A wartime calculation problem
ENIAC, short for Electronic Numerical Integrator and Computer, was built at the University of Pennsylvania’s Moore School of Electrical Engineering. Its purpose was practical and military. The U.S. Army needed firing tables for artillery: charts that tell gunners how to aim to hit a target at a given distance under given conditions. Producing one table meant calculating many trajectories, and doing that by hand or with desk calculators was slow, even with large teams of human calculators working on it.
The Army funded the project in 1943. John Mauchly, a physics instructor, proposed the idea of an electronic calculator, and J. Presper Eckert, a young engineer, led much of the hardware design. Army officers, including Herman Goldstine as a liaison, supported the work. Dozens of engineers, technicians and others contributed over the following years.
What the machine was like
ENIAC was large. It filled a big room and is commonly described as using about 18,000 vacuum tubes and weighing roughly 30 tons. Those figures vary a little between sources, but the scale is not in doubt. It also drew a lot of electrical power and needed constant maintenance, because tubes burn out.
Two features stand out:
- It was electronic. Its arithmetic was done by switching tubes rather than by moving parts, which made it far faster than the relay and mechanical calculators that came before it.
- It was decimal. Unlike most later machines, ENIAC represented numbers in decimal form rather than in binary numbers.
It was also programmable in a way that was very different from today. To change the task, operators set switches and plugged cables into panels, in some ways like an old telephone switchboard. That could take a long time, even though the calculation itself, once set up, ran quickly. The instructions were not held in the machine’s memory the way they are in a modern computer.
The people who programmed it
Much of the setup work was done by a group of women who had been hired as human calculators and then selected to work on the new machine. They are usually named as Kay McNulty, Betty Jean Jennings, Betty Snyder, Marlyn Wescoff, Fran Bilas and Ruth Lichterman. With little documentation to go on, they worked out how to turn a mathematical problem into a pattern of cables and switch settings. For many years their contribution was overlooked in popular accounts, and historians have since done a good deal to correct that. Their story is part of the wider history of the early programmers of the electronic era.
Completion and public debut
The machine was completed in 1945, too late to affect the war. Early work included calculations connected to secret projects at Los Alamos, and ballistics calculations continued after the war. The Army presented ENIAC to the public in February 1946, and newspapers covered it with wonder, sometimes calling it a giant brain. It was later moved to the Army’s Aberdeen Proving Ground in Maryland, where it stayed in use for several more years until it was shut down in the mid-1950s.
How its ideas spread
ENIAC’s influence came in several ways, and it helps to separate them.
Proof that large electronic machines could work. Many experts had doubted that a machine with so many tubes would run reliably. ENIAC showed that, with care, it could, which encouraged funding and further projects.
The design that came next. While ENIAC was being built, the Penn team and consultants, including John von Neumann, began discussing a successor, EDVAC, with instructions stored in memory. A 1945 draft report on it was widely circulated, and the ideas in it are now often grouped under the stored-program concept. Credit for those ideas is shared and sometimes disputed among several people, which is typical for this period. ENIAC itself was later altered, around 1948, to run in a stored-program style using its function tables, mainly for convenience.
Teaching others. In 1946 the Moore School ran a summer lecture course that brought engineers and scientists from various institutions to learn about electronic computer design. Several people involved in building later machines in the United States and Britain were connected with it.
Commercial spin-off. Eckert and Mauchly left the university and started a company to build computers for sale, which led to the UNIVAC line. That story belongs to the early commercial computers.
What ENIAC was not
Popular accounts sometimes call ENIAC the first computer, full stop. That claim is too simple. Konrad Zuse’s relay machines in Germany, the Atanasoff-Berry machine in Iowa and Britain’s Colossus codebreaking machines all came before or around the same time, each with different capabilities. ENIAC was general-purpose and electronic at a large scale, and that is its real distinction.
A well-known 1970s court case over its patent ended with a ruling that the patent was invalid, in part because of earlier work by others, including Atanasoff. The decision is often cited in debates about who deserves credit, though it was a legal ruling rather than a final word from historians. The broader lesson, discussed under the question of who invented the computer, is that this was group work across several countries.
Why it still matters
ENIAC’s direct design was a dead end: later machines were binary, stored their programs and were smaller and more reliable. But it showed what electronic speed could do for science and engineering, trained the people who went on to build its successors, and helped convince governments and companies that the investment was worthwhile. The path from ENIAC to the machines on our desks is indirect, but it runs through the people and ideas that came out of that project.