News

Happy (belated) birthday, ENIAC

Not the first digital computer, but the best known

Jon Peddie

All projects have four birthdays. They are: inception, funding, completion, and announcement, and maybe a fifth—first sale. The Electronic Numerical Integrator and Computer (ENIAC) was conceived and built-in Philadelphia, Pennsylvania, and completed in 1945. The project’s inception can be traced to August 1942, when Dr. John W. Mauchly produced a seven-page memorandum—The Use of High-Speed Vacuum Tube Devices for Calculation. The concept was embraced, and a project was initiated to build a machine that could generate gunnery tables for artillery being developed by the U.S. during WWII. In 1943, the project was funded by the U.S. Army Ordnance Department and the Ballistic Research Laboratories. Dr. John W. Mauchly, a physicist, and J. Presper Eckert, Jr., a young electronics engineer, were assigned to the project. Completed in 1945, it wasn’t until February 14, 1946, that ENIAC was shown to the public.

 It was always a headline grabber. From the number of vacuum tubes it used (17,000 to 18,000 as binary switches), the iconic photographs of a giant calculating machine (and the spiritual model for dozens of science-fiction stories and movies), to the high-profile projects, it was used on like the hydrogen bomb and Army enlistment ads. Also, there was heavy promotion of its status by the University of Pennsylvania and Mauchly and Eckert, who parlayed their work and helped establish the Univac Corporation.

But it wasn’t the first.

Picking the first of anything is fraught with definitional arguments. That’s illustrated in a table in Wikipedia’s History of computing hardware, which I encourage you to look at. The following table is part of a larger table included in the Wikipedia article.

The forerunners of our modern computers were developed in the 1940s. Earlier computers were based on the decimal system, they were mechanical, and usually fixed function. Historians have also come to use the term Turing Complete to define the capabilities of computers. A machine can be said to be Turing Complete, or computationally universal if it can be used to simulate any Turing Machine. For more on this, Alan Turing defined the challenge of computing in his paper “On Computable Numbers, with an Application to the  Entscheidungsproblem.”

Computer Shown Workin Binary Electronic Programable Turing Complete
Zuse Z3 May 1941 Yes No By punched film stock Yes
Atanasoff-Berry Computer Summer 1941 Yes Yes No No
Colossus December 1943/ January 1944 Yes Yes Partially, by rewriting No
Harvard Mark I – IBM ASCC 1944 No No By punched paper tape Yes
ENIAC 1944 No Yes Partially, by rewriting Yes
Small-Scale Experimental Machine (Manchester Baby) 1948 No Yes By Function Table ROM Yes

 

Mr. Zuse (yes, that’s his real name) was a civil engineer and tinkerer. In the mid to late 1930s, using relays, he built the first programmable digital computer in his parent’s flat. 

Ing. Konrad Zuse (courtesy of ArtMechanic (CC BY-SA 3.0)

 

It was a binary machine and contained floating-point number capability. Ing. Zuse used punched film to program the machine. Zuse’s decision to use switches, which are binary—required the machine to use binary math. 

Zuse Z1 replica in the German Museum of Technology in Berlin (courtesy of BLueFiSH.as (CC BY-SA 3.0)

 

In 1946, IBM took an option on Zeus’ patents. The Zuse Institute Berlin is named in his honor.

Between 1939 and 1942, Professor John Atanasoff designed the first vacuum-tube digital computer.

Clifford Berry (1918–1963) (© www.computer-enthusiast.com) and John Atanasoff (1903–1995)  (courtesy of Iowa State University)

 

Graduate student Clifford Berry helped construct the prototype of the ABC (Atanasoff-Berry Computer) at Iowa State College.

The ABC Atanasoff–Berry computer (courtesy of Iowa State University)

 

In 1942, John Atanasoff went to Washington, D.C., to do physics research for the U.S. Navy. The ABC machine was later dismantled.

Meanwhile, the British developed another vacuum-tube digital computer during the war to crack German encryption codes and gave it the best name ever—Colossus

A Colossus Mark 2 computer being operated by the Women’s Royal Naval Service (WRNS). The slanted control panel on the left was used to set the “pin” (or “cam”) patterns of the Lorenz. The “bedstead” paper tape transport is on the right (The National Archives, United Kingdom)

 

The U.K. government destroyed the Colossus machines in 1945 to maintain secrecy. Their existence only became generally known in the 1970s.

If you think this is fascinating, I encourage you to buy my book, The History of Visual Magic in Computers: How Beautiful Images are Made in CAD, 3D, VR, and AR. And if you wonder how all this information came about, thank the lawyers.

In an attempt to gain a dominant position in the emerging computer market, Mauchly and Eckert filed for a patent in 1947. However, U.S. patent, 3120606, wasn’t granted until February 1964. Then, during a 1973 landmark federal court case, Honeywell v. Sperry Rand, it was discovered that Mauchly had visited Atanasoft in June 1941. That was almost a year before Mauchly published his paper (which was said he started writing on the train home from Iowa). As a result, the U.S. patent office voided the patent. The court ruled to put the invention of the electronic digital computer in the public domain. It provided legal recognition to Atanasoff as the inventor of the first electronic digital computer, even though Zuse really was first. Maybe the distinction was the use of vacuum tubes.

Home of the first digital computer, Neukirchen Germany

 

In 1937, Zuse submitted two patents that proceeded a von Neumann architecture. In 1938, he finished the Z1 which contained some 30,000 metal parts. On 30 January 1944, during a British air raid, the Z1 and its original blueprints were destroyed, along with his parents' flat and many neighboring buildings. The home was reconstructed and stands today.

ENIAC was a useful machine up till October 1955. While working on the hydrogen bomb at Los Alamos National Laboratory, von Neumann became aware of the computer and ran the first test problems for the bomb. ENIAC was also instrumental in developing the Monte Carlo method which later played a major role in ray tracing.

The original contract amount was $61,700; but like all government projects, the final cost was much higher—almost $500,000, which would be approximately $ 7,266,166 today. Imagine what the project would be bid at today if such a project were proposed—$7.3 million would seem cheap and unrealistic.

As good as it was, the ENIAC wouldn't be any good at mining Ethereum.