In this biography
Quick facts
- Born
- 1953
- Birthplace
- Manchester, England, United Kingdom
- Country
- United Kingdom
- Nationality
- Not independently established
- Known for
- ARM processor design and SpiNNaker
- Early collaboration
- BBC Micro and ARM with Sophie Wilson
- University project
- SpiNNaker
Why this life matters
He co-designed SpiNNaker to model spiking neural networks across many processors.
Timeline
Key moments
Doctorate and Acorn
Furber completed his Cambridge PhD and entered Acorn’s hardware group.
Manchester appointment
He moved into university computer-engineering research.
Architecture account
A joint paper documented SpiNNaker’s system design.
From mathematics to hardware
Furber was born in Manchester in 1953. He studied mathematics at Cambridge, graduating in 1974, and completed a PhD in aerodynamics in 1980. He then worked in Acorn’s hardware-development group.
This route was not a conventional computer-science training sequence. It combined mathematical reasoning with experience turning a design into an operating physical system.
Sources for this section: [1]
A partnership at Acorn
Furber worked with Sophie Wilson on the BBC Micro and the ARM processor. The Computer History Museum recognizes their joint contributions to both projects. Furber’s Acorn period lasted until 1990.
A processor’s visible instruction set and the circuits that execute it have to fit together. Accounts that credit only the instruction designer or only the hardware designer miss the negotiation between those requirements.
Sources for this section: [1]
Manchester research and a different scale
In 1990 Furber joined Manchester as a professor of computer engineering. His later research included asynchronous processors and SpiNNaker. In a 2024 interview he discussed retiring after a university career spanning more than three decades.
The SpiNNaker architecture paper was a seven-author collaboration involving researchers at Manchester and Southampton. It describes ARM-based cores linked by a custom packet fabric, designed for large-scale neural simulation and tolerant of the practical problems of a very large machine.
Simulation is a purpose, not a biological identity
Furber’s account of SpiNNaker explains the ambition to use many processors to investigate brain-inspired computation. It is a research instrument built around simplified neural models, not a claim that a computer has become a human brain.
The contrast with Acorn is instructive. A compact processor must implement useful instructions efficiently; a simulation machine must also distribute a large pattern of events. Counting cores does not by itself establish biological accuracy or the adequacy of the model being simulated.
Sources for this section: [2]
Quick answers
Frequently asked questions about Steve Furber
- Does SpiNNaker’s use of packets make it an Internet model?
- No. Its packet fabric is designed for communication inside a neural-simulation machine; that purpose differs from a public data network.
Keep exploring
Lives connected to this story
- Sophie Wilson
Read the instruction-set collaborator.
- Donald Davies
Compare packet communication in a different kind of system.
Sources and further reading
Explore the evidence
These references support the biography and offer further reading. Official records, first-person accounts and independent reporting provide different perspectives; image records document attribution and licensing.
- 1. Computer History Museum, 2012 FellowSteve Furber
- 2. Steve Furber interviewed in Coding Your Future, University of Manchester teaching projectSteve’s Story
- 3. Steve B. Furber, David R. Lester, Luis A. Plana, Jim D. Garside, Eustace Painkras, Steve Temple and Andrew D. Brown, IEEE Transactions on Computers 62(12), 2454–2467, 2013Overview of the SpiNNaker System Architecture
