The show had a guy who said that the total computing ability of the spacecraft computers was ‘somewhere between a digital watch and a cell phone. Probably closer to the digital watch.’ So, nothing of substance to add.
ISTR that the AGC implemented a lot of its programming in hardware, not in software.
There’s some interesting information about the Apollo computing systems at http://www.clavius.org/techcomp.html. It says that a lot of the number crunching was done Earth-side.
A trivial point – slide rules in the Apollo 13 were an anachronism. NASA had switched over to hand-held calculators by then.
What’s the name of that show? I remember that exact quote, but I can’t remember which show it was from.
(It made me think, the Apollo computers used to be compared to pocket calculators. I suppose calculators have far surpassed Apollo technology by now!)
The capsule’s computer used a wire recorder-type system, and it was, IIRC, more like a punch card sytem in operation than what most of us would consider to be an actual “program.” (i.e. you turned it on, and the computer ran through the motions without getting feedback from other systems.) On the Apollo 11 mission, they had problems with the computer in the LEM on the way down and had to reboot it. No idea if Buzz did it by giving it the ol’ 3 fingered salute, or if he just punched it.
Now that we have the hardware out of the way, could someone cook up the necessary ballistic calculations on Excel?
They had a bunch of IBM 360s at JSC for the heavy-duty number crunching. The USAF had a CDC 6600 at the Cape for handling real-time trajectory data from many sources, C-band radars, optical trackers, launch vehicle telemetry and antenna pointing data.
Given the size, speed and reliability of computers at that time, the dominant design philosophy was to keep as much complexity as possible on the ground, instead of on the spacecraft. Ground based systems are relatively cheap, and you can add redundant systems for enhanced availability.
When they compare a digital watch to a spacecraft computer, they are talking about transistor-count (complexity), not processing power or memory.
When I first went to work at China Lake they had just installed an IBM 701 computer. I suspect my Radio Shack programmable hand held had more capacity, except for memory. And in the 701 the memory was mostly in the form of paper punch-cards and magnetic tape.
This is a huge range. The processing in even 10 year old cell phones is quite high. When I started working in the cell phone biz the phone chips I worked on had a 17 MHz ARM 7 processor with 2 DSP processors running at about 10 MHz plus a fair amount of specialized signal processing.
You sure can. I did this exact thing in my sophomore year at University 
Weird favor to ask: if you still have it, could you e-mail me a copy of it (in my profile)? Many moons ago* I took an astrophysics class, (Orbital Mechanics) which fascinated me. But that was years ago, and I’m still fascinatedly curious.
Tripler
*Pun intended
I MIGHT still have it on my computer at home or one of my old school floppies. if i can find/extract it, I’ll send it to ya.
He discovered the docking radar was on as well as the ground radar, and using up so much RAM they were getting overflow errors. He turned the docking radar off and the problem went away.
Thanks!!
D’oh! :smack:
I just remembered, I wrote that ballistic calculator in FORTRAN, not Excel! My bad. If you still want it tho, I’ll still look for it.
(FORTRAN, man I feel really old now. They were still teaching it to engineering students in 1994, honest!)
Alas, I don’t remember.
Dover Publications sells inexpensive copies of textbooks used by NASA and the Air Force on the subject you might find interesting.
Wow, I have all but one of those books. My copy of Fundamentals of Astro is so well-worn it’s held together by duct tape.
God i’m such a nerd.
Another factor that lent itself to the “do the calculations on the ground” design of the Apollo missions was the design of memory.
Intel had just barely come into existence and began marketing a 2K IC memory chip in 1966. NASA would have been freezing the technology for the mission right around this time if their current practice is any indication, and I doubt they would have risked such a high profile and dangerous mission on a startup and their new technology.
The IBM 360s they were using had magnetic core memory, which, while some very good miniaturization was achieved over time, mag-core hardly followed Moore’s Law in terms of physical size.
NASA stuck with magnetic core memory right up through the Space Shuttle.
NASA had, maybe, but that doesn’t necessarily mean that all of the engineers in NASA had. Slide rules remained popular for a while.
I dont know, but I have a sperry 1616 univac ballistic computer off a sub in the barn … we could plug it in and give it a smoke test=)