ARM microcontrollers have become ubiquitous in embedded systems and IoT devices. Their low cost, low power consumption, and performance have made them a popular choice for designers. But how did ARM become such a dominant force in the microcontroller market? Let’s take a look at the origins and evolution of ARM technology.
The Beginning of ARM
ARM stands for Advanced RISC Machine. The history of ARM microcontrollers begins back in the 1980s at Acorn Computers Ltd in Cambridge, England. Acorn was known for making home computers like the BBC Micro and Acorn Electron. As Acorn worked on developing a successor to the BBC Micro, they decided they needed a new processor design.
At the time, CISC (Complex Instruction Set Computing) processors like the Intel 80×86 dominated the market. But a small team at Acorn believed a RISC (Reduced Instruction Set Computing) architecture held greater potential. RISC processors use simpler instructions that can execute more efficiently. This gives better performance at lower clock speeds, resulting in power savings.
In late 1983, Acorn initiated a project they called the Acorn RISC Machine. The goal was to produce a low-cost RISC processor that could outperform CISC processors. A team led by Sophie Wilson and Steve Furber designed the first ARM processor. It had about 25,000 transistors (orders of magnitude fewer than complex CISC chips at the time). The ARM1 chip ran its first program in April 1985.
The Rise of Advanced RISC Machines
After the ARM1 came the ARM2 in 1986. Acorn used ARM processors in its Archimedes computer systems. However, Acorn realized the ARM had more potential beyond its own computers. In 1990, Acorn spun off ARM into its own company called Advanced RISC Machines Ltd.
Advanced RISC Machines partnered with chip manufacturers like VLSI Technology to license and produce ARM processors. Early customers included Apple for the Apple Newton PDA and DEC for an early ARM-powered PC. But a major inflection point came when Advanced RISC Machines partnered with Nokia in the early 1990s to use ARM processors in mobile phones.
As cell phones took off in the 1990s, the efficient performance per watt of ARM processors made them the perfect fit for battery-powered devices. ARM sales skyrocketed from just $18 million in 1993 to $387 million in 2000. The ARM architecture also found success powering embedded systems like automotive microcontrollers and hard disk drives.
The Intel Challenge and ARM’s Big Break
By the early 2000s, ARM was the dominant player in embedded processors. But Intel saw an opportunity to leverage its manufacturing leadership and move into the embedded market. In 2001, Intel launched the XScale ARM-based processors. However, Intel failed to gain much traction and sold the XScale business to Marvell in 2006.
Intel’s stumble left an opening for ARM to become the standard architecture for mobile devices. When Apple launched the iPhone in 2007, they chose an ARM-based processor tailored for mobile. Other smartphone makers quickly followed suit. The iPhone revolutionized the mobile landscape and ARM rode the wave, powering the Android explosion as well.
The performance, power, and cost advantages of the ARM architecture have made it nearly impossible to displace in the mobile world. As of 2021, ARM processors power 95% of smartphones, 35% of embedded devices, and 15% of servers. ARM went from a tiny British startup to the standard architecture of the mobile computing era.
Acquisition by SoftBank and Arm’s IPO
The success of ARM attracted attention from investors. In 2005, ARM relocated its headquarters to California to grow its US business. That same year, ARM was acquired by the Japanese telecom company SoftBank. Under SoftBank, ARM continued accelerating growth through both organic initiatives and acquisitions.
In 2010, SoftBank took ARM public with an IPO listed on the London Stock Exchange and NASDAQ. The IPO valued ARM at around $5 billion and allowed employees and early investors to cash out. But SoftBank maintained ownership of the majority of ARM shares. Public ownership fueled further growth at ARM, with its market cap reaching $24 billion by 2016.
Acquisition by Nvidia
ARM’s growth and dominance in mobile computing made it an attractive asset. In September 2020, Nvidia announced it would acquire ARM from SoftBank for $40 billion. The deal closed in early 2022, though it faces regulatory scrutiny in the UK, EU, US, and China.
Nvidia plans to infuse cash into ARM to accelerate its R&D efforts. Its goal is to expand ARM technology into new markets like data centers, the Internet of Things, self-driving cars, and AI. However, the acquisition raises concerns that Nvidia could restrict access to ARM’s intellectual property that is licensed widely across the semiconductor industry.
The Evolution of ARM Processor Designs
Over its history, ARM developed successive generations of processor designs to meet evolving needs:
- ARM1 to ARM3 – Early simple designs focused on low cost and power for embedded systems.
- ARM6/7 – Improved performance to run early 90s PDA devices like the Apple Newton.
- ARM7TDMI – Introduced the Thumb instruction set, enabling higher code density for better performance.
- ARM9 – Superscalar core with enhanced signal processing capabilities suited for GSM cell phones.
- ARM11 – Higher clock speeds up to 600 MHz for smartphones and media players.
- Cortex-A Series – High performance application processors for mobile computing.
- Cortex-R Series – Real-time processors for embedded control systems that require responsiveness.
- Cortex-M Series – Microcontroller-oriented designs optimized for cost and energy efficiency.
The Cortex-M series of ARM processor designs ignited the rise of ARM microcontrollers. Cortex-M0, Cortex-M3, and Cortex-M4 cores power most of the cheap, energy-efficient MCUs on the market. Their combination of low cost, low power, and solid performance has led to the dominance of ARM in the 32-bit microcontroller market.
The Open Ecosystem of ARM
A key factor in ARM’s success has been its open ecosystem model. ARM makes its core IP available for licensing by semiconductor companies. This creates competition and drives innovation among chip vendors. Over 500 silicon partners license ARM’s processor IP today.
Vendors like Microchip, NXP, STMicroelectronics, and others take ARM’s core and add their own customizations and peripherals. This creates an enormous diversity of ARM chips tailored for every application imaginable. It allows small companies access to cutting-edge IP and enables faster time-to-market.
The ARM ecosystem democratized embedded system and IoT product development. It fueled innovation in small microcontrollers all the way up to advanced application processors powering smartphones. This open model was key to ARM’s meteoric rise.
The Future of ARM
ARM finds itself at an inflection point under new owner Nvidia. Much of ARM’s future trajectory will depend on Nvidia’s ability to grow the ecosystem while maintaining ARM’s commitment to open licensing practices.
On the technology front, ARM is pushing performance higher with designs like the Cortex-A78 for mobile applications. It’s also optimizing power and cost efficiency for target markets like AI, self-driving vehicles, and the Internet of Things. And it is expanding beyond processor IP to offer entire system-on-chip solutions.
With over 130 billion chips shipped to date based on its IP, ARM remains one of the great technology success stories. From its origins in a small British startup to powering the global mobile computing revolution, the history of ARM microcontrollers illustrates how the right technology at the right time can transform industries and enable rapid innovation across the world.