Harvard Mark I

The Harvard Mark I, or IBM Automatic Sequence Controlled Calculator (ASCC), was one of the earliest general-purpose electromechanical computers used in the war effort during the last part of World War II.

One of the first programs to run on the Mark I was initiated on 29 March 1944 by John von Neumann. At that time, von Neumann was working on the Manhattan Project, and needed to determine whether implosion was a viable choice to detonate the atomic bomb that would be used a year later. The Mark I also computed and printed mathematical tables, which had been the initial goal of British inventor Charles Babbage for his analytical engine in 1837.

According to Edmund Berkeley, the operators of the Mark I often called the machine "Bessy, the Bessel engine", after Bessel functions.

The Mark I was disassembled in 1959; part of it was given to IBM, part went to the Smithsonian Institution, and part entered the Harvard Collection of Historical Scientific Instruments. For decades, Harvard's portion was on display in the lobby of the Aiken Computation Lab. About 1997, it was moved to the Harvard Science Center. In 2021, it was moved again, to the lobby of Harvard's new Science and Engineering Complex in Allston, Massachusetts.

Origins

The Harvard Mark I began as an idea in 1935 by Howard Aiken, a student at Harvard University. He wanted to build a machine that could solve hard math problems automatically. He shared his idea with IBM, and they agreed to help build it. Work started in 1939.

Because of World War II, Aiken had to pause the project for a time, but IBM kept working on it. The machine was finished in early 1943 and sent to Harvard in 1944. It was used to help the US Navy with calculations and was an important step in the history of computers, even though it wasn’t the very first working computer. The Mark I brought many ideas from Charles Babbage’s Analytical Engine to life, adding new features too.

Design and construction

The Harvard Mark I, also called the IBM Automatic Sequence Controlled Calculator, was made from switches, relays, rotating shafts, and clutches. It had 765,000 electromechanical parts and used hundreds of miles of wire. The machine was very big — about 51 feet long, 8 feet high, and 2 feet deep, and it weighed around 9,445 pounds.

It used a long drive shaft and an electric motor to power its calculations. The Mark I was built by IBM for Dr. Howard Aiken at Harvard University. An industrial designer named Norman Bel Geddes designed its outer case.

Operation

The Harvard Mark I could store 72 large numbers and perform basic math operations quickly. It read instructions from special paper tapes and processed them one after another.

Early on, people had to control the machine's steps by hand, but later updates let it follow more complex paths automatically. Some of the first people to work with this machine became leaders in computer science.

Instruction format

The Harvard Mark I used special tapes to control its operations. These tapes had 24 channels split into three groups of eight. Each group held important information: the first group told the machine where to store the result, the second group told it what numbers to use, and the third group told it what operation to perform. All these instructions were written in binary code, a simple system of zeros and ones.

Contribution to the Manhattan Project

John von Neumann worked on important calculations for building the atomic bomb as part of the Manhattan Project. In March 1944, he brought two mathematicians to Harvard to write a program for the Harvard Mark I computer. They used it to study how the first atomic bomb could be made to explode using a method called implosion.

The Mark I computer helped make these calculations much more precise than older methods. It could work with many more decimal places, which meant scientists could get better results for their important work. Von Neumann believed this implosion design would be very useful for future atomic bombs.

Aiken and IBM

Howard Aiken announced the Mark I in a press release, listing himself as the sole inventor. He mentioned only one IBM engineer, James W. Bryce, even though several others had helped build it. This upset IBM chairman Thomas J. Watson, who attended the dedication ceremony reluctantly on August 7, 1944. Because of this, Aiken decided to build future machines without IBM's help, and the machine became known as the "Harvard Mark I". Later, IBM built its own Selective Sequence Electronic Calculator to test new technology and showcase its work.

Successors

The Harvard Mark I was followed by several more machines. The Harvard Mark II came next in 1947 or 1948, improving on the Mark I but still using electromechanical relays. The Mark III/ADEC in 1949 used mostly electronic components like vacuum tubes and crystal diodes, but it still had some mechanical parts, such as rotating magnetic drums for storage. The Harvard Mark IV in 1952 was completely electronic, using magnetic-core memory instead of mechanical parts.

The Mark II and Mark III were given to the US Navy base at Dahlgren, Virginia, while the Mark IV was built for the US Air Force but stayed at Harvard. The original Mark I was taken apart in 1959, and parts of it are now on display in the Science Center at the Harvard Collection of Historical Scientific Instruments. In July 2021, it was moved to the new Science and Engineering Complex in Allston. Some pieces of the original machine are at IBM and the Smithsonian Institution.