iPhone 7/Plus Geekbench score is lower than average with good battery health

[rui no onna]

People are getting this throttling thing entirely wrong.

Throttling basically reduces your phone's CPUs peak output. The thing is, no phone runs at 100% CPU 24x7. Almost 80% of time it's going to be running underclocked because texting or snapchat or facebook or Safari doesn't require the full power of the CPU. Throttled phone or not, this tasks would not be affected because CPU is never running at the full speed. It's the intensive apps like photo editing, video editing, gaming that would be affected because the CPU is "capped".

It's like driving a car, you don't utilize your cars 100% potential. You go pedal to the metal like once in a blue moon. So if that your car's engine RPM is capped at 5k, your normal driving which includes driving in 1500-2500 RPM range is not going to be affected.

Depends on how extreme the throttling goes. I've seen some who get throttled all the way down to 600 GB4 scores on the iPhone 6 and 6s. That would be similar to running an iPhone 5 on iOS 11 except even slower.

That said, agree, current throttled performance of iPhone 7 units is actually still pretty good. 2600 is around (unthrottled) iPhone 6s speed and is actually still faster than the iPad Air 2. If it were my phone, I'd just wait until it gets worse (~1500 single core) or December 2018 whichever comes first before I get the battery replaced at $29.

 

[Hal~9000]

That said, agree, current throttled performance of iPhone 7 units is actually still pretty good. 2600 is around (unthrottled) iPhone 6s speed and is actually still faster than the iPad Air 2.

Too bad that’s definitely not the type of performance that customers buying the 7 paid for

Glad I got an 8+ with 11.1.2 which isn’t infected with Apples throttling virus “feature” like the 7 and below. The 8+ works perfectly so I don’t plan on updating it any further beyond 11.1.2 since Apple said they were planning on bringing their “feature” to the 8/X in a future iOS update (like they did with the 7 and 11.2).

 

[C DM]

Can you elaborate on this? How do you calculate that the 60% of the battery life has been consumed when the capacity of the battery is down by 12%?

It seems that it's in reference to 60% of usable (stable) life, where full 100% of that is the first 20% of battery's overall health.

 

[rui no onna]

Too bad that’s definitely not the type of performance that customers buying the 7 paid for

Glad I got an 8+ with 11.1.2 which isn’t infected with Apples throttling virus “feature” like the 7 and below. The 8+ works perfectly so I don’t plan on updating it any further beyond 11.1.2 since Apple said they were planning on bringing their “feature” to the 8/X in a future iOS update (like they did with the 7 and 11.2).

It actually kinda is.

Most people probably only ever use the low power cores on the iPhone 7 and that's actually at iPad Air 2 speed (~1800). Most users are likely not aware that their iPhone 7 scores ~3500 peak in Geekbench 4. As long as the phone handles Twitter, Facebook, Instagram, SnapChat, etc just fine, they don't really care whether the phone scores 2600 or 3500 in GB4.

Heck, all my desktops and laptops run at ~1GHz almost all the time only ramping up to max speed (or Turbo if power and thermals allow) when video encoding. Even most of my PC games don't tax the CPU enough to run at full-bore. The GPU is usually a bottleneck.

Granted, I'm quite interested in seeing Metal and GFXBench scores for the throttled iPhone units. From what I read, the throttling due to battery slows down the GPU unlike Low Power mode which only affects the CPU.

However, regardless of perceived performance, this is definitely bad PR for Apple and even people who don't care about benchmarks and don't see any performance degradation might rethink buying another iPhone on their next upgrade.

 

[Hal~9000]

It actually kinda is.

Most people probably only ever use the low power cores on the iPhone 7 and that's actually at iPad Air 2 speed (~1800). Most users are likely not aware that their iPhone 7 scores ~3500 peak in Geekbench 4. As long as the phone handles Twitter, Facebook, Instagram, SnapChat, etc just fine, they don't really care whether the phone scores 2600 or 3500 in GB4.

Heck, all my desktops and laptops run at ~1GHz almost all the time only ramping up to max speed (or Turbo if power and thermals allow) when video encoding. Even most of my PC games don't tax the CPU enough to run at full-bore. The GPU is usually a bottleneck.

Granted, I'm quite interested in seeing Metal and GFXBench scores for the throttled iPhone units. From what I read, the throttling due to battery slows down the GPU unlike Low Power mode which only affects the CPU.

However, regardless of perceived performance, this is definitely bad PR for Apple and even people who don't care about benchmarks and don't see any performance degradation might rethink buying another iPhone on their next upgrade.

I’d have to disagree.

It’s not really a “kinda”. Either it is or it isn’t. For example, I payed for the A11 in my 8+, not the A10 of the iPhone 7. Even if the majority of my time is being powered by the low power cores... that doesn’t change the fact that I paid to have the speed of the A11 when I need it, and not throttled down to the speed of the A10. Trying to justify it by saying “well customers won’t even notice” doesn’t make it right to throttle a customers product to the point that the newest generation they paid for initially now has the performance of the previous generation without being crystal clear about it ahead of time.

Imagine I bought a hybrid car because they advertised it as averaging 65mpg... then a year later they put out a software update that lowers the efficiency down to 40mpg. Of course they could try and justify it and say “Oh well... most people don’t care about whether or not they average 40 or 65 mph. Most customers will be too dumb to realize it anyways!” That doesn’t make it right.

If you paid for a particular product, you should get said product. At the very least Apple needs to be upfront during iPhone keynotes about their chip performance and advise customers before they buy their newest products that while their newest chip is super fast... over time it will be just as slow as the previous generations unless they replace the battery before a certain degradation point.

 

[Hieveryone]

Next thing you know Apple may be doing ads like

“Have you been using your iPhone for 6 months? Or worse almost a year? It’s time to get an iPhone the way yours was when you paid through your nose a few months ago. Now selling iPhone at full speed at your local Apple store. See a genius today!”

 

[z970]

Next thing you know Apple may be doing ads like

“Have you been using your iPhone for 6 months? Or worse almost a year? It’s time to get an iPhone the way yours was when you paid through your nose a few months ago. Now selling iPhone at full speed at your local Apple store. See a genius today!”

*"Now selling iPhone at full speed for one more year before THAT gets throttled at your local Apple store."

 

[Ledsteplin]

Geekbench scores are pretty meaningless. A difference between 3295 and 2633 isn't much. And a lot depends on what you're doing on the phone. Your next test might show different. But it's certainly not worth the worry. If you have 88% battery life, you're in good shape. Apple has not slowed down your CPU.

 

[Hieveryone]

*"Now selling iPhone at full speed for one more year before THAT gets throttled at your local Apple store."

Hahahaha at the bottom it might have one those disclaimers like

“Subject To Throttling Real Soon. You Know How We Do”

 

[Newtons Apple]

Geekbench scores are pretty meaningless. A difference between 3295 and 2633 isn't much. And a lot depends on what you're doing on the phone. Your next test might show different. But it's certainly not worth the worry. If you have 88% battery life, you're in good shape. Apple has not slowed down your CPU.

This!

You can run 10 different CPU test apps and they will all give you different results. Same for testing the battery. Too many variables involved.

 

[Hal~9000]

This!

You can run 10 different CPU test apps and they will all give you different results. Same for testing the battery. Too many variables involved.

If the CPU test apps can’t be trusted, then Apple themselves should be 100% transparent and with the future iOS update they promised... come out with their own CPU test + Battery capacity diagnosis tools that are 100% consistent and show exactly what type of performance the customers iPhone is putting out.

No excuses Apple!

 

[Newtons Apple]

If the CPU test apps can’t be trusted, then Apple themselves should be 100% transparent and with the future iOS update they promised... come out with their own CPU test + Battery capacity diagnosis tools that are 100% consistent and show exactly what type of performance the customers iPhone is putting out.

No excuses Apple!

I agree 100%!

At least we would all be on the same page. Something that would track the battery from one day to another to get a true look at its performance.

Maybe Apple is working on something for us, time will tell.

 

[z970]

Hahahaha at the bottom it might have one those disclaimers like

“Subject To Throttling Real Soon. You Know How We Do”

Throttlegate.

 

[Hieveryone]

I’ve noticed the same thing. Different battery apps will give different results sometimes. If not all the time lol.

So it’s hard to know anything.

But technical stuff aside

As a regular dude putting up serious dollars

I don’t want my iPhone slow. Now it’s on apple to figure out how.

 

[rui no onna]

I’d have to disagree.

It’s not really a “kinda”. Either it is or it isn’t. For example, I payed for the A11 in my 8+, not the A10 of the iPhone 7. Even if the majority of my time is being powered by the low power cores... that doesn’t change the fact that I paid to have the speed of the A11 when I need it, and not throttled down to the speed of the A10. Trying to justify it by saying “well customers won’t even notice” doesn’t make it right to throttle a customers product to the point that the newest generation they paid for initially now has the performance of the previous generation without being crystal clear about it ahead of time.

Imagine I bought a hybrid car because they advertised it as averaging 65mpg... then a year later they put out a software update that lowers the efficiency down to 40mpg. Of course they could try and justify it and say “Oh well... most people don’t care about whether or not they average 40 or 65 mph. Most customers will be too dumb to realize it anyways!” That doesn’t make it right.

If you paid for a particular product, you should get said product. At the very least Apple needs to be upfront during iPhone keynotes about their chip performance and advise customers before they buy their newest products that while their newest chip is super fast... over time it will be just as slow as the previous generations unless they replace the battery before a certain degradation point.

How many Apple customers actually know what chipset runs on their iPhone? MacRumors forum members know and care but we tend to be more technically inclined. We are not the norm.

I highly doubt a significant portion of the 200 million iPhone buyers yearly chose the iPhone because of the processor. In my experience, people usually opt for the iPhone because of overall experience - ease of use, battery life, app support, ecosystem, customer service (#throttlegate, etc. withstanding, they usually have pretty good regular customer service). I reckon for majority of users, the only performance expectations are that the UI is sufficiently fluid and the device runs their apps quickly enough.

Now these performance expectations certainly weren't met in certain cases. An iPhone that's throttled enough to score just 1000 in GB4 would deliver a pretty poor experience and I believe there have been some 6/6+ and 6s/6s+ scores posted that have gone all the way down to 600. At what age and whether Apple diagnostics tested the battery as good, I have no idea. That's also something that needs to be addressed. If battery health causes a device to throttle, those batteries shouldn't be testing as good per Apple diagnostics.

Yes, I do agree, Apple needs to be more upfront about their power management and I'd like to have advanced power management in the settings (I use custom power profiles on my Windows laptops). Actually, I'd really prefer to have slightly thicker and heavier phones with higher capacity batteries. From the kernel distribution graph posted by Geekbench, there are already a significant number of iPhone 7's getting throttled after just one year (granted, that population's probably quite biased towards the technically inclined).

Mind, I did actually did buy the iPhone 7 because of the A10 Fusion. Not because of peak performance but because the High Efficiency cores would mean longer battery life. Still running iOS 10.3.3, though, so currently unaffected by #throttlegate.

 

[z970]

How many Apple customers actually know what chipset runs on their iPhone? MacRumors forum members know and care but we tend to be more technically inclined. We are not the norm.

I highly doubt a significant portion of the 200 million iPhone buyers yearly chose the iPhone because of the processor. In my experience, people usually opt for the iPhone because of overall experience - ease of use, battery life, app support, ecosystem, customer service (#throttlegate, etc. withstanding, they usually have pretty good regular customer service). I reckon for majority of users, the only performance expectations are that the UI is sufficiently fluid and the device runs their apps quickly enough. Now these performance expectations certainly weren't met in certain cases. An iPhone that's throttled enough to score just 1000 in GB4 would deliver a pretty poor experience and I believe there have been some 6/6+ and 6s/6s+ scores posted that have gone all the way down to 600. At what age and whether Apple diagnostics tested the battery as good, I have no idea. That's also something that needs to be addressed. If battery health causes a device to throttle, those batteries shouldn't be testing as good per Apple diagnostics.

Yes, I do agree, Apple needs to be more upfront about their power management and I'd like to have advanced power management in the settings (I use custom power profiles on my Windows laptops). Actually, I'd really prefer to have slightly thicker and heavier phones with higher capacity batteries. From the kernel distribution graph posted by Geekbench, there are already a significant number of iPhone 7's getting throttled after just one year (granted, that population's probably quite biased towards the technically inclined).

Mind, I did actually did buy the iPhone 7 because of the A10 Fusion. Not because of peak performance but because the High Efficiency cores would mean longer battery life. Still running iOS 10.3.3, though, so currently unaffected by #throttlegate.

Isn't 10.3.3 affected by #throttlegate? I'm pretty sure #throttlegate first happened in iOS 10 and it only got brought to light in iOS 11's life.

#throttlegate

 

[Hal~9000]

How many Apple customers actually know what chipset runs on their iPhone? MacRumors forum members know and care but we tend to be more technically inclined. We are not the norm.

I highly doubt a significant portion of the 200 million iPhone buyers yearly chose the iPhone because of the processor. In my experience, people usually opt for the iPhone because of overall experience - ease of use, battery life, app support, ecosystem, customer service (#throttlegate, etc. withstanding, they usually have pretty good regular customer service). I reckon for majority of users, the only performance expectations are that the UI is sufficiently fluid and the device runs their apps quickly enough. Now these performance expectations certainly weren't met in certain cases. An iPhone that's throttled enough to score just 1000 in GB4 would deliver a pretty poor experience and I believe there have been some 6/6+ and 6s/6s+ scores posted that have gone all the way down to 600. At what age and whether Apple diagnostics tested the battery as good, I have no idea. That's also something that needs to be addressed. If battery health causes a device to throttle, those batteries shouldn't be testing as good per Apple diagnostics.

Yes, I do agree, Apple needs to be more upfront about their power management and I'd like to have advanced power management in the settings (I use custom power profiles on my Windows laptops). Actually, I'd really prefer to have slightly thicker and heavier phones with higher capacity batteries. From the kernel distribution graph posted by Geekbench, there are already a significant number of iPhone 7's getting throttled after just one year (granted, that population's probably quite biased towards the technically inclined).

Mind, I did actually did buy the iPhone 7 because of the A10 Fusion. Not because of peak performance but because the High Efficiency cores would mean longer battery life. Still running iOS 10.3.3, though, so currently unaffected by #throttlegate.

Thanks rui no onna for the thoughtful responses and being able to see both sides of the argument. Appreciate you take the time to post them as they def add good perspective to the discussion

Isn't 10.3.3 affected by #throttlegate? I'm pretty sure #throttlegate first happened in iOS 10 and it only got brought to light in iOS 11's life.

#throttlegate

iOS 10.2.1 and above throttles the 6, 6s, and SE.

Only iOS 11.2 and above throttles the 7.

8 and X are currently not infected... for now.

 

[z970]

iOS 10.2.1 and above throttles the 6, 6s, and SE.

Only iOS 11.2 and above throttles the 7.

8 and X are currently not infected... for now.

That's right. iForgot.

I just realized that the moment you posted.

 

[newellj]

Also regularly totally drain your battery, then fully charge it. This should improve the battery's health and charge, even slightly.

I agree with much of what you wrote, but this ^^^ is very damaging to lithium batteries.

 

[rui no onna]

Isn't 10.3.3 affected by #throttlegate? I'm pretty sure #throttlegate first happened in iOS 10 and it only got brought to light in iOS 11's life.

#throttlegate

As mentioned, Apple introduced throttling for the iPhone 7 on iOS 11.2 hence, I'm unaffected.

 

[z970]

I agree with much of what you wrote, but this ^^^ is very damaging to lithium batteries.

Really? I was sure it was supposed to improve the charge time... If I recall, that's what Apple said to do. ...Somewhere...
[doublepost=1515100773][/doublepost]

As mentioned, Apple introduced throttling for the iPhone 7 on iOS 11.2 hence, I'm unaffected.

Right. As I previously mentioned, I remembered that too late.

 

[newellj]

Really? I was sure it was supposed to improve the charge time... If I recall, that's what Apple said to do. ...Somewhere...
[doublepost=1515100773][/doublepost]

Right. As I previously mentioned, I remembered that too late.

People suggest it to "calibrate" the software battery meter, but it's bad for lithium cells. Sample technical discussion: http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

 

[z970]

People suggest it to "calibrate" the software battery meter, but it's bad for lithium cells. Sample technical discussion: http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

Oh.

OK.

 

[Applejuiced]

I haven't seen this discussed, but it's possible that the SoCs in the 5S and earlier simply didn't apply the same demands to the battery that their more powerful successors did. I'm not saying that's a fact, it's just a theory, but I think it's plausible. Silicon tech keeps advancing significantly, battery tech hasn't.

Then either way its a design defect with hardware or battery.
If Apple puts out a product with a battery that cannot meet voltage demands and needs to slow down the processor by 50% then that is a big problem.
[doublepost=1515213761][/doublepost]

I’d have to disagree.

It’s not really a “kinda”. Either it is or it isn’t. For example, I payed for the A11 in my 8+, not the A10 of the iPhone 7. Even if the majority of my time is being powered by the low power cores... that doesn’t change the fact that I paid to have the speed of the A11 when I need it, and not throttled down to the speed of the A10. Trying to justify it by saying “well customers won’t even notice” doesn’t make it right to throttle a customers product to the point that the newest generation they paid for initially now has the performance of the previous generation without being crystal clear about it ahead of time.

Imagine I bought a hybrid car because they advertised it as averaging 65mpg... then a year later they put out a software update that lowers the efficiency down to 40mpg. Of course they could try and justify it and say “Oh well... most people don’t care about whether or not they average 40 or 65 mph. Most customers will be too dumb to realize it anyways!” That doesn’t make it right.

If you paid for a particular product, you should get said product. At the very least Apple needs to be upfront during iPhone keynotes about their chip performance and advise customers before they buy their newest products that while their newest chip is super fast... over time it will be just as slow as the previous generations unless they replace the battery before a certain degradation point.

Hey you bought and paid for a V12 Ferrari but the manufacturer later decides that they will slow it down to a Toyota Echo speeds.
Who needs 800 Horsepower anyway? Nobody uses all that power.
You can get from point A to point B with 108 Horsepower and your gas tank will last longer. See they did you a favor with this new "feature".
Same type of logic as you are arguing with the poster above lol

 

[Abs_p]

Then either way its a design defect with hardware or battery.
If Apple puts out a product with a battery that cannot meet voltage demands and needs to slow down the processor by 50% then that is a big problem.
[doublepost=1515213761][/doublepost]

Hey you bought and paid for a V12 Ferrari but the manufacturer later decides that they will slow it down to a Toyota Echo speeds.
Who needs 800 Horsepower anyway? Nobody uses all that power.
You can get from point A to point B with 108 Horsepower and your gas tank will last longer. See they did you a favor with this new "feature".
Same type of logic as you are arguing with the poster above lol

Only difference is the slow down is not as big as 800 to 108,but 300 to 220-230 and that too only at high end of the RPM. Basically if you buy a Ferrari and always drive it in a neighborhood, Ferrari capping your max power won't affect you, it's only when you drive it on a track than the capping is noticeable. Because BHP is what the engine produces at WOT(google what it means) not how that power is applied and at what RPM.

When you use an analogy, use it right because then you come across as misinformed.

Apple is smoothening out the peaks NOT slowing the phone down completely across the board. So your texting, FaceBook, Candy crush wont be affected anyway.