31–40 of 118 posts
Apart from what others already said (performance per watt) this PR fails to acknowledge that M1 also has a built-in Neural Engine cores for accelerating machine learning, which is going to become the main feature of many apps. Indeed, M1 is unique in the respect that you get CPU, GPU and ML cores in a single chip, with incredible performance and minimal power utilization.
Earlier quoted context omitted.
Maybe you're thinking of FJCVTZS, which is intended to accelerate a common operation in JavaScript engines? That's not unique to M1, though; rather added to a relatively recent ARM ISA revision. See https://news.ycombinator.com/item?id=25422168 I think there may be other optimizations related to Apple's reference counting and type tagging models for Objective-C and Swift.
FJCVTZS I read that out loud, thinking I was back at the optometrist.
Apple has already won. Even if Intel catches up in efficiency, it doesn't matter. Even if they're 30% faster, it doesn't matter. Apple is in charge of their own destiny with Apple Silicon. That's invaluable.
Earlier quoted context omitted.
Moving to a completely proprietary platform just to spite one x86 vendor is cutting off your nose to spite your face ;)
I thought arm was a lot more open than x86 based tech?
x86 has its roots in the PC, which was an open and relatively standardised platform.
ARM's rise to popularity came later, and there are literally thousands if not tens of thousands of different SoCs with an ARM core, which have nothing in common but the core. All the locked-down mobile devices are in this category, and their SoC manufacturers have been extremely reluctant to release much in the way of documentation.
It might be faster in some areas, but not in all, and certainly not at the same power budget. Marketers gotta market, though.
I don't want to leave ARM because I don't want to support Intel either way. So, there!
Moving to a completely proprietary platform just to spite one x86 vendor is cutting off your nose to spite your face ;)
Apple has already won. Even if Intel catches up in efficiency, it doesn't matter. Even if they're 30% faster, it doesn't matter. Apple is in charge of their own destiny with Apple Silicon. That's invaluable.
Earlier quoted context omitted.
Depending on the workload: does it need to? Most people I've seen use laptops near chargers, or at least within 2 hours of the nearest charging opportunity. Sometimes people travel and need more, but I don't think it's that common to need eight hours of charge at maximum performance. Don't get me wrong, the M1 is an engineering marvel and as Apple's competitors move to a similar node size I hope they too will get the…
If Intel can come up with a laptop that's faster than the M1 at two thirds the price with the only downside being that you only get four hours of battery out of it with all cores clocked to their limit I don't see how Intel can come up with a macbook air equivalent at a price of $350.
Apart from what others already said (performance per watt) this PR fails to acknowledge that M1 also has a built-in Neural Engine cores for accelerating machine learning, which is going to become the main feature of many apps. Indeed, M1 is unique in the respect that you get CPU, GPU and ML cores in a single chip, with incredible performance and minimal power utilization.
Unique in what sense? Didn't Samsung make similar arm cpu/you/ml chips for Google's pixel 6 line?