Author Topic: Ken Shirriff - Arma Micro Computer which was also used in Atlas  (Read 1616 times)

Offline Remes

  • Full Member
  • ****
  • Posts: 446
  • Germany
  • Liked: 263
  • Likes Given: 146
Rewind to the first digital computers used in Atlas. The early Atlas were guided by ground based RF stations. Later on the Arma computer were used.

http://www.righto.com/2024/02/the-first-microcomputer-transfluxor.html

Given, that it had a 1/2s Cycle time I would use it was used for the higher level guidance equations and e.g. attitude control was done with analog circuitry.

Last year I was in the Computer History Museum in San Francisco and it really blew my mind how they took whatever was there and made the computers they made and achieved the goals they were set. The discussed Arma computer is from 1962 and remember: Roughly in 1960/61 more or less the basic concept for the AGNC was decided.

Also an interesting developer is mentioned:

https://en.wikipedia.org/wiki/Wen_Tsing_Chow
From Wikipedia:
Quote
He would continue working throughout the 1960s and early 1970s to develop and advance missile and spacecraft digital computers and guidance systems technology beyond the state of the art - working at the Aerospace Corporation in the Gemini and Minuteman programs and at IBM in the B-1, B-52, Saturn V and Skylab programs, and of course, in the development of the AP-101 digital computer used in the Space Shuttle Computer Complex.

Chow, uniquely, worked on the guidance computers and guidance systems for every major United States Air Force ICBM and NASA crewed space program from the very beginning with the Atlas, through Titan, Gemini, Saturn, and Skylab, to missiles and spacecraft still in service today, Minuteman and the Space Shuttle.
Seems to me like the John Young of Computer Science.

In the aftermath I found
https://ntrs.nasa.gov/api/citations/19840008166/downloads/19840008166.pdf
Centaur D-1 A Guidance/Software System
But didn't have time to read it so far.

Offline leovinus

  • Full Member
  • ****
  • Posts: 1154
  • Porto, Portugal
  • Liked: 909
  • Likes Given: 1794
Thanks for the pointers and interesting reading. As I am working on space-related IBM FORTRAN and assembler listing from the 60s (and might start a thread on that at some point), I was wondering whether there is still assembler software around for Arma? Printed listings maybe? Card decks? And how where these machines debugged and tested anyway?

As you mentioned one NTRS report, maybe you'll like these as well. They probably need to be FOIAd unless there are copies anywhere.

19650060544 ,The design and performance of the arma strapped-down inertial navigation system, 1964,
19650064304 ,Atlas missile b-2 pod - aig section investigation of acoustic and mechanical vibration levels in the arma computer, 1964,
19650080232 ,Arma inertial guidance system configuration and data evaluation for re-entry vehicle systems program, 1962,
« Last Edit: 03/12/2024 09:12 pm by leovinus »

Offline LittleBird

  • Full Member
  • ****
  • Posts: 1185
  • UK
  • Liked: 350
  • Likes Given: 589
Last year I was in the Computer History Museum in San Francisco and it really blew my mind how they took whatever was there and made the computers they made and achieved the goals they were set. 

Fascinating post, thanks.  Were there any specifically space-related exhibits in that museum ? It's a place I'd like to get to one of these days, but Mountain View is a fair distance from SF and I have yet to do so. I have fond memories of the old Boston Computer Museum in the 90s:

... though if one is in the Valley there is also the Hiller aviation museum https://www.hiller.org/ which I have been to and which was well worth it.
« Last Edit: 03/13/2024 01:35 pm by LittleBird »

Offline Blackstar

  • Veteran
  • Senior Member
  • *****
  • Posts: 15477
  • Liked: 8066
  • Likes Given: 2
Last year I was in the Computer History Museum in San Francisco and it really blew my mind how they took whatever was there and made the computers they made and achieved the goals they were set. 

Fascinating post, thanks.  Were there any specifically space-related exhibits in that museum ? It's a place I'd like to get to one of these days, but Mountain View is a fair distance from SF and I have yet to do so. I have fond memories of the old Boston Computer Museum in the 90s:

... though if one is in the Valley there is also the Hiller aviation museum https://www.hiller.org/ which I have been to and which was well worth it.


The computer museum is well worth a visit. I was there around 2018 and was surprised at how interesting it was. I don't remember any specific space-related exhibits, although they probably have a few artifacts. The opening section of the museum includes a really nice walk through of the early era of computing, starting with the mechanical systems, then up to scam and I think they even have some equipment from SAGE.

One really interesting aspect of the museum is that they have at least one (maybe more) big demonstration rooms with 1950s/1960s era computers and a technician explains how they used punch cards and how the processing worked. I thought it was a great display that countered some of the technological hubris that we have today when we look back and think "that stuff was so primitive." The technician explained how that machine dramatically improved manual tasks like processing payroll. I think even a young person who is immersed in their phone would understand how and why computers were developed decades ago.

Offline LittleBird

  • Full Member
  • ****
  • Posts: 1185
  • UK
  • Liked: 350
  • Likes Given: 589
Last year I was in the Computer History Museum in San Francisco and it really blew my mind how they took whatever was there and made the computers they made and achieved the goals they were set. 

Fascinating post, thanks.  Were there any specifically space-related exhibits in that museum ? It's a place I'd like to get to one of these days, but Mountain View is a fair distance from SF and I have yet to do so. I have fond memories of the old Boston Computer Museum in the 90s:


The computer museum is well worth a visit. I was there around 2018 and was surprised at how interesting it was. I don't remember any specific space-related exhibits, although they probably have a few artifacts. The opening section of the museum includes a really nice walk through of the early era of computing, starting with the mechanical systems, then up to scam and I think they even have some equipment from SAGE.

One really interesting aspect of the museum is that they have at least one (maybe more) big demonstration rooms with 1950s/1960s era computers and a technician explains how they used punch cards and how the processing worked. I thought it was a great display that countered some of the technological hubris that we have today when we look back and think "that stuff was so primitive." The technician explained how that machine dramatically improved manual tasks like processing payroll. I think even a young person who is immersed in their phone would understand how and why computers were developed decades ago.

Talking of walk throughs ... and indeed young people ...


Offline Remes

  • Full Member
  • ****
  • Posts: 446
  • Germany
  • Liked: 263
  • Likes Given: 146
Were there any specifically space-related exhibits in that museum ?

Exhibits change of course. When I was there, they had an opened AGNC hanging on the wall together with a dsky. Furthermore an Autonetics D-17B Computer (without the inner part, was it a drum memory or a gyro, not sure). Some control Panel/computer for Mir/Salyut? (Not sure, I was so obsessed with the AGNC that forgot later to look at that piece).

Maybe it's not so much the space artifacts. But the museum really shows the computer from the beginning. Including the Hollerith machines, mechanical/electric machines able to sort, add, provide basic operations (that e.g. have been used in the Manhatten project to do scientific calculations). It shows what time there was when they build a non critical Computer for Gemini and a mission/crew critical computer for Apollo. With integrated circuits, with an operating system, with an interactive panel, ... That is everything else than a given.

While I was taking many pictures, someone asked me if I have any connections to that computer. And I can't help myself, but the guy (he was their with his daughter) looked like Andy Weir. I'm very much face blind, but unless proven otherwise, for me it was him.

Offline Blackstar

  • Veteran
  • Senior Member
  • *****
  • Posts: 15477
  • Liked: 8066
  • Likes Given: 2
While I was taking many pictures, someone asked me if I have any connections to that computer. And I can't help myself, but the guy (he was their with his daughter) looked like Andy Weir. I'm very much face blind, but unless proven otherwise, for me it was him.

I think Weir lives in that area.

My memory is a bit faded, but I remember that the museum displays started to lose focus around the 1990s. At that point it seemed to be kinda a collection of consumer electronics devices and there was not much of a narrative then. Of course, that was a few years ago and maybe they've updated that section of the museum by now.

Offline Remes

  • Full Member
  • ****
  • Posts: 446
  • Germany
  • Liked: 263
  • Likes Given: 146
I was wondering whether there is still assembler software around for Arma? Printed listings maybe? Card decks?

Even for apollo the software was not really lying around in a museum or another institution. After the moon program they even sold the original tapes which were used to record the tv signal in the receiving station in Australia.

In some of these videos:
https://www.youtube.com/playlist?list=PL-_93BVApb59FWrLZfdlisi_x7-Ut_-w7
you will see, how they take down the AGNC in the Computer History Museum to read out the core memory. (Core memory keeps it content for a very long time). Also they received a printed copy of the source code from Don Eyles. He wrote the landing routine for the LM. He wrote also a book on the topic of software for the AGNC: "Sunburst and Luminary: An Apollo Memoir" Meeting other people from the Apollo program (Edwin Hall e.g.) they would ask for core rope memory modules to read them out and to get the various versions.

Quote
And how where these machines debugged and tested anyway?
I have no specifics for Arma. But in that time often the software was run on "bigger" computers which emulated the smaller one. With the possibility to make printouts. That's why the printers were so fast even back then. Just to release the data and go on with the next batch job.

https://in.pinterest.com/pin/501658845988070822/
For Apollo, they had a core rope simulator connected to the AGNC allowing for fast software changes. They had a LM simulator with a complete Hardware in the Loop setup. Cameras automatically guided over a model of the moon surface would provide image signals which were shown on screens outside the LM windows.

Even without hardware breakpoints you can always mask all interrupts and enter a while loop, which dumps your core memory to a tape or printer.

I worked on hard real time control systems. We also used to toggle pins for interrupts and other events. Debugging with output would mess up the DSP (ADMC401, programmed in assembler).

For the AGNC, as it was also handling telemetry, small amounts of data could be exchanged here. In "Curous Marc" videos they also showed a big debugging connector which provides access to the memory bus and other interriors.

Offline Remes

  • Full Member
  • ****
  • Posts: 446
  • Germany
  • Liked: 263
  • Likes Given: 146


That is the video with the Computer History Museum AGNC. Mike talks about halting the AGNC and then going in via the debug connector.

The CHM has also IBM 360 or parts of them, which were used in the RTCC - Real Timer Comptuer Complex. The ground based computer which did a lot of mission work.

One really interesting aspect of the museum is that they have at least one (maybe more) big demonstration rooms with 1950s/1960s era computers and a technician explains how they used punch cards and how the processing worked. I thought it was a great display that countered some of the technological hubris that we have today when we look back and think "that stuff was so primitive."

Yes, this "we flew with slide rules to the moon" doesn't really apply. They had computers. Not in abundance. But in the most important places they had some. The F1 engine was tested under the supervision of computers.

And when there was no CFD software, they made tests. When they had no FEM, they made tests. Fuel slosh? Test and measure. I think they solved the most basic problems of big manned rockets in that time. Yes, materials today are better, you can do more simulation and less testing (not no testing), computers are faster smaller and more reliable. But then you have emails, Microsoft Teams/Slack, mobile phones, ...

Offline LittleBird

  • Full Member
  • ****
  • Posts: 1185
  • UK
  • Liked: 350
  • Likes Given: 589
Were there any specifically space-related exhibits in that museum ?

Exhibits change of course. When I was there, they had an opened AGNC hanging on the wall together with a dsky. Furthermore an Autonetics D-17B Computer (without the inner part, was it a drum memory or a gyro, not sure). Some control Panel/computer for Mir/Salyut? (Not sure, I was so obsessed with the AGNC that forgot later to look at that piece).

Maybe it's not so much the space artifacts. But the museum really shows the computer from the beginning. Including the Hollerith machines, mechanical/electric machines able to sort, add, provide basic operations (that e.g. have been used in the Manhatten project to do scientific calculations). It shows what time there was when they build a non critical Computer for Gemini and a mission/crew critical computer for Apollo. With integrated circuits, with an operating system, with an interactive panel, ... That is everything else than a given.

While I was taking many pictures, someone asked me if I have any connections to that computer. And I can't help myself, but the guy (he was their with his daughter) looked like Andy Weir. I'm very much face blind, but unless proven otherwise, for me it was him.


Indeed, sorry if my tone sounded like I am not interested in computers or if I thought they were OT. In fact I am interested in them and have visited several computer museums in my time, Id be more than happy to get to more. I was also privileged to be a teenager in early years of pocket scientific calculators and (pre IBM) PCs. I remember how both Rockwell and HP for example would use their connections to space in advertising calculator s some examples are scattered through these threads.

[Edit: and re "Yes, this "we flew with slide rules to the moon" doesn't really apply. They had computers. Not in abundance. But in the most important places they had some. The F1 engine was tested under the supervision of computers" you are of course so right. In fact it was more a case of "we flew to the moon and that was one of the reasons we then got the electronic slide rule" ;-) .  I've added a couple of examples of how the space connection was used to sell calculators back in the day, though the two I'd really like to find are elusive online. One is an HP ad showing the Shuttle landing with the slogan "when performance must be measured by results", the other  would be one of the Rockwell print ads for their scientific calculators, like the 64RD "Scientific Slide Rule" shown below. For a while at least these hopefully played up the space angle, though alas they didn't live up to their billing-I so wanted to like them ;-) ]
« Last Edit: 03/18/2024 03:06 am by LittleBird »

Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
0