Apple Downplays Differences Between iPhone 6s Chips, Blames Aggressive Synthetic Benchmarks
Listen up folks, any battery life woes you might have heard about or experienced yourself with the iPhone 6s and iPhone 6s Plus are exaggerated and not indicative of real-world performance. Scout's honor. How do we know that? Because Apple said so, that's how. Case closed and let's move on, right? Not so fast.
Let's pull the train back in the station before heading down the tracks, shall we? It was discovered that Apple's been dual sourcing production of the custom A9 System-on-Chip (SoC) found in the latest iPhone 6s models to Samsung and TSMC. No big deal, except there are reports that TSMC's larger 104.5mm2 chip is both a little bit faster and a lot more battery efficient than Samsung's 96mm2 part. Some reports have iPhones running TSMC hardware lasting as much as two hours longer than iPhones with Samsung's silicon inside.
That's not an insignificant difference, but if you ask Apple -- and TechCrunch did -- the benchmarks being used to test battery life are not indicative of real-world usage. Furthermore, results based on benchmarks that "spend an unrealistic amount of time at the highest CPU performance state" are "misleading," Apple says.
"Our testing and customer data show the actual battery life of the iPhone 6s and iPhone 6s Plus, even taking into account variable component differences, vary within just 2-3 percent of each other," Apple said.
Apple has a point, real-world usage doesn't typically entail stressing a smartphone's CPU until the battery is drained. In fact, gauging battery life is one of the most difficult things to do in the land of benchmarks, because really, what constitutes real-world usage? It varies by person to person, which is why here at HardwareLogic we often run two tests -- a stress test to mimic a worst case scenario, and a lighter weight web browsing test. Real-world usage for most people will probably fall somewhere between the two.
None of this means Apple gets a Get Out of Jail Free card. Yes, Apple has a point about benchmarking, but stress testing isn't the only place people are seeing significant differences in battery life. It's a point of frustration for users who are noticing considerable less run time on their Samsung-based iPhone 6s and iPhone 6s Plus models compared to their previous non-s versions. Among them is a trusted (and experienced) colleague who reports having needed to charge his Samsung-based iPhone 6s Plus after making two phone calls and using GPS (with the display turned off) for around 30 minutes.
I can't speak to any of those claims personally -- my iPhone 6s Plus, which I'm in the process of reviewing (stay tuned!) sports a TSMC-built A9 chip inside and is getting excellent battery life that's comparable to the iPhone 6 Plus I upgraded from. However, the unflattering claims are out there, countered by Apple's attempt to downplay the issue.
Like our battery benchmarks, the truth is probably somewhere in between.
Let's pull the train back in the station before heading down the tracks, shall we? It was discovered that Apple's been dual sourcing production of the custom A9 System-on-Chip (SoC) found in the latest iPhone 6s models to Samsung and TSMC. No big deal, except there are reports that TSMC's larger 104.5mm2 chip is both a little bit faster and a lot more battery efficient than Samsung's 96mm2 part. Some reports have iPhones running TSMC hardware lasting as much as two hours longer than iPhones with Samsung's silicon inside.
That's not an insignificant difference, but if you ask Apple -- and TechCrunch did -- the benchmarks being used to test battery life are not indicative of real-world usage. Furthermore, results based on benchmarks that "spend an unrealistic amount of time at the highest CPU performance state" are "misleading," Apple says.
"Our testing and customer data show the actual battery life of the iPhone 6s and iPhone 6s Plus, even taking into account variable component differences, vary within just 2-3 percent of each other," Apple said.
Apple has a point, real-world usage doesn't typically entail stressing a smartphone's CPU until the battery is drained. In fact, gauging battery life is one of the most difficult things to do in the land of benchmarks, because really, what constitutes real-world usage? It varies by person to person, which is why here at HardwareLogic we often run two tests -- a stress test to mimic a worst case scenario, and a lighter weight web browsing test. Real-world usage for most people will probably fall somewhere between the two.
None of this means Apple gets a Get Out of Jail Free card. Yes, Apple has a point about benchmarking, but stress testing isn't the only place people are seeing significant differences in battery life. It's a point of frustration for users who are noticing considerable less run time on their Samsung-based iPhone 6s and iPhone 6s Plus models compared to their previous non-s versions. Among them is a trusted (and experienced) colleague who reports having needed to charge his Samsung-based iPhone 6s Plus after making two phone calls and using GPS (with the display turned off) for around 30 minutes.
I can't speak to any of those claims personally -- my iPhone 6s Plus, which I'm in the process of reviewing (stay tuned!) sports a TSMC-built A9 chip inside and is getting excellent battery life that's comparable to the iPhone 6 Plus I upgraded from. However, the unflattering claims are out there, countered by Apple's attempt to downplay the issue.
Like our battery benchmarks, the truth is probably somewhere in between.
