Apple Called Out By Geekbench Over Slowing Down iPhones

It’s been a bit of an urban legend that Apple deliberately slows down iPhones so that you’ve be more likely to buy one. Thanks to Geekbench, there may, and I say MAY be proof that this may not be an urban legend.

Geekbench benchmarked the performance of iPhone 6 and iPhone 7 devices running on different versions of iOS to see how the kernel density changes as the operating system is updated. What they found is that the iPhone 6’s score for iOS 10.2.0 appears ‘unimodal’. Meaning that it doesn’t change in performance. However when it studied the iPhone 6 running iOS 10.2.1, the phone’s performance peaked at the average score and several other peaks around some of the lower scores. It reported that there was even more of a disparity when the iPhone 6 was tested on the latest version of iOS 11.2.0. For the iPhone 7, the scores were pretty much identical across iOS 10.2.0, iOS 10.2.1, and iOS 11.1.2. However, when using iOS 11.2.0, the graph showed more peaks again, suggesting the performance is impacted as the device ages.

Here’s what Geekbench says is the reason behind this:

First, it appears the problem is widespread, and will only get worse as phones (and their batteries) continue to age. See, for example, the difference between the distribution of iPhone 6s scores between 10.2.1 and 11.2.0.

Second, the problem is due, in part, to a change in iOS. The difference between 10.2.0 and 10.2.1 is too abrupt to be just a function of battery condition. I believe (as do others) that Apple introduced a change to limit performance when battery condition decreases past a certain point. Why did Apple do this? kadupse on Reddit offers the following explanation:

Many iPhone 6s devices were shutting down unexpectedly, even after the battery replacement program (Which many people weren’t entitled to use). Because degraded batteries last much less and end up with a lower voltage Apple’s solution was to scale down CPU performance, it doesn’t solve anything and is a bad experience… but it’s better than having your device shutdown at 40% when you need it the most.

Assuming that this is accurate, this is craptastic. Users expect their shiny iPhone to run at full power all the time. But if it can’t, the phone has to tell the user that it’s scaling back the performance to save battery life or whatever other reason it needs to scale back performance. By not telling the user that this is happening really makes Apple look like they’re doing something underhanded. Now to be clear, I don’t know if they are or not. But they need to explain this right the hell now. If they don’t, Apple’s bad year where the “it just works” mantra went out the window will simply get worse and carry on into 2018.


4 Responses to “Apple Called Out By Geekbench Over Slowing Down iPhones”

  1. […] today I posted a story about a report from Geekbench showing evidence that Apple was slowing down iPhones via software as […]

  2. […] Let’s recap the iPhone slowdown saga. Geekbench was able to prove that Apple was via software slowing down older iPhones with aging batteries. Apple then confirmed […]

  3. […] of Apple products that seems to highlight that their QA process is in free fall. Oh yeah, Apple was caught slowing down iPhones with aging batteries which they tried to explain away, but got targeted with […]

  4. […] Apple products that seems to highlight that their QA process is in free fall. Oh yeah, Apple was caught slowing down iPhones with aging batteries which they tried to explain away, before offering up […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: