The Apollo computer

[kinetic]

I have always read that the Apollo guidance computer
was no more powerful than a typical calculator.

Here is an interesting few links about the AGC in the Command and
Lunar modules. Core rope memories and early software languages
some of the more interesting details to me :)

http://en.wikipedia.org/wiki/Apollo_Guidance_Computer
http://www.netjeff.com/humor/item.cgi?file=ApolloComputer

It mentions that a full minute of guidance inputs were ignored on descent
to the moon because of the '1201' and '1202' alarm story :)

Steve

[RobF]

Wow. Thanks Steve - fascinating reading. I wasn't aware of any of that.

Also interesting that few of the pages describing the 1201/1202 alarms compare the speed of the Apollo computer 0.0475 MHz with first IBM PC (4mHz) and "today's" (1994!) PCs at the blinding speed of 66 MHz....:lol:

[kinetic]

Hehhehe ...yes Rob , 66 Mhz, that was funny:thumbsup:
I have seen those core rope memory cards years ago, probably
25-30 years ago when I pulled apart an old punch card reader
from my brother's work as a teenager.
I remember being fascinated that a little ceramic bead would hold
a bit of memory :)

Steve

[CraigS]

Off topic a little, but I notice Buzz Aldrin is coming to Australia shortly (for presentations etc).
He tells tales of the notorious incident Steve mentions above.

Must've been a bit hairy at the time, given the historical importance of what unfolded in the following next few hours.

Cheers

[multiweb]

Must have been a b||ch to change those two small lithium batteries during descent with the full gear mittens on hey? ;)

[RobF]

Interesting too that the smoothing polynomials for the course corrections could theoretically have plotted a course through the ground and back up again under certain circumstances (which fortunately never happened). Also that the initial landings had severe engine surging problems due to 0.2 sec delay programmed even though engine had been upgraded to <0.1 sec delay (without being documented/communictaed). If they had programmed the documented 0.3sec delay the surging would have been catastrophic.....

[RobF]

Yes. The first big chemistry analyser I ever worked on (Technicon SMAC) had a 60s computer - hand wound ferrite cores for the memory! Now that's a computer! :)

[CraigS]

Even scarier story was the Apollo 13 orbit corrections done on the way back from lunar orbit. They had to do it with no guidance computer !

Man, those guys were lucky !!
(And tremendously skilled in saving their bacon !).

Cheers

[Omaroo]

It's a very common misconception that the Apollo guidance system wasn't as powerful and was simpler than a modern pocket calculator. Nothing in this world is further from the truth. It was a horrendously complex feat of electro-mechanical/digital bespoke engineering designed to guide a spacecraft through three-dimensional space based principally on simple inertia.

One of the best documentaries around is a series called, simply, "Moon Machines". They're all on YouTube and are so well worth the watch. The set in particular you should see is titled "The Navigation Computer" (Zuke969):

http://www.youtube.com/watch?v=pjVxhIQ14Qc

Watch all parts, you'll be left with an impression somewhat different to that which we're asked to swallow.

My old man was part of the endeavour too. He was a top IBM engineer back in the 50's, 60's and 70's. Here's myself and my younger brother Steven sitting on the console of an IBM System/360 Model 40 during that time (April 1967) at an overseas posting. I can't remember what part of the project he officially played, but it was an exciting time nonetheless. :thumbsup:

[stephenb]

Thanks for posting Steve :thumbsup:

Great pics Chris :thumbsup:

As Chris said, Buzz will definately be in Australia for two events only. A private event for Omega on October 5, then as a guest at the Australian Geographic Award Dinner the following night.

[Barrykgerdes]

I think computer programmers need to learn from the efforts of the builders and programmers of those early computers. With the limitations in memory, processors etc they learnt to write an effective program in Bytes (sometimes nibbles).

These days programmers need Gigabytes to write programs that are often no faster or or more effective than the old primatives.

My first computer experience came with a weapons computer that controlled everything effectively on a warship with 4K of magnetic core RAM and dip switch programming (or a mylar tape reader)

Barry

[CraigS]

Very interesting doco there, Chris.
I've only watched two episodes but certainly a trip down memory lane !

Somewhere in my travels, I've seen one of the Apollo Astronauts talking about the central operations centre (Houston) telemetry displays/systems.

In it, he said that the monitors that the control personnel were viewing were actually TV screens fed by closed circuit cameras which were pointed at a massive display board. Regularly transmitted telemetry data from the spacecraft was actually taken from printers and copied manually, in handwriting, onto the display board (like a whiteboard) and then transmitted by the TV cameras to the control staff displays/monitors !

Not as high tech as the spacecraft navigation/computer system … (which in itself, was not critical .. as evidenced by Apollos 12 & 13 .. and the onboard 'sextant').

It all worked though, (somehow) !

Cheers

[DavidU]

Great thread ! thanks:thumbsup:

[kinetic]

Thanks Chris for those videos....wow, what a great insight.
The inertial Nav system spun me out, that they had so much faith
in it to let it fly the plane by itself.
Touched on so many areas to give you a real understanding of how
they all came together.
How funny is the quote from one of the MIT engineers when he told
his wife what he was involved in:
'I'm writing the software that will send men to the moon'
She says: 'please don't tell our friends' (software is such a new
concept it sounds insignificant and boring) :)

classic

thanks again Chris, thoroughly enjoyed those links.

Steve

[Omaroo]

A pleasure fellas. Watch the rest of the series too - it all goes to show you just how silly the conspiracy theorists are. There were hundreds of thousands of very smart people involved in the Apollo (and preceding) programmes.

[Hans Tucker]

Why not build your own.

http://www.galaxiki.org/web/main/_blog/all/build-your-own-nasa-apollo-landing-computer-no-kidding.shtml

Seriously..I was actually contemplating this when I first came across this link.

Currently I am reading and highly recommend Moon Lander - How We Developed The Apollo Lunar Module by Tom Kelly (Project Manager)..after that I will read Digital Apollo by David Mindell which deals with the topic of this thread.

[supernova1965]

Not only can you build your own Apollo Comp you can also build your own virtual Solar system at the bottem of this web page link:eyepop:

[RobF]

Yes - thanks Chris - really enjoyed that link.
Fascinating your father was involved too, wow!

[gary]

Some may find the following footnote interesting.

Within minutes of landing, one of the first tasks the astronauts performed
was a star sighting in order to re-establish the attitude of the LM's on-board
IMU (Inertial Measurement Unit).

Aldrin performed a two-star alignment using Rigel and Gamma Cassiopeiae,
the later of which the astronauts referred to by the name "Navi".

The origin of the name "Navi" is interesting and is discussed in the Apollo 15
Lunar Surface Journa (http://www.hq.nasa.gov/alsj/a15/a15.postland.html)l.



In honor of Grissom, White and Chaffee, our own Argo Navis uses the name
Navi in its list of alignment stars and northern hemisphere observers routinely
select it to align their scopes just as Aldrin used Navi to align the IMU
whislt on the Moon in July of 1969.

[gary]

Hi Chris,

Great picture!

The School of Elec Eng & Computer Science at UNSW had a model 50 which
had been replaced by the PDP-11's running UNIX when I first arrived as an
undergraduate in the late 70's.

When the head of Computer Science, Prof. Murray Allen, retired in the 80's
the School presented him with the front panel of the 360. I still have a large
core memory board from it.

The major milestone of the all-encompassing architecture of the 360 was
featured in Allen's lectures on the history of computing.

I remember seeing the NSW TAB's IBM in operation in Harris Street in the late 60's
or early 70's. Came with its own resident IBM engineer.