Apple Silicon
Apple has seemingly done the impossible. In 2020 Apple announced its newest creation: Apple Silicon. Specifically, the M1 is an ARM-based system on a chip (SoC) as a central processing unit for its Macintosh computers line. But the M1 is more than that. The M1 introduces a unified memory architecture or UMA allowing high bandwidth and enabling low‑latency memory between SoC components.
Hold on. Let us rewind a bit and remember that computers could increase either performance or power efficiency but not both at the same time. The reason for this is simple, the greater the performance, the greater the power requirements. Want to improve power efficiency and nearly always speed will take a hit.
Apple has combined the following components into a single SoC: Central Processing Unit (CPU), Graphics Processing Unit (GPU), Neural Engine, Cache, Fabric, and memory (DRAM). However, this time the CPU architecture is an ARM also known as AArch64. The new M1 has an eight-core CPU with four high-performance cores and four high-efficiency cores. The GPU has eight cores, and the Neural Engine has 16 cores allowing 11 trillion operations per second, making it excellent at machine learning. The M1 maximizes battery life topping out at 17 hours of web-browsing or 20 hours of video playback. So how did Apple do it?
Lots of practice…
Every iPad iteration provided incremental improvements to Apple’s silicon. And it started with the A7 in the iPhone 5S. So while Intel made mediocre improvements to their silicon, offering only modest speed improvements, Apple has been tirelessly ironing out a truly innovative processor.