Apple's Alleged Throttling of Older iPhones With Degraded Batteries Causes Controversy

[I7guy]

Apple's probably hoping the issue will die down if they remain silent
[doublepost=1513340579][/doublepost]
There were also many Note 7 which didn't explode. Doesn't mean a recall wasn't needed and there was no issue

Sure you don’t know anything about what is going on except the 10 people in this thread.
[doublepost=1513341180][/doublepost]

What version of iOS?
Proof of Geekbench scores?

Show us some evidence or belt up.

Evidence is the senses I was given at birth. Next.
[doublepost=1513341278][/doublepost]

You don't even have a geekbench score as proof which is what the underlying issue is.

For a subset of people. Others are claimed no issue, which is why you can’t draw a firm conclusion, at least imo.

 

[dysamoria]

There was already scientific testing on the “slowing down your old phone when new ones are released” and the conclusions are clear: Apple does not slow down older phones. Period. There is no legitimate debate after those tests were conducted.

"Scientific"... ha ha ha ha... That PR article was all about GPU and CPU speeds, not about iOS performance, which is what people are REALLY complaining about that triggered that PR article to be produced. Without the intellectual honesty to address the bloating of the OS, why would you trust the conclusions of the article?
[doublepost=1513344778][/doublepost]

The difference is that computers ship with processors that don't throttle below their base clock. Intels turbo boost can bump clocks up if conditions allow, but you are assured of a baseline performance standard.

My understanding is that they throttle down when temps get high, which seems to be a well-known, but under-acknowledged issue with MacBooks (and iMacs?), because they're too small to dissipate the heat they generate while running at their proper speeds for a long time (working on a CPU/GPU heavy project). While this isn't related to battery (unless they're hot from charging), the point is that there is no assurance of any baseline performance.

 

[MagicBoy]

Evidence is the senses I was given at birth. Next.

That's the funniest thing I've heard today. That's not an independently verifiable fact.
[doublepost=1513345210][/doublepost]

my iPhone 6 has about the same scores as your low-end. with 95%+ i get around 1100...then it drops to between 600-839/840

what a difference. what was the price to replace the battery?
I was thinking of upgrading some time late next year though.

£79 UK.

Not cheap, but worth it for another couple of years out of the iPhone, and for the peace of mind of a job done properly with official parts. Removing a glued in battery without the proper tools wasn't worth the risk for me.

 

[dysamoria]

And of course the people drawing emojis for Apple aren’t battery engineers that have been pullled off the iPhone.

No, but Apple IS putting a ludicrous amount of resources into such pointless features as chat animations/emoji. Development time costs money, even if the feature is inane and pointless. It probably costs less, admittedly, to develop animated emoji that responds to user's faces than it does to engineer batteries, but the point is that all this inane crap still takes resources away from other necessary work (like, oh... FIXING the iOS 7 bugs STILL in the most current iOS today!!). When users see inane crap being added to iOS while suffering other problems that go without being properly resolved, FOR YEARS, or when system performance degrades, yeah, they can legitimately complain about Apple's development priorities.

 

[I7guy]

That's the funniest thing I've heard today. That's not an independently verifiable fact.
[doublepost=1513345210][/doublepost]

£79 UK.

Not cheap, but worth it for another couple of years out of the iPhone, and for the peace of mind of a job done properly with official parts. Removing a glued in battery without the proper tools wasn't worth the risk for me.

Since we’re into “verifiable”, all of this is certainly anecdotal at best. Whatever your opinion is not mine.

 

[MagicBoy]

Since we’re into “verifiable”, all of this is certainly anecdotal at best. Whatever your opinion is not mine.

£79 is a fact. I'll happily post the receipt from Apple to prove the cost, date, location and the parts used. I've already posted before and after GeekBench results.

 

[newellj]

This CPU throttling is happening to my iPhone 6, and I just had my battery replaced this past June 2017 (at an Apple Store). Using the "CPU DasherX" app, I note that with battery charge at 90%-100%, the CPU frequency is 1400 MHz, and drops to 1127 MHz when it hits 89%. Single/Multi GeekBench scores drop in proportion (almost) to the CPU throttle -- 1561/2689 to 1334/2292. I can definitely "feel" this 15% hit in performance.

It seems fairly obvious that they updated iOS to slowly throttle the CPU as the battery discharges and ages, which solves two problems... 1) Customer problem: prevent user phones from randomly shutting down at 30-40% 2) Apple problem: replacing 100 gazillion iPhone 6/6s batteries. Two design birds with one stone. Apple does have the best 1st-world-problem solvers in the world!
[doublepost=1513207591][/doublepost]
Yes, that's true on my iPhone 6 (w/ 6 month old battery). The CPU throttles as the battery discharges.

Battery capacity may be an issue. Battery charge level isn't, at least not on the iPhone 7. I ran GeekBench 4 on my iPhone 7 from ~85% charge to 17% charge this week and the GB results were essentially unchanged. Do you have a Coconut Battery or other measure of your iPhone 6's battery capacity?

 

[Tzerlag]

FWIW

Two two-year old iPhone 6S. One (A) supplied from the initial 6S battery recall (and lightly used) the other (B) built after the recall date range and moderately used.

Both fully charged.
According to battery life app A was at 1% wear, B was at 24% wear.
According to CPU Dasher64 A was at 1848 MHz, B is was 600 MHz.

Running various apps (YouTube. Mail, maps/directions, iMessage, StarWalk, weather, camera,...) I find no appreciable difference in performance.
With one exception, B was marginally slower to sync and run Gaia GPS.

I then simultaneously updated both to iOS 11.2.1
B was noticeably faster each step of the way and completed rebooting about 5-10 seconds faster than A.
That surprised me considering the CPU measurements.

According to CPU Dasher64 (and the opinion of many posters) iPhone B at 600 MHz is “crippled”.

I’ll admit the B battery doesn’t last as long, even on standby, as the A battery but for my purposes, I don’t see much degradation in actual performance. Granted, I didn’t try playing any games.

Also, while CPU Dasher64 is consistent, BatteryLife is not. The battery wear determination changes wildly on B, seemingly depending on overal charge and recent usage.

 

[Jmausmuc]

FWIW

Two two-year old iPhone 6S. One (A) supplied from the initial 6S battery recall (and lightly used) the other (B) built after the recall date range and moderately used.

Both fully charged.
According to battery life app A was at 1% wear, B was at 24% wear.
According to CPU Dasher64 A was at 1848 MHz, B is was 600 MHz.

Running various apps (YouTube. Mail, maps/directions, iMessage, StarWalk, weather, camera,...) I find no appreciable difference in performance.
With one exception, B was marginally slower to sync and run Gaia GPS.

I then simultaneously updated both to iOS 11.2.1
B was noticeably faster each step of the way and completed rebooting about 5-10 seconds faster than A.
That surprised me considering the CPU measurements.

According to CPU Dasher64 (and the opinion of many posters) iPhone B at 600 MHz is “crippled”.

I’ll admit the B battery doesn’t last as long, even on standby, as the A battery but for my purposes, I don’t see much degradation in actual performance. Granted, I didn’t try playing any games.

Also, while CPU Dasher64 is consistent, BatteryLife is not. The battery wear determination changes wildly on B, seemingly depending on overal charge and recent usage.

A Geekbench comparison would be very interesting...

 

[magbarn]

Battery capacity may be an issue. Battery charge level isn't, at least not on the iPhone 7. I ran GeekBench 4 on my iPhone 7 from ~85% charge to 17% charge this week and the GB results were essentially unchanged. Do you have a Coconut Battery or other measure of your iPhone 6's battery capacity?

It's not just capacity we're talking about here but how much amperage a battery can put out without dropping voltage significantly. As a lipo wears out, it's ability to handle load decreases. I have some lipo batteries that when I drain them slowly I can get the close to full capacity, but if I drain them at higher loads (that they're fully rated for when new), their voltage drops much quicker. If you want the technical term for this, it's IR or internal resistance.

 

[newellj]

It's not just capacity we're talking about here but how much amperage a battery can put out without dropping voltage significantly. As a lipo wears out, it's ability to handle load decreases. I have some lipo batteries that when I drain them slowly I can get the close to full capacity, but if I drain them at higher loads (that they're fully rated for when new), their voltage drops much quicker. If you want the technical term for this, it's IR or internal resistance.

That could certainly be the case. I was trying to make the point, not very clearly, that simple charge level isn't the problem, or at least it isn't with the iPhone 7.

 

[dumastudetto]

I just tested an old iPhone 6s Plus running the latest ioS 11.2.1 with a less than stellar battery and the CPU benchmarks were above what Geekbench says is average for the device.

So I don't think this is a real issue at all.

Nothing on iMore from Rene Richie or Daring Fireball so I think this is a bit of a fantasy dreamed up by someone on reddit.

 

[newellj]

FWIW

Two two-year old iPhone 6S. One (A) supplied from the initial 6S battery recall (and lightly used) the other (B) built after the recall date range and moderately used.

Both fully charged.
According to battery life app A was at 1% wear, B was at 24% wear.
According to CPU Dasher64 A was at 1848 MHz, B is was 600 MHz.

Running various apps (YouTube. Mail, maps/directions, iMessage, StarWalk, weather, camera,...) I find no appreciable difference in performance.
With one exception, B was marginally slower to sync and run Gaia GPS.

I then simultaneously updated both to iOS 11.2.1
B was noticeably faster each step of the way and completed rebooting about 5-10 seconds faster than A.
That surprised me considering the CPU measurements.

According to CPU Dasher64 (and the opinion of many posters) iPhone B at 600 MHz is “crippled”.

I’ll admit the B battery doesn’t last as long, even on standby, as the A battery but for my purposes, I don’t see much degradation in actual performance. Granted, I didn’t try playing any games.

Also, while CPU Dasher64 is consistent, BatteryLife is not. The battery wear determination changes wildly on B, seemingly depending on overal charge and recent usage.

Intersting. All I want to say here is that my experience with the battery wear metrics from BatteryLife have also varied very widely on the same device, and I have also seen that it correlates highly to overall charge. Coconut Battery seems to be a great deal more reliable.
[doublepost=1513363484][/doublepost]

I just tested an old iPhone 6s Plus running the latest ioS 11.2.1 with a less than stellar battery and the CPU benchmarks were above what Geekbench says is average for the device.

So I don't think this is a real issue at all.

Nothing on iMore from Rene Richie or Daring Fireball so I think this is a bit of a fantasy dreamed up by someone on reddit.

Wouldn't that be remarkable???

 

[magbarn]

I just tested an old iPhone 6s Plus running the latest ioS 11.2.1 with a less than stellar battery and the CPU benchmarks were above what Geekbench says is average for the device.

So I don't think this is a real issue at all.

Nothing on iMore from Rene Richie or Daring Fireball so I think this is a bit of a fantasy dreamed up by someone on reddit.

Kindly connect that 6S plus to any mac computer and post pics of the coconutbattery report on the phone. I would like to see how really bad the battery is on total available capacity. Some screenshots of the geekbench scores would also be nice.

 

[BugeyeSTI]

Battery capacity may be an issue. Battery charge level isn't, at least not on the iPhone 7. I ran GeekBench 4 on my iPhone 7 from ~85% charge to 17% charge this week and the GB results were essentially unchanged. Do you have a Coconut Battery or other measure of your iPhone 6's battery capacity?

My 2 year old 6S has identical GB results no matter what battery percentage also. My battery health is 93% on coconut battery...

 

[Jmausmuc]

My 2 year old 6S has identical GB results no matter what battery percentage also. My battery health is 93% on coconut battery...

93% is still pretty good...

 

[BugeyeSTI]

93% is still pretty good...

The phone has been rock solid, easily my favorite iPhone to date. Sucks to see others that have batteries that haven’t held up or are legitimately defective..

 

[JosephAW]

My friend's iPhone X battery life has been terrible. I showed him how to turn off Touch to wake, 3D Touch, and background applications. It's better now.

 

[newellj]

My 2 year old 6S has identical GB results no matter what battery percentage also. My battery health is 93% on coconut battery...

My wife's 6, which I thought was a year and a half old but maybe is 2 1/2 years old, reads 88% of design capacity and 739 cycles on CoconutBattery. I haven't tried to run GB on it.

My 7, purchased in October 2016, reads 98.4% of design capacity and 78 cycles. GB4 reports very consistent, very high CPU benchmarks all the way from 85% to 17% charge level.

 

[WannaGoMac]

My wife's 6, which I thought was a year and a half old but maybe is 2 1/2 years old, reads 88% of design capacity and 739 cycles on CoconutBattery. I haven't tried to run GB on it.

My 7, purchased in October 2016, reads 98.4% of design capacity and 78 cycles. GB4 reports very consistent, very high CPU benchmarks all the way from 85% to 17% charge level.

Your batteries are in very good shape. People reporting this issue claim the battery was far below yours in % design capacity. That is the common thread so far, not the current charge level. As the battery ability to store charge declines, iOS slows the CPU to compensate

 

[dumastudetto]

Kindly connect that 6S plus to any mac computer and post pics of the coconutbattery report on the phone. I would like to see how really bad the battery is on total available capacity. Some screenshots of the geekbench scores would also be nice.

I will later. I am confident the battery is in pretty poor health though based on how quickly it loses charge. It's well below 50% of peak performance.

 

[Jmausmuc]

My wife's 6, which I thought was a year and a half old but maybe is 2 1/2 years old, reads 88% of design capacity and 739 cycles on CoconutBattery. I haven't tried to run GB on it.

My 7, purchased in October 2016, reads 98.4% of design capacity and 78 cycles. GB4 reports very consistent, very high CPU benchmarks all the way from 85% to 17% charge level.

78 cycles in a year?

 

[newellj]

78 cycles in a year?

That's what it says. Fourteen months, actually, since it was bought in October and this is December. Daily driver, and used with AW since early October this year. The reason why is probably because I basically never turn on the cell radio, and Apple's power management for wifi and BT is excellent.

 

[alphaod]

That's strange. I have had the opposite experience with my Samsung hence why I haven't upgraded since my contract ended two months ago. The iPhone 5S I had before it though, was disturbingly slower towards the end of the 2 years I had it.

I'm surprised to hear this. I've had similar issues with every Samsung Note and Galaxy S I've owned (Note 1 to 5 and S to S7 Edge).

Currently I use my S7 Edge that I'm using day-to-day. The phone lags all the time when I'm switching apps, and every time I reboot it, it takes about 10 minutes before I can use it normally. It gets ridiculously slow when I try to update the apps on Google Play.

And I don't have any such problems with any of my other Android devices. I have several older ones that still run smooth even on newer OSes.

 

[newellj]

My wife's 6, which I thought was a year and a half old but maybe is 2 1/2 years old, reads 88% of design capacity and 739 cycles on CoconutBattery. I haven't tried to run GB on it.

My 7, purchased in October 2016, reads 98.4% of design capacity and 78 cycles. GB4 reports very consistent, very high CPU benchmarks all the way from 85% to 17% charge level.

For what it's worth, my wife's 6 turns out to be a December 2014 phone, 1090 days old, according to CB. So, considering its age and the number of cycles, 88% of original battery capacity is very respectable.