Industry Voices – Entner: ARM-ing Apple Silicon is a giant leap for mobile

apple m1 chip
Roger Entner explains how the M1 SoC from Apple helps developers with a common CPU instruction set across Mac and iOS devices, which could lead to an integrated platform for servers as well. (Apple)

 

Roger Entner

On November 10, Apple introduced its first computers using ARM processors on what it calls Apple Silicon. Since the most recent generation x86 CPUs in the Mac Book Air were slower than what the A14 SoC (System on Chip) in iPhones and iPads provides, it came as no surprise that the new computer centric SoCs, called M1, were even faster.

With this step, Apple insources another part of its supply chain and brings it under its control. This follows the long and successful Apple strategy of integrating software and hardware to achieve a competitive advantage. While this is a significant step for Apple, it is a giant leap for the mobile industry.

Free Daily Newsletter

Interesting read? Subscribe to FierceElectronics!

The electronics industry remains in flux as constant innovation fuels market trends. FierceElectronics subscribers rely on our suite of newsletters as their must-read source for the latest news, developments and predictions impacting their world. Sign up today to get electronics news and updates delivered to your inbox and read on the go.

ARM-based processors are the most popular processors in the world. Virtually every mobile smartphone runs on an ARM-based RISC (reduced instruction set computer) instruction set (I am fully aware of the irony of duplicating words for a reduced instruction set) for its CPU (central processor unit). The advantage of ARM-based chips over other processor types like x86 computers that use a hybrid of CISC (complex instruction set computer) and RISC is that ARM-based chips use less power and therefore generate less heat. The advantage of CISC-based computers is they typically run faster using more power and heat. Why is that so important?

The biggest problem in device design is heat. The brains of the device, regardless if it’s a smartphone or a computer, emits heat when it runs. The faster it runs, the more powerful it is, the more electric power it consumes, the hotter it gets. There are two ways to deal with the problem: Find a way to get rid of the heat or find a way to generate less heat.

In computers, the way to get rid of heat is by using fans to take the heat from the central processor and even graphics cards and blow the heat away from the processors. Since water can absorb more heat more quickly, high end computers are being water-cooled like cars. Desktop computers use large fans to achieve that goal. With laptops, consumers want small, light, portable, silent, powerful machines with long battery life when unplugged. With an x86 processor, these are typically mutually exclusive goals. Powerful laptops use a lot of power which create a lot of heat and are loud because they have small fans that run at high speeds to get the heat away from the central processor.

 Apple laptops have prioritized thinness, long battery life, and silence over performance. The pinnacle of how these design priorities have conflicted with the need for power are the high end MacBook Pros. The processors were simply too powerful and thereby heat emitting to run at full speed. They were regularly heat/power throttled to the point that the investment in a faster processor was pointless because it never ran at speeds that were faster than the slower versions. Furthermore, powerful compute capabilities also mean they need a lot of electricity and therefore the battery becomes large to ensure sufficient time without being plugged into a wall.

ARM-based processors, which include “Apple Silicon” A-series chips and Qualcomm Snapdragon chips, have become very competitive with x86 chips at the entry- to mid-level laptop speed segment using substantially less power and thereby less heat and longer battery life. This makes it possible to make substantially thinner and longer lasting laptops doing average workloads like browsing, watching videos, text processing, presentations or spreadsheets. The Snapdragon-powered laptops are very good with the applications built natively for ARM processors on Windows. The moment they have to use emulation mode, problems start with more well-known applications having fewer issues and the more obscure the application, the more problems occur due to legacy issues. These difficulties have led to a modest market adoption.

We might even see now the 20-year-running prediction that next year we will see significant ARM server adoption as the biggest hurdle for it is going away. Linus Torvalds, the creator of Linux explains it as follows: “If you develop on x86, then you're going to want to deploy on x86, because you'll be able to run what you test ‘at home.’” (And by "at home" I don't mean literally in your home, but in your work environment.) What Linus Torvalds means here is that the development platform is of critical importance. That is why Apple showed how easy it is to now build iOS applications on the new Apple laptops because both devices use the same CPU instruction sets.

Without a development platform using the same CPU instruction sets as the deployment platform, development is very hard. Huawei is trying to create a developer platform for its ARM-based Ascend AI clusters and Kunpeng general purpose processors. Not even the $3 billion in incentives were enough to get a developer platform off the ground. But where Huawei has failed, Apple has a very good chance to succeed. With almost 15% market share in the United States and a cult following, Apple should be able to create enough of an ARM computer universe to create a robust developer platform.

When we have a robust development platform on ARM Apple computers, programmers will have a much easier time by having an integrated developer platform for mobile applications in general, Apple Mac applications specifically and server applications as an extension. The implications for edge computing are significant as it makes ARM-based edge computing a reasonable proposition. It will be interesting to see if we will see ARM-based servers emerging that focus on machine learning and AI with other companies now succeeding now that Apple has entered the game -- by providing a development platform and succeeding where Huawei has struggled.

Roger Entner is the founder and analyst at Recon Analytics. He received an honorary doctor of science degree from Heriot-Watt University. Recon Analytics specializes in fact-based research and the analysis of disparate data sources to provide unprecedented insights into the world of telecommunications. Follow Roger on Twitter @rogerentner.

"Industry Voices" are opinion columns written by outside contributors—often industry experts or analysts—who are invited to the conversation by Fierce staff. They do not represent the opinions of Fierce.

Suggested Articles

In June, Su stood out for heartfelt commentary on social injustice after the killing of George Floyd. She challenges industry to do more.

New chip will be incorporated in coffee mug-sized device to cost $199 next year

In a nutshell, Edge ML is a technique by which smart devices process data locally using machine and deep learning algorithms.