The Eckert Tapes: Computer Pioneer Says ENIAC Team Couldn't Afford to Fail -- and Didn't

The all-electronic system made its debut 60 years ago. In interviews taped in 1989, co-inventor J. Presper Eckert discusses the technology behind ENIAC and debunks some myths.

There are two epochs in computer history: before ENIAC and after ENIAC. While there are controversies about who invented what, there's universal agreement that the Electronic Numerical Integrator and Computer was the watershed project that showed all-electronic digital computing was practical. ENIAC was unveiled Feb. 14, 1946, after nearly three years of development at the University of Pennsylvania's Moore School of Electronics. The two men most responsible for its success were J. Presper Eckert and John W. Mauchly, who together went on to build commercial systems such as Univac and also founded one of the companies that merged to form Unisys Corp.

Eckert died in 1995. I recorded two days of interviews with "Pres" in 1989, when he was 70 years old. My father was Eckert's best friend -- as a child, I played with his children, and I visited him regularly as an adult. I sat on the interview tapes for many years but decided to transcribe them for ENIAC's 60th anniversary and release the text publicly. Excerpts from the interviews follow:

Eckert was chief engineer on the ENIAC project.
1pixclear.gif
Eckert was chief engineer on the ENIAC project.
1pixclear.gif
How did calculating machines work before ENIAC? Well, a person with a paper and pencil can add two 10-digit numbers in about 10 seconds. With a hand calculator, the time is down to 4 seconds. The Harvard Mark 1, an electromechanical computer, could add two 10-digit numbers in 0.3 seconds, about 30 times faster than paper and pencil. The ENIAC was the first electronic digital computer and could add those two 10-digit numbers in 0.0002 seconds -- that's 50,000 times faster than a human, 20,000 times faster than a calculator and 1,500 times faster than the Mark 1. For specialized scientific calculations, it was even faster.

So it's a myth that ENIAC could only add, subtract, multiply and divide. That's a calculator. ENIAC could do three-dimensional, second-order differential equations. We were calculating [artillery] trajectory tables for the war effort. The trajectory tables were calculated by hundreds of people operating desk calculators -- people who were called "computers." So the machine that does that work was called a computer.

There's a story that ENIAC dimmed the lights in Philadelphia when it was in use. That story is total fiction, dreamed up by some journalist.

Did the military guys working on ENIAC salute the machine? Another ENIAC myth.

How many tubes did ENIAC use? ENIAC had 18,000 vacuum tubes. The tubes were off-the-shelf; we got whatever the distributor could supply in lots of 1,000. We used 10 tube types but could have done it with four; we just couldn't get enough of them. We decided that our tube filaments would last a lot longer if we kept them below their proper voltage -- not too high or too low. A lot of the circuits were off-the-shelf, but I invented a lot of the circuits as well. Registers were a new idea. So were integrator circuits.

Are any of your circuits still in use in personal computers? No, but that's true of any first invention. Edison's original light bulb bears no resemblance to a modern bulb. They do the same thing but with totally different components. Same with the computer. What did survive were the concepts, not the hardware. The idea of a subroutine was original with ENIAC. Mauchly had this idea based on his knowledge of the inner workings of desk calculators. On Mark 1, if they wanted to do a calculation over and over, they had to feed the same tape in over and over. We invented ways to run the same subroutine without any mechanical input. The idea of using internal memory was also original with ENIAC.

There's a story that some guy was running around with a box of tubes and had to change one every few minutes. Another myth. We had a tube fail about every two days, and we could locate the problem within 15 minutes.

Was ENIAC programmable? Yes and no. We programmed the machine by plugging wires in from place to place. That's not hard-wired; it's not software; it's not memory. It's pluggable programming. And we had switches to set the functions.

How old were you? We signed the contract on my 24th birthday, May 9, 1943. What prepared you for building an electronic computer? Remember, in that era, Philadelphia was "Vacuum Tube Valley." Radios and televisions were predominantly made in Philadelphia. I worked on primitive television at Farnsworth [Television] back as a teenager, and at Penn I had been working on various radar problems trying to measure the time for a pulse to go out and come back. All this is a good lead-in for building an electronic computer.

Was it you, or was it the times? Well, I may have been uniquely prepared. I was very good in math, and I was fascinated with all electronics. I was designing electronic gadgets as a kid. Maybe I had the right fusion of interests. But every inventor stands on the pedestals built by other people. If I hadn't done it, someone else would have. All that any inventor does is accelerate the process. The main thing was we made a machine that didn't fail the first time. If it had failed, we might have discouraged this line of work for a long time. People usually build prototypes, see their errors and try again. We couldn't do that. We had to make it work the first time out.

When you were working on ENIAC, did you have any inkling that these things eventually would be laptop-size and that everyone would own one? No one had any idea the transistor and chip technologies would come along so quickly. It is shocking to have your life work reduced to a tenth of a square inch of silicon.

ENIAC, which debuted Feb. 14, 1946, contained 18,000 vacuum tubes.
ENIAC, which debuted Feb. 14, 1946, contained 18,000 vacuum tubes.

Randall is a professor of communication at the University of the Virgin Islands and director of the school's new computer communication laboratory.

Join the discussion
Be the first to comment on this article. Our Commenting Policies