I always charge my iPhone/iPad overnight when low, back to 100%.
No problems and iPad is from 2015.
Another urban myth....
Same here. I quit worrying about batteries years ago.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
BEWARE! iOS 14.5.1 is Throttling!
- Thread starter skardvin
- Start date
- Sort by reaction score
Same here. I quit worrying about batteries years ago.
At this point I've tried just about every test I can, including force quitting everything, not force quitting, combining that with reboots, etc.
As I mentioned in my previous post, the one constant seems to be that a reboot fixes it, sometimes, but not always, and even if it does fix it, it may start to slow down while the phone is on, or it might not.
I'm going to guess that everyone reporting that a reboot, wipe/restore, or 14.6 beta fixed their problem didn't continue to check to see if it was still happening.
Another interesting thing I've noted is that if my phone is on all day and I check both the CPU and Compute (graphics) Geekbench scores, sometimes the CPU scores will remain at their proper level but usually the Compute score is about 200 points lower or more, but is occasionally at the right score range.
Are you running Geekbench tests under the same conditions, i.e. battery level, number of apps open in the background. network strength, or even some process at the iOS level which is not user controlled?
Normal usage of the iPhone under varying conditions will likely give a slightly different scores.
Normal usage of the iPhone under varying conditions will likely give a slightly different scores.
I've tried with all types of conditions. There's no rhyme or reason to it other than whether or not the phone was rebooted recently. But, if I get a "proper" score I'll follow it up immediately because if there is no issue, it should be the same immediately after, but since this bug or whateveer it is, the scores can drop even immediately after getting a proper score.
Also I'm ignoring scores that are only slightly off the average or slightly different, but I've used Geekbench for years and typically under nominal conditions I would never see a score more than 100 points lower than their "average" score. With this 'bug" I am seeing them dip well over a thousand points, and sometimes the dual core score is 50% of the average score.
This appears fixed on the 14.6RC.
As I mentioned before I managed to restore performance on 14.5.1 and it stayed at max performance as I didn’t reboot. Yesterday I installed the beta profile so i could get the 14.6RC due to the Bluetooth fix as my headphones were being stupid and when it rebooted after the profile install the performance tanked. 14.6 then downloaded and installed overnight and this morning I ran a Geekbench and my performance isn’t back to normal, in fact slightly faster than the average according to Geekbench.
As I mentioned before I managed to restore performance on 14.5.1 and it stayed at max performance as I didn’t reboot. Yesterday I installed the beta profile so i could get the 14.6RC due to the Bluetooth fix as my headphones were being stupid and when it rebooted after the profile install the performance tanked. 14.6 then downloaded and installed overnight and this morning I ran a Geekbench and my performance isn’t back to normal, in fact slightly faster than the average according to Geekbench.
I am curious, have you checked to be sure that optimized charging is disabled? The reason I ask is that it is enabled by default. If your battery is still at 100% all this time, you must have hit the battery lottery. Lol
why would you not turn on optimized charging? Lithium batteries are happiest when they are below 80% and that's why Apple implemented it. It minimizes the time the batteries sits at 100%. There's nothing wrong with leaving your phone plugged in overnight, but using optimized charging could potentially increase the number of cycles your battery is capable of.
I've had it on ever since it was available and it's never consistently worked for me. I always plug in my phone at the same time as well
I hear ya. On paper, the feature is awesome because it has the potential of increasing battery lifetime considerably, but in practice, it doesn't work as well as intended. I actually have my charger on a timer that shuts off at 7am. And if I forget one time to put it on the charger overnight or do anything random with my charging pattern for a couple of days, it may not work at all.
I wish they could have just implemented a "make sure it's charged by <user-chosen-time>" feature instead. Instead of this "iPhone will learn your patterns" crap because even though it's supposed to be "AI", it's a pretty stupid AI. For my electric car, I can set a simple schedule when I want the car charged by, and it works fine.
It’s true that implementing something like that, or even something like « stay at 80% between xx:xx to xx:xx » would be good haha
I don’t believe that it should, unless there is a bug. From my understanding, this only engages when charging to pause the charge at 80% until a pre-determined time dictated by the machine learning from the users charge history.
Ah, ok. I keep it on, but I've yet to see it ever actually do this.
Last edited:
Yes, it's disabled. I manually disable it every time an update re-enables it.
I really never had any issues with my iphone batteries before this feature. I just use it and charge it overnight. My phone is about 1.5 years old and still going strong. I usually trade in/sell my phones around the 2 year mark anyway.
Ok, so it seems now that if the phone boots and I get a "normal" (within 100 points of average) Geekbench score, it'll continue to stay that way unless I reboot the phone and it stops for some reason.
However, more concerning is the GPU "Compute" scores. Even if I'm getting consistently "normal" CPU scores, the GPU/Compute score is almost always lower, as in several hundred points or even several thousand lower than "normal". Sometimes it will initially be "correct", but if I follow it up with a second test immediately, it will be lower.
Would any of you care to run three Compute/GPU tests in a row and report your results?
The lower Compute score even when I'm getting correct CPU scores aligns with how my phone has been acting - still fast, but visibly choppy.
However, more concerning is the GPU "Compute" scores. Even if I'm getting consistently "normal" CPU scores, the GPU/Compute score is almost always lower, as in several hundred points or even several thousand lower than "normal". Sometimes it will initially be "correct", but if I follow it up with a second test immediately, it will be lower.
Would any of you care to run three Compute/GPU tests in a row and report your results?
The lower Compute score even when I'm getting correct CPU scores aligns with how my phone has been acting - still fast, but visibly choppy.
If i run the compute test one after another without pause, then the score will drop slightly for each test. But if i complete the first test then force close geekbench and run another 2nd test then the score remain consistent 9300++ for both test. So i guess something wrong in the apps
Honestly, just keep the feature on. There is absolutely no rationale for turning it off. At worst, it does nothing. At best, you may squeak some additional cycles out of the battery.
It's a proven fact that Lithium Ion batteries are more strained at higher state of charge. Keeping the battery at 80% can actually double the number of cycles of the battery's lifetime. Let's say the iPhone battery is rated at 500 full charging cycles if you charge to 100%. If you actually restricted the battery to 80%, you would get up to 1000 cycles! https://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries
So again, keep the feature on. It makes NO sense to turn it off unless it does something strange.
If your phone is 1.5 years and still going strong, it may be even stronger if you had the feature on.
The ONLY caveat for me is that the feature doesn't do anything more than half the time, but it doesn't affect my usage of the phone at all.
Doing an iTunes Synch - not backup - deletes it. I don't know that there is another way. I literally just finished doing this. Still leaves 18.5GB of Other on my 12 PM.
Unfortunately, for the optimized charging to actually work, one needs to be on a consistent charge schedule and it may take a few weeks for the AI to learn enough to engage the feature. It works on my end but that is because I normally go to bed around the same time and I only charge it overnight typically every other day.
After updating to iOS 14.6 RC i can confirm that the CPU throttling is fixed, at least for me. I can actually feel the difference in speed and animations. The whole phone feels more fluid than on iOS 14.5.1.
I also had one problem with 14.5 and 14.5.1 on my iPhone 12 Pro - ram management. Sometimes it felt like i was using a phone with 2gb of ram, in some cases it was even worse than using my iPhone X with 3gb of ram. The apps were constantly reloading and the phone kept like maybe 3-4 apps open. Now, after updating i think it's back to normal when i was on 14.3 and 14.4.
I also had one problem with 14.5 and 14.5.1 on my iPhone 12 Pro - ram management. Sometimes it felt like i was using a phone with 2gb of ram, in some cases it was even worse than using my iPhone X with 3gb of ram. The apps were constantly reloading and the phone kept like maybe 3-4 apps open. Now, after updating i think it's back to normal when i was on 14.3 and 14.4.