I owned an iPhone 6 for two years. During the second winter I spent with it, whenever I'd go for a run in the cold, the iPhone would shut down at some point mid-run, even if I'd charged it fully right before I left my house. There was no way to turn it back on without connecting it to power. It was beyond frustrating.

Now I know why it was happening. My iPhone, which I was using to track my run via GPS, stream music from Amazon or Spotify over cellular, and relay social-media "cheers" to me as audio notifications via the Nike+ Running app — not to mention all the other everyday background stuff an iPhone does — was, at some point, pushing the processor hard enough to demand a lot of current from the battery. And my battery, which was over a year old at this point, simply couldn't handle it, certainly not in the cold. It had a panic attack and shut down.

Apple directly addressed this problem when it released iOS 10.2.1 in January 2017. That was too late for me — I had moved on to the iPhone 7 by then — but presumably thousands of early morning runners with old iPhones in the Northern Hemisphere were grateful.

That is, until this week.

After Reddit users kicked off a discussion about slower iPhone performance in older models, Geekbench founder John Poole appeared to confirm what everyone thought: Many of the older iPhones score lower (sometimes much lower) in benchmarks than a brand-new model. And it's clear that specific iOS updates were introduced to limit performance, at least in certain conditions. In a statement emailed to Mashable on Wednesday, Apple essentially confirmed this was the case.

This seems outrageous — as Poole says in his post, "We expect battery capacity to decrease as batteries age, we expect processor performance to stay the same."

Let's holster the pitchforks for a second, though. The "certain conditions" are the detail this hinges on. Apple is clear that the OS will scale back processor performance so it won't tax the battery with "peak current demands," and that it only does so in cold conditions, when the battery has a low charge, or when they batteries are old.

Let's also be clear what benchmarks are: a set of tests to push a processor to its limits to assess how it compares to other phones and chips out there, not an assessment of overall experience. Most day-to-day activities rarely push your iPhone to "peak."

What will you experience if you do? The iPhone will spread its power demands out over a slightly longer period of time, which could mean encountering some lag. In my morning-run example, I probably would have heard a song pause for a few seconds, or the Nike+ app might have been slow to respond. God help me if I asked Siri for anything in that moment.

But compared to an out-of-the-blue shutdown? I'll take it. I'd agree with Matthew Panzarino at TechCrunch that this isn't "throttling" per sesince it only scales things back at peak conditions. Let's call this what it is — performance limiting. The point, though, is that ultimately Apple did the right thing: This is a feature, not a bug.

Certainly, Apple could have been more transparent about this, and it could stillbe more transparent. (I'd like to know what percentage Apple considers a "low" battery, and how many charging cycles constitutes an "old" battery.) While Apple's culture of secrecy is still brilliant marketing, in situations like this, it doesn't help. As this ordeal shows, Apple can't change the laws of battery chemistry, but it could change the conversation with its customers — if it would just have one in the first place.