A circuit board from the Saturn V rocket, reverse-engineered and explained
righto.comThis reminds me of this short piece about debugging a live Saturn V (http://www.zamiang.com/post/debugging-a-live-saturn-v), and also a detailed video about the Saturn V's Launch Vehicle Digital Computer (https://www.youtube.com/watch?v=6mMK6iSZsAs) — it had 112 KB of dual-redundant hand-woven magnetic core memory.
(Links borrowed from my weekly newsletter about the space industry called Orbital Index https://orbitalindex.com — check it out if you like this kind of nerdery.)
> it had 112 KB of dual-redundant hand-woven magnetic core memory.
Purely out of curiosity, do we know the amount of memory a modern orbital rocket like the Falcon 9 has?
You might find this question interesting:
https://space.stackexchange.com/questions/9243/what-computer...
> The Falcon 9 has 3 dual core x86 processors running an instance of Linux on each core. The flight software is written in C/C++ and runs in the x86 environment.
It's unusual that x86 and Linux, which are generally not considered to be reliably and robust under an extreme environment, are used here. But since it's developed by SpaceX, it makes sense - move fast and break things, use off-the-shelf commercial systems as the basis to reduce costs.
Anyway, I think it should be more interesting to compare it with a modern rocket that uses a more specialized computer system, VxWorks comes to mind.
It's a risk assessment tbh; if they can put in more redundancy instead of fault-tolerant hardware at a fraction of the cost then it'll be cheaper for them.
I mean compare it with mainframes vs cloud computing; with the latter, you use off-the-shelf hardware and build your software in such a way that you will randomly lose machines, BUT because cloud computing you'll automatically spin up a new machine in that case.
640K
> This reminds me of this short piece about debugging a live Saturn V
Fantastic link, thank you for sharing.
This guy's blog is ridiculous(ly good). Browse the rest of it, you'll be glad you did. I'm glad I did.
http://www.righto.com/2014/10/how-z80s-registers-are-impleme...
http://www.righto.com/2020/03/inside-titan-missile-guidance-...
http://www.righto.com/2016/10/simulating-xerox-alto-with-con...
Thanks! I'm glad you're enjoying the blog. I'm here if you have any questions.
Do you plan on revisiting your analysis of power bricks with regard to current models. I know it's not as exciting as Apollo era computers but I found it intriguing.
I probably won't repeat my charger analysis since I'm unlikely to find anything new and interesting. There are other sites that are doing detailed reviews of chargers now.
My new favorite YouTuber, CuriousMarc, has a whole series of videos about restoring the Apollo Guidance Computer (AGC).
His AGC Playlist: https://www.youtube.com/watch?v=2KSahAoOLdU&list=PL-_93BVApb...
His channel is full of fascinating retro-computing and EE videos.
Just to keep everything straight, the AGC that we restored is a totally different computer from the LVDC/LVDA that this board is from. The AGCs were on the Command Module and the Lunar Module that went to the Moon's surface, while the LVDC was onboard the Saturn V rocket.
The AGC was one of the very first computers to use integrated circuits, while the LVDC used hybrid modules. The LVDC used triple-redundant circuits with voting while the AGC was not redundant. The LVDC was a 26-bit serial computer, while the AGC was a 15-bit computer. The LVDC was built by IBM, while the AGC was built by MIT and Raytheon.
It's interesting that the two computers were different in so many ways.
the audio is often so bad tho, I can't understand what they are saying half the time.