#iPhoneSlow: What the analysts and experts have to say


[background music]

Rene Ritchie: I'm Rene Ritchie and this is "Vector." Today, we're doing a special edition, a last-minute roundtable to talk about a headline. I don't want to call it a breaking story because it's about 11 months old, but in the last 48 hours, it's gotten a lot of attention.

That is Apple's decision to more aggressively manage power, to prioritize battery life over peak processor performance on iPhone SE, 6, 6s, and recently iPhone 7.

I wanted to bring together a bunch of really, really smart people so that we could discuss as much science and as much data, share our opinions, and help everybody to understand more about the story. Here we go.

Joining me today we have John Poole from Geekbench. How are you, John?

John Poole: I'm well, thank you.

Rene: We have Jerry Hildenbrand from Android Central. Hey, Jerry.

Jerry Hildenbrand: Hey, Rene. How are you?

Rene: I'm doing very well, thank you. We have industry analyst, Ben Bajarin.

Ben Bajarin: Hi.

Rene: How's it going, Ben?

Ben: [laughs] Good.

Rene: Carl, let me pronounce your name right. Is it Carl Howe?

Carl Howe: That's exactly right.

Rene: I have not had the privilege of speaking to you before. What's your one-line bio?

Carl: I'm a former industry analyst. Now I go by long-time Apple watcher.

Rene: Recovering industry analyst, almost? [laughs]

Carl: Exactly.

Rene: We're here to talk about the current controversy, might be the right word for it, around Apple and what they're doing with power management, battery life, and SoC performance on iPhones, primarily iPhone SE, 6, 6S, but as of iOS 11.2, also iPhone 7.

This actually started about a year ago when people were seeing battery drain and actual power downs on iPhone 6 and iPhone 6S. Apple had a series of media briefings with top-level people in product marketing.

Some people get the wrong idea about product marketing. There are product marketing people in some companies who are glorified sales people. At Apple, you're expected to be able to run the full engineering roadmap. They can, and they do, really well.

Talking to Apple marketing is a very intense experience. They went down what was happening, why it was happening, and gave a statement saying that they were going to put a system in place that would prevent these power peaks.

To make it quick, some phones that were having battery wear, for example, they'd been dropped repeatedly or they had been placed in sources of heat, they'd done something that had damaged the lithium ion and they were aging prematurely. If there was a peak power demand, they would shut down to protect themselves.

Apple put out a code that would change the way power management worked, in order to prevent that from happening. What power management means quite often is you have to balance battery life. Battery life is a currency in mobile. You pay for it with everything that you do. Everything costs you in terms of battery. One of the things that cost is the CPU and the GPU.

Apple was being much more careful with how they allow them to ramp up and for how long they allowed them to ramp up. Fast forward to a week ago or so, and someone on Reddit was having issues with their iPhone 6 running slow and went and had the battery changed. Because it was a brand new full battery, healthy battery in it, suddenly the performance kicked right back to maximum.

They were astonished and they said how great the experience was. I think John, at that point, is that where you got involved?

John: That's when we started digging into it. We've been hearing reports about this sort of behavior, users complaining their phone was slow. In past years when this has happened...especially because we see that uptick that, "My old phone is now the old and busted model so it must be slow."

We'd see an uptick in users coming to us and saying, "My phone is slow. What can I do?" Normally, there's some explanation for it, low power mode being the most popular one. It could have been maybe the phone was hot because it had been left in the sun or something like that. There's usually an obvious issue there that could...

Rene: Well also...and not to interrupt you. There's also everyone gets the new iOS update all at the same time, every region, every carrier. When you see Google trends for iPhone slow, it's always maxed out in September and it gets headlines, where with Android and other vendors the updates are intermittent and spread out throughout the year. You have tiny spikes but you never see the critical mass you see with iPhone.

John: Exactly. We're used to a certain amount of that traffic. What was different this year was that it was far more than we've seen in previous years, and far more sustained. It wasn't that brief spike around the launch. We were getting a constant ping from our users saying, "I ran Geekbench. My numbers are lower. What do I do?"

Our stock line was, "Well, we don't know. It could be an issue with your phone. Try resetting it. Take it into Apple. See if they have any feedback." That sort of thing. We were scratching our heads over it.

When I saw the Reddit post and they said, "Oh, I replaced the battery and then everything got better," I thought, "Is this the systematic effect? Is this an issue that's affecting other large numbers of phones?" That's when we dove into the database to pull up results and start doing that analysis.

Rene: Carl, could you give us a little primer on lithium-ion battery technology, and how it works?

Carl: Sure. The thing that I think a lot of people don't understand about lithium-ion batteries is that they're not your father's double A battery. Lithium is very volatile. It's very reactive. In some sense, you're carrying around a little chemical reactor in your pocket.

Actually, one of the interesting things about lithium-ion batteries is, if you don't treat them well, they self-destruct. Witness last year Samsung, etc.

We in the electrical engineering world, we spend a lot of time trying to make non-ideal things ideal. There's a processor added to most lithium-ion batteries that manages its operation. They try to keep it in the good zone where it just delivers power as you would like it to do.

But at the end of the day, it's pretty non-ideal. It does all kinds of crazy things. We don't charge it properly. It has a lot of variation in voltage and current according to temperature.

Rene: It doesn't like [inaudible 6:27] .

Carl: Yeah, there's all kind of things that go wrong. Age is another one. You've got this sort of operator sitting there watching over it. Unfortunately the chemical reactions degrade over time. It's got to do the best job it can to deliver what the processor wants. Frankly, the processor isn't that interested in your problems. It just has to make do.

With all this controversy, the analogy I always make is this is a little bit like running a power plant in the middle of the summer. You've got those days when everybody turns on the AC. If it's in excess of the power you can deliver, the guys in the control room, that little processor that's running the battery, have a choice. Do you want to have a brownout or do you want to have a blackout?

If you want to just keep ramping stuff up, yeah, we can just turn it down. The thing will turn off because I have to protect the battery. The battery is going to blow up and cause fires and things like that if I don't do the right thing. So I have a choice of either slowing things down or turning things off. I think that's the controversy.

John: That's an excellent analogy that I wish I had thought of. I want to tell you that.

Rene: Ben, I know John said that he was seeing an increase in interest on this. What I love about your work is that you always bring numbers to the game. Are you seeing anything in either sentiment analysis or satisfactional analysis that gave you any indicator?

Ben: There are two things. We've seen customer satisfaction maintain a really high number. We break that down by model as well. When we ask the question, we can look at that by model of device that they own. We don't see much change even as devices get older. People remain still high levels of satisfaction, north of 97 to 99 percent depending on the model. We haven't seen that even go down.

I think what's interesting is when we look at also the things that drive upgrade cycles for Apple, battery is actually not the top of the list. Depending on the model, like with Plus owners for example, it's fifth on the list. With non-Plus owners, it's fourth on the list.

It's interesting because then I look at that and say, "Well, whatever." I look at that stat. I'm not convinced that Apple users have a battery pain point. I wonder if this work is something where Apple is doing a good job in helping your battery last all day through all sorts of things that they're doing besides just the silicon optimization and the software side. That that's just not the same pain point.

Because to be honest with you, in Android world, we see battery a much, much higher driver of top things of interest driving new purchases.

Those are the things where it doesn't seem to me like it's something that the market has come back and really, this is your average consumer, your normal customer, is getting either angry about or feeling that pain point about because again, like I said, satisfaction remains high even amongst older devices. Battery's just not top of the list.

The thing I was pointing out too on Twitter yesterday I think really, really interesting is Apple has this unique position in that the life of their devices is longer than almost all other smartphone vendors, not just because their devices do actually last a long time which again is an argument that it's not necessarily planned obsolescence.

But that also there's a big secondary market for iPhones whether or not someone gives that to a friend or family member or sells it on the market or turns it back into their carrier who then sells it to corporate accounts.

Those things live for a really long time. Apple kind of has this problem, as Carl pointed. Lithium ion is a limited science. We also know the limitations of that science. Over time, it just simply degrades. If you have devices that are on the market for four-and-a-half to five years, Apple has to manage that.

On top of having to manage that, they still want to bring the cutting edge technology of iOS to those old devices, which, by nature, also consume quit a lot of CPU and GPU clock cycle. They're just in a different position. I think this is the strategy that they went to handle it. Everyone can knock whether that's right or wrong. I think we all agree if that's what's going to happen, they should tell people.

You'd be hard pressed, I think, to think that a consumer is going to consciously say, "I'd really rather my phone just turn off at 30 percent," thinking that they have a whole day and they actually don't, versus Apple doing what they can to make sure that if they're going to hold onto that phone for as long as they are and/or give that to someone else, that it is actually going to last a day and still have a good experience.

Like I said, they have to do something that other vendors don't, and this is the approach that they took to sort of solve this unique problem.

Rene: One of my favorite episodes of "The West Wing" was the lame duck congressman where they tried to force him to come back and vote for nuclear disarmament. They said, "This is what your constituents want."

He said, "I've been briefed on this for a decade, and I barely understand the issue. You can't tell me that 30,000 people really understand every nuance and subtlety about this nuclear arms treaty." I think that's one of the things. We like, as consumers, to feel empowered, but we're also existing in this really interesting time of social media.

I never mistake social media for reality, but it is played out on a very social stage. You saw that with Samsung last year. I think we're seeing that with Apple now. How do you react to that?

Jerry: Yes, this is a convenience thing. We don't want our phone to shut off at 30 percent. There's also a big user safety issue here. As Carl explained, a lithium-ion battery can have a spectacular fail, and nobody wants that to happen in your pocket.

A phone is kind of unique in the way they work. The CPU or the GPU, they go from what we call sleeping or idle to 100 percent almost instantly. That hits a battery hard. It needs a whole lot more energy than even a new battery can deliver to make that possible.

Once you get some age on the battery and it can no longer hold that much capacity to provide what you're used to, you have to do something lest the battery fail in some way. We know what can happen sometimes when a battery fails.

Somebody at Apple said, "We don't want this to blow up on somebody. Let's figure something out." Whether this was the right way to figure it out or not is up for debate, but there is a safety issue here.

Rene: John, it's interesting that we saw this with the A8. We saw this with the SE, the 6, and the 6 Plus. Apple hasn't implemented this on older devices, and I've seen this. I don't know if Ben has experienced this similar. I'll go out to game, and I'll see a bunch of people on iPhones and on Android phones. People on newer iPhones are OK. People on anything older, for Android mostly they're connected 24/7 to power packs. People with iPhone 5S and earlier, they do get...

It's like 60 percent, and they're like, "I'm ready to game," and then their phone shuts off, especially in the cold now. [laughs]

Carl: Yeah, I've got a point about that. I think we often look at these devices and we think that that percentage is a real measurement. The reality of this is, again, going back to my control room analogy, those guys keep records about how your battery performs. By the way, the batteries are not uniformly all the same.

They actually keep a lot of records, and they kind of create a model for what your battery capacity and your battery ability to deliver current is. That's what generates that percent number. It's not a direct reading off a battery. It's a little piece of software.

I think what we often sort of, "Oh, it went from 40 percent to 0 and it shut off." Actually what happened is that the last time I ran the model it said you had about 40 percent left. Then something happened that put the battery into a state that said, "Oh, things are going south here. Again, we're kind of getting into blackout territory. What do you want to do?"

In the older models, they didn't really have an option to tell the processor, "Hey, reduced the load," so they just would shut down. That model is actually important because it's the thing that generates the percentage. That's not a really measured number. It's a modeled number.

Ben: One of the things that's interesting too is -- I don't remember how, Rene, you would remember, how this came out -- we all have found out Apple is essentially making a power management controller now, custom designing it.

I wonder if part of this has to do with it as well. Obviously, not just the radios but just things having to do with the frequencies that's being put through based on workloads. This might be another thing that they're hoping to address, which could address the battery as well.

Rene: My understanding is they've been doing that for a lot longer than they've been talking about they've been doing it.

Carl: Absolutely.

Rene: There was this change, right, where Apple went from...As they've been making their custom CPUs and GPUs, they've changed the characteristics.

There was a really good article by Andrei -- I'm going to butcher his last name -- Andrei Frumusanu at AnandTech where he was talking about some of these architectural choices in the CPU, Jerry and how we're doing things with phones that we never were doing before.

When they first announced it, it was like, "You can surf the Web, you can check your email," and now you're downloading constantly on Instagram while you're Snapchatting, Facebooking, and running VR and AR games and all these different things. When you're trying to design SoCs for this, Apple famously went wide and slow, not thin and fast, but they're still susceptible to spikes and to other things like this.

Jerry: Yeah, on a hardware side, like you said, we want to do more with these tiny computers in our hands that they really weren't designed for in the beginning. Some things have to change. We also want them to go to sleep more so our battery lasts even longer when we're not trying to do things with them.

Rene: The race to sleep.

Jerry: Yeah, the gap between not being used and how they're being used is bigger than anybody is even thinking about here.

Rene: They try to offload. They try to have coprocessors like sensor fusion hubs and all these things so the main processor doesn't have to... [laughs] Even now they're putting arms inside of it. It's a constant battle.

Carl: I think one of the most interesting things about battery technology today, and one of Apple's unique assets here, is this ability to design their own silicon. They have control over all the pieces in the device. I actually wrote some research on this before I left the analyst community. I looked at all the different types of phones, how long they lasted, what the size of their batteries were, and I ran models on them, and all that sort of good stuff.

The most interesting thing is that Apple is in a unique position in that it can have the battery tell the phone to slow down. That is not a common function in most worlds. Mostly your battery acts as a slave to the master processor, and it just has to do whatever the processor asks for, and if it can't do that, then something happens.

The idea that the battery can actually go and say, "You know, I've got this much capacity. You might want to think about slowing things down or changing your consumption pattern so that we can continue," because otherwise, again, we're going to go to black out in about 30 seconds.

That feedback loop is actually really interesting. When I looked at the data, in fact, Apple has a huge advantage in terms of how many cycles they get out of milliwatt hour in their battery, simply because it's designed as a system. All the pieces talk to each other.

Rene: There's so much to break down in here. I have this problem that whenever I try to explain Apple, I get accused of being an apologist. I don't want to be overly sensitive to that, so that I stop explaining what's happening, but I just want to make it clear to everyone that we're trying to discuss the facts of this.

You are free to make any decision you want after that. If at the end of this, you think that Apple made absolutely the wrong choice for the wrong reasons, and just plain blew it, my only hope is that you're hating smart instead of hating dumb. That's the best result, for me, out of all of this.

Ben, there is a lot to do with sentiment here, where there is that perception that iPhones are slow. Whether it's right or wrong, there is that perception, and this plays into it.

Apple, very specifically, in their statement said that we're doing this to improve the user experience. The reaction to that was, "Apple's admitting that they're forced obsolescence." [laughs]

Ben: Again, if that's the case, then people wouldn't hold onto them for three years, like they do. People would be like, "Oh man, I can't use it."

This would be an interesting question, I don't know if you guys have benchmarked this before, but how does a iPhone -- like the ones that you guys tested -- age-wise, how does that benchmark in terms of performance against a phone of similar age? Just to say like, "It's slower, but is it still dramatically slower than the competitive product similar age-wise, or is it still actually faster than one, age-wise?"

Again, there are software updates that come along with that. There are all sorts of nuances in there that come along with that.

It's interesting to say like, "Well, how big of an impact is on the user experience?" Then, again, somebody who holds their phone for three to three-and-a-half years, how much do they do with technology? Are they really feeling the pain?

I think these are, again, interesting parts to think about when you think about how this is impacting the customer experience, but I'm not worried about this impacting sentiment.

The Note fires almost didn't even affect Samsung's sentiment around their customer.

Rene: Yeah, we have no attention span and huge brand affinity. [laughs]

Ben: Your phone can catch on fire, and your OK, then, OK, Apple is slowing down your phone a little bit once it gets over a certain age, and it's still usable. I just don't see that being the same issue.

Jerry: Agreed.

A great example is you brought up the Note 7. The fact that the storage facility that had all the recalled Note 7s actually caught fire because of those batteries. It's on the Web, you can find...somebody reported this, but it barely made a blip across all the tech blogs. I knew right then, "Well, this isn't a big deal."

Rene: [laughs] It seems like there's a few issues here to break down. One is that Apple always has to make choices.

Often, they're damned if they do, damned if they don't. I think that's fine. I think that's Apple's job is to make hard choices, and it's their job to deal with the fall out of those choices. They're worth almost a trillion dollars. They can absolutely handle this.

But the choice here was they were facing a situation where people would suffer from drain or shutdown, random or unexpected shutdown of the phones. Not random, it's really more unexpected from a user point of view. Or they could start doing things like power management. I think power management was never really well explained.

Carl, when Apple explains power management, and I did a poor job of explaining this too, people often understand it in the context of thermal. The phone can only run so fast, so hot for so long, before they have to start the processor down, or if it's outside in Arizona, it'll geek the screen, and then it'll give you that warning, for example, to protect the processor.

Apple widened that to apply to cold conditions, wear conditions, old battery conditions, spike conditions. It's not unheard of, in fact, it seems to me it's rather typical that processors are throttled in smart phones.

Carl: But the processor's usually throttled based upon other data, such as what applications are running. You might throttle based upon a sensor, as you were saying, but the idea that the batter controller has the ability to send a slow down is unique.

Is it a master/slave relationship or is it a discussion among peers, I think, is the real difference. In this case, it's more of a peer relationship. Since battery capacity is kind of the life blood of mobile, I think it was actually one of the smartest things they did early on when they sort of said, "You know, we're going to actually manage this in a way that treats it like the vital resource that it is."

I say, it turns out a little bit more like a peer, whereas in most other systems, and frankly there aren't that many other operating systems. But I used to work with the BlackBerry guys. They were paranoid about batter life, so they were one of the other companies that actually did everything possible to maximize battery life.

That was really getting through the day, as opposed to "I'm going to deal with old batteries." As my son reminded me earlier today, BlackBerry's advantage was, "Oh, just buy more batteries."

Rene: Yeah. [laughs]

Carl: That kind of went away when they started sealing the batteries inside the device.

They had to do something, and I say, I think they don't really have a peer at the moment in terms of how they manage their resource, that life blood of the mobile device.

Rene: What do you think about the decision they made here, Jerry? I think in some ways it was a decision that hurt them, because if your phone is draining or your phone is turning off, as a consumer, you'll think battery. But if your phone is ramping down to protect battery life, suddenly you're thinking performance.

Previously, and still with iPhone 5S, I would go to Apple, and I've actually done this, and say, "My battery's not working," and they test it, and they replace the battery. But it seems like people don't know that's the issue now.

Jerry: My only real problem with this gait, I hate that word, is that I used to build robots who sorted parts.

Rene: [laughs]

Jerry: The big pens, just random parts, and it was an amazing fun job, but you didn't have to be elegant. We use the word "elegant" with Apple a lot.

This is not an elegant solution and part of that is because nobody knew. Even if they could have known, they didn't remember seeing it, or they didn't ever see it when it first came to light a year ago. It's just, it feels very sketchy.

"Sketchy" and Apple get put together so often, unfairly, and I don't think it's fair in this case, but it just doesn't feel like an elegant solution, even though I can't argue with what they did. I kind of like what they did.

I stress the safety issue, because I did go through the Note 7, and it was 23 hours a day of non-stop back and forth with users and with Samsung, and trying to get everybody up to speed so they knew that their phone was or was not going to blow up on them. That was a nightmare.

Apple avoided this nightmare and I think that was really wise. It showed a lot of foresight to do it before it happened. It just feels sloppy.

Rene: Is that it, too, John? Because Apple added the disclosure. Since iOS 10.2.1, I think that was the right one, they added something to settings that would tell you your battery needs servicing.

You pointed out that on, I think it was your wife's phone, it never popped up, and a lot of other people are saying they never saw it either.

John: Right, the disclosure on my wife's phone, I've never seen, and to be fair, this is my old phone. I usually have the latest version of the iPhone and she has the previous version.

The thing that struck me with my wife's phone was that it's about two years old. It was purchased September 2015, so it's just over two years old. She had the random shut down issue back in November, December of last year. The 10.2.1 update fixed it, and we didn't realize it at the time. I just noticed she stopped complaining about it.

Then she came to me, actually after the Reddit post, and I was starting to dig into this, and she said, "My phone is really slow."

Looking at the Geekbench numbers for it, I mean it wasn't even flagship - people like to talk about how far ahead iOS is, and the iPhone compared to the Android handsets - but we're talking mid-tier, creeping into low-tier Android performance. It was noticeably sluggish. It was a really horrible experience.

Even today, I believe she's on 11.1.2, so fairly recent update. Going into settings, and looking under the battery, there's no notification that the battery's wrong, even though she's running at approximately 40 percent performance. The battery is in such a state and the processor is so limited that she's getting a fraction of the performance out of the phone that she should be.

I think if this was a discussion involving five-year-old phones and five-year-old batteries, then that would be one thing, but I think it's surprising since these issues started, just outside the year warranty period, I think that's where this discussion becomes sort of interesting and tricky because how long should we expect a phone to last?

Apple's fix, you know, as other folks have pointed out, Apple's fix is probably the best they could do, given the circumstance they're in, but should they be designing their phones differently? Should there be larger capacity batteries so they're not throttling the processor right away, relatively speaking, in terms of the lifetime of the device?

Two, the fact that Apple hasn't really been transparent about this issue. Those notifications aren't showing up in settings saying, "Hey, it's time to get your battery serviced," or something like that. People just sort of look at their phones and say, "It's slow."

It's only been since I published those charts that it's really been taking out of the tech sphere, talking about it in Reddit in anecdotal anecdotes, I guess, [laughs] is the best way of putting it, and actually being able to point to a dataset of hundreds of thousands of devices, saying, "No, this is a problem. This is what it's affecting," that it's sort of become a national conversation, and Apple has actually said, "Yeah, this is what we're doing."

I think if Apple had been a lot more transparent about it from the get go, it would have made a huge difference. Talking about the Note 7, Samsung got in front of it pretty quickly, saying, "Yes, there's a problem." They did the recalls and they seemed to have fixed that issue.

As others have mentioned, their perception has not really been changed.

For some people, it feels like Apple sat on this for a year and didn't say anything, even though they were slowing down phones. I think the reasons were good why they were doing that.

We can have a larger conversation as to pose this as a design fault with the iPhone, but I think the decision to introduce this limit was sensible. I think the decision not to talk about it at all, aside from a vague press briefing that really most iPhone users aren't going to hear about, I think that wasn't a wise move.

Rene: Ben, I want to throw it to you one more time before you have to go. To John's point, I wrote about this on December 12th, over a week ago, and nobody cared. Nobody picked it up. Nobody really commented on it. It went, I guess, just everybody's radar, and now it's this huge thing.

Apple did those briefings a year ago, and nobody seemed to notice. A lot of outlets reported it. I probably wasn't as clear. I didn't explain it as well as I could, but nobody really noticed it.

It seems like it's a confluence of cold weather coming again, which is what caused it last year, iOS 11, which added further burden to it, and the batteries being 11 months older. It's like people don't care until they feel the pain maybe.

Ben: Yeah. Again, I'm still sort of interested in what level of pain is being felt. The cold one was a unique one. They're having to manage this, like I said, in really interesting ways. Obviously there are things that they can do because they designed their own silicon and now that they'll do much more in the power management side they can be more efficient with that.

I think the reality is, especially somebody who holds onto their phone for that long, may not be your heaviest of users. While they stay at slow, again, it's not unusable. Again, I think the fact that they do last as long as they are...

That's one of the other things that I think is interesting as a solution because people are saying, "Well, just tell them you should replace your battery." At that point, if you told a grandma or your non-techy mother or mother-in-law, "Go replace your battery," they could freak out over that. They might just say, "I just need to buy a new phone."

Rene: Mine would just grab a screwdriver.

Ben: If you could. That just feels more like a forced upgrade, even though it's expensive to do so, than just letting you hold onto that phone and let your battery last. That's what you really want. It's really hard. Everybody agrees that Apple should be telling people.

The question is, when you've got a huge base of really non-techy, normal computers, how do you tell them this? What do you say? What's the right way to communicate this that they'll understand and not become anxious or worry about their phone if it's not needed?

Rene: I know you have to run. Ben Bajarin, you're @BenBajarin on Twitter?

Ben: @BenBajarin.

Rene: Where can people find your writing?

Ben: I write at Techopinions.com, which is the site that I run. Sometimes that shows there. Anyway, Twitter is the easiest way to find me. I talk about everything there.

Rene: Awesome. Thank you so much, Ben. I hope to have you back soon. Thank you for taking the time.

Ben: Yeah, my pleasure. Talk to you later.

Rene: John, you mentioned this. Jerry, I'm curious about your opinion too. I think it would have been better. Again, I don't think I did a particularly good job explaining what Apple was doing with 10.2.1. They certainly didn't give it a stage, for example.

I also think that, because people don't care about this stuff until it affects them, either nobody saw it back then or nobody remembers it. There needs to be some sort of persistent disclosure for it. Maybe it wasn't implemented the best way possible.

I know at AnandTech they mentioned that maybe it was just a dumb counter and not something that was dynamically associated with the battery. I don't know if that's true or not. The fix and the implementation are often two different things, right?

John: Before we go too far into what AnandTech is saying, I think, actually, Andrei got it wrong. I think this is actually a little more sophisticated than just a simple counter. I, unfortunately, don't have the data to go out and track down Andrei and say, "No, this is wrong." It is something worth looking into.

I think, really, when they came out with the fix for 10.2.1 they had that obviously problem. They sort of substituted an obvious problem for a non-obvious limitation. They didn't clearly, at the time, say that. As I said, my wife's phone was being affected by these sudden shutdowns.

I thought it was because of some of the apps she was running so I went out and bought her a battery pack. With the coincide of the 10.2.1 release, the problem was solved, and I didn't really think about it again. I read Apple's disclosure at the time. I thought, "Oh, OK. That's interesting." I didn't put two and two together.

I don't want to toot my own horn here, but if I'm not putting two and two together, I think it's unreasonable to assume most users are going to put two and two together.

Jerry: Absolutely. That's what I meant by it was a sloppy way of doing it and something that Apple is better at when it comes to other things.

Putting out the press release that they did and talking about the way they monitor the battery and the operating system now compensates for the lack of capacity that older batteries have, that makes somebody like me interested. I'm like, "Oh, that's great," but most people, as we've said, they...

Rene: They wouldn't even read it.

Jerry: Right. If they did read it and understand, they wouldn't care until it affected them. I think you hit that right in the head. We know Apple can do better. It just bothers me when they don't do better. As I said, I like their solution.

Until they invest the same amount of money in LG CAM that they did in the display division and find additives that make batteries safer and last longer, you have to do something through a combination of custom hardware and software.

This was step one. It was a great solution at the time, and I'm glad that they're working on something even better. Just the way that it was done and put across to users really bothers me.

Rene: Carl, one of the things that is always weighed is they could have left it the way it was, which is how previous iPhones work and which is how a lot of other phones work. That is it will drain or shut off if it has a spike. They chose to do the throttling instead.

Apple does do updates for phones going back many years, but some people really don't like that. They find that it does slow down their phone or the CPU can't support all the new features they're adding.

But if Apple didn't offer updates for as many phones going as far back as they did, they'd probably be accused of deliberately withholding features to force people to upgrade. I think there's no neat way out of that narrative. If you're going to be hit by it, all you can do, maybe, is the best choices you can make.

Carl: Yeah, it's kind of a lose/lose. They can do what they always did, and then people will criticize them for their phones shutting off. There was quite a to-do about that. Then if you throttle them, then people who run benchmarks say, "Oh, my phone is being throttled." I'm not sure most people actually care about benchmarks.

Nonetheless, you can't really win at that. This idea that, "Oh, if they were only more transparent, life would be better," I'm not quite sure what the form of that transparency would be. There's lots of disclaimers in the little safety booklet you get with your phone. Nobody ever reads those, so that's not going to work.

If I can make an analogy to some other industries, I used to own a fairly high-end luxury car. Being the geek I am, I would read the manual and find out things about the car that I think a lot of people don't realize.

One of the things I discovered is that under some conditions the onboard computer can detect a fault, which basically suggests that there's something wrong with the car or some part is aged out of lifetime or whatever, and it goes into limp home mode.

Limp home mode is one that doesn't let you go over, I think, 30 or 35 miles an hour. It's probably in kilometers. Basically, it cripples your car and puts it in the state where the only thing you can do is drive it to get it repaired.

This is kind of the same story. Not widely publicized, and only geeks who read the manuals knew about it, but I'm sure people who ran into it by accident because of a failure probably weren't terribly happy. For some reason, though, Apple gets held to the standard that, "Oh, we can't do what everyone else does. We have to have some magical unicorn fly down and fix your phone for you."

As I say, I'm not sure what I'd do. It does seem like a lose/lose. They're clearly damned if they go the blackout route. They're damned if they go the slow down route. Yes, they could design "better devices." Unfortunately, they'd be bigger devices is usually the way you end up there.

Of course, they've made a lot of their reputation and people love them because they don't make giant dinner plates that you hold to your ear.

Rene: That's one of the things that people say. "This wouldn't happen if Apple just doubled the battery size." Even if we think about a world where lithium ion is perfect, it does not affect RF and it does not affect thermal, you can just pump it in like you did frosting, for example, you would still add weight.

Ben is not here anymore for his study, sadly, but weight is a usability issue. People often talk about the thinness of a device, but the thinness is a reaction, a side effect of the lack of weight, the lightness of the device.

That absolutely plays into the appeal because, first of all, big, heavy devices traditionally don't sell as well. People go into the store. They pick up two phones. If one is super heavy, they just leave it on the counter. They won't even entertain it because, a couple Nokia fans aside, they don't want to carry huge bricks around with them. [laughs]

Also, if it's really heavy, you can't hold it up for a long period of time. People even complained about Galaxy Notes and iPhone Pluses that if they're on an airplane or at a doctor's office and they're reading books, watching movies, and playing video games that they can't hold it up for as long as they want to.

It doesn't really matter how good their battery life is because you've hit a usability road block. None of these things are easy from an engineering point of view, but again, I'm super happy that Apple is held to a ridiculously high standard. I wish every company was.

I remember on Android Central, Jerry, when we had the Galaxy Note 7 thing, the worst reaction we got was Samsung people, people who loved Samsung, telling us to shut the hell up and to give them back their phones. [laughs] It was almost the inverse of what we're seeing. [laughs]

Jerry: Yeah, I still have people that I hear have kept their Note 7 and have found ways to keep it running and have altered the software so the carriers can't identify it. I think that's insane.

Rene: John, you mentioned this earlier, some of the design choices in these current generation SoCs and how that is maybe influencing some of the choices you have to make later.

John: It could be that with Apple's obsession with single core performance. You see now with the iPhone X you're getting basically getting a laptop in a cell phone factor. With that computing performance comes a very high power draw.

You don't see this sort of performance on Android phones, which I think is part of the reason why you don't see this issue on Android phones. The iPhones tax the battery to a degree that just...With a battery this size, I'm not surprised now, in hindsight, seeing that they're having this issue where the battery is just simply not able to keep up with the processor after 16, 18, 24 months of use.

What I'm left wondering is, "There are being tradeoffs made. Are these the right tradeoffs they should be making?" That's a hard problem to solve and a hard question to ask because really only Apple knows what the sweet spot is they're targeting and how to best get about that.

Speaking from a personal perspective, I know with the phone that, going back to my wife's phone, seeing this sort of degradation within two years, it's disappointing. I understand that lithium ion isn't perfect and that batteries age, but to have a phone be at 40 percent of its performance, I don't really view that as acceptable.

I view this as really a shortcoming. I know people are saying, "Well, what can Apple do to notify users? You don't want to freak users out."

I think there's a very happy middle ground between an in-your-face alert that says, "Hey, go to the nearest Apple store immediately," and alert that, if a user notices there is a problem and they're trying to figure out that problem, then they can, just by poking around settings and having that discreet message say, "Your battery may need to be serviced."

I think there's a happy medium there that could be explored because right now, up until we published that article earlier this week, we had our users coming to us and saying, "My phone was slow, so I downloaded a benchmark just to see if basically am I being crazy. Can I get some sort of objective measure of the performance of my phone?"

Low and behold, the benchmark confirms their suspicion. Then they come to us and say, "My phone is slow. What can I do?" We have no idea because we didn't build the phone. We didn't design the phone. We just kind of have to go off of whatever is with the sudden shutdown issue or other things like that.

We can kind of lean on what we've heard and distill that down to the user, but in this case until this happened, the analysis happened and Apple made their statement, we were guessing. We basically just said, "Contact Apple."

I know some people went to Apple and said, "My phone is slow. What's going on?" Apple ran their diagnostics and said, "No, everything is fine."

Rene: I bought two iPhone 6s because I originally bought a black one and then I had to buy a gold one because, of course, I had to buy a gold one. Jerry understands.

Jerry: Yes.

Rene: I've given those away. My sister has one of them. My mother has one of them. When this broke, I went and tested both of them, and they're both fine. I used them every day for half a year each. Actually, no, I lent a friend of mine one of them. They were both used consistently since almost launch day and then the week after launch day, and they're still being used every day. They're absolutely fine.

It might be that just the battery health is much better and the system of power management hasn't kicked in. I think that's part of the problem now. We have so many millions of devices in the field. It's almost like a butterfly effect. They're so susceptible to initial conditions and to millions of little variations. People have such different experiences with them.

John: I don't know if their level of communication is right. I don't think anybody does. One of the things we do know about Apple, though, is they're not afraid to be different. They're also not afraid to say, "And if you don't like our device, you should buy a different one maybe."

The research I did on batteries, I will point out the Android folks have much bigger batteries. As was pointed out, they're not running close to the limits of those batteries. Those are designed more for peak. You pay more in weight. You pay more in size. If that's really the thing that suits your usage best, buy one of those. There's no reason why you shouldn't.

Even in Apple's own line, you can get a bigger battery. You buy a plus phone. Those have bigger batteries. They're not old enough, I think, yet to really know whether this factor shows up there, but my guess is they're not going to have nearly the incidents that we're seeing in the normal iPhone 7, iPhone 8, iPhone X. That's because they just have bigger batteries and they're not running quite so close to the edge.

Jerry: Battery life, we tend to think of as a measurement of time, and it's not. It's a measurement of capacity. Charging cycles affect it. Temperatures affect it. A lot of things, the way you use your phone affects your battery capacity.

Rene: Now quick charge, Jerry, right?

Jerry: Right.

Rene: Some of the quick charge technology.

Jerry: Oh, quick charge, I hate it. I wish it had never appeared. If you need a longer charge, leave it plugged in longer. Don't over charge it and excite the battery too much. That's a whole other podcast.

Rene: People are using quick charge over night when they're sleeping for eight hours anyway. [laughs]

Jerry: The person who plays Pokemon Go three days a week every week is going to have a very different experience with their iPhone 6S than you are, or I am, or any of us who work 40 plus hours at a desk and aren't out beating that battery up.

Rene: Or just Snapchatting all the time while they're walking around using geofilters and downloading media constantly.

Jerry: Right, that affects battery life, not the number of months.

Rene: The communication thing is always interesting to me, and I use this example, Jerry. I know you can relate to this. It's like when there was a brouhaha about iMessage, people switching to Android and not getting their messages and Apple should alert them. I had to point out what would the alert be. "Hi, have you switched away from iMessage today? Hi, have you switched away from iMessage today?"

You'd have to every day say, "No, I'm still using iMessage. No, I'm still using iMessage." Then if you ever did switch away, you would not see that message. It would have to wait like, "Oh, are you just on vacation? Is the phone in for servicing? Are you not using it for a few days? Oh, we switched off your iMessage."

Some of these things it's really easy to say how they should be handled. I made a couple of proposals. For example, I said I thought maybe Apple should let your phone shut down once and then when it boots back up give you a screen that said, "Your phone shut down unexpectedly. We can manage the power for you if you want to prevent this from happening again."

Then you press a button. You're taking to settings, and you have a toggle for it. It says, "This might affect performance." You're super aware with that, but I got a bunch of responses on Twitter of people saying, "Well, how is it going to know which way it shut down? Are people going to make informed decisions about this?

"Will they understand the performance? Will they just turn it off and then complain about the battery life or it turning off again?" These are often really, really complicated engineering problems to solve. I'm glad I don't have to make those decisions, John.

John: As an aside, I just pulled out my iPhone 6 and went to the battery thing. Actually, I hadn't known about that little thing. It does say, "Your iPhone battery may need to be serviced."


Jerry: I have a question. I haven't had an iPhone in a while. I have to admit that. Since I went to work for Android Central, my time is kind of wrapped around a Nexus or a Pixel phone. What does Apple charge you if you walk into an Apple store and your AppleCare is done, and you want the battery changed?

Rene: I believe it's $79 if you're out of warranty.

John: That's what I remember.

Jerry: I figured it was somewhere around $100, so it's even better than I thought. Why is this even a problem? You can spend $79 and get another two years out of the phone you like if that's what you want to do, or you can go to Google Play and play $900 and buy another Android phone. I don't see the controversy here.

John: I think one of the issues that some of our users experienced was that they went into Apple, sort of between the Reddit post and when this became a big thing, and they said, "My phone is slow." Apple simply flat out refused to change the battery, passed the in-store diagnostics, and they were basically turned away.


Rene: It varies by Genius Bar. One of those things is they have a wide discretion. If you're polite but firm, they can often do things for you. I don't know the right way to put this. It's not uniform across Apple stores. I would recommend go to a different Genius Bar or a different Apple store if you can or come back later, and you can often get those things done. [laughs]

John: Yeah, and if it's left up to the discretion, I have a feeling it's not going to be left up to discretion...


John: I think with all the attention this is attracting, they're going to have a slightly different policy. That was the issue that some of our users were saying. Simply, they knew what the fix was, and they couldn't do it.

Rene: It's 80 percent. If your battery is at 80 percent health or more, Apple, by default, they will not replace it. If it's under 80 percent, they will replace it. That's just the threshold that they've set internally.

Jerry: That's reasonable to me. I don't see a problem there unless your phone is only a month old and the battery capacity is only 80 percent.

Rene: Then they'll swap it immediately because they'll do it under AppleCare.

Jerry: That's hard for me to understand all of the uproar. I do admit that's probably because I have six phones on my desk and 15 on the shelf behind me. Grabbing another phone is not an issue, so I've forgotten what it is like to not have a phone that works well. I fully admit that, but there is a solution if this is bothering you and it's relatively cheap.

Rene: The last thing I want to address is this idea of...Because iPhone slow has been a meme for a long time, and it's multi factor. Like I said, sometimes it's just as simple as Google trends because every iPhone is updated at the same time everywhere in the world.

That's not the case with Android phones. Sometimes it's new features being added to an operating system. You'll play with those new features, like those iMessage bubbles or new Map transit directions or something, and then the screen is on longer. The radios are firing longer. You're putting more drain on the battery. But if you don't get those new features, then you think Apple is withholding them.

But if you get them, sometimes it does impact at least initially performance. When you first upgrade, there's a lot of...Spotlight is reindexing. A bunch of libraries are being moved around. There's a lot of overhead. It gets hot because all the radios are firing to update bit code for different...All sorts of things can happen.

There's a whole bunch of reasons why a person, if they don't really explore it deeply, can think Apple is slowing down my phone. But the idea of built in obsolescence, and John Gruber spoke to this well, has always seemed weird to me because Apple's one and only goal is to sell iPhones. You've already bought this iPhone. They want you to buy the next one.

Just as a company, their business is prioritized around selling you the next iPhone. They want to have as good an experience as possible. Obviously, they can't just give you a free iPhone to ensure that but as good an experience as possible that your next phone will be an iPhone.

At the engineering level, again John said this well, Apple engineers are not the highest paid in the Valley. They're there because they believe what Apple is doing is beneficial in many ways. If they were ever told to artificially slow down a phone, they would just start quitting.

Once I asked about this, I got these impassioned messages with curse words, Jerry, that you would appreciate going, "F no. We'd never do that. We'd riot. There'd be people walking down the streets." I think to an engineer it's borderline offensive to suggest that at least when engineers of that caliber that they would ever do anything harmful to their user base.

Jerry: Sure.

Rene: But we still have this narrative out there, and I think this story played into that narrative quite heavily.

Carl: I would quote that famous philosopher, Taylor Swift, who says, "Haters are going to hate."

Rene: [laughs] Go ahead, Jerry.

Jerry: I was going to say I remember the iPhone 4S, or was it the 4 in Siri?

Rene: 4S.

Jerry: 4S, it worked just great on phones that were jail broken, but all the sudden the official version was not able to get this new feature. Users rioted.

Apple has to give you more every time they update the OS, which causes this sort of thing.

Rene: This year, it was an Animoji or Portrait Lighting being withheld, when really it would just redline the processor in real-time.

Carl: One thing I will point out, that's kind of interesting to think about in the future. You are now in the era of the iPhone subscription. You can sign up to get an iPhone every year.

This may, if you want to get into conspiracy theories, you could argue that this is simply a way to avoid this problem altogether. If I can rely on you getting a new iPhone every year, I'm going to design the battery to probably...It'll last at least a year. It'll last longer than that, but they don't have as much motivation to make it last forever.

Rene: Well, they do want to resell it into the emerging markets, but still.

John: One can argue, when you trade it in, it then gets refurbished.

Rene: Yes. Oh, that's a good point.

John: It probably gets a new battery.

Jerry: I would really love to see you put that on Reddit so I could read the reactions to that theory.

Rene: [laughs]

Carl: Yes, I know, I would be roasted alive, but nonetheless, I do think one of the things we have to think about, if we analyze this market, is the concept that phones becoming a subscription has been a topic that's been bandied around for quite a while.

Apple is kind of the first company that actually did it.

Rene: Phone as a service.

Carl: Phone as a service, right.

Things may change in the future because of phone as a service. You may get different devices if you buy phone as a service. Hard to know, but interesting to speculate about how the industry could change.

Rene: Just to wrap up, I'd like to go around the room and see where you'd like Apple to go from here. From my part, I don't think I'd revert the system. I'd like to see it improve to the point where they can manage the communication between the battery and the processor well enough that people would see minimal impact, unless battery health was so degraded that there was nothing else they could do. At that point, there'd be a wicked obvious notification that you needed to take it in for servicing.

One would trip the other, so you couldn't have one without being absolutely aware of the other. I think that everything Apple is doing, they're already using machine learning, Jerry, to figure out which processes you're learning when and where to pre cache and all these things. I think we're heading there.

What about you, John?

John: It's hard for me to speculate because we very much focus on measuring performance. We don't do that much in building performance when it comes to hardware.

If I had my genie-in-a-bottle magic wish thing, I'd love to see Apple find a way to solve this, just as you said, where the impact on users was minimal. I know people like to sort of feel like, "Ah, bench marks, who runs them?"

Rene: [laughs]

John: For our users, a lot of people who came to us said, "I didn't use Geekbench until I thought I had a problem." They're not really concerned whether their phone is the fastest. They're concerned whether their phone is working well.

I think that's a very legitimate and very valid concern. If this sort of fix continues into the future, what I'd love to see is some way of minimizing the impact on the usability of the phone.

Rene: What about you, Jerry?

Jerry: I would like to see Apple solve this with hardware and not have to worry the user about it at all. They're building their own SoC completely, and they're doing a marvelous job.

Work with LG CAM, somebody, to build their own lithium-ion battery charge controller, tie that with the hardware, and control the way the CPU ramps its start, and underload and stretch the battery out as long as it can without affecting the user experience very much, and instead of an issue people are seeing with two-year-old phones, you don't see this until the phone is five or six years old. Then you can accept it.

Rene: I think they're doing a lot of that already, but I think it may not be to the point where they can manage those spikes yet. Maybe we can get there. Carl, what about you?

Carl: Yeah, I think Jerry is on the right track, but I'm not sure we're going to see huge leaps in battery performance. Batteries are chemistry. Chemistry is pretty well understood. It improves maybe a percent a year if you're lucky.

I think most of our improvements are going to come through software, through being more clever about how we manage that vital resource. I'm afraid that I think we're going to see more of this to manage the performance over time.

Jerry started, I think, with this is a safety issue. I think you have to think a little bit about if you're in a car in dark space, your tire is flat, and you're trying to call a tow truck, I think it's a lot better to have your phone run slow than to have it shut down.

Rene: Absolutely.

Carl: I think we're going to see more of this behavior of degraded performance, but I don't think we're going to see a wholesale revolution in battery life. I think probably the way out is the phone is a service and you just always use a new phone.

Rene: Thank you all so much for joining me. I know this was last minute, and I appreciate you spending the time with us. John, if people want to learn more about you, more about Geekbench, where can they go?

John: Geekbench.com. We like to keep it nice and simple. You can also follow me on Twitter at JFPoole, not that I tend to post much. A lot of it tends to be off-topic, but whenever we find something new and exciting I tend to post it there.

Rene: I've been following you for a long time, and I enjoy it thoroughly.

John: I'm glad somebody does.

Rene: [laughs] What about you, Jerry?

Jerry: You just come to my house, and we'll have a cookout.

Rene: [laughs]

Jerry: The best place to find me is Twitter. I'm on a sabbatical right now. Just too much politics. I can't stand it, so I had to take a break. It's @GBHIL. Like John, most of what I have to say is off topic, but I'm more than willing to jump back on topic with you if that's what you want.

Rene: Also you write super smart things at Android Central.

Jerry: Well, I try.

Rene: You do, you succeed. Carl, where can I find you?

Carl: I'm on Twitter @CDHowe. Again, I'm maybe going to be commenting more on big data, which is what I teach nowadays, but always happy to talk about Apple topics and certainly enjoyed my time doing that.

Rene: Awesome, instant follow. You can find me @ReneRitchie on Twitter. You can email me Rene@imore.com. Let me know what you think about this story. Please let me know your opinion. That's it. We're out.


Rene Ritchie

Rene Ritchie is one of the most respected Apple analysts in the business, reaching a combined audience of over 40 million readers a month. His YouTube channel, Vector, has over 90 thousand subscribers and 14 million views and his podcasts, including Debug, have been downloaded over 20 million times. He also regularly co-hosts MacBreak Weekly for the TWiT network and co-hosted CES Live! and Talk Mobile. Based in Montreal, Rene is a former director of product marketing, web developer, and graphic designer. He's authored several books and appeared on numerous television and radio segments to discuss Apple and the technology industry. When not working, he likes to cook, grapple, and spend time with his friends and family.