Live data from Hacker News

Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

text.npr.org

201–210 of 266 posts

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#201
post #108

Earlier quoted context omitted.

A high power workload on a phone might require ~5-10W over 100ms, i.e. ~0.5 joules. At 4V capacitor voltage, you would need ~30mF capacitor. Not likely to fit in a phone. This assumes you’re powering the phone solely off the cap. Even with a battery you’d still need something in the mF range to make up for the battery’s degradation.

The first data sheet I pulled up has an option for 47mF at 4.5V in 20x15x2.3mm of supercap. So it would fit if you really wanted.

Okay the reply got deleted but I already did all this math so I'm going to post anyway.

> [only able to use the cap from 4V to 3V]

> (C * 1V * 1V)/2

> [you'll need 1000mF]

Wrong equation. That's the energy from 0V to 1V. Each incremental volt stores more power than the last. 0V-3V is 9x that amount, and 0V-4V is 16x that amount, so 3V-4V is 7x that amount, and you'll need 140mF.

But also it shouldn't be connected directly in parallel to the battery; so I'd think of it more as 4.2V to 2V of supply. That would mean you need 74mF.

So put in two and you'll more than hit the target.

Even better, if you expect the battery to contribute some, you only need the one.

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#202
post #192

Earlier quoted context omitted.

Why would I want the phone to turn off sooner?

Why would I want phone with worse performance?

Presumably, you had a phone before your current one. If so, you evidently have a preference between slower phone and no phone at all.

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#203
post #192

Earlier quoted context omitted.

Why would I want phone with worse performance?

Presumably, you had a phone before your current one. If so, you evidently have a preference between slower phone and no phone at all .

Actually I had a desktop computer before my current phone. So my preference is faster, even if it spends more time on the cord.

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#204

Earlier quoted context omitted.

Hi, Battery engineer here at an electric vehicle company. The crux of it is this: when you draw high current from a small battery, the voltage will sag (1). The manufacturer has an on-board battery management system that MUST turn off the battery if it goes below a certain voltage threshold. As the battery ages, the sag effect increases significantly, up to 2-3x. Thus, when your phone clearly indicates say 20% batter…

> Thus, when your phone clearly indicates say 20% battery remaining, if a high load were to be applied, the last 20% would be inaccessible because the battery sag from the spontaneous high load would cause a reboot. So, I treat the battery readout as basically a scale of "high / medium / low". This works well with the empirical fact that a 20% readout might in fact last longer than a 50% readout. What I don't underst…

Disclaimer: I've had to deal with batteries a lot in my career so I'm familiar with their complex nature. I am not an expert though, so I'm sure there are a million more confounding variables that complicate things further.

> What I don't understand is that most other people seem to treat the percentages as meaningful. They certainly aren't based on "time until a recharge is needed", which is what people want. What are we notionally seeing percentages of?

So yes it kind of is based on "time until recharge". The problem is batteries are more complicated than that. I think a lot of people tend to view batteries as like a power supply that has a certain capacity you consume. This isn't an accurate view.

First capacity isn't static, it depends on temperature. Lithium batteries generally have thermal cut-off circuity so temperature compensation is pretty standard to add in.

Second batteries have internal resistances which affects its behavior. Internal resistances naturally rise as batteries age, or more accurately go through recharge cycles. This changes voltage sag behavior as GP noted.

Third, as GP mentioned, batteries don't supply a constant voltage. A lot of people realize battery voltage drops as a battery is used, which is true in a very simple sense. A cell can have a fully charged open circuit voltage for 4.2, but it quickly goes down to 3.7 where it spends most of its cycle around. Note that I specified an open circuit voltage. The second you apply a load the voltage will drop in relation to the load applied. This voltage sag and the subsequent recovery is not instantaneous either, it takes time to reach equilibrium states.

So in essence a battery percentage is simplistically portraying the capacity remaining in a battery, but isn't - nor can't - portray the fact that some of that capacity is unusable depending on loading conditions combined with safety cut-offs to protect the battery. Apple's solution was to change the load so there was less unusable capacity.

I'd like to re-iterate that I'm not an expert here. Batteries are complicated electrochemical devices.

Edit - Let me illustrate a hypothetical analogy. Let's say you have a car with a fuel tank and a poorly designed fuel pump. You fill up the tank with ten gallons, and start driving. The fuel pump heats up as you press on the accelerator, the harder you accelerate the more it heats up. However the fuel acts as a heat sink for the fuel pump. After driving for a while you have two gallons left. At this point you go full throttle and suddenly the engine cuts out because the thermal protection circuitry on the fuel pump kicked in. Your fuel gauge still reads two gallons, the problem is you simply can't consume it at full throttle because the pump won't operate under those conditions. You can still drive at maybe half throttle though.

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#205
post #176

Earlier quoted context omitted.

Apparently Apple doesn't agree that this is covered under warranty since they have denied wrongdoing (at least last I saw as of the $500M settlement, can't access the latest one).

Isn’t battery a consumable and thus not covered by warranty by default? (Unless it has a manufacturing defect)

Is it reasonable to consider an element of a device a consumable if it is not user replaceable?

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#206

Worth mentioning here is that this wasn't just about the "sneaky" fix for battery degradation, it was about doing this without notice , and having internal tooling that would deny a replacement (even at full retail price) if the battery was above an arbitrary health threshold (but still low enough to cause throttling). Before Apple changed their tune, if you read about this, and went into an Apple store for a replace…

Hi, Battery engineer here at an electric vehicle company. The crux of it is this: when you draw high current from a small battery, the voltage will sag (1). The manufacturer has an on-board battery management system that MUST turn off the battery if it goes below a certain voltage threshold. As the battery ages, the sag effect increases significantly, up to 2-3x. Thus, when your phone clearly indicates say 20% batter…

Look, you can continue to stan for Apple in the comments if you like, but even they admitted they were in the wrong to the tune of a hundred million dollars.

What's the old sales motto again .. "Sometimes the customer is actually right"?

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#207

Earlier quoted context omitted.

Hi, Battery engineer here at an electric vehicle company. The crux of it is this: when you draw high current from a small battery, the voltage will sag (1). The manufacturer has an on-board battery management system that MUST turn off the battery if it goes below a certain voltage threshold. As the battery ages, the sag effect increases significantly, up to 2-3x. Thus, when your phone clearly indicates say 20% batter…

Then they should show battery at 0% or 1%, so we can go and update the battery not the device itself. Instead they decided to throttle CPU silently. And oh it conveniently timed with new product launches and iOS version releases, so your then flagship device now "feels" sluggish compared to the new models.

It either "feels sluggish" or shuts itself down when the power draw peaks. Those are the only two options, and neither looks good compared to a new phone with a new battery.

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#208
post #207

Earlier quoted context omitted.

Then they should show battery at 0% or 1%, so we can go and update the battery not the device itself. Instead they decided to throttle CPU silently. And oh it conveniently timed with new product launches and iOS version releases, so your then flagship device now "feels" sluggish compared to the new models.

It either "feels sluggish" or shuts itself down when the power draw peaks. Those are the only two options, and neither looks good compared to a new phone with a new battery.

How about i realize my battery drains "quickly" now, i will go to the store & Apple advices to keep the phone but change the battery for 50 bucks. My device still runs as smooth as it was once the "flagship". But it didn't happen and my device is so slow now (as if the CPU is tired of running) and a new "flagship" is released with 2x the speed and feels so smooth but costs $1000. Hence the lawsuit.

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#209

Earlier quoted context omitted.

I agree that it's possible they weren't being malicious. The part that gets me is that they didn't decide to instead allow folks to set a max battery charge to extend their battery life. It's pretty widely accepted that charging a battery to 100% over-and-over will cause battery degradation. It seems fair to me that decisions that are hidden from consumers and obviously degrade the consumer's experience should negati…

It actually makes most sense that they weren't being malicious. Half the posts on this thread don't even understand what the problem is. Namely throttling to prevent the device from shutting off as if it was some kind of anti-consumer feature when it was actually pro-consumer.

I can’t speak for everyone on this thread, but the combination of “hiding” and “degrading” are what make this anti-consumer. If your car wouldn’t go over 30mph and the manufacturer wouldn’t disclose why, how would you feel?

From the article:

>Apple, the most valuable company in the world, acted deceptively by hiding the shutdown and slowdown issues, according to the court filing.

>"Many consumers decided that the only way to get improved performance was to purchase a newer-model iPhone from Apple," Arizona Attorney General Mark Brnovich wrote...

Re: Apple Agrees to Pay $113M to Settle 'Batterygate' Case over iPhone Slowdowns

#210
post #114

I think it's clear Apple intentionally* degrades software quality on older iphone/mac models in order to drive purchase of newer hardware. I've been considering switching away from Apple hardware, not because of hardware quality, but because of software. * By pushing software updates optimized to new hardware with obvious regressions on older models.

But this case shows the exact opposite. They updated phones to last longer without unexpected shut-downs. If they didn't do that update, many people who experienced these shut-downs outside of warrenty would change their old phone for a new ones. You can usually live with a slightly slower phone when the battery is low, but a phone that randomly dies on you? Not so much.

To be clear, Apple still messed up, because they didn't communicate this at all. They could have also put in a slightly larger battery to begin with.

But in the tech world, Apple isn't the worst in pushing people to upgrade hardware, which is illustrated by the fact that people use iPhones longer before upgrading than Android users. That's the one upside of Apples crazy App Store margins... they really don't need to push new hardware on you to make a shitload of money (while for Android that money goes to Google rather than hardware manufacturers)

Apple could do better - WAY better - but I think it's a bit naive to think that this could possibly change, and is worth complaining about. The only way it would, if there was somehow significant money in spending lots of time optimizing OS updates for old devices. But most people like upgrading their phones anyway. For Apple to change, consumers must change first. I don't think there's much hope for government regulation here either. It's a problem that's too complex to regulate.

Post reply on HN