing was done on a Nokia Lumia 900 with the following parameters;
-Data (mostly LTE) and Wi-Fi ON – Bluetooth OFF
-8 Active Background Tasks (3 internal)
-Location Services ON
-5 Email Accounts syncing (3 push, 1 every 15 minutes, 1 every 2 hours)
Test 1 – As I normally keep both of these services turned on, with Use Cellular Data disabled, I couldn’t use any previous tests (18% over 8 hours or 2.25% per hour) as a baseline. For this test I turned on all Feedback and Phone Update features. RESULTS: 99% at12:07AM– 81% at 8:06AM – Difference: 18%, or 2.25% drain per hour. Dang, that sounds familiar.
Test 2 – For this test, I disabled all of the toggles for Feedback and Phone Update, but left everything else the same. RESULTS: 99% at 12:05AM – 71% at 8:05AM – Difference of 28% or 3.5% per hour. Wow! How could that be. Didn’t I turn these settings off. It is possible though that I skewed the results. I normally set my phone down in the same place on my desk and leave it undisturbed during the testing period. But for this test I set it down right next to a power distribution thingy (about 12” from its normal spot) that sits under my monitor. Probably had zero effect, but as the results were so out of whack I decided to throw this result aside and try again.
Test 3 – I repeated Test #2 above, with both services turned off, but this time setting my phone down in it’s normal place. RESULTS: 99% at 12:27AM – 75% at 8:30AM – Difference of 24% or 3% per hour. What! Almost the same result as the suspected skewed test above. This was certainly unexpected and made no sense to me. More testing required. Btw, I did perform a test on 5/20 (Sunday) with all services turned back on, but overslept so could not get a good graphic for the screenshot. But Battery Meter indicated a 17% drain for the 8 hour test period. That’s about right.
Test 4 – Ok, so I turned everything back on again, hoping to repeat the results from Test 1. RESULTS: 99% at 12:07AM – 83% at 8:10AM – Difference of 16% or 2% per hour. A little less than the above test, but this is probably a good time to talk about rounding. As Battery Meter only displays round numbers and I don’t have any idea how Windows Phone displays their data (does 99.4 = 99 or does 99.9 = 99) all of my results have a variance of +/- 2%. So the 18% vs. 16% result could effectively be almost identical.
Test 5 – I needed to better understand what was going on when these services were disabled, so after turning off Feedback and Phone Update, I soft reset my phone (first time in two weeks) just prior to starting the test. RESULTS: 99% at 12:17AM – 81% at 8:18AM – Difference of 18% or 2.25% per hour. Very interesting. So with a soft reset, effectively clearing out the memory, I got the same result as having these services turned on. A Microsoft bug perhaps, where the OS will continually try to send feedback even though the service has been disabled. Don’t know. More testing required.
Test 6 – For my final try, I turned these setting back on, but also performed a soft reset just before starting the test. RESULTS: 99% at 12:14AM – 83% at 8:14AM – Difference of 16% or 2% per hour. Pretty much the same result as in Test 4, and due to rounding, no different than any other result, with Feedback and Phone Update turned on.
It should be noted that as I have no idea how either of these services function, my results could be seriously skewed. For example, Feedback might only work when the phone is awake, or might only transfer data after several days of collection. Without real lab tests, and access to the OS, it would be impossible to determine exactly how much energy these services consume. One thing is for certain though. The two high drain results were more than a coincidence.
Second Truth: Turning both Feedback and Phone Update ON have virtually no effect on battery performance, and presumably transfer very little data (as they don’t impact the battery at all). But, turning these services OFF without following up with a soft reset (Power off/on cycle) will result in excess battery drain. Who would have thunk. Next up, Battery Saver.