Skip to content

Evolution of the ARM Architecture

When people talk about the history of ARM, they usually mean the chip architecture that powers virtually every smartphone, tablet, smartwatch, and modern Mac on the planet today.

What started as a tiny side project in a Cambridge house during the early 1980s turned into the most widely deployed computer instruction set architecture in human history, with over 250 billion ARM-based chips produced to date.

Did you know?

More than 250 billion ARM-based chips have been produced to date.

What began as a custom CPU project for Acorn's internal workstation needs evolved into the standard RISC architecture powering iPhones, Android devices, Apple Silicon, AWS Graviton, and Raspberry Pi

1981: BBC Literacy Project Launched

Acorn Computers won the landmark BBC contract and released the BBC Micro Model B, which went on to capture over 80% of the British educational market throughout the 1980s.

Far removed from modern GUI-driven operating systems, the Acorn MOS was a lean, command-line OS tailored for 8-bit architecture powered by the MOS 6502 processor. Software was primarily loaded via standard cassette tapes.

Where rival systems of the era were closed off, the BBC Micro was engineered for extensibility. It featured a rich array of expansion ports that enabled hands-on physical computing—allowing students to interface with light sensors, robotic arms, and weather stations in much the same way makers use Raspberry Pi HATs today.

The modern maker philosophy—centered on GPIO pins, breadboards, and hardware experimentation—traces its lineage directly back to the BBC Micro.

1983: The Birth of ARM

In 1983, Acorn wanted a faster processor for their next-generation workstations, but commercial CPUs off the shelf were too slow and power-hungry. Designers Sophie Wilson and Steve Furber decided to design their own 32-bit RISC processor from scratch.

They called it the Acorn RISC Machine (ARM).

The ARM1 chip was tested inside a modified BBC Micro as a second processor. That tiny 32-bit architecture evolved into ARM Holdings.

1990: Spinout & Apple Partnership

Apple wanted an ultra-low-power chip for its handheld Newton PDA. Acorn, Apple, and chip maker VLSI Technology spun off the processor division into a new joint venture: Advanced RISC Machines Ltd (ARM).

1993: The IP Licensing Breakthrough

Realizing they couldn't compete with giants like Intel in manufacturing, ARM adopted a radical business model: don't build silicon, license the designs. Companies like Texas Instruments licensed the ARM7 core, integrating it into mobile phones (including iconic Nokia handsets) and the Nintendo Game Boy Advance.

2008–2011: Smartphone Explosion & Cortex

ARM reorganized its chip family into the Cortex ecosystem (Cortex-A for smartphones, Cortex-R for real-time systems, and Cortex-M for microcontrollers). As Apple introduced the iPhone and Google launched Android, ARM cores became the universal standard for mobile computing.

2012: The Raspberry Pi Model B

The Raspberry Pi Foundation launches the $35 Model B (named in homage to the BBC Micro Model B), powered by an ARM processor.

Cambridge researcher Eben Upton noticed that incoming Computer Science applicants went from hobbyist tinkerers who knew assembly to kids who had only ever used web browsers. To solve this exact problem, he co-founded the Raspberry Pi Foundation—deliberately naming their low-cost device as a modern continuation of the BBC Micro's mission.

The ARM-based SoC (System on Chip) at the heart of Raspberry Pi traces its design heritage directly back to Acorn's workshop.

2013–2020: 64-Bit Era & Apple Silicon

In 2013, Apple surprised the industry with the 64-bit ARMv8 A7 chip in the iPhone 5S. By 2020, Apple transitioned its entire Mac lineup from Intel processors to custom ARM-based Apple Silicon (M1/M2/M3/M4), proving ARM could deliver desktop-class performance while using a fraction of the power.

2021-Present: Cloud & AI Expansion

ARM chips moved far beyond phones into cloud data centers (e.g., AWS Graviton), Windows laptops (Snapdragon X Elite), and automotive AI systems, making ARM a core pillar of modern computing infrastructure.

Why ARM Outpaced the Competition

Traditional x86 desktop chips (Intel/AMD) prioritized raw clock speed and backward compatibility, relying on complex decoding logic. ARM prioritized energy efficiency and licensing flexibility, letting manufacturers integrate custom hardware (like graphics processors or neural engines) right alongside the CPU core on a single piece of silicon (System-on-Chip, or SoC).

At the heart of the technical divide between ARM and x86 is their underlying philosophy of instruction set design:

  • ARM is a RISC (Reduced Instruction Set Computer) architecture. It uses simpler, uniform instructions that execute in a single clock cycle whenever possible.

  • x86 is a CISC (Complex Instruction Set Computer) architecture. It provides a vast library of complex instructions, where a single instruction can perform multiple low-level operations (like fetching from memory, adding, and storing back).

Instruction Decoding & The "Decoder Tax"

Because x86 instructions vary in size (anywhere from 1 byte up to 15 bytes), the CPU's decoder unit has to do heavy lifting just to figure out where one instruction ends and the next begins. Modern x86 processors (like Intel Core or AMD Ryzen) actually convert complex x86 instructions into smaller micro-operations internally before executing them.

ARM instructions are uniform in length (32-bit or 64-bit word sizes). This makes decoding dramatically simpler and requires far fewer transistors, saving chip space and electrical power.