The Selective Sequence Electronic Calculator
This ‘electronic brain’ advanced pure science research, mesmerized the public and set IBM on a path to a whole new future in computers
Thomas Watson Sr. and Francis (Frank) E. Hamilton at the SSEC dedication in 1948

IBM unveiled the Watson Scientific Computing Laboratory in 1945 with a promise to support scientists using the computational horsepower of its best tabulators. Meanwhile, the company was also envisioning a whole new machine, one that could crunch big numbers at speeds previously unmatched in mechanized calculations. By removing time as a hindrance to math, IBM hoped to expand the horizons of pure science exploration.

The Selective Sequence Electronic Calculator (SSEC) would build on the breakthroughs IBM had achieved a year earlier with the Automatic Sequence Controlled Calculator, also known as the Mark I, the largest electromechanical calculator ever built and the first automatic digital calculator in the US. The SSEC would be “very much faster and more compact” than the Mark I, an IBM management summary explained, but its importance went well beyond functional upgrades.

The SSEC reflected the bold ambitions of modern scientific inquiry, IBM’s first CEO Thomas J. Watson Sr. said, and honored the advancements in engineering that made building the machine possible.

Built for science, not business

Wallace J. Eckert, a Columbia University astronomy professor, became the first director of the Watson Lab, which had been built on the school’s campus. With the cooperation of his colleagues, he defined most of the SSEC’s specifications and capabilities. The machine would be constructed to perform addition, multiplication, subtraction and division using high-speed recurring electronic pulses. It would include some form of internal storage for accumulating mid-process data. There would be other storage on punched paper tape. A master sequencing panel and punched card machines would perform input and output functions.

Eckert made it clear that the SSEC’s principal function would be to serve science, not business, arguing that “it would be a pity if this superb development were commercialized to such an extent that pure science were purely coincidental.” The Mark I, he noted, had failed to live up to its potential because it was sold to the US Navy for secret projects.

Watson Sr. roundly supported a public-minded mission for the machine. A plaque quoting Watson hung in the SSEC’s future home at IBM headquarters at 590 Madison Avenue, in New York City. It read: “This machine will assist the scientist in institutions of learning, in government, and in industry to explore the consequences of man’s thought to the outermost reaches of time, space and physical conditions.”

This machine will assist the scientist to explore the consequence of man’s thought to the outermost reaches of time, space and physical conditions Thomas J. Watson Sr. IBM’s first CEO
A mechanical marvel in midtown Manhattan

The company designed, built and placed the SSEC in operation in two years. The U-shaped assemblage of glass and metal cabinets went on display in a 60-foot-by-30-foot showroom at IBM headquarters, where pedestrians passing by could view its array of popping lights and whirring paper tape punches.

At launch, the SSEC had the largest internal storage capacity of any computer in existence. It could store 160 digits in its 12,500 vacuum tubes, 3,000 digits in its 21,000-plus electronic relays, and 400,000 digits on tape. Still, it was not a computer in the modern sense. It had no memory for software storage, and it was programmed using punched paper tape.

But the SSEC was fit for the audacious purpose for which it was designed: quickly doing basic math operations at a large scale. It could multiply two 14-digit numbers at a rate of 50 per second, divide the same two numbers 33 times per second, and produce 28-digit sums or differences at a rate of 3,500 per second. The SSEC was 250 times faster than the earlier Mark I.

This speed would prove valuable in many vocations in the coming years, but some of the earliest applications involved calculating the vibrations of ship hulls, the flow of oil through the ground, and the orbits of major planets.

The SSEC was faster than the earlier Mark I: Multiply  

It could multiply two 14-digit numbers up to 50x faster

Divide  

It could divide the same two numbers up to 33x faster

Sum  

It could produce 28-digit sums or differences up to 3,500x faster

Moon math

At the New York City dedication for the SSEC on January 27, 1948, the project’s chief engineer, Frank Hamilton, outlined the SSEC’s first big project: calculating the position of the moon over time. Each position required a vast array of inputs: 165,000 digits; 1,100 lines of instruction expanded to 10,350 lines through reference tables; 11,000 additions and 9,000 multiplications; and 2,000 references to a hundred-value sine table.

The company would put the SSEC through its paces by calculating the moon’s position at six-hour intervals over the past and ensuing 100 years — a series of calculations that would have taken a million mathematicians a lifetime to achieve. Calculating each position required more than 20,000 operations and an evaluation of more than 1,600 terms, which the SSEC could process in seven minutes. The entire project would take a couple of years.

Eckert, an early innovator in mapping lunar motion, used the SSEC to create a new, more accurate 20-year lunar ephemeris — a book of tables showing positions of the moon over time. Issued jointly by the Nautical Almanac Offices of the US and the UK, Eckert’s ephemeris became the official standard in every astronomical laboratory around the world. It formed the statistical foundation for later moon missions.

 

A media blitz for the ‘electronic brain’

For a machine designed specifically for scientific applications, the SSEC became the talk of the dinner table. It aroused the public’s interest in the potentially life-changing possibilities of technology. IBM promoted the machine’s capabilities through the media, which marveled at this new “electronic brain” while ruminating on its potential to make workers, mathematicians specifically, “as obsolete as the dodo bird.” Newscaster Alma Dettinger of WQXR radio in New York said, “They’re saving work for everybody nowadays. Pretty soon we’ll have nothing to do at all.”

IBM distributed more than 750 press packets among radio outlets, newsrooms and periodicals with a reach of almost 200 million people. Watson, for his part, was careful to describe the SSEC as a way to save time, not replace jobs. He referred to it as “a small tool” to help “great scientists” benefit mankind.

The SSEC would prove to be a vital bridge to IBM’s entry into commercial computing. No sooner was the SSEC humming away on scientific research than the company began moving on to its next iteration of computing, this one targeted for government and industry. Spurred on by the start of the Korean War in 1950, Thomas Watson Jr. formed an executive committee to define specifications for a new “Defense Calculator” that could be used for aircraft design, nuclear development and munitions manufacturing.

Watson Jr. recognized that the company was now in the electronics business and that “we’d better move pretty fast.” Actual design of the Defense Calculator, or model 701, began in February 1951. By the time assembly of the first production machines began in June 1952, the company already had 10 commercial orders in hand, including from the US Weather Bureau and General Electric.

The first production machine shipped from Poughkeepsie to the IBM Technical Computing Bureau at IBM headquarters in New York, where it would operate in the space previously occupied by the SSEC. The 701’s first public showing was on April 7, 1953.

During its five-year run as one of the world’s most powerful and best-known “electronic brains,” the SSEC stimulated the public’s imagination about the possibilities of computer technology, and fulfilled its purpose solving the toughest math problems of the day. It paved the way technologically and prepared society emotionally for the coming computer age.

For a machine designed specifically for scientific applications, the SSEC became the talk of the dinner table
Related stories Wallace Eckert

He created new machines to support science and explore unknown territory

The origins of IBM Research

IBM has always provided its scientists the freedom to explore and discover solutions to society's greatest problems

The IBM 700 series

Establishing IBM’s reputation in the mainframe market — and the electronics industry at large