Killing apps in the background is fun, right?
Well, if you're an app user and rely on your app to do the job while the phone is closed... not exactly. Your phone manufacturer however, may not agree to you at all. For them being able to advertise their "great battery time" is of utmost importance. So the shameful apps (all that use GPS) that drain battery "and make us (the phone manufacturers) look worse" need to be killed whatever they do and and whatever you (as a user) paid for those apps to do.
The bad news? If you're here, chances are high that you are a trail runner and you are using a GPS app (like Spectro.Life) to track your activity. These apps are all subject to being killed on some Android phones as it is very unlikely you are running with your phone looking at the screen all the time. Once these apps go to the background... they're doomed. Unless... give us a minute, we'll get to the point. But first, you need to know this has nothing to do with Spectro.Life or your other apps, this has to do with your phone manufacturer. Sorry!
Do all Android phones kill apps the same way?
No. Based on dontkillmyapp.com not all Android phones behave the same way. Some are more intrusive, while the default AOSP Android implementation does not interfere in a toxic way like the ones from the top of the list. See below:
For a complete list see dontkillmyapp.com. For demonstration purposes we chose Huawei P30 Lite to walk you through the setup process.
How do I know if my phone is set up correctly?
We suggest you install the DontKillMyApp mobile app from Android Play Store and start the shortest benchmark of 1 hour.
As you can see from the picture below in the first step (1) we have performed all the configuration steps shown in https://dontkillmyapp.com/huawei, even though some of the them were not applicable, option was not present in the phone's menu, the result was still not sufficient. When running the benchmark using the DontKillMyApp, the result still showed an overall 54%, and even worse results in the work and main categories.
We needed to go to the second step (2) and apply the developer options, installing tools to remove the pre-installed PowerGenie app from the Huawei phone. We know such operations are not easy at all for most of the people reading this blog. That's why we call this out and suggest you ask for help from more technically versed friends. Sorry, we cannot do anything about this just point you to the right resources. After this step, you can see the benchmark gave 100%.
So, what is the difference between correct and no setup?
We've done a run (1) using Huawei P30 Lite without any setup and we compared it with the same run on an iPhone. By the way the iPhone did not need any setup, it works out of the box. Afterwards we performed the setup and did another run (2) and compared the run against the same one on iPhone. Here is what we can observe:
First run:
The activity did not finish, we needed to manually close it from the support icon, 'Finish Activity' action
The run started in the background and the operating system instantly cut the GPS
From kilometer 5 to 8 the app run in the foreground, the runner looking at the screen and GPS was acquired correctly
From kilometer 8 till the end the operating system killed the app once it went back to background
On the other hand on the iPhone things went well from the start to finish, GPS was constant, accurate, timings on each segments were correct, hence performance drop was also correctly calculated to be 12%
Second run:
The activity finished correctly both on Huawei and iPhone and on time
There was constant GPS signal from start to finish on both platforms
A small offset can be seen between Huawei and iPhone in the start time and in the segment timings without significant effect on the run
A larger effect can be seen on the steady pace and performance drop calculation. Let's quickly check the reason
The reason for the difference
As you can see in the picture below, the GPS signal on the left (Huawei) compared to the one on the right (iPhone) is much worse. Sometimes there is a deviation, sometimes it just jumps over geofences. But let's make this a topic for a different post.
One thing we could conclude for sure: iPhone not only worked out of the box, but the accuracy of the GPS is much higher as well. Chances are high you are in fact an Android user and this may not go well with you. Sorry for that! What we can promise on the flip-side is that we work very hard so Android users of Spectro.Life experience our services in similar quality as our iPhone users.
