Extracting code from the first NASA computer flown in space

[TerraHertz]

http://www.orrazz.com/2016/09/a-hacker-from-south-africa-just-rescued.html
A Hacker From South Africa Just Rescued The First NASA Computer In Space


Apollo Rope Memory Modules (Part 1 - Introduction)
Francois Rautenbach   Aug 22, 2016  ( youtube  watch?v=WquhaobDqLU )
Extracting the software from the Rope Memory modules of the long lost Apollo Guidance Computer used in Flight AS-202.

[Tomorokoshi]

Awesome project and story! Complete with clicking relays and flashing lights!

[Alex Eisenhut]

I didn't see this in the video.

https://en.wikipedia.org/wiki/Gemini_Guidance_Computer

This was the first computer flown in space by NASA.

[FrancoisR]

I have to admit that I really don't know that much about the Gemini Guidance Computer, except that it was designed and built by IBM. Wikipedia indicates that it was designed using the same RTL technology as that used in the Apollo Guidance Computer (AGC), but I have not seen any other definitive proof of this. It would be really interesting to get more info about this.

What I do know is that IBM was very much against using integrated circuits when MIT first proposed using this for the AGC. There are many documents where IBM attacked this decision, claiming that it was too unreliable. IBM actually proposed a different design using transistors and there is another document where MIT responded, showing that the IBM design was far less powerful than what was needed to land on the moon. IBM created enough doubt with NASA that they got the contract to develop the Apollo Abort Guidance System (AGS), which was a much simpler (and slower) transistor based computer of a similar size and weight. By that time, MIT already had two years experience designing with integrated circuits and the AGC3 was already functional. IBM was so insistent during the 1960's, on only using transistors in their designs that their Mainframe computers only started using TTL components in 1972.

The Gemini Guidance Computer was definitely used in flight before the AGC, but production AGC computers were available and used in training at least 18 months before the Gemini computer. The guidance computer used in the Minuteman II missiles was actually introduced at roughly the same time as the Gemini guidance computer, but also trailed the AGC development by 12-18 months.

The only reason why the AGC flew after the Gemini and Minuteman guidance computers were because of delays on other parts of the Apollo project. The AGC was definitely ahead of the pack by at least a year even though it was an order of a magnitude more complex and much faster. It was also much closer to a general microcomputer than the other special-purpose designs. It was used in 1972 for the first fully electronic fly-by-wire test onboard a modified F8 Crusader. It has also been used in US Navy's Deep Submergence Rescue Vehicle for navigation tests. It influenced the design of many other notable computers e.g. the PDP-11 and later general purpose microcomputers. You wouldn't be too far off, comparing it with a modern embedded microcontroller like the PIC. In my opinion, it is much closer to a general purpose microcomputer than any other computer built during the 1960's.

The title of the post is a bit ambiguous, so to clear it up. There is no doubt that the AGC was the first computer built using integrated circuits. It was also a general purpose computer that could do much more than just navigation. It was not the first computer in space, but the first computer built with microchips (that also happened) to fly in space.

[FrancoisR]

Alex's post about the Gemini Guidance Computer had me stumped a bit. Wikipedia indicated that RTL chips were used in the Gemini GC, which didn't made sense at all. My gut instinct told me that it was unlikely that IBM would have designed anything using integrated circuits in the early 1960's.

NASA's web site "Computers in Spaceflight: The NASA Experience" has the following to say about the Gemini Guidance Computer:
"The machine consisted of discrete components, not integrated circuits."
http://history.nasa.gov/computers/Ch1-2.html

Anybody care to update the Wikipedia page?

[TheBay]

Some things I don't get about this...

Firstly how did he get his hands on all this, why is he doing this. Sorry but scrap metal and FBI just leaving him alone yeah right...

And with a really cheap psu and hobby grade equipment. Surely this should be done in a proper lab somewhere with decent equipment etc. Not to mention how this all seems to be stored, hardly any precautions to stop it getting damaged physically or electrically. And wouldn't this be classified?

[SeanB]

NASA is not military, it is all meant to be freely available to everybody, seeing as it was paid for with taxpayers money in the USA. Thus you can see the archives they have, and even get the images and all the data off all the missions for free off the servers, at full resolution. Only things not available are stuff where they did not archive it, either the tapes were reused or the archives were lost with time.

[TheBay]

NASA is not military, it is all meant to be freely available to everybody, seeing as it was paid for with taxpayers money in the USA. Thus you can see the archives they have, and even get the images and all the data off all the missions for free off the servers, at full resolution. Only things not available are stuff where they did not archive it, either the tapes were reused or the archives were lost with time.

Now that makes it all plausible, only reason I thought it was all classified was due to having visited NASA a couple of years back and sort of given that impression while I was there.
However maybe that is done to make you feel more excited about the whole thing. Not to mention what we see in films

Guess it's more a case of them being irresponsible with this data/equipment, that should have all been archived and preserved surely, I wonder if there are any issues with intellectual property on that data or would that have lapsed now.

Hats off to the guy then 

[Alex Eisenhut]

Well, I think the first digital computer to fly in space was the Autonetics D-17 Minuteman guidance computer. Discuss? 
Perhaps the first analog computer in space was the V-2's. 

[FrancoisR]

The Minuteman II rocket first flew sometime in 1965 while the Apollo block I computer only flew on 25 August 1966. However, development on the AGC started in 1961 and production versions of the AGC were available in 1964, almost a year before the Minuteman II computer. Work on the Minuteman II guidance computer started late in 1962, about 12-18 months after the AGC.

The AGC was the first computer that used integrated circuits. The only reason it flew after the Minuteman was because of delays and constraints on other parts of the Apollo project. Technically it was functional and was used in training almost a year before the much simpler Minuteman guidance computer. The August 1966 date was the first time the computer was launched into space and not the first time the computer was used.

[JoeN]

And I thought all of this stuff went into the big warehouse shown at the end of Raiders of the Lost Ark.  I guess our government isn't quite as competent as Spielberg imagines.   

[rrinker]

 That's only for the stuff they don't want us to know about. All those things you never heard of? Ever wonder why? Because THEY don't want you to know.

 Even these so-called "Freedom of Information Requests". If you know what to request, then it's something they really in the end don't mind letting you know about, because you know it already exists, or you wouldn't be requesting information. The stuff they really don't want you to know, you don't even know to ask about.

[TerraHertz]

And I thought all of this stuff went into the big warehouse shown at the end of Raiders of the Lost Ark.  I guess our government isn't quite as competent as Spielberg imagines.   

Wasn't that the (fictional) British government? The US government (real or fictional) is entirely different.

"TOP men working on it."

[apelly]

"TOP men working on it."

Who?

[EEVblog]

"TOP men working on it."

Who?

TOP, men.

[apelly]

 

[Towger]

I don't suppose NASA still has not found the lost film footage they shot on the moon.

[Alex Eisenhut]

The Minuteman II rocket first flew sometime in 1965 while the Apollo block I computer only flew on 25 August 1966. However, development on the AGC started in 1961 and production versions of the AGC were available in 1964, almost a year before the Minuteman II computer. Work on the Minuteman II guidance computer started late in 1962, about 12-18 months after the AGC.

Fine, but the Minuteman I used the D-17B and flew in 1962.

https://en.wikipedia.org/wiki/D-17B

https://en.wikipedia.org/wiki/LGM-30_Minuteman#Minuteman-I_.28LGM-30A.2FB_or_SM-80.2FHSM-80A.29

Why bring up the Minuteman II?

"The AGC was the first computer that used integrated circuits."

http://www.ti.com/corp/docs/company/history/timeline/defense/1960/docs/61-first_ic.htm

I don't suppose NASA still has not found the lost film footage they shot on the moon.

Double negatives are hard to parse.

[rdl]

Sigh. What to believe? Company propaganda or an encyclopedia that anyone can edit...

If TI integrated circuits were actually being used in the Minuteman-I, which was already in service by 1962, they should try to get all those Wikipedia errors fixed.

[FrancoisR]

The TI article is accurate but it omits a few things that are mentioned in other documents.

TI tried very hard to sell their integrated circuits to a number of big users, including MIT. It was only in 1962 that TI convinced the USAF to try integrated circuits. By that time the group at MIT who built the AGC had already been working with Fairchild Micrologic RTL devices for about 12-18 months and the AGC4 prototype was quite far down the path.

Btw, Fairchild introduced the first Micrologic RTL devices in the spring of 1961 and MIT immediately ordered 1000 devices. The price was much more reasonable than the TI quote of $60,000 for 60 devices! TI also failed to deliver on time and a few orders with TI were cancelled because of this. From 1962 onward, TI did much better and became the second source to Fairchild.

The Minuteman I guidance computer did *not* use integrated circuits. The first big order (100,000 devices) that TI delivered to Autonetics (who built the Minuteman GC) was in late 1964 and by 1965 TI were delivering 15,000 devices per week. By that time the AGC had already been in use for about a year.

[rdl]

The TI article seems to reference both/all Minuteman programs together as one. It seems it would be more correct if they had specified Minuteman-II.

The demo and tour of the tiny computer helped tip the scales in TI's favor and made it possible to start applying integrated circuits to defense equipment in 1961 and to Minuteman and other programs in 1962.

However, the Minuteman-I, using the D-17B computer, was already in service by 1962.

[oldway]

They say that the memory of AGC could not work. 



I think it is bullshit. 

What's your opinion about this video ?

[arlipscomb]

Videos of parts of schematics that jump around to music don't tell me anything.

[rrinker]

 So if the rope memory modules could not work, how is he proposing the missions actually flew? The computer obviously worked, and was used. Did they use alien memory technology and hid it by making up these magnetic core ropes which were familiar technology at the time? Something in his premise is wrong, I can feel it, something fundamental, but I'll have to watch again at home when I can actually pay attention to the whole thing.

[Len]

So if the rope memory modules could not work, how is he proposing the missions actually flew?

The moon missions were faked, doncha know.