At this year’s Worldwide Developers’ Conference, Apple confirmed the reality of ARM-powered Macbooks. The announcement did not catch many off-guard as we had already seen a handful of clues and leaks before.
Rumors began to surface in 2018 when analysts from Bloomberg hinted that the tech giant was working on a special project dubbed ‘Kalamata.’ It was all about the plan to switch from Intel processors to custom-designed ARM chips by 2020.
Project ‘Kalamata’ started to gain credibility in 2019 when Apple hired a former top chip engineer at ARM Holdings. The move ignited and kept the rumor mill blazing, with the obvious guess being that something was cooking in the company’s ARM department.
For the better part of 2019, the chitchat kept on going, thanks to frequent teasers by reliable leaker Ming-chi Kuo and a series of never-ending speculations from Apple-centric blogs.
Most rumors turned out to be true. We eventually witnessed the launch of Apple’s first ARM-based processor for Macs – the Apple M1 – on November 10. The M1 came with a whopping 16 billion transistors and exceptional integration of CPU, GPU, Neural Engine, and other components on a single chip package.
Apple also announced its first ARM-based Mac devices based on the M1: a MacBook Air, A Macbook Pro 13-inches, and a Mac Mini.
By switching to ARM chips, Apple is basically arming MacBooks to compete more fiercely against non-armed laptops.
For starters, ARM CPUs offer several advantages over X86-based processors. Since they are built on Reduced Instruction Set Computing, they have a quick response time and lower latency. This means faster operations.
Another Plus with Arm chips is they consume less power than their X86 counterparts, meaning Apple can offer a relatively long battery life on ARM-based MacBooks than what conventional laptops offer. A real-world battery test by Toms Guide revealed that the M1-based MacBook Pro and MacBook Air could hit 16 hours and 14 hours, respectively. That is way above what the Dell XPS 13 (11 hours) and Intel-based Macbook Air 2020 (9 hours) can manage.
Compared to other CPUs, ARM processors are also relatively cheaper to create and deploy as they do not need expensive equipment to produce. Not only do ARM processors incorporate cheaper circuits but also fewer transistors. This makes it easy for companies like Apple to fit them in compact devices. In fact, the compact nature and high power efficiency of ARM CPUs are some of the main reasons why smartphone manufacturers prefer them. There is no doubt that the compactness of the M1 was an incentive towards creating a thinner and lighter MacBook.
Multiprocessing capabilities of ARM architectures are also worth considering. Though ARM CPUs are generally known for their low power applications, the addition of multiple processors and accelerators on single chips has drastically improved their performance. Apple’s M1 comes with eight cores. Four of these are specifically dedicated to high-power performance, while the rest are designed for low-power applications. Customization allows Macs to run more efficiently while nibbling as little power as possible.
Perhaps the most remarkable thing about the M1 is its integration of multiple systems on a single chip. The chip is equipped with a CPU, GPU, neural engine, and a built-in storage controller, among other systems for handling encryption, image processing, and sensing. The 8-core CPU handles all basic computing tasks, such as browsing and document processing, and can hit peak speeds of up to 3.5 GHz. That’s extraordinary for an ARM-based CPU, but that’s all thanks to advanced semiconductor processes, which allowed Apple to employ a 5nm process in the design.
Just like the CPU, the integrated GPU has 8-cores. It boasts 128 total compute units, which deliver 2.6 teraflops of throughput in graphics performance.
Performance in machine learning- related tasks is accelerated by a 16-core neural engine, which is capable of computing 11 trillion operations per second. This will definitely come in handy for users who regularly perform machine learning tasks, such as voice recognition, video analysis, and image processing.
Apple has optimized Mac OS Big Sur to fully unlock the processing power of the M1, including Core ML for machine learning and developer technologies.
With Apple Silicon, MacBooks are more set to fulfill the computing needs tomorrow, which to some extent, involve some aspects of intelligence and AI computing. The integrated Neural Processing Unit of the M1 chip gives the MacBook line-up a leap ahead of other laptops when it comes to training and deploying AI models.
And with AI moving from the Cloud to the Edge, now is the right time for Apple to add AI computing capabilities in products. It was a smart move, both from business and consumer viewpoints.
Unifying iPhones, iPads, and Mac Platforms
In 2019, we learned about Project ‘Marzipan,’ an initiative to help developers build apps that would work across iPhones, iPads, and Macs. That project is being realized with the M1 chip. Now iPhone and iPad apps can run on the Mac.
The transition to Apple Silicon revolutionizes not only the MacBook line of products but also Apple’s business model. Transitioning to in-house chips means more versatile, cheaper, and smarter MacBooks that adapt quickly to the needs of the modern user while giving the company a competitive edge in the notebook market.