Bench Talk

“I Believe We Should Go to the Moon.”

With these words, President Kennedy launched the space program.

Back in 1961, these were words that today would sound like, “I Believe We Should Go to Mars.” This was a difficult goal, when a single google query today “sets in motion as much computing as it took to send Neil Armstrong and eleven other astronauts to the moon. Not just the actual flights, but all the computing done throughout the planning and execution of the 11-year, 17 mission Apollo program.” This is hard to believe, but it’s in a post from the official Google search blog.

Computing for the Apollo space missions was spare, by today’s terms. There was one Apollo Guidance Computer (AGC) on the Command Module and one on the Lunar Module for all missions but one. The AGC was the first computer that used integrated chips; it was the first embedded computer. The first version of the AGC used 4,100 single three-input NOR gates by Fairchild Semiconductor. It had a real-time operating system and ran at a blazing 40KHz with a 16-bit word length. A second version of the AGC was designed in 1966 and used 2,800 dual three-input NOR gate ICs. Modern engineers would cry if they had to go to the moon with this kind of computing power. Certainly there was no room for commenting the source code! When I was a kid and complained about walking a mile to school in the rain, my dad would tell stories of walking 5 miles to school in the snow “uphill both ways.” So next time you hear a programmer complain about not enough memory… just mention the Apollo 13 Mission had 64Kb on the AGCs that the astronauts used while in space. I wouldn’t think my garage door opener trustworthy if it had 64Kb, but trusting the lives of 3 crew members to 64Kb? The astronauts were test pilots and many wanted to fly it themselves. Indeed, in the event of an AGC failure, they would have been able to by following pre-dawn curves for attitude, velocity, and g-loading.

David Scott, astronaut for the Gemini 8, Apollo 9, and Apollo 15 missions described the importance of the AGC. He said, “When you come back from the moon, you really have to hit the corridor. If you have a basketball and a baseball 14 feet apart, where the baseball represents the Moon and the basketball represents the Earth and you take a piece of paper edgeways, the thinness of the piece of paper would be the corridor that you would have to hit when you came back. That’s only position. You have to hit it with the proper velocity, too. You have to have a good computer, and when you are approaching the reentry corridor you are thinking about that because you only have one chance.”

Going to Mars may sound like a nerd’s dream, but put it in perspective to what we had when we went to the moon. I believe we can go to Mars. Do you?

Sources:

You can download a PDF of the approximately 1,700 page “AGC Program Colossus by NASA” of 1966 here (warning, this is a large file.): https://authors.library.caltech.edu/5456/1/hrst.mit.edu/hrs/apollo/public/archive/1701.pdf

The Apollo Expeditions to the Moon: https://history.nasa.gov/SP-350/toc.html

A transcript of Dave Scott’s remarks on June 10, 1982 can be found via PDF download here: https://klabs.org/history/history_docs/ech/agc_scott.pdf

Major units of the CM Guidance, Navigation, and Control System, from The Apollo Guidance Computer: A Designer’s View by Eldon Hall, Designer of the AGC. (Source: https://ed-thelen.org/comp-hist/TCMR-V02.pdf)