News for hackers. Stuff that matters.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by bgentry on Monday March 02, 2020 @01:38PM (#22467306)
It's shocking to me that even in a technical audience like HN, so many people don't understand the fairly simple story that you've outlined here. Look around this thread and see how many people spread the narrative like "Apple intentionally slowed phones down" or even "Apple tried to extend battery life at the cost of performance". Reiterating the summary:

- Some iPhones in this generation began randomly shutting down as their batteries aged.

- While looking into this, Apple determined it happened because certain aged batteries could no longer deliver enough current to sustain peak workloads (computers use more current when performing more intensive tasks). This is probably due to Apple underprovisioning the battery capacity in this generation, but that's not something they could fix in existing devices.

- Apple added a mechanism to detect when these random shutdowns occurred on a specific device due to insufficient power from the battery.

- Once the random shutdown had been detected on a specific device, that specific phone would instead throttle its CPU when the system was under heavy load to prevent the random shutdown from occurring.

In short, your phone was not affected unless your battery had already degraded and was experiencing power-related random shutdowns. The changes Apple made to deal with this were an unequivocally user-friendly change to work around a hardware shortcoming and keep peoples' phones working well for longer than they otherwise would have.

The mistakes Apple made were:

- Designing the phone hardware with barely enough battery capacity, such that some moderately old phones might not be able to keep working properly when the batteries decayed.

- Not being transparent with the hardware design flaw above, such as proactively replacing any customer whose device was being affected by this flaw while still under warranty.

- Not going far enough with the software changes to initially add something like the battery health info they later added in response to this PR fiasco.

That's about it, I think?

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by chooseaname on Monday March 02, 2020 @03:24PM (#22468520)
> The mistakes Apple made were:

> - Designing the phone hardware with barely enough battery capacity, such that some moderately old phones might not be able to keep working properly when the batteries decayed.

Indeed! This was (and still is) very much a design flaw. People like to say, all phones have this issue, but where are the examples? Also, look at the iPads which are just scaled up iPhones. They don't have this issue because they don't suffer the same design flaw.

Apple put a bandage on the issue and people throw praise at them.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by manfredo on Monday March 02, 2020 @03:36PM (#22468653)
This is like saying my car's gas tank is a design flaw because I want to drive farther than my tank will allow. Increasing battery capacity means increasing the volume of the battery, which results in a phone that isn't as sleek. This isn't a design flaw, this is a tradeoff between form factor and battery capacity - just like how car designers have to make a tradeoff when selecting a fuel tank capacity.

The examples are, literally, every device that uses a lithium ion battery. Take any phone or tablet, repeatedly charge and discharge the battery fully (better yet, do so in a hot climate) and eventually it will not put out sufficient voltage. This will either trigger CPU throttling, or your device will start crashing. The iPhone was the most widely publicized example, but this is a fundamental issue with battery chemistry that all battery powered devices are subject to.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by chooseaname on Monday March 02, 2020 @03:43PM (#22468718)
Other phones don't have the issue because they didn't match a tiny battery with an over sized CPU. To use your analogy (which still isn't adequate at explaining the issue), this is like paring a top fuel dragster with a tiny fuel tank.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by manfredo on Monday March 02, 2020 @04:33PM (#22469267)
> Other phones don't have the issue because they didn't match a tiny battery with an over sized CPU.

It's not a question of if it's a question of when. All battery powered devices are subject to the limits of battery chemistry. Some phones have a bigger safety margin than others, but use the battery for long enough and it will put out less voltage than the device requires. As other commenters pointed out, other phones do indeed hit this limit.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by chooseaname on Monday March 02, 2020 @10:13PM (#22471478)
> It's not a question of if it's a question of when.

That's a good point and Apple's phones failed much, much sooner than any other phones. Thus the reason we don't hear of the others.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by manfredo on Monday March 02, 2020 @11:58PM (#22471911)
Plenty of cheap <$300 phones crap out after a year or two of use. You don't hear of them because they're not Apple.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by chooseaname on Tuesday March 03, 2020 @08:50AM (#22474389)
Anecdotally, I have several of those cheap ones that are still going great. (moto g 1st gen for example). They make great music/video/game players for the kids.

Apple designed about a years worth of battery capacity into a phone most people expected to keep 2 or more years. And clearly they put too powerful CPUs into the phones because they "fix" is to throttle them!

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by oflannabhra on Monday March 02, 2020 @04:07PM (#22468985)
Actually, lots of Android phones, including Nexus phones, have this problem, but none of them had a software fix shipped.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by davidgould on Monday March 02, 2020 @07:54PM (#22470792)
My Android phone started crashing mysteriously under load. I was so fed up after a while I switched to iPhone. I now assume that the degraded battery without throttling was responsible for the crashes.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by edmundsauto on Monday March 02, 2020 @03:59PM (#22468912)
There is no correct answer here (no arguing matters of taste), but consumers seem to have voted for the sleek design.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by chooseaname on Monday March 02, 2020 @10:16PM (#22471492)
No, they didn't. Apple never gave a choice. If you want an Apple iPhone you get what Apple wants you to have.

The non-Apple choice is ... well, we all know the trade-offs there.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by gentleman11 on Monday March 02, 2020 @04:50PM (#22469429)
We did not. I can’t buy an iPhone 5 anymore that can still run iOS properly, so I had to get one of these huge but thin things
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by colejohnson66 on Monday March 02, 2020 @05:10PM (#22469577)
Apples newer phones are actually thicker. They stopped the “make it thinner each year” thing.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by dilippkumar on Monday March 02, 2020 @10:43PM (#22471607)
> People like to say, all phones have this issue, but where are the examples?

Are there any other phones out there that are supported with software updates as long as the iPhone? If there is an Android phone that's been used for 3+ years, it will be ba good data point.

Afaik, Android phone OEMs don't try to make a device last as long.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by to11mtm on Monday March 02, 2020 @06:16PM (#22470135)
I've had Android and Windows phones in the past where replacing the battery fixed random app crashes/reboots.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by SenHeng on Monday March 02, 2020 @10:52PM (#22471649)
I had a 2nd gen Nexus 7 that after 1.5 years of use would intermittenly shut down even though battery capacity was around 60%.

It's the same issue.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by donarb on Monday March 02, 2020 @02:22PM (#22467812)
Anyone who has ever owned a smartphone knows that the battery degrades over time, the battery drains faster requiring more frequent charges. So besides the slowness, a user would already know that the battery was getting old. Why in the world would they buy a new phone when they could just buy a new battery?
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by AlisdairO on Tuesday March 03, 2020 @10:12PM (#22480900)
Because most people didn't realise that an older battery would cause their phone to slow down with zero notification. They just see that they have an old device that's slow, and assume it's too old to perform well with modern software.

The experience of old hardware feeling slow after a bunch of use and software updates is common. The experience of hardware slowing down with no notification when the battery is getting old is not. I'd actively prefer random shutdowns over non-notified slowdowns-at least the former indicates that tech support is required.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by klingonopera on Monday March 02, 2020 @01:51PM (#22467461)
> it happened because certain aged batteries could no longer deliver enough current to sustain peak workloads

Where exactly is the difference in this to a new battery that's not fully charged?

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by cdumler on Monday March 02, 2020 @02:26PM (#22467846)
There is a difference between energy density and power density. Energy density is how much charge is stored in the battery. Power density is how much electricity can be delivered per period of time at a point of charge.

A battery may have 2,000 milliwatts of charge, but can only deliver 500 milliwatts per hour; thus, it could run at full blast for ~4 hours or half at ~8 hours if the power density is not dependent on the charge. Batteries normally have a power curve, the amount of power you can draw depending on the charge in the battery. When you over draw the power on a battery, the tendency is for the voltage to quickly drop.

Cycling the battery and age both play in lowering the energy and power densities, but they are not one-to-one correlated. This is why you'll see people who say that their battery will last to 40% and die. The battery's voltage is used to gauge how much charge is left in the battery; however, at that level, the power density has dropped such that any significant draw on the battery causes the voltage to plummet and the phone shuts off.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by lopmotr on Monday March 02, 2020 @07:03PM (#22470484)
> 2,000 milliwatts of charge > 500 milliwatts per hour

It's bad enough journalists confusing watts and power and energy but better not in the middle of a post explaining the difference between energy and power!

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by thebruce87m on Monday March 02, 2020 @02:41PM (#22468045)
My understanding is that the internal resistance of the battery increases as the battery ages.

You basically have a voltage divider - with some of the resistance being the load (phone electronics) and some being the internal resistance of the battery. The voltage from the battery is split across these resistances. Normally the load gets almost all of the voltage.

When the battery is old and the peak current demand comes through the battery “sees” this as a lower resistance. The larger internal resistance causes a bigger voltage drop internally in the battery and the load therefore gets a lower voltage. When this is lower than the phone needs you get a brownout/reset.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by peterwoerner on Monday March 02, 2020 @03:21PM (#22468497)
Resistance increases due to microcracking (fatigue) of the copper leads from thermal cycling.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by lopmotr on Monday March 02, 2020 @10:51PM (#22471644)
That can't possibly be true. There are only be on the order of 1000 day/night charge/use cycles in the life of a battery and that number is too small for high-cycle fatigue to appear. Low cycle fatigue can happen but only with extreme strains where it's obviously deformed.

Also, if you could make batteries last longer just by using more finely stranded wire, somebody would have done that.

Not only that, but a crack doesn't lower resistance by much because the length of the current path is so short that the narrower cross-section hardly affects resistance over the whole length of a conductor.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by davidgould on Monday March 02, 2020 @07:57PM (#22470812)
Is really the case? an you link something so I can find out more?
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by esmi on Tuesday March 03, 2020 @01:09AM (#22472237)
That shouldn’t be that case for a well designed battery pack.

An unavoidable issue with all lithium ion batteries is plating during charging. This process increases the output impedance of the battery reducing its ability to deliver power to a load, regardless of how much energy is inside it.

https://www.sciencedaily.com/releases/2014/09/140903105638.h...

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by TheSoftwareGuy on Monday March 02, 2020 @01:58PM (#22467547)
the difference is that a brand new battery would be able to deliver enough current to sustain the peak workloads, even when at 1% battery remaining. An aged battery would not be able to do this.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by clairity on Monday March 02, 2020 @02:32PM (#22467926)
my experience contradicts the notion that random shutdowns are directly correlated to battery degradation.

just over a year ago, the then 3-year-old battery on my 6s was experiencing very occasional instant shutdowns, usually when the battery indicator was under 10%.

so i got a new (presumably undegraded) battery through the free replacement program, thinking it would improve things, but random shutdowns are worse and now my battery indicator goes to 1% for hours at a time.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by mark13 on Monday March 02, 2020 @04:39PM (#22469335)
This was my experience too with a 6s. Even with a brand new battery I'd get maybe 3 hours of use out of my phone (and I'm a pretty light user) and furthermore it would often shut down registering as much as 30% remaining. I bit the bullet and upgraded to a new iPhone 11 a few months ago, and I now regularly end the day with 70% battery life, with no change to my phone habits.

So, at least anecdotally, there's more to the story here....

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by fsh on Monday March 02, 2020 @03:53PM (#22468858)
This isn't about understanding the story, it's about whether you believe what Apple is saying. Is there any proof that the original secret throttling mechanism actually analyzed battery health and wasn't just a timer?
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by ineedasername on Monday March 02, 2020 @06:44PM (#22470357)
Indeed. We don't know if Apple just throttled the CPU on every device, even those that might have been rarely used and still have plenty of battery life. But the main problem here is Apple's lack of transparency on the issue.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by tjoff on Monday March 02, 2020 @02:26PM (#22467849)
What about intent?

They covertly slow down your phone yo mask that it is defective and needs a new battery.

This way they avoid having to deal with warranty and bad press.

Or design a phone with removable battery or at least a decent battery.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by manfredo on Monday March 02, 2020 @03:40PM (#22468688)
> They covertly slow down your phone yo mask that it is defective and needs a new battery.

So Apple should have juts let iPhones with old batteries become unusable?

> Or design a phone with removable battery or at least a decent battery.

Increasing battery size means increasing phone size - and Apple wants the iPhone to be sleek. Removable batteries also have certain design implications that increase the bulk of the device. Apple's products are at the top end of the market, where form factor is more important than serviceability. Mercedes' are more laborious to service than Toyotas. iPhones are more laborious to service than cheaper devices.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by tjoff on Monday March 02, 2020 @04:16PM (#22469088)
It is a faulty device, they should be honest about that.

A better battery would also lead to a phone that lasts longer. And that's bad if you are apple.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by manfredo on Monday March 02, 2020 @04:30PM (#22469238)
A "better battery" would have been a larger battery, which makes the phone bulkier. This isn't a good or bad choice, it's a tradeoff between form factor and battery life. This is a good choice for customers that want bulkier phones with longer battery life. But it would be a bad choice for customers that want a slim phone.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by zepto on Tuesday March 03, 2020 @11:42AM (#22476090)
The software fix also leads to a phone that lasts longer.

Bad for Apple? No.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by tjoff on Tuesday March 03, 2020 @01:11PM (#22477102)
If you consider the phone usable, sure... But the number one reason to buy a new phone over an old one is because it is slow.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by zepto on Tuesday March 03, 2020 @07:48PM (#22480233)
Sure but that’s an effect of new phones being a lot faster because of advances.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by tjoff on Thursday March 05, 2020 @05:27AM (#22492423)
That is a very charitable viewpoint in the context where apple are accused of slowing down devices.
Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by pergadad on Monday March 02, 2020 @02:15PM (#22467727)
I'm wondering where this story/version of the story comes from. What you're describing I've not heard anywhere else and I can't see any reference. Looking as an outside observer there is no way to tell whether this is just some story apple launched to make it sound reasonable/escape the lawsuits, is fully made up by someone else, or is actually the truth. From what I've read this issue was affecting all smartphones of certain generations, not just specific ones.

In any case the way they went about it and just pretended all is fine certainly is worth investigating. And as shifty as Apple has been in the past with such things - see planned obsolence, lack of possibility to replace parts/batteries, the throw-away airpods that even apple itself can't fix - it just does not sound to me like apple would really make user friendly decisions. Their only concern has (from all the evidence I've seen) been to protect profits. Even when apple was forced to do battery replacements they still charged for it.

Therefore I honestly doubt this story unless some third party can corroborate it.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by powowowow on Monday March 02, 2020 @02:37PM (#22467997)
The 'Battery Health' screen shows the device specific aspect of the implementation.

There's a section titled 'Peak Performance Capability' which will say "This iPhone has experienced an unexpected shutdown because the battery was unable to deliver the necessary peak power. Performance management has been applied to help prevent this from happening again." if that specific device had an unexpected shutdown related to battery health.

That's how it worked both before and after the controversy; the only thing that changed was that Apple added the battery health UI along with an option to disable the performance management features.

Re:Apple to pay up to $500M to settle U.S. lawsuit over slow... by mint2 on Monday March 02, 2020 @02:37PM (#22467988)
I have an iPhone se and it’s actually shut down once due to the battery and turned on the performance capping. Honestly I haven’t noticed a significant difference. But that’s trivia, my main point is next.

Batteries are consumable products. That they degrade over use is a given.

No phone company states the expected useful life of their batteries in a way people can understand or the expected loss of performance due to degradation.

I don’t fault Apple for this alone. Apple seems no worse than anyone else. The entire industry is shady when it comes to batteries.