Hacker News new | ask | show | jobs
by neohaven 2326 days ago
Imagine if, to avoid damage to engine parts, engines would keep their heat limits in safe zones to not blow up when friction goes up. Imagine if, when you don't tune up your car, it starts running, factually, with less horsepower.

You do lose X horsepower every 1000 miles. You get it back with your tuneup. Which involves changing the oil.

You lose X MHz every Y% of battery discharge capacity lost. You get it back with your tuneup. Which involves changing the battery.

You don't need to detune maliciously to get the effect Apple is getting here. The system just needs to self-calibrate to a depleting battery... Which is super good for the battery because it won't actually get worse as fast. Discharge cycles, especially close to the discharge limit, puts strain on cells, which have voltage sag closer to end of life.

3 comments

When a vehicle overheats and goes into limp mode the car tells you. Changing a vehicles oil does not increase the horsepower in a measurable way like the percent apple lopped off as batteries aged. If that was the case oil changes would be a much bigger deal.

Again, this is all semantics. The whole point is apple never told people they were throttling. And you've got two groups saying "Oh they are so nice, they did it so you could keep using your phone, how nice of them" and then the "my phone is slow I guess I need to buy a new one".

Did Apple turn the "check engine" light on?
I remember a few messages saying my battery was degraded/degrading and needed replacement a few months before my iPhone 6 started dying at 20+% battery because of battery issues. Post-patch, it was slow, but worked.
What's annoying about this particular issue is that the problem that users don't like isn't Apple's battery saving measure, or even being lied to, it's phone performance slowing over time. That is what users care about. The battery peak load degrading over time is a factor, but even with battery replacements phones will feel more sluggish over time because the NAND is a disposable part. Fill up a higher percentage of your total storage? Feels slow. Take videos for a year? Feels slow. There is no reasonable oil change for this and it's something that is not being talked about anywhere.
Is there any reference at all for this?

My iPhone X is 3 years old, and feels just as fast as when it was new.

Reference for what, exactly? There are no references that highlight the issue of storage performance in iPhones. I haven’t even found a review that looks at storage performance of iPhones in the past three generations. Brand new performance, not even performance over write cycle. That is the issue I’m raising.

If you want a reference that proves that NAND performance degrades over write cycles: then I will only list one, but there are hundreds, at least.

https://doi.org/10.1109/RELPHY.2008.4558857

Sure - entropy.

But is there any evidence that translates into an appreciable impact on device performance?

Perhaps it doesn’t, and this is why nobody is talking about it.