Space 4 November 2026 4 MIN READ
The Apollo Guidance Computer
36,864 words of memory woven by hand, and a landing that survived five program alarms because the software knew what to drop.
THE SHORT ANSWER
The Apollo Guidance Computer had 36,864 words of woven read-only memory and 2,048 words of erasable memory. During the Apollo 11 landing it raised alarms 1201 and 1202 because a radar interface was stealing memory cycles. The software was designed to drop minor jobs and restart, so the landing continued.
On this page
The numbers, and a warning about units
The Apollo Guidance Computer that flew the Lunar Module had 36,864 words of fixed memory, which could not be changed in flight, and 2,048 words of erasable memory. Draper Laboratory’s own reliability report tabulates exactly those figures for the Block II version. Each word was 16 bits: 14 data bits, a sign bit and a parity bit.
In bytes, the Smithsonian gives 72 kilobytes of read-only and 4 kilobytes of read-write memory, while MIT News counts the same capacity differently, at about 36 KB and 2 KB. Words are the safe unit.
Other figures vary by source. NASA’s history of spaceflight computing says the clock ran at one megahertz and puts the Block II at 24 by 12.5 by 6 inches, 70.1 pounds and 70 watts. MIT lists 70 pounds, 55 watts and 0.97 cubic feet, which Draper’s 1972 report matches. A memory cycle took 11.7 microseconds, and one addition 23.4 microseconds, according to the Apollo Flight Journal.
Software you could hold
The fixed memory was core rope: a bit was stored by a wire. A wire threaded through a tiny magnetic core read as a 1, and a wire that bypassed it read as a 0. The ropes were woven at Raytheon, mostly by women hired for their experience in the textile industry, and a finished rope could not be edited. Any software change meant weaving a whole new one.
Margaret Hamilton led software development at the MIT Instrumentation Laboratory. The Smithsonian records that her team jokingly called the rope weavers “LOLs”, for little old ladies, and mysterious minor bugs “Funny Little Things”. Her title for overseeing code production and the threading of code into ropes was “Rope Mother”.
Program alarms during the descent
During the Apollo 11 descent, the computer showed the 1202 program alarm, then a 1201, then more. MIT’s letter to NASA of 4 August 1969 counts five Executive overflow alarms in the landing, in the order 1202, 1202, 1201, 1202, 1202.
What the codes meant: the computer’s scheduler had run out of work areas. It lit the program light, stored the alarm code and jumped to a restart routine that stopped everything and restarted only the essential jobs. In the source code the comment for the 1202 alarm reads “No more core sets”, according to American Scientist.
The cause, in MIT’s diagnosis, was a flood of spurious counter pulses from the rendezvous radar interface, each stealing one memory cycle. The designers had required a tolerance of about 10% lost cycles during landing; the radar interface took about 15%. American Scientist puts the pulses at up to 12,800 per second. A contributor to the Lunar Surface Journal explains the cause differently, so it is described here as MIT described it in 1969.
A decision in Houston
Steve Bales, the guidance officer in Mission Control, judged that the overflow would not jeopardise the landing. He said “Go” on the first 1202 and on each of four further alarms before touchdown. Houston knew the computer would recover because MIT had tested the restart capability extensively. After each alarm, the computer restarted the important work, steering the descent engine and driving the display, and dropped the spurious radar jobs.
Hamilton’s own recollection, in a 1992 oral history for the National Air and Space Museum, is that she was at Draper Laboratory, not in Mission Control, when the alarms began. She says they meant a process overload that the software was designed to survive.
Why it matters
The famous alarm was the machine doing what it was designed to do. A computer with only a few kilobytes of memory landed on the Moon because it knew how to drop unimportant jobs when it was overloaded.
Our take: design for the bad day. The Apollo computer was not powerful. It was built so that when it was overloaded, it kept doing the most important thing.
Do you agree?
Take it to Chevlume: write your view, add your sources and let others answer. The closed alpha is by invitation.
Keep reading
The Voyager Golden Record
What is on the disc launched into interstellar space in 1977, how an alien is meant to play it, how long it may last and the arguments around it.
SPACE / DISCOVERYMars Climate Orbiter: Pounds and Newtons
One unit mix-up put a Mars orbiter about 170 km too low. What the NASA board found about the people and process failures behind it.
SPACE / DISCOVERYAriane 5 Flight 501: Thirty-Seven Seconds
How reused software and one unprotected number destroyed a new rocket in 1996, and what the official inquiry recommended.
Sources
- Apollo 11 Lunar Surface Journal: Program Alarms · NASA History (Apollo Lunar Surface Journal) · Accessed
- Reliability History of the Apollo Guidance Computer (Eldon C. Hall) · Charles Stark Draper Laboratory via NASA Technical Reports Server · Accessed
- The "Rope Mother" Margaret Hamilton · Smithsonian National Air and Space Museum · Accessed
- Behind the scenes of the Apollo mission at MIT · MIT News · Accessed
- Computers in Spaceflight: The NASA Experience, Ch. 2 - The Apollo guidance computer: Hardware (J. E. Tomayko) · NASA History Office · Accessed
- The Apollo On-board Computers (Phill Parker, Spaceflight, Oct 1974) · Apollo Flight Journal (NASA History) · Accessed
- MIT and Apollo (AeroAstro magazine) · MIT Department of Aeronautics and Astronautics · Accessed
- Exegesis of the 1201 and 1202 Alarms Which Occurred During the Mission G Lunar Landing (AG# 370-69) · MIT Instrumentation Laboratory (letter to NASA Manned Spacecraft Center) · Accessed
- Moonshot Computing · American Scientist (secondary source) · Accessed
- Apollo 11 Lunar Surface Journal: The First Lunar Landing · NASA History (Apollo Lunar Surface Journal) · Accessed
- Remembering Margaret Hamilton Through Her Own Words · Smithsonian National Air and Space Museum · Accessed
Published
Discuss on X