Quote:
Originally Posted by Yellefan
8 and I have no regrets-- I'm not sure how they do it, but what I'm doing would definitely require more ram on my old intel machine, yet it seems I don't need it with this machine. It's crazy, not sure how they do it...
|
It's the advantage of hardware optimized for software (and vice versa) rather than generic software running on generic hardware. They can optimize the processor to be good at the things their code needs it to do.
But it means the performance is specific to the way Apple writes their software. If you were to run Windows or Android on your Macbook's M1 it wouldn't be any faster because Google and Microsoft have different software development philosophies that the chip isn't optimized for.
Quote:
fun fact: retaining and releasing an NSObject takes ~30 nanoseconds on current gen Intel, and ~6.5 nanoseconds on an M1.
…and ~14 nanoseconds on an M1 emulating an Intel
|
The (geeky) benchmarks above are from an Apple engineer on Twitter and are really quite telling if you know a little bit about how Apple builds software (
link). Retaining and releasing an NSObject is core to Apple's philosophy of writing code, so your computer is doing it just about all the time. The new M1's are built to do this specific task quickly and efficiently. That they can emulate an Intel chip and then still do this task faster than the Intel chip itself is
insane and goes a long way to explain the performance improvements. When you can do these core tasks so well, you free working memory for other purposes, the battery lasts longer, and everything feels snappier.
These chips are a really exciting development for Apple desktop users (their mobile processors have been running circles around the competition for a long time now) and I can't wait to get my hands on one in the higher spec'd machines when they are eventually released.
