There are several android phones that compete with the iphone 13 pro performance wise, while running a much less optimized os.

Likewise in the desktop world, the single core performance of the M1 is on par with top of the line chips, and M2 will be the same, and intel and amd work fine on linux.

Apples decision to keep a locked box has nothing to do with performance, solely to do with keeping people in the ecosystem for revenue.

I'm aware benchmarks don't tell the whole story, but Geekbench shows the iPhone 13 Pro[0] significantly outperforming the Samsung Galaxy S22 Ultra[1] - about 60% faster multicore (presumably thermally limited?) and ~80% faster singlecore.

[0]: https://browser.geekbench.com/ios_devices/iphone-13-pro [1]: https://browser.geekbench.com/android_devices/samsung-sm-s90...

The "coupled with" was important. They hit a spot on the performance/responsiveness/power-use graph that others don't, and a lot of that's due to software, not just hardware. It's easy to get incredible battery life if you accept poor performance, or to have great performance by sacrificing battery life, but Apple does the extra work to achieve both. Kinda like how BeOS used to feel way smoother than Windows or (GUI) Linux even when running on far worse hardware.

Yes, although I think a lot of the issue is generic hardware drivers that don’t necessarily configure things correctly, and it looks like the Asiago linux people are paying a lot of attention to power management, so it may end up having decent battery life too.

So you're saying that BSD has better power management than Linux? Would love to see some real research and analysis on this.

It's not a BSD vs. Linux thing. It's the entire software stack including user space, e.g. how much is the CPU woken up by silly background tasks doing useless things. That thing can be fixed regardless of the underlying kernel if people care enough. The total investment in desktop Linux is tiny compared to what Apple puts into macOS.

The vast majority of power management is less background services while active.

It's mostly about reducing idle power consumption sleep mode, and closed lid hibernation.

Most battery loss in mobile devices can be attributed to two things.

The screen, and how much is lost when the device is idle in your pocket/backpack.

My claim is that macOS and iOS have far better power management than Linux or Windows, or Android, respectively. Possibly that's directly due to their BSD heritage, but I doubt it.

Apple's kernel isn't merely one of the BSD derivatives. And of course their userland is mostly proprietary.

No, macOS does its own power management - the base kernel may be BSD, but macOS has vastly more work in it than simply shipping a basic BSD system.

The base kernel is not really BSD.

https://wiki.freebsd.org/Myths#FreeBSD_is_Just_macOS_Without...

>Darwin - which consists of the XNU kernel, IOkit (a driver model), and POSIX compatibility via a BSD compatibility layer - makes up part of macOS (as well as iOS, tvOS, and others) includes a few subsystems (such as the VFS, process model, and network implementation) from (older versions of) FreeBSD, but is mostly an independent implementation. The similarities in the userland, however, make it much easier to port macOS code to FreeBSD than any other system - partially because a lot of command-line utilities were imported along with the BSD bits from FreeBSD. For example, both libdispatch (Grand Central Dispatch in Apple's marketing) and libc++ were written for macOS and worked on FreeBSD before any other OS.

It’s pretty BSD. Sure Mach is under there, but that doesn’t mean it “is” Mach. There’s a lot more original work that doesn’t belong to either ancestor.