We're working on a graphics-intensive game for iOS, and we are very concerned about battery drain and thermals. While it's hard to control the user's conditions (ambient temperature, insulating case, battery lifecycle, etc.), we know we can't ignore these issues and we want to make sure we're in line with general expectations.
We have found it very challenging to get consistent measurements for the different thermal conditions, such as how long it takes to reach T2 from T1. Room temperature has huge effects. So, we thought a more consistent metric may be battery drain: how much % is used after 15 minutes of play, for example? If we were to focus on this as our metric, would this also take care of heat generation? Specifically, if we make a change and do an A/B test, and found that battery drain was the same, can we safely infer that our heat behavior is also the same? Or, if we did another AB test and found that it drained the battery faster, can we assume the change would also increase phone temperature faster?
We have of course observed oddness with the battery percentage, but with some caveats in mind, it seems like something we can measure much more easily and consistently.
Any insights in this would be much appreciated!
