Reading the Manual for ENIAC, the World’s First Electronic Computer
Sometimes you have to take a long look back to realize just how much things have changed. And if you looked around our modern-day, cloud-enhanced web this month, you’d find several sites sharing memories about the launch of the ENIAC computer in 1946 — and of all those unstoppable mid-century engineers who tirelessly made it work.
ENIAC (Electronic Numerical Integrator and Compiler) was the world’s very first fully electronic general-purpose computer. Smithsonian magazine once called it “the room-size government computer that began the digital era.” And last week the I Programmer site shared a link to an original operating manual for ENIAC, originally published 75 years ago this month.
It’s dated June 1st, 1946 — it was published by the school of engineering at the University of Pennsylvania in Philadelphia — and the manual’s page at Archive.org show it’s been viewed just 2,309 times. (“There are no reviews yet,” reads the boilerplate on the site. “Be the first one to write a review.”)
The archive identifies it as part of “the bitsavers.org collection” — a project started by a software curator at the Computer History Museum, with over 98,500 files and more than 4.7 million text pages.
So what can we glean about the ENIAC’s moment in history from the manual which documents its operation?
It seems like the machine was temperamental. For example, it warns that the DC power should never be turned on without first turning the operation switch to “continuous.”
“Failure to follow this rule causes certain DC fuses to blow, -240 and -415 in particular.”
But the consequences are even worse if you opened the DC fuse cabinet when the D.C. power was turned on. “This not only exposes a person to voltage differences of around 1,500 volts but the person may be burned by flying pieces of molten fuse wire” (if one of the fuse cases suddenly blew). In fact, the ENIAC was actually designed with a door switch shunt that prevented it from operating if one of its panel doors was open, “since removing the doors exposes dangerous voltage.” But this feature could be bypassed by holding the door switch shunt in its closed position.
In a video shared by the Computer History Archives Project, chief engineer J. Presper Eckert Jr remembers that it was rare to go more than a day or two without at least one tube blowing out.
And in addition to potential shocks, dust was another potential hazard. “Dust particles may cause transient relay failures,” the manual warns, “so avoid stirring up dust in the ENIAC room.”
“Also, if any relay case is removed, always replace in exactly the same position in order not to disturb dust inside the case.”
Eniac Operating Manual A look at the ENIAC Operating Manual published 75 years ago this month reveals just how difficult it was to program the world's first electronic general-purpose computer and how frustrating it must have been for its programmers. https://t.co/nTb4vwp9bD
— I Programmer (@Iprogrammerinfo) June 8, 2019
The ENIAC used an IBM card reader, but that had its own issues too. At one point the manual actually recommends against having the same number in every column of a punchcard, since “this weakens a card increasing the probability of ‘jamming’ in the feeding mechanism of the IBM machines.”
Despite these limitations, ENIAC was a remarkable piece of technology. The manual includes intricate drawings and detailed diagrams of its racks, trays, cables, and wiring. But most important are the front panel drawings, which “show in some detail the switches, sockets, etc. for each panel of each unit.”
“They contain the essential instructions for setting up a problem on the ENIAC.”
ENIAC’s panels were equipped with neon lights corresponding to things like the “denominator flip-flop” and the “divide flip-flop.” The manual includes footnotes that carefully explain under what circumstances each light will be lit.
“The square root of zero is perhaps the easiest test to repeat on the divider-square rooter…”
It’s not until page 28 that it explains that turning on the start switch “starts the initiating sequences for the ENIAC, turning on the DC power supplies, the heaters of the various panels, and the fans…” And it also turns on a little amber light.
“When this sequence has been completed, showing that the ENIAC is ready to operate, the green light goes on…”
There were gates for a “constant transmitter” (which transmits to an “accumulator”), and its circuitry included “program pulse input terminals” — for add pulses and subtract pulses. And the machine also included two “significant figures switches.”
“When 10 or more significant figures are desired, the left-hand switch is set to 10 and the right-hand switch set so that the sum of the two switch readings equals the number of significant figures desired.”
There are tantalizing glimpses of how it all works together. The manual recommends a complicated test to make sure all the hardware is working properly. It involves a card with the value P 11111 11111, which gets input into the machine’s “accumulator” 18 times. The mathematical result — 19,999,999,998 — apparently exceeds the range of the accumulator, so the expected result is actually M 99999 99998. Then a card with the value P 00000 00001 is transmitted to the accumulators exactly twice — which instead of twenty billion (20,000,000,000) should give the value P 00000 00000.
“Note that this test assumes that the significant figure switch is set to ’10’…”
‘A Breathtaking Enterprise’
In Smithsonian magazine, technology writer Steven Levy remembers living in Philadelphia in the 1970s and renting an apartment from a man named J. Presper Eckert Jr. “It was only when I became a technology writer some years later that I realized that my landlord had invented the computer.”
In the early 1940s, Eckert had been a graduate student in the school of engineering who became the ENIAC’s chief engineer. A professor had proposed electronic calculations for munitions trajectories to help the American military during World War II.
Levy calls it “a breathtaking enterprise. The original cost estimate of $150,000 would rise to $400,000. Weighing in at 30 tons, the U-shaped construct filled a 1,500-square-foot room. Its 40 cabinets, each of them nine feet high, were packed with 18,000 vacuum tubes, 10,000 capacitors, 6,000 switches and 1,500 relays… Two 20-horsepower blowers exhaled cool air so that ENIAC wouldn’t melt down.”
By the time they’d finished building it — World War II was over.
But there was still work to do. The Atomic Heritage Foundation site reports that ENIAC was used to help perform the engineering calculations for the world’s first hydrogen bomb (along with two other more-recently developed computers). “It took sixty straight days of processing, all through the summer of 1951.”
Levy cites an Army press release describing ENIAC as a “mathematical robot” that “frees scientific thought from the drudgery of lengthy calculating work.”
A recent documentary called The Computers reminds modern-day viewers that the ENIAC’s original programmers were all women — Kay McNulty, Betty Jennings, Betty Snyder, Marlyn Wescoff, Fran Bilas and Ruth Lichterman.
There’s now also a site called the ENIAC Programmers Project that shares a brief overview of the documentary with more information. During World War II, the U.S. military had put together a team of nearly 100 women, trained in mathematics, who were calculating complex ballistic-trajectory equations. Six of them were selected to program the ENIAC.
Back in 1996, the IEEE Annals of the History of Computing ran a profile of “The Women of ENIAC,” interviewing 10 of the women who’d worked with the computer during its 10-year run.
The poster for the documentary describes them as “six women lost from history who created technologies that changed our world.”
The ENIAC was eventually left behind by ever-faster and ever-cheaper computers. “By the time it was decommissioned in 1955 it had been used for research on the design of wind tunnels, random number generators, and weather prediction,” remembers an ENIAC web page at Oak Ridge National Laboratory.
And even though ENIAC was decommissioned in 1955, 50 years later it was reassembled for a humble ceremony in Philadelphia, Levy remembers.
“Vice President Al Gore threw a switch and the remaining pieces clattered out the answer to an addition problem.”
According to Levy, the ENIAC’s chief engineer later groused “How would you like to have most of your life’s work end up on a square centimeter of silicon?” But Levy sees another way to look at it. “[T]he question could easily have been put another way: How would you like to have invented the machine that changed the course of civilization?”
Yet legacies aside, it also seems like it was a real thrill just to have been a part of the work itself.
“I’ve never seen been in as exciting an environment,” remembers Jean Jennings Bartik in the film. “We knew we were pushing back frontiers.”
And more than 60 years later, she also still remembered that the ENIAC computer “was a son-of-a-bitch to program.”
- How Margaret Hamilton, before leading the team who developed the Apollo in-flight software, wrote the programs for Edward Lorenz to establish Chaos Theory.
- Man builds a convincing chat-bot using a deep neural network trained on seven years of messages from Facebook.
- npm stops a scheme to steal cryptocurrency with malicious updates to an upstream code library.
- Quest for an Iron Man suit led the military to connect humans with machines in a revolutionary “data-driven, cognitive overhaul.”
- How French vintners are using drones.
- 20 years later, people are still playing the online game Everquest.