14
$\begingroup$

Were the consoles simple display terminals or were they computers? If terminals were they tied into a computer for control purposes? If a change in trajectory was necessary was it performed from the console? Thanks. Any informative links would be appreciated

$\endgroup$
3
  • $\begingroup$ Do you mean the Apollo-era monochrome CRT displays we see in historical photos and videos, something like these or these or these? I think that the way that data was transmitted to and then displayed on these screens from a big computer elsewhere has been described in detail in som some previous questions and answers here. $\endgroup$
    – uhoh
    May 4, 2020 at 0:12
  • 2
    $\begingroup$ Obligatory Simpsons reference: frinkiac.com/meme/S05E15/896127/m/… $\endgroup$ May 4, 2020 at 13:49
  • $\begingroup$ or these $\endgroup$
    – uhoh
    May 4, 2020 at 23:38

4 Answers 4

22
$\begingroup$

Since you have tagged the question , I will explain the Apollo era consoles. Sources are from this Ars Technica article, which interviewed flight controller Sy Liebergot.

The flight controllers' consoles were "dumb"—they contained no computing elements and just displayed what came in from the mainframe.

In the Apollo era, the consoles were simply displays; they did not have any computing equipment inside. Signals originated inside the spacecraft, were processed into a telemetry signal which was transmitted to the Manned Space Flight Network, received by several dishes on Earth, switched through the Goddard Space Flight Center in Maryland, then downlinked to the Real Time Computing Center on the first floor of mission control in Houston. These real-time computers then formatted the data into something like TV channels that were sent to the consoles in the mission control room. The flight controllers could tune their monitor screens to whatever channel they needed. The results were also recorded on tape, which the flight controllers could order played back. Many consoles had indicator lights, which were also controlled by the real-time computers.

There were also support team in the "back rooms", which could talk on voice channels. The consoles had switches to choose which voice channels to listen to or talk on. The consoles could also place telephone calls to other rooms or even to the public telephone system. Each console had one set of green/yellow/red switches to indicate status to the Flight Director.

Most flight controllers simply monitored information and made suggestions. Only a few consoles could actually do something directly to the spacecraft. Only the CAPCOM (and rarely the Flight Director) could directly talk to the spacecraft. The Flight Director and Flight Dynamics Officer had "ABORT REQUEST" buttons, which would turn on a flashing red light in the spacecraft, prompting the astronauts to perform an abort. The GUIDO could uplink commands and data to the CM and LM computers, if the astronauts turned on the uplink. Everything else had to be done by the astronauts in the spacecraft.

More details are available in the linked article.

$\endgroup$
3
  • 1
    $\begingroup$ The implication being that if something happened that caused the astronauts to lose consciousness or communications, there would be basically nothing that ground control could do for them. $\endgroup$ May 4, 2020 at 20:00
  • $\begingroup$ Thnx for the great article!!! $\endgroup$
    – PaulObo
    May 9, 2020 at 20:16
  • $\begingroup$ Some more things could be commanded from the ground: - Range Safety Officer could command the booster to disable itself (triggering an abort), and then to destroy its fuel tanks, - CAPCOM could cause an alarm to sound in the spacecraft (to wake the crew up in an emergency during their sleep period), - someone could command limited amount of changes to the radio/antenna configuration, telemetry bitrate, and to the state of Data Storage Equipment (a tape recorder used when no radio contact with MSFN was possible). See UP TLM in history.nasa.gov/afj/aoh/aoh-v1-2-08-telecoms.pdf $\endgroup$ May 9, 2021 at 13:05
13
$\begingroup$

Were the consoles simple display terminals or were they computers?

They were simple display terminals throughout the Mercury/Gemini/Apolo era, and even well into the Shuttle era.

Do not be blinded by what is commonplace today. It's very important to realize that a state-of-the-practice laptop computer of today has a lot more computational capability, a lot more data storage capability, and a lot more throughput capability than did the very most powerful supercomputer from the Apollo era.

So-called "smart terminals" appeared a couple of years after the 1969 Moon landings. These "smart terminals", such as the IBM 3270, could buffer keyboard input (up to a limit) until the user hit the enter or return key, and could receive and then display hundreds of characters at a time rather than a single character at a time. These smart terminal capabilities of the early 1970s do not qualify as a computer.

1971 also saw the release of the very first microprocessor chip, the Intel 4004. This first microprocessor chip was supplanted in a few years by the first usable microprocessor chips, the Intel 8080 and then the Zilog Z80. It was these two devices that gave birth to the minicomputers, and later, personal computers.

These key 1971 milestone events in computing happened a couple of years after the Apollo 11 Moon landing -- too late to be of benefit to the Apollo problem. The architecture of the Apollo-era mission control center was in place well before the Apollo 11 Moon landing.

NASA later did switch from the mainframe-based architecture of the Apollo era to a distributed architecture, but that switch started in the mid to late 1980s, as a research project. The minicomputer-based Real Time Data System became operational in 1991, over two decades after the first Moon landing.

$\endgroup$
1
  • $\begingroup$ Thnx. I’m an old fart. I cut my teeth on the 8080 z80 and 6800! Appreciate the answer $\endgroup$
    – PaulObo
    May 9, 2020 at 20:18
8
$\begingroup$

Not only were the terminals not computers, they weren't even dumb terminals. They were just TV monitors that could be connected to any of various TV channels. The contents of each TV channel was generated by a computer that wasn't in the control room. The computer could command a console to attach to a TV channel in response to the console operator's request for a different display format, or the console operator could attach the display to a specific channel:

During Earth-orbital flights of the Apollo program, a complement of 28 computer-driven TV channels was sufficient. However, expansion to 36 channels was required for the Apollo lunar-landing missions.

Two control modes were provided for the TV display system. The capabilities are discussed in the following sections.

Display request mode. - An individual console may request a display format; the computer generates the data, formats the display, outputs the data to the next available computer-driven TV channel, and automatically connects that channel to the TV monitor of the requesting console. The concept of a computer-driven channel assignment was based on a "first come, first served" basis as opposed to consigning channels to displays or consoles. Initially, difficulties were experienced when all channels were being used and additional critical displays were needed. To provide positive control, a new display format was added to identify which display format was on each channel and to identify which console had requested the display. Through the use of this display, the flight control team could determine which display formats to release to the proper channels.

Channel attach mode. - In the channel attach mode, a console requests a given TV channel and receives whatever data are on that channel. This concept of sharing TV channels has been very workable and has resulted in the need for significantly fewer channels.

Apollo Experience Report - Flight-Control Data Needs, Terminal Display Devices, and Ground System Configuration Requirements by Richard A. Hoover, JSC, May 1974, p. 5

$\endgroup$
0
$\begingroup$

Technology of the early sixties did not support the manufacture of compact stand alone display terminals connected by a serial data transmission line. There were only small scale integrated circuits containing a few gates only.

More than 10 years later medium scale integrated circuits suitable for those terminals were available:

  • microprocessors for control
  • display memories to store display data
  • character generators to store pixel data for letters and numbers
  • timing circuits to generate pixel, character, horizontal and vertical frequencies and address

Therefore available TV monitors were used for displays connected to central controllers for 28 or even 36 channels. This way only one timing generator could be used for all display channels together. Also a central display memory and character generator read only memory used for all displays.

If the same display data screen could be used by several flight controllers, a single display channel could be connected to several TV monitors.

The central display control channels required much less electronic components than local controllers for each single display.

$\endgroup$
3
  • $\begingroup$ How did the master display controllers work? I can imagine that if one could stream data to a display controller at 243 times the horizontal refresh rate (about 3.82Mhz), had 36 delay lines that could delay data that was sent at that speed by 243 cycles, and had seven 64-byte ROMs for character shapes, one could use that circuitry with some multiplexing on the output to drive 80x24 displays. Would that be anything like what machines of that era actually did? $\endgroup$
    – supercat
    May 5, 2020 at 23:04
  • $\begingroup$ @supercat integrated 64-byte ROMs did not exist, they may have used discrete diode matrix ROMs. The use of delay lines is plausible, may be they used rotating magnetic drum memories too. Magnetic core memories were too slow for a video display. $\endgroup$
    – Uwe
    May 6, 2020 at 10:37
  • $\begingroup$ By "64-byte ROM" I meant a board with a diode row-decode matrix, feeding 64 transistors which would in turn feed rows of diodes for a column matrix. Having a separate ROM for each row of a character would mean that if different terminals were outputting lines at different times, one could have each terminal fed from a separate ROM without increasing the total amount of ROM circuitry. I hadn't thought about magnetic drums, but that would be a perfect technology for this since they need to output data at a very regular repeating pattern. $\endgroup$
    – supercat
    May 6, 2020 at 14:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.