- ACORNS ARM1-CHIP STARTED A 40-YEAR COMPUTING ARROW
- Arm -chips now drives over 300 billion units all over the world and count
- 99% of smartphones are running on arm and there is growing adoption in IoT, Sky and AI workloads
In April 1985, a small team of Acorn Computers in Cambridge, UK, began to consider what a processor could be. Engineers Sophie Wilson and Steve Furber developed Arm1 (it originally stood for advanced RISC machines), a modest chip with only 25,000 transistors, to operate the BBC micro and drafted a 32-bit processor that emphasized reduced instruction set for faster, more effective calculation.
The low power consumption of the design was partially driven by practical limitations, namely the need to run in cheaper plastic packaging. Arm2 soon followed, incorporated into Acorn Archimedes, the first RISC-based home computer. Arm3 introduced a 4KB cache and further improved the performance.
After spin-off from Acorn in 1990, Arm Ltd. Founded as a joint venture between Acorn, Apple and VLSI. An early commercial success was Apple Newton, followed by widespread adoption in mobile phones such as Nokia 6110, which contained Arm7TDMI.
Looking at the future
Arm6, introduced in 1991, brought full 32-bit treatment and a MMU, the key to operating GSM mobile phones. In 2005, the Armv7 architecture debuted with the Cortex-A8 processor, which brought SIMD (Neon) support and ran many early smartphones.
In 2011, the Armv8 introduced 64-bit support and became the basis for Sky, Data Center, Mobile and Car Calculation. Features such as Sve and Helium pushed performance and AI capabilities further.
The launch of Armv9 2021 marked the architecture’s shift to AI-centric workloads. The introduced scalable vector extension 2 (Sve2), scalable matrix extension (SME) and confidential calculation architecture (CCA).
These features made it suitable for everything from smartphones with advanced image processing to AI servers that handle generative workloads. SME accelerates generative AI and MOE models, while SEM2 brings improved AI capacity to general calculation.
Arm’s Compute Subsystems (CSS), based on Armv9, now serves client, infrastructure and car markets. By integrating CPUs, interconnections and memory interfaces, these CSS platforms support rapid development of specialized silicon.
From the original Arm1 with only 25,000 transistors to today’s Armv9 CPUs packing 100 million gates, the architecture has consistently driven computing forward for four decades. Arm-based chips are now driving over 300 billion units around the world, from small embedded sensors to full-scale data centers.
With 99% of smartphones running on arm and growing adoption in IoT, Cloud and AI workloads, architecture must continue to scale thanks to its energy efficient design and flexible license model.
Looking ahead, there have been growing rumors that arm could move beyond license and to chip production, something that would put it in competition with its greatest customers. This speculation was recently intensified after the acquisition of Ampere Computing, Arm’s only independent server chip supplier, by Softbank, Arms Japanese owner.
