Apple on iPhone Battery Performance: You're Measuring It Wrong

Apple issued a statement Thursday responding to reports of different battery performance in iPhone 6s and iPhone 6s Plus models. Some testing and real-world usage has found that devices powered by an Apple A9 chip made by TSMC get as much as 50 minutes more battery life than devices powered by A9s made by Samsung. Apple told TechCrunch, however, that its internal data shows a modest 2-3 percent difference as a whole, and that benchmark testing procedures are not an accurate way to measure battery life.

To put that more succinctly, Apple says some folks are testing it wrong.

iPhone 6s Hooked up to a Car Battery

Apple designed the A9 chip in-house, but manufacturing is handled by third party fabs. In this case, some processors were made by Samsung using a 14nm process, while others were made by TSMC using a 16nm process. As discussed on Thursday's TMO's Daily Observations, some people have reported longer battery life in devices with the TSMC-made chip.

Apple responded to the stories, telling TechCrunch:

With the Apple-designed A9 chip in your iPhone 6s or iPhone 6s Plus, you are getting the most advanced smartphone chip in the world. Every chip we ship meets Apple’s highest standards for providing incredible performance and deliver great battery life, regardless of iPhone 6s capacity, color, or model.

Certain manufactured lab tests which run the processors with a continuous heavy workload until the battery depletes are not representative of real-world usage, since they spend an unrealistic amount of time at the highest CPU performance state. It’s a misleading way to measure real-world battery life. Our testing and customer data show the actual battery life of the iPhone 6s and iPhone 6s Plus, even taking into account variable component differences, vary within just 2-3% of each other.

There is something to be said for Apple's reasoning here, starting with the fact that Apple has access to a much larger data set than everyone else. Millions of Apple customers allow anonymized diagnostic data to be set to Apple, and the company sees a far bigger picture than the rest of the world.

To that end, it's a safe bet that Apple is telling the truth when it says it sees a 2-3 percent variance between its new iPhone models. It's also quite true that the vast majority of users aren't pegging their CPUs nonstop for hours at a time, making measurements taken with such methods irrelevant to most real-world usage.

On the other hand, some users do use their devices nonstop for hours at a time with activities that do peg their CPUs and the graphics chips, and maximize RAM usage. Many games on the iPhone can peg even this newest device.

For those users—subset though they are—benchmark-based tests are relevant.

On TDO this morning, we also discussed the myriad of issues that can go into every individual's real-world usage. Not all apps are created equal, and not all apps behave in the background like they should. I'm looking at you, Facebook, in particular.

The point being that aggregate data matters. If Apple is seeing a 2-3 percent performance difference, it's a reasonable outcome. That doesn't make it any less annoying for those outliers who are putting their devices through more stress and seeing poorer performance than they might if they simply had the larger chip from TSMC.

Of course, that invites renewed debate on Apple's obsessive pursuit of thinness at the expense of battery life, so have it.

Image made with help from Shutterstock.