While some aspects of this story are of purely anecdotal interest, others shed light on some ARM design decisions, which were taken in an unusual set of circumstances to meet specific goals, now seen to meet the demands of an innovative and exciting market place requiring good performance and low power consumption, balanced with low cost.
British readers will probably be familiar with Acorn Computers Ltd, its products and its history of phenomenal success in the UK computer market of the early 1980s. Other readers may not have had access to as much information on the vibrant home computer market in the UK then, or to Acorn's record for technical innovation.
The story starts with the original development of the ARM processor, and ends with the establishment of ARM Ltd as a global force in the microprocessor industry. In between, it sheds some light on various design decisions which were taken in the genesis of the ARM design.
The smaller size of the UK market (compared to the US) also ensured that even the most successful companies could not achieve the size of American rivals, affecting their ability to invest in research and development and to ride out the ups and downs of the market for personal and home computers.
Acorn's initial success was sealed when the British Broadcasting Corporation (BBC) commissioned a new home computer model from the company to be sold as the BBC Microcomputer, to tie in with a public computer education programme shown on BBC television in the UK.
The release of the BBC Micro in 1982 caught the crest of the home computer wave in Britain, and the BBC name gave Acorn's design added credibility compared with competing machines from the many other developers in this market. Sales exceeded all expectations: original estimates by the BBC and Acorn were that at best tens of thousands of units would be sold. In fact, to date nearly two million BBC Micro-compatible computers have been sold by Acorn, and it quickly grew from a small company with tens of staff into a medium-sized company employing hundreds with an annual turnover of tens of millions of pounds.
The BBC Micro was based around the 8-bit 6502 processor from Rockwell, the same chip that powered the Apple II. Initial models featured colour graphics and 32 kbyte of random access memory. Data was stored on audio cassettes; hard and floppy disk drive interfaces were also available, and Acorn was an early proponent of local area networking with its Econet system. Another important feature of the BBC Micro was its capacity to accept a second processor attached via an expansion port known as the Tube. Connectivity, interoperability and networking were familiar concepts to many BBC Micro users long before they were established in the rest of the personal computer world, via such options as the Tube. This required a degree of interoperability between host and second processor, as well as Acorn's Econet local area networking standard.
The team tried all of the 16- and 32-bit processors then on the market but found none to be satisfactory for their purposes; in particular, the data bandwidth was not sufficiently greater than that offered by the 6502 to justify basing the next generation of Acorn computers upon them. Processors were tested by building BBC Micro `second processor' units based upon them, and it became clear that no chip would be found to fit the very precise requirements on which the Acorn design team had settled.
Acorn has always been renowned for the calibre of its research and development staff. It was able to pick the cream of graduates from Cambridge University, home of a highly regarded computer science faculty, as well as attracting staff from around the world.
To them, designing a processor from scratch to meet their carefully specified criteria was an obvious thing to do. Acorn's phenomenal success with its 8-bit computers had created a research and development environment where staff could afford to pursue advanced projects which would not necessarily result in immediately saleable products, and were actively encouraged to do so.
For example, Sun developed the SPARC RISC chip and architecture for its own computer workstations, while notable RISC processors from established chip producers include Intel's i860 graphics processor and AMD's 29000, which has mainly been used as a graphics accelerator or in printers. However, both Sun's and MIPS' efforts were based on earlier research efforts at Stanford and Berkeley universities respectively, while Acorn's project was effectively begun from scratch, although reports on the Berkeley and Stanford research were read by the Acorn team and were part of the inspiration behind designing a RISC processor.
One of the reasons the ARM was designed as a small-scale processor was that the resources to design it were not sufficient to allow the creation of a large and complex device. While this is now presented as (and genuinely is) a technical plus for the ARM processor core, it began as a necessity for a processor designed by a team of talented but inexperienced designers (outside of university projects, most team members were programmers and board-level circuit designers) using new tools, some of which were far from state-of-the-art. With these restrictions on design and testing, it is hardly a surprise that a small device was developed.
While the ARM was developed as a custom device for a highly specific purpose, the team designing it felt that the best way to produce a good custom chip was to produce a chip with good all-round performance.
The design team worked in secret to create a chip which met their requirements. As described earlier, these were for a processor which retained the ethos of the 6502 but in a 32-bit RISC environment, and implemented this in a small device which it would be possible to design and test easily, and to fabricate cheaply.
First the instruction set was specified by Wilson, based on his knowledge gained as the author of much of the original software for the BBC Micro, including its BASIC interpreter. The important initial decisions were to use a fixed instruction length and a load/store model. Other design decisions were taken on an instruction by instruction basis.
An event-driven simulator was designed, also in BASIC, which allowed the support chips, the video controller VIDC and memory controller MEMC (which both had slightly more complex timing requirements), and the I/O controller IOC, to be designed and tested. A development of this simulator, since rewritten in Modula-2 and then in C and known as ASIM, is still used by both Acorn and ARM Ltd. for design and testing today.
There was a great deal of excitement at and confidence in the new chip. The ARM was used internally at Acorn and by Acorn developers when it was made available as a second processor add-on for the BBC Micro; this device used the ARM1 as an additional coprocessor and accelerator for the 6502-based BBC micro. In fact, this second processor was used to improve the performance of the simulation tools the team had designed to finish the support chips and also to develop the next ARM processor.
The second processor add-on also enabled third-party developers to start working with the processor and contemplating the development of software to exploit its advanced features. The purpose of releasing the second processor was to ensure that when a complete ARM-based system was released, potential users and developers had some experience of ARM and were not deterred from developing application software for it by the novelty of the technology and the lack of wide support for it in the market.
This addition would facilitate real-time digital signal processing, which was to be used to generate sounds, an important feature of home and educational computers.
A coprocessor interface was also added to the ARM at this stage, which would enable a floating point accelerator and other coprocessors to be used with the ARM. Even after all these additions the ARM2 maintained its small die size and low transistor count; the die was 5.4 mm square and the transistor count around 25 000. This second device was also improved by being fabricated in a 2 \xb5 m process. That this was an extraordinary achievement, and that the ARM is an unusual processor in terms of size/performance, is shown more clearly in Figure 1.1 which shows the relative die size of the ARM and other processors
Now the leading names in the computer market were IBM, producers of clones of its personal computer, and Apple. Compatibility with existing computers, and particularly the IBM standard, was of increasing importance, as was the ability to run market-leading application programs, especially those aimed at the growing business market.
Unlike Acorn, Apple had adopted an off-the-shelf 32-bit processor, Motorola's 68000, and so it was able to bring a 32-bit computer, the Macintosh, to the market in 1984, although it was some time before it gained full acceptance by the business community. Apple too went through a stage when its technical resources and designs were unsurpassed but not translated into success in the marketplace.
A financial crisis enveloped Acorn, and led to it being taken over by one of Europe's leading computer and office equipment manufacturers, the Italian giant Olivetti Ing et Cie, which apparently bought up Acorn in 1985 for its share of the UK computer market, without knowing that its research labs housed the first samples of a new family of RISC processors.
Deep within the advanced research and development labs in Cambridge, and at the research lab that Acorn had established in Palo Alto, California, Acorn staff were also designing an office automation system using the ARM processor. This system was a long-term goal of Acorn's co-founder, Dr Hermann Hauser.
A new operating system, known as ARX, was being developed to run on the processor, but progress was slow and Roger Wilson has described it as `a black hole', at least as far as programming resource was concerned. However, the need for Acorn to release a new product to reach its existing market in education, small businesses and the home meant that this project was abandoned and a home computer, the Archimedes, was launched in 1987 as the first commercial product using the ARM, featuring an 8 MHz version of the ARM2 and the three support chips MEMC, VIDC and IOC, an input/output controller and a simple operating system.
To answer some of these criticisms, software emulators were launched with the machine, which allowed Archimedes users to run most PC and BBC Micro software, but it took two or three years for a credible amount of application software native to the ARM and Archimedes to be developed.
Since then Acorn has refined and improved its computer models and confirmed its position as a leader in the British home computer and educational computing market. A wide range of software is available to these users, most of it developed by small companies loyal to Acorn since the early 1980s, and including applications intended for home, business and education use. Because of Acorn's dominant position in the UK educational computing market, the range of programs suitable for use in the classroom is probably at least as large as that for any other computer.
The purpose of designing the original ARM chips, ARM1 and ARM2, had been to develop a processor capable of offering better-than-acceptable performance in low-cost personal computers. The next step was to expand the design so that it offered the kind of performance expected of a high-end personal computer, or workstation. Intel- and Motorola-based personal computers were already offering performance which perceptibly outstripped that of ARM-based systems.
Acorn's partner in building the chip, VLSI Technology Inc., was to develop further markets for the ARM processor and its support chips, while Acorn continued to develop personal computers based on the chip.
The inclusion of the cache and its control circuitry led to a much higher transistor count of around 300 000, but this was still a highly compact device; so much so that problems occurred trying to find an IC package capable of accommodating the tiny ARM3 die.
In 1989 the ARM3 was launched at the significantly increased clock rate of 25 MHz. Acorn's desktop computers using this chip were first launched in 1990, although third parties were selling ARM3 chips on upgrade boards for ARM2-based computers in 1989. The first of these was Aleph One Ltd, a small company based in Bottisham, the next village to ARM Ltd's current home.
VLSI Technology Inc. was having some success in convincing other companies to use the ARM, particularly as an embedded processor. Some companies incorporated ARM into their products; others took samples of the chip to use in their research. One of these was Apple.
Interest in the ARM family was growing as more designers became interested in RISC and the ARM's design was seen to match a definite need for high-performance, low power consumption, low-cost RISC pro-cessors. In conditions of greatest secrecy an agreement was reached between Acorn, VLSI Technology Inc. and a company which had expressed an interest in the ARM for some time now, Apple.
The ARM development team moved out of the building they had long occupied at Acorn's Cambridge headquarters. Newly-appointed managing director Robin Saxby, former MD of European Silicon Structures (usually referred to as ES2), chose a converted 18th century barn in the picturesque Fenland village of Swaffham Bulbeck, ten miles outside Cambridge, as ARM's new home.
ARM Ltd was founded with a clear mission to continue the development of the ARM processor and to facilitate its use by system developers, whether as a standalone processor or as a macrocell with custom logic or other ARM components added to it to make a custom chip.
ARM Ltd was also to license its designs to chip foundries who would sell the chips, giving ARM Ltd a royalty, rather than establish its own fabrication facilities. VLSI Technology, which had built all previous ARM chips as well as custom logic devices for both Apple and Acorn, was the first licensee.
A two-digit number denotes a self-contained chip consisting solely of this device and the minimum necessary interface and test circuitry, for example ARM60 and VIDC20. A three-digit number denotes a device which integrates the processor macrocell with other standard ARM macrocells and/or custom logic, for example ARM250 and ARM610.
ARM Ltd's first major commission was to design a CPU for Apple suitable for use within a hand-held personal organizer device. This device became known as ARM600, from which the ARM610 used in Newton was later derived. At the same time ARM Ltd's software team developed the ARM Cross Development Toolkit, a suite of software which allowed designers working on a range of platforms to use ARM development tools, assembler, compilers, and debugging and emulation programs.
Hardware evaluation kits were also produced to enable designers to test the ARM6 processor and to begin to develop operating system and support software for use with their own designs before the availability of finished systems. ARM Ltd developed the PIE (Platform Independent Evaluation) Card, which allowed system designers to test their ideas on an ARM card attached to a host machine running the Cross Development Toolkit.
ARM Ltd has taken steps to raise its profile within the merchant microprocessor market, with staff making regular presentations at conferences worldwide. A new visual image was adopted, with the ``ARM-powered'' label to be attached to any systems using ARM processors.
The standards of the 1980s are now themselves starting to look like old technology, and the quest for a new generation of information and leisure technology products has provided immense opportunities for companies like ARM Ltd with timely products.
Late in 1992 a new venture, The 3DO Company, announced that it too had designed the ARM (in this case, ARM60) into its product, a CD-ROM based leisure computing box to be known as the Interactive Multiplayer. 3DO and its licensees plan to ship productsorganizer, both hardware and software, during 1993. A wide range of leisure and commercial software developers signed up to work with the 3DO format, offering it a good chance of success in a market dependent on both the delivery of technology and the availability of attractive software. 3DO did not plan to manufacture ARM-based hardware itself, but to encourage its hardware licensees to produce a range of products conforming to the standards it defined. Among its licensees are Japanese electronics giant Matsushita.
Both these product types, electronic personal organizers and leisure computing devices, require powerful processors at a cost low enough that the end-product is still competitively priced for a consumer market. Hand-held portable organizers require this computing power to be delivered in a compact form and without heavy power consumption, so that the unit can be small and run from batteries. ARM Ltd's processors are ideal for this and the growth of this market represents a major opportunity for ARM and its customers.
The embedded controller market has traditionally focused on 8-bit microprocessors, but the growing complexity of many control requirements in sophisticated products indicates a need to move to more powerful processors. The ARM and its variants offer manufacturers the opportunity to move directly to 32-bit controllers at low cost and with a great deal of flexibility for designing custom controllers.
Potential applications for custom embedded controllers using ARM macrocells include real-time controllers in the automotive market. Potential applications include engine management systems and entertainment systems controllers.
The ARM has had previous successes as an embedded controller. Cambridge (England) robotics company Microrobotics has used various ARM devices as the basis of its microcontroller system used for applications as diverse as controlling animatronics puppets and complex event lighting systems. British company Rediffusion Simulation uses the ARM in its Commander flight simulator.
Other companies around the world are planning to use the ARM as a controller for arcade computer games, high-speed data communications, videophones, fuzzy logic controllers, and data-logging and test equipment.
From its earliest days within Acorn, ARM Ltd has worked closely with VLSI Technology, Inc., its first partner and the first manufacturer of ARM devices.
In the UK, GEC Plessey Semiconductors was signed as an ARM foundry and partner in January 1992. Plessey now produces a range of ARM standard parts. It is also the foundry for the ARM250, a custom processor developed for Acorn out of standard macrocells and a small amount of custom circuitry.
Establishing a relationship with a major Japanese manufacturer was a key component of ARM's strategy, and this was achieved in March 1993 when the Sharp Corporation of Japan signed a deal to manufacture and market ARM processors and associated products. Sharp already has a relationship with Apple which is expected to result in products based on Apple's Newton technology, to which Sharp is contributing.
At around the same time ARM Ltd strengthened its claim to be a truly global company by receiving a significant investment from Japanese investment house NIF. ARM Ltd's investors now include European companies, in the form of Acorn (and through it Olivetti), US companies Apple and VLSI Technology, and NIF in Japan.
Shortly after these agreements were signed, Texas Instruments was added to the list of ARM partners, with the intention of using ARM macrocells as the basis of custom embedded controllers.
ARM Ltd now has offices in California and Japan in order to maintain a close relationship with licensees and their major customers, and to promote existing ARM devices and the company's ability to produce new ones to future customers. It is likely that ARM will continue to establish relationships with new partners around the world.
As the market for ARM devices has grown and the requirements of potential customers have developed and become more sharply defined, so too have ARM Ltd's design objectives. The ability to develop custom processors and controllers quickly from its library of standard macrocells has always been there, but this is now being formalised in the QuickDesign system, which was launched at the COMDEX exhibition in November 1992. As the name implies, the purpose of QuickDesign is to create a custom part from standard parts as quickly as possible, and to show how these can be interfaced with custom technology developed by ARM Ltd or the customer working in partnership to produce a timely and low-cost product.
ARM Ltd's design objectives are now clearly stated as developing processors which use RISC design principles to meet the following goals.
The ARM processor has always differed from other commercially available RISC processors in that it is intended to meet a price/performance ratio rather than to be the most powerful processor available. Acorn's computers have always been aimed at the middle of the market, so the processor designed to power them was too. ARM processors are not the most powerful, but offer an extremely good price/performance ratio compared to other processors, at about a dollar per million instructions per second (MIPS) in the case of ARM6.
It also provides some measure of confidence that future developments of the ARM processor family will appear on schedule, so that system designers need not worry that their new designs will be held up while vital components are developed and debugged. ARM Ltd's own mythology is that virtually all the chips they have designed have worked first time; a row of champagne bottles, each opened to celebrate the arrival of working silicon, lines the staircase at ARM Ltd's barn to bear witness to this.
This has proved a critical key to the success of ARM processors. Unlike many other processor designs, the ARM was easily re-implemented in static form rather than the usual dynamic CMOS. This, along with the small die size, reduced power consumption, making ARM processors ideally suited for power consumption-critical products such as portable computers. Furthermore, it allows the clock to be stopped, a useful powersaver in portable designs.
The ARM610, commissioned by Apple, is one example based on macrocells, which includes the 32-bit ARM6 processor core, a 4 kbyte cache, a write buffer and a memory management unit. Even with all these additional components, the end result is a much smaller package than familiar processors such as the 80386.
Acorn Computers has also enjoyed the fruits of commissioning a custom chip from ARM which effectively combined the original ARM2 four chip set on to a single device, the ARM250. This process was carried out from the original concept to volume production in 12 months, resulting in a single device with a sixth of the footprint, one third the power consumption and half the cost of the devices it replaced.
The term Reduced Instruction Set is applied to a great many processors and it is not obvious that at the extremes of the category they have much more in common with each other than they do with CISC devices. RISC techniques are often employed in extremely large and complex devices such as the i860, where the size and complexity of the chip means that advantages it gains from using RISC techniques are very different from those gained by the ARM processor.
Why was the ARM from its inception designed as a reduced instruction set processor? At the time the first ARM chip was being designed, RISC was a relatively new concept and CISC processors were still being developed which offered growing performance.
RISC's advantages were originally propounded as being:
The 6502, to which Acorn's designers looked when designing the original ARM, had a short and simple instruction set which lent itself well to RISC. RISC was a sensible option for the design team to consider; all three of the above points suggested it as a suitable choice when designing the chip. The team's resources in terms of staff, time and development tools were limited, and the requirement was for a processor which would be cheap to make and sell but still offered sufficiently high performance that computers based on it would perform as well as or better than comparable personal computers.
The advantages of RISC that have attracted further users to the ARM chip set appear mainly to be its delivery of high performance for low cost, in a compact package which takes up little space and consumes little power. While processors fulfilling this set of requirements may have been a small market niche a few years ago, it is now a highly competitive and fast-growing area of the computer market, and ARM Ltd and its processors are placed well to compete within it.
While at its launch the ARM and systems based on it were seen as being ahead of their time, the current vogue for all things RISC has led to an increased interest in the ARM. This, combined with changing market conditions influencing Acorn, led to the ARM design team being established as ARM Ltd, with investment from other partners including Apple Computer, and to the redesign of the ARM itself to exploit its benefits still further.
From being a single design aimed at a particular project the ARM is now a set of highly customizable processors and supporting macrocells suitable for use in a wide range of applications but targeted at systems requiring high performance from a compact device with low power consumption.