Manchester computers

Long ago, people built big computers.

These machines were very tall. They helped solve hard math problems. One was called the Baby. It was the first to store its own plans. This helped it work. Another machine was called Atlas. It was the fastest in the world. It was very powerful. These machines helped us learn how computers work today. They were very special.

For thirty years, a small team worked at the University of Manchester. They built a series of new computers. This team was led by Tom Kilburn.

The first machine was called the Manchester Baby. It ran its first program on 21 June 1948. This program used 17 instructions to find a math answer. The Baby was the first stored-program computer. This means it could keep its own instructions in memory. The Baby was very big. It weighed almost one long ton. It used 550 valves. Valves are small parts that control the flow of electricity.

Later, the team built a machine called Atlas. It was finished in 1962. At that time, it was the most powerful computer in the world. Atlas was very special because it used virtual memory. This is a way to give each user a lot of storage space. Atlas also used an operating system. An operating system is a set of rules that helps the computer work. These early machines helped make the computers we use today.

For thirty years, a small team at the University of Manchester changed how we use technology. Led by Tom Kilburn, they created a series of innovative electronic computers between 1947 and 1977.

These machines were not just tools for math. They were meant to test new ideas like the Williams tube. This was an early way to store computer memory using cathode-ray tubes. The team wanted to see if computers could solve hard math problems. Their work led to the world's first stored-program computer and the first transistorised computer. They even built the fastest computer in the world during the early 1960s.

One way these machines worked was by using parts called thermionic valves. These valves controlled electricity to run programs. The first machine, the Manchester Baby, used 550 of these valves. It was a huge machine that weighed almost one long ton.

On 21 June 1948, the Baby ran its very first program. This program had 17 instructions to find a specific math answer. It took 52 minutes to finish the task. Later, the team used transistors instead of valves. Transistors were smaller and used much less power. This helped make computers like the Manchester TC.

The history of these computers is full of famous names and big steps. After the Baby succeeded, the UK government asked a firm called Ferranti to build a commercial version. This became the Ferranti Mark 1.

In 1954, a machine called Meg began running programs. It was smaller and faster than the older Mark 1. By 1962, the team finished a massive machine named Atlas. Atlas was incredibly powerful for its time. It was even faster than four IBM 7094 computers combined. This machine helped create many ideas we still use today.

There are many important facts to remember about the Manchester series. The Manchester Mark 1 was operational by April 1949. The Final Specification version of that machine used 4,050 valves. It used 25 kilowatts of power to run.

The Atlas computer was commissioned on 7 December 1962. It could give each user one million words of storage space. Later, the MU5 project began with a large grant of £630,466 in 1968. By 1971, the staff for the MU5 project grew to 60 people. These machines were built in places like Manchester and Chilton near Oxford.

You can see how these old machines link to your life today. Many parts of the Atlas computer are kept in museums.

The Atlas computer pioneered the use of an operating system. An operating system is the set of rules that manages a computer. Atlas also used virtual memory to help users. Modern computers use these same basic concepts to work quickly. Even the design of the MU5 helped shape newer computers like the ICL 2900 series. Every time you use a computer, you are using ideas born in Manchester.

Between 1947 and 1977, a small research team at the University of Manchester changed computing forever. Led by Tom Kilburn, they developed a series of innovative electronic computers. These machines were not just tools for solving math problems. They were experimental platforms used to test groundbreaking ideas. The team aimed to prove the practicality of the Williams tube. This was an early form of computer memory that used cathode-ray tubes, or CRTs. Their work led to several world firsts. They created the first stored-program computer and the first transistorised computer. They also built the world's fastest computer in 1962.

The first machine in this series was the Manchester Baby. It was designed as a test-bed for the Williams tube memory. Work on the Baby began in 1947. On 21 June 1948, it ran its first program. This program used 17 instructions to find the highest proper factor of 218. The calculation took 52 minutes to complete. The machine was quite large and heavy. It weighed almost one long ton. To function, it used 550 thermionic valves. These valves included 300 diodes and 250 pentodes. The Baby consumed 3.5 kilowatts of power during operation.

Following the success of the Baby, the team developed the Manchester Mark 1. This machine was intended to be a more practical computing facility. In October 1948, the UK Government Chief Scientist, Ben Lockspeiser, saw a prototype. He was so impressed that he started a contract with the firm Ferranti. This partnership produced the Ferranti Mark 1, the first commercial general-purpose computer. The Mark 1 introduced index registers. This is a feature that is commonplace on modern computers today. The Final Specification version of the Mark 1 was fully working by October 1949. It contained 4,050 valves and used 25 kilowatts of power.

As the team gained experience, they shifted focus toward scientific computing. In 1951, they began designing a machine with a floating-point unit. This resulted in the Meg, or the megacycle machine, in 1954. The Meg was smaller and faster than the Mark 1. Ferranti later produced a commercial version called the Ferranti Mercury. This version replaced the Williams tubes with more reliable core memory. Around this same time, researchers worked on the Manchester TC. This project aimed to build a smaller, cheaper computer using transistors. The first version became operational on 16 November 1953. It used 92 point-contact transistors and 550 diodes.

While transistors were efficient, they were not always reliable at first. Early transistors in the Manchester TC caused failures about every 90 minutes. Reliability improved when the team began using junction transistors. The 1955 version of the TC used 250 junction transistors and 1,300 solid-state diodes. This version consumed only 150 watts of power. This was a massive reduction from the older valve-based machines. The design of the TC was later used by Metropolitan-Vickers. They built the Metrovick 950 using junction transistors. Six of these machines were completed starting in 1956.

In 1956, development began on a machine called MUSE, or the microsecond engine. The goal was to reach processing speeds of one million instructions per second. Ferranti joined the project, and the machine was renamed Atlas. The first Atlas was commissioned on 7 December 1962. It was the most powerful computer in the world at that time. Atlas was equivalent to the power of four IBM 7094 computers. It pioneered the use of virtual storage and paging. This allowed each user to have one million words of storage space. Atlas also introduced the Atlas Supervisor, a very early modern operating system.

The MU5 project was the successor to the Atlas system. In 1968, the Science Research Council gave the university a grant of £630,466. By 1971, the project staff grew to 60 people. The MU5 processor used associative memory to speed up data access. This was helpful because the processor cycle was five times faster than the main memory. The MU5 also used a specialized operating system called MUSS. This system was highly adaptable and could run on different processors. The MU5 was fully operational by October 1974. Its design heavily influenced the ICL 2900 series of computers.

The legacy of the Manchester computers connects directly to modern technology. Many of the concepts used in the Atlas, such as virtual memory, are still used today. The design of the MU5 helped shape the development of the ICL 2980. Even the project's influence reached into the 1980s with the MU6 series. These machines were built in locations like Manchester and Chilton near Oxford. Today, parts of the Atlas computer are preserved in museums. The work done by this small team created the foundation for the digital age.