The Moon and a Computer

by Richard Passov

“…If I were to say my fellow man that we shall send to the moon 240,000 miles away from the control center station in Houston a giant rocket more than 300 ft tall the length of this football field made of new metal alloys some of which have not yet been invented capable of standing heat and stresses more than have ever been experienced fitted together with more precision than the finest watch carrying all of the equipment needed for propulsion, guidance control, communication, food and survival on an untried mission to an unknown celestial body and then return it safely to earth re-entering at speeds of over 25,000 miles per hour causing heat about half that of the surface of the sun and do all this and do it right … then we must be bold.”

—President John F. Kennedy

On May 25th, 1961, President Kennedy gave a State of the Union address to a joint session of Congress.  Though only five months into his term, Kennedy had reached a low point. The prior month witnessed Yuri Gregorian become the first man to orbit earth and the failed invasion at the Bay of Pigs.

After acknowledging Russian resistance to a nuclear test ban treaty, Kennedy asked Congress to fund a rapid deployment force, a programing effort to counter Soviet and Chinese propaganda and a nation-wide effort to build fallout shelters to “…insure against an enemy miscalculation…”  

It was only in the final minutes that he turned to space.   The time had come, he said, for “…a great new American enterprise, which may hold the key to our future on earth.” That enterprise was to send a man to the moon and back before the end of the decade “…in full view of the world.”  And so the Apollo Space Program was launched.

Rather than assume the chairmanship of the Space Council, established as part of the 1958 National Aeronautics and Space Act and which Eisenhower tried to disband, Kennedy saw it as a parking spot for his troublesome Vice President.  Before taking office, he asked for an amendment to the Act to allow Lyndon Johnson to assume the chairmanship.

Soon into office, facing a pressing need for a Cold War Victory, Kennedy asked Johnson whether there was any “…space program which promises dramatic results in which we could win.”  The question behind the question was:  Could the U.S. beat the Soviets to the moon?  Johnson turned to Werner von Braun, the former Nazi then leading the fledgling US efforts in rocketry.  Von Braun gave just a “…sporting chance…” to beating the Russians to a manned orbit of the moon, especially as the Soviets appeared “…willing to waive certain … safety features.”

When it came to a soft moon landing, Von Braun held a different view. Since getting men to the moon and back would require a “…performance jump by a factor of 10 over present rockets…” Von Braun reasoned that the U.S. could avoid simply aiming for the next step in the race, which it was sure to lose, and instead “…with an all out crash program…” beat the Soviets to “…a first landing of a crew on the moon, including return capabilities, of course.”

*          *          *          *          *          *          *          *          *          *          *

The Apollo Moon Landings have been described as the single greatest technological achievement of mankind.  The thimbleful of technology that guided the Apollo crafts, created at the very beginning of our digital era, contains the foundation of modern computer architecture.

The first computer science degree program at an American university began in the year following the start of the Apollo program. Whether Apollo mission clocks would be analogue or digital was vetted at the highest levels.

At a laboratory on the MIT campus, a small core of mathematicians and engineers created the software for the Apollo Guidance Computer. They solved the math of guidance, encoded their solution, then built the machine that ran their code.

In 1970, the Lab published a 337-page boast:

“The great achievement of this Laboratory was to supply the design for the primary hardware and software necessary to solve the Apollo guidance, navigation and control problem.

The quantum leap in technology nurtured by the Apollo program has been and should continue to be of immensely significant benefit to this country–socially, economically and in terms of its national esteem. It is the responsibility of all those who contributed to the proud achievements of Apollo to convince their countrymen of the directions this nation ought to follow in implementing these newly gained -and hard fought for-advances.”

What did the Lab design that took the Apollo crafts to the moon?

Two identical Apollo flight computers … one each in the command and lunar modules. The computer is the nucleus of the system in each spacecraft. Its computational capabilities are enormous for its size of only about one cubic foot and 70 pounds and for its operating power of approximately 70 Watts…   Apollo Guidance Navigation and Control MIT report, sec (3.6)

Just what were these capabilities?

All of the programs available to the astronauts were stored in 70k bytes of fixed memory.  Your cell phone has two million times more memory.

The Apollo Guidance Computer surveyed in two-second increments and landed four miles from its target.  As you back down your driveway a rear-view computer scans hundreds of times per second.

The Guidance Computer managed a state vector that kept track of position and velocity.  A new scan of environmental data updated four data points.  Calculations were performed and velocity and thrust angles adjusted.

The GPS in your cell phone communicates with satellites stationed in geosynchronous orbit above the earth.  Where you are, where you’re going, who else is also going, everything there is to know about the collective paths and all of the opportunities to shop and eat at your final destination – all brought to you by asking SIRI.

To initialize their state vector, just as a 17thcentury seafaring captain would, the Apollo Astronauts looked into the heavens through a navigational porthole and used a sextant to take a visual measure of the angle between two stars.

Before Apollo, ‘computers’ used for flight guidance were ‘analogue’.  Think of a grandfather clock as an analogue computer.  The Apollo computers were the first large-scale use of the then untested technology of integrated circuits in silicon.   At its peak, the Apollo computers consumed 70% of the world’s output of Integrated Circuits.

No software project of the scale required to support the Apollo missions had ever been contemplated. Project managers brought in military planners and invented the concepts of ‘Software Engineering’ and ‘Software Testing.’

Software has bugs. Hardware has bugs. Combining software and hardware magnifies the impact and complexity of bugs. And yet the Apollo computers had to remain ‘up’ at all times, no matter what.   So the concept of ‘Fault Tolerant’ was invented.

Intense discussions at the highest levels, sometimes lasting weeks, were driven by the need to save “…50 to 100 words of memory…” – to conserve a resource that today costs 1/10,000,000thof a penny.

Over the better part of a decade, two key criteria governed the functionality designed into the computer: 1) Whether the task was absolutely necessary to the mission and 2) whether it was absolutely clear that the computer bested humans in performing the task.

What fifty years ago was the technological frontier is today a simple toy. And yet that toy, combined with a deep sense of mission, allowed each person in a 400,000 strong army to see their contribution take the astronauts to Tranquility Base and back.

Sending humans where none had ever gone was essential to conquering the hearts and minds of men and women around the globe.  But at the cost of tremendous complexity, astronauts added little scientific advantage.

No astronauts meant no life support, no momentum shuffling about as the capsule swung toward the moon.  Caring for astronauts required that every part had to fit and work seamlessly lest the whole world watch a spectacular failure.

We may get to bigger challenges but the role of the individual in the challenge will never be the same.

*          *          *          *          *          *          *          *          *          *          *

There is no shortage of books on the Apollo Project.  Astronauts have described their experiences.  Eldon Hall wrote of his fight to build the Guidance Computer. Hugh Blair Smith added a witty chronicle of his contributions as a nascent software engineer while various historians have described the impact of the landings.  Don Eyles, writing from the vantage point of a young mathematician working on the heart of the computer, best captures the wonder of the mission and the spirit of the times:

“That was where we came in – we who wrote the flight code.  The program was a piano roll punched with music we composed, with its characteristic rhythms and polyphones, but with a melodic line that might sometimes surprise us as the internal logic, based on the immediate situation, selected among the alternate pathways that we made possible.  Our computer was a piano player that could improvise.  When it flew, our experience would be that of a composer attending the performance of a work he created, not listening like a member of the audience but living the music internally.”

Don Eyles, “Sunburst and Luminary, an Apollo Memoir.”

March, 2018 Fort Point Press

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Tweet about this on Twitter
Share on Reddit
Share on LinkedIn
Email this to someone