Author: Xiaozao Jun
Source: Fresh Date Classroom (ID: xzclasscom)
In 1978, a company named “CPU” quietly emerged in Cambridge, England.
Don’t be surprised, this CPU is different from the one we often refer to in computers.
The full name of this CPU company is Cambridge Processor Unit, which literally means “Cambridge Processor Unit”.
The founders of the CPU company were an Austrian physicist named Hermann Hauser and his friend, a British engineer named Chris Curry.
Hermann Hauser and Chris Curry
After the establishment of the CPU company, it mainly engaged in the design and manufacture of electronic devices. Their first order was to manufacture a microcontroller system for slot machines.
This microcontroller system was developed and called Acorn System 1.
Acorn System 1
In 1979, as the business gradually got on track, the company changed its name to Acorn Computer Ltd.
What does Acorn mean? An acorn. Yes, it’s that thing the squirrel has been chasing in Ice Age.
The logo of Acorn Computer Ltd features an acorn.
The reason for the name Acorn has an interesting story: they wanted to be listed before Apple in the telephone directory…
After Acorn System 1, they developed System 2, 3, 4, and a consumer-oriented cassette computer – Acorn Atom.
Acorn Atom
By 1981, the company encountered a rare opportunity – the BBC intended to broadcast a series of programs to improve computer literacy across the UK and hoped Acorn could produce a computer to accompany it.
This plan was very ambitious, and the UK government also participated (half of the purchase cost would be funded by the government), and once the computers were procured, they would enter every classroom in the UK.
After taking on this task, Acorn got to work. However, they quickly realized that their product’s hardware design could not meet the demand. At that time, the trend in central processing units was shifting from 8-bit to 16-bit. Acorn did not have suitable chips to use.
Initially, they planned to use 16-bit chips from National Semiconductor and Motorola. However, after evaluation, they found two flaws:
First, the execution speed of the chips was somewhat slow, and the interrupt response time was too long.
Second, the price was too high; for a £500 computer, the processor chip accounted for £100.
Therefore, they decided to approach the then-dominant Intel, hoping the company would provide some design materials and samples of the 80286 processor. However, Intel coldly rejected them.
Disheartened, Acorn decided to take matters into their own hands and start designing their own chips. (Does this plot sound familiar?)
At that time, Acorn’s research and development staff found a study on a new type of processor from the University of California, Berkeley – the Reduced Instruction Set, which happened to meet their design requirements.
Based on this, after years of hard work, computer scientists Sophie Wilson and Steve Furber from Cambridge University finally completed the microprocessor design.
Sophie Wilson and Steve Furber
The former was responsible for the instruction set development, while the latter was responsible for chip design.
They named this chip the Acorn RISC Machine.
This is where the famous “ARM” acronym comes from.
Acorn is the company name, Machine means machine, and what does RISC mean?
As mentioned earlier, they based their chip on the “Reduced Instruction Set” technology. RISC stands for Reduced Instruction Set Computer.
Note! Warning for non-technical readers ahead, please skip this part…
Let me explain what the “Reduced Instruction Set” means.
It is a concept relative to “Complex Instruction Set (CISC)”.
Early processors were all CISC architecture (including Intel processors), and over time, more and more instruction sets were added. Because compiler technology was not sophisticated at that time, programs were often written directly in machine code or assembly language. To reduce program design time, single instructions and complex operations were gradually developed. Designers only needed to write simple instructions and leave the execution to the CPU.
However, later it was discovered that out of the entire instruction set, only about 20% of the instructions were frequently used, accounting for about 80% of the entire program; the remaining 80% of the instructions only accounted for 20% of the program. (The typical 80/20 principle)
Thus, in 1979, Professor David Patterson from the University of California, Berkeley proposed the idea of RISC, advocating that hardware should focus on accelerating commonly used instructions while utilizing common instructions to compose more complex instructions.
In simple terms, CISC has strong task processing capabilities and is suitable for desktop computers and servers. RISC simplifies the types of CISC instructions, formats, and addressing methods to achieve energy-saving and efficient results, suitable for portable electronic products such as mobile phones, tablets, and digital cameras.
The first processor chip developed was named ARM1.
Let’s compare ARM1 with the then Intel 80286 processor (also known as 286):
It can be seen that ARM1 and 80286 each have their strengths.
However, in the same year, in October 1985, Intel released the 80386. In the face of the 80386, ARM1 could only be beaten.
Intel 80386
32-bit, 275,000 transistors, with a frequency of 12.5MHz, later increased to 33MHz.
It was obviously unrealistic for ARM to compete directly with the performance of the x86 series. ARM deliberately chose a different design route from Intel – while Intel continued to move towards high-performance x86 designs, ARM focused on low-cost, low-power research and development.
Getting back to the BBC’s computer request.
As mentioned earlier, the BBC proposed its demand in 1981, and if they waited until 1985 for ARM1 to come out, wouldn’t that be too late?
Therefore, before ARM1 was released, Acorn had already provided a solution to the BBC.
At that time, Acorn’s computer temporarily used the MOS 6502 processor (an 8-bit microprocessor developed by MOS Technology).
MOS 6502 processor
This computer was initially named Proton, later renamed BBC Micro.
BBC Micro
By 1984, about 80% of schools in the UK were equipped with this computer. Acorn completely established its presence in front of the British public.
Later, after the ARM processor was developed, it was used in the subsequent models of the BBC Micro.
ARM chip inside the BBC Micro computer
After ARM1, Acorn launched several more series, such as ARM2 and ARM3.
Time continued to move forward.
In 1990, Acorn established a company named ARM to collaborate with Apple.
The logo is quite simple
Note that here, ARM refers to the company name, not the chip name. The full spelling of this ARM is different: Advanced RISC Machines.
The previous chip name:Acorn RISC Machine
The current company name:Advanced RISC Machines
ARM is a joint venture; Apple invested £1.5 million, chip manufacturer VLSI invested £250,000, and Acorn itself contributed £1.5 million in intellectual property and 12 engineers.
Despite this, ARM’s start was relatively humble. Their initial office was located in a barn…
However, the interior of the barn was quite nice.
In the first few years after its establishment, ARM’s performance was mediocre, and engineers were anxious about potential layoffs.
In this situation, ARM decided to change their product strategy – they stopped producing chips and instead licensed their chip design plans to other companies, adopting an open “Partnership” model.
Unexpectedly, this model opened a new era for ARM.
Note! The following paragraph is crucial!
ARM adopted an IP (Intellectual Property) licensing business model, charging one-time technology licensing fees and royalties.
Specifically, ARM has three licensing methods: Processor, POP, and Architecture Licensing.
Processor licensing refers to licensing partner manufacturers to use ARM-designed processors; they cannot alter the original design but can adjust product frequency, power consumption, etc., based on their needs.
POP (Processor Optimization Pack) licensing is an advanced form of processor licensing where ARM sells optimized processors to licensed partners, facilitating the design and production of guaranteed performance processors under specific processes.
Architecture licensing allows ARM to license its architecture to partner manufacturers, enabling them to design processors according to their needs (for example, Qualcomm’s Krait architecture and Apple’s Swift architecture were designed after obtaining ARM’s licensing).
Thus, licensing fees and royalties became ARM’s main sources of income. In addition, there are revenues from software tools and technical support services.
For semiconductor companies, how much are the licensing fees and royalties? One-time technology licensing fees range from $1 million to $10 million, while royalty rates generally range from 1% to 2%.
It is precisely this licensing model that greatly reduced ARM’s R&D costs and risks. It formed an ecosystem centered around ARM through a risk-sharing and benefit-sharing model, making low-cost innovation possible.
After ARM proposed this cooperation model, they began to experiment –
In 1991, ARM licensed its products to GEC Plessey Semiconductor in the UK.
In 1993, ARM licensed its products to Cirrus Logic and Texas Instruments (TI).
The collaboration with Texas Instruments brought significant breakthroughs for ARM and established its reputation, confirming the feasibility of the licensing model.
Subsequently, more and more companies participated in this licensing model and established cooperative relationships with ARM, including Samsung, Sharp, and others.
Based on this, ARM strengthened its determination to adopt the licensing model and began designing more cost-effective products.
In 1993, Apple launched a new handheld computer product – Newton, which used the ARM6 chip developed by ARM.
Apple Newton Message Pad
Now regarded as the ancestor of PDAs and smartphones
Unfortunately, due to overly advanced technology and some user experience flaws, it failed to gain market acceptance and ended in failure.
But ARM accumulated experience and continued to improve its technology. Soon, ARM welcomed its golden opportunity – the advent of the mobile phone era.
ARM first welcomed a major customer – Nokia.
At that time, Nokia was advised to use Texas Instruments’ system design for its upcoming GSM phone, which was based on ARM chips.
Due to memory space issues, Nokia initially rejected ARM.
To address this, ARM specifically developed a 16-bit custom instruction set to reduce memory usage.
Thus, the Nokia 6110 became the first GSM phone to use an ARM processor, achieving great success upon its launch.
Nokia 6110, many people may have used or seen it
ARM then launched a series of chips like ARM7, licensing them to over 165 companies. With the explosive popularity of mobile phones, ARM profited immensely.
On April 17, 1998, the rapidly growing ARM Holdings was listed on both the London Stock Exchange and NASDAQ.
ARM listed on NASDAQ
After ARM went public, Apple, in the post-Jobs era, gradually sold its shares in ARM, investing the funds into the development of iPod products.
Given Apple’s researchers’ familiarity with ARM chip architecture, the iPod continued to use ARM chips.
Apple iPod
As is well known, under Jobs’ leadership, the iPod achieved tremendous commercial success.
But that was not the end; in 2007, a truly groundbreaking product appeared.
That was the iPhone.
The launch of the Apple iPhone completely revolutionized mobile phone design and opened a new era.
The first-generation iPhone used an ARM-designed chip manufactured by Samsung.
The iPhone’s hot sales and the rapid rise of the App Store completely bound global mobile applications to the ARM instruction set.
Then, in 2008, Google launched the Android system, also based on the ARM instruction set.
Thus, the smartphone entered a rapid development phase, and ARM established its dominant position in the smartphone market.
In the same year, ARM chip shipments reached 10 billion units.
In 2011, even the traditional Wintel alliance (Windows + Intel) announced that the Windows 8 platform would support ARM architecture.
Intel must be regretting its decisions.
In fact, it was Intel that gradually paved the way for ARM to occupy its current irreplaceable position. In the 1990s, Qualcomm wanted to collaborate with Intel, but Intel deemed the mobile phone market too small and rejected the partnership. Later, Apple also wanted to collaborate with Intel for the first-generation iPhone, but Intel rejected the partnership for the same reason. As a result, the mobile device market was handed over to ARM, and Intel can’t reclaim it now.
In June 2010, Apple expressed its intention to acquire ARM for $8.5 billion, but the ARM board rejected the offer.
On July 18, 2016, Masayoshi Son, who had invested in Alibaba, and his SoftBank Group acquired ARM Group for £24.3 billion (approximately $30.9 billion).
Masayoshi Son
As a result, ARM became a wholly-owned subsidiary of SoftBank Group. However, SoftBank stated that it would not interfere with or influence ARM’s future business plans and decisions.
Lastly, Xiaozao Jun feels it is necessary to briefly introduce ARM’s product system. This part is indeed a bit confusing, and many industry insiders easily get confused.
Before the ARM11 chip, the relationship between each chip and its architecture was as follows:
There are actually more chip models than these
After the ARM11 chip, starting from the ARMv7 architecture, ARM’s naming convention changed.
The new processor families are named Cortex and divided into three series: Cortex-A, Cortex-R, and Cortex-M. Haha, did you notice that the three letters are A, R, and M again?
-
Cortex-A series (A: Application)
Designed for the growing consumer entertainment and wireless products, used in applications with high computational demands, running rich operating systems, and providing interactive media and graphics experiences, such as smartphones, tablets, automotive entertainment systems, digital televisions, etc.
Cortex-A series
-
Cortex-R series (R: Real-time)
Designed for systems requiring real-time operations, aimed at deep embedded real-time applications such as automotive braking systems, power transmission solutions, and high-capacity storage controllers.
Cortex-R series
-
Cortex-M series (M: Microcontroller)
This series targets the microcontroller field, mainly for cost and power-sensitive applications such as smart metering, human-machine interface devices, automotive and industrial control systems, home appliances, consumer products, and medical devices.
Cortex-M series
In fact, in addition to the three major series mentioned above, there is also a security-focused Cortex-SC series (SC: SecurCore), mainly used for government security chips.
Alright, after discussing so much, let’s summarize!
In short, ARM’s current position is due to both external opportunities and internal strategic factors.
They chose a path that is completely opposite to Intel. Intel has always adhered to a heavy asset, closed full industry chain business model, while ARM adopted a light asset, open cooperation and win-win model.
For ARM, the success of its partners also means its own success. Every company that conducts business with ARM has established a “win-win” symbiotic relationship with ARM.
ARM’s partnership community includes over 1,200 partners
ARM’s DNA of low power consumption coincided with the explosive growth of mobile devices, ultimately leading to its glory.
In the upcoming Internet of Things era, it is foreseeable that ARM is likely to achieve even greater success.
I wonder what everyone has learned from ARM’s story. Here are my insights:
1. When pushed to the limit, people can do anything.
2. Your perceived weakness may actually be your strength.
3. If you can’t compete with your opponent, find a way to unite more friends.