How the Integrated Circuit Took Us to the Moon
Вставка
- Опубліковано 13 лип 2024
- The NASA space project advanced the technological progress of the human race by leaps and bounds. Examples of these revolutionary developments abound. But one extra-special item was in its computer.
The Apollo Guidance Computer or AGC was one of the first to use silicon-based integrated circuits.
In this video, we are going to look at how the silicon integrated circuit supercharged the AGC and guided us - literally - to the moon.
Links:
- The Asianometry Newsletter: asianometry.com
- Patreon: / asianometry
- The Podcast: anchor.fm/asianometry
- Twitter: / asianometry
Bro don't be apologizing about making videos about America or any country for that matter. Like you don't know what kind of people are on the internet. You make great content that's educational.
Great content indeed, and all the little details of the connection between technological and political developments make your vid's so much more interesting!
Keep the good stuff coming, ignore those who are more interested in yelling then in learning something.
This kid of factual reporting adds a lot of important context to technological development around the world. It is OK to stray a bit when adding context for everything else.
Absolutely agreed! Plus America isn't all that bad 🤷♂️
I'm predicting a new channel, Americanometry
@@jpierce2l33t I'm American, I'm not ashamed of that.
My dad was an engineer at Fairchild in the 1960s. When Bob Noyce and others quit it rocked the industry. A year later, he was approached about coming over to what would become Intel. He didn't and regretted that decision for the rest of his life
i can see that your dad's decision, cost him many millions, if not hundred millions....
Too bad about your Dad's decision. Perhaps he could have solved the page register problem in the 8088 series before it evolved into the Pentium and forever crippled the computer industry.
@@darylh8657 I’m intrigued. What is this issue?
For all things Apollo IT related, like the inner workings of the AGC and how the communication system worked, the CuriousMarc channel is incredible. They even have original Apollo hardware that they are restoring and reverse engineering. Some Soyuz stuff, too.
They had a whole series on repairing and actually executing code on an actual Apollo AGC. Highly recommended.
I agree the AGC restoration was fascinating now they are working on the complex microwave communications system.
There is also a video from another channel (which I don't recall but the title is "The Ultimate Apollo Guidance Computer Talk") with a live lecture on the DETAILED design of the AGC.
"They even have original Apollo hardware" and apparently super expensive military grade computers were available for free at the local junkyard...
This video is showing the picture of Saturn 5 computer and its size, its job in comparison to the job of what computer of a moon lander woud need to do was simply trivial.
For 50 years there was this talk that we are not going back to the Moon because there is no point, China anounced that they also are planing to land on the Moon and instantly the narrative changed to "we need to have another Moon mission and land there again(before China)".
Why exactly Space-X rockets strugled so much for so long to land an empty rocket(no internal mass change to wory about)?
AGC was doing that with not a single mistake of any kind and Space X rocket that was at least millions times faster(powerful) was not able to do it for so long and so many time even with the modern engine gimbals that are much more capable that the one avaible half century ago(not to mention that the landing space was perfectly flat and not full of some random huge rocks).
@@Bialy_1 ermm. What are you trying to say? That the AGC was awesome, or are you one of those moon hoax people?
These days "computer chip" "microprocessor" and "IC" are often used pretty much interchangeably in a lot of contexts. Back in the AGC days, the integrated circuits had so few elements that they really were more like just a generic handful of bare components. All the chips in AGC were just identical gates, like metaphorical lego bricks you could use to build anything. In a very real sense, the chips are not the AGC. The wiring diagram of the AGC is the computer. The chips themselves don't do math, have no instruction set, etc. The interconnections of the generic gates is really what defines all the capabilities of the computer. It's very counterintuitive for a modern computer user. If you use a modern HDL like Verilog to program an FPGA, each IC in the AGC would be below the simplest element you can directly control, despite the fact that it was the most complex element that could be manufactured at the time. AGC is such a neat little artifact of a very brief moment in time when IC's were truly bleeding edge, and it was still normal to build a computer completely from scratch for every new project. No legacy. No backward compatibility. No existing code reuse. About five minutes later, Moore's law started to really have noticeable impact, and the industry was off to the races in a whole new world.
"All the chips in AGC were just identical gates, like metaphorical lego bricks you could use to build anything." yea right, anything that was not in need to make any significant calculations in real time. The Saturn V "computer" was this huge setup and was doing hardly any job in comparison to what AGC would need to do.
Even sensors basic data processing would be to much for that computer in real time, not to mention calculating in real time trajectory corections and everything else that was claimed to be in official AGC specification.
From wikipedia: Big lie(große Lüge)->a lie so colossal that no one would believe that someone "could have the impudence to distort the truth so infamously."
Scholars say that constant repetition in many different media is necessary for the success of the big lie technique, as is a psychological motivation for the audience to believe the extreme assertions.
It is funy that even educated people with degree in IT can easly swallow all that amazing almost magical AGC capabilities that was aparently completly forgoten when Space X was trying to land an empty rocket on a specialy prepared landing spot(not a completely unknown and unprepared rocky surface).
So how many year more before NASA will be ready to go back there(as it was announced suddenly when China announced its Moon program)?
> These days "computer chip" "microprocessor" and "IC" are often used pretty much interchangeably in a lot of contexts
Really? That's weird.
I'd say that "computer chip" shouldn't be used. And ICs can be anything. Microprocessors are implemented using ICs, but not every IC is a processor. Microprocessors are digital circuits, typically having a clock. ICs can be analog or digital, but not every digital circuit is a processor. One can build a "microprocessor" from discrete transistors but it wouldn't be much "micro" then.
@@Bialy_1 Are you an Apollo landing denier?
When I was in military repair, the first IC’s I saw were in 1966. The really secret stuff was encased in black epoxy so it couldn’t be back engineered. The first transistors were these cans
sticking up off the board. To fix them we used to flick the side of the transistor can with our fingernail, started them right up.
Not true, i can watch videos on YT where they are showing real working Apollo computers from that time and according to the story this super expensive, super high tech military grade computers were completly unprotected and unacounted by ayone so some random people were able to find them even on some South Africa yunkyards.
@@Bialy_1 He's talking about the military, and you're talking about NASA. Two separate things.
I'm not going to yell at you for doing videos about America... In fact, I'll only thank you!
I’m a retired electrical engineer in the US. Just found your channel and you’re really good. Thanks for the work.
At NASA's spending peak in the 60s, they were procuring ~60% of all IC manufactured in the US. I'd say that pushed development of microchips quiet a bit.
Not as much as you think, the AGC didn't use IC tech, as designed using late 50/60's stuff. By the time it was in actual use, whole CPU's on a chip where being produced, 1000's if times smaller then the agc and more powerful too
Well, the end of the video suggests that they were buying 60% of very little. Asianometry chose to emphasize Fairchild's risk as the thing that really mattered, and I tend to agree.
The fact that the military and NASA were buying them face them the confidence that they could produce them, and gave them an idea about their reliability. So, in that way, it did contribute.
I don't thing that's enough to say they pushed the industry. The industry didn't really pick up as a result of their purchases; even after NASA was done with them, they still weren't getting commercial buyers.
I agree on all points. I think the big takeaway is to not downplay NASA's role. If NASA had not been buying so many IC's would Fairchild (and others) lasted long enough to take a risk that did ultimately kickstart the industry.
Things to watch:
Brian Troutwine's The Charming genius of the Apollo Guidance Computer (one of my favourites, great info about how the AGC emulated a better computer, _the_ better computer that was required to make it to the moon in fact)
Moon Machines, the Apollo guidance computer episode specifically but watch ask of them, they're all up on UA-cam
CuriousMarc's 24 part series on refurbishing and running a real vintage AGC.
The video "light years ahead"
Thanks for the trip down memory lane. Glad you mentioned not only the AGC but also the LVDC. An interesting piece of trivia while the LVDC was triple reductant there was no redundancy on the AGC other then tremendous effort put into insuring reliability. If memory serves (I'm winging it) neither the AGC or LVDC experienced post launch failures, I think the lightning strike took out some of the telemetry signal processing gear.
Having grownup at the end of the vacuum tube era (valves for our friends on the other side of the pond) I am in awe at the tremendous advancement in electronics over the last 50 years.
There was a software failure causes by a systemic design error on Apollo 11's AGC.
However, the AGC's 'OS', the scheduler and the everyone resident applications of my memory serves correct, was designed from scratch to keep the mission alive in the event of software failure.
As it happened, the executive dumped all precesses, reloaded the 'keep then alive' processes and restored their memory set (last known good location, pitch, roll, climb and velocities)
Go watch Brian Troutwine's presentation on the AGC and learn the story of the 1202 alarm.
@@MostlyPennyCat Not to get too much in the weeds but that error indicated CPU overload. The software was designed to deal with that problem by dumping lower priority tasks. So I'd argue it was not an AGC failure. Wasn't the root cause of the overload because the checklist had the astronauts turn on the radar too early?
@@tomschmidt381
Ah, well there's a thing, isn't it?
If you wrote your software to cope with failure, did it actually fail?
There's an RTOS rabbit hole there and it will swallow you whole!
Ultimately the fault was shown to be systemic, an AC line was run too near to the return cable from the rendezvous radar. So even when it was switched off the AGC received 100s of responses a minute and was incapable of ignoring them.
They filled up the executive and bang, 1202, no slots.
Interesting I was not aware of the wiring problem, I thought it was simply an improper procedure. Thanks for the info.
The funny thing there's presently development of IC technology applied to valves.
You should do a video on core memory. The idea that the nav systems on Apollo did what they did on core memory still blows my mind
"Software woven into wire: Core rope and the Apollo Guidance Computer", is the name of one of many articles about it.
Great summary. Before I wasn't aware that Texas Instrument's integrated circuits had internal gold wire connections and only Fairchild's planar process paved the way. Thanks for making that clear.
Your content is so wide and varied, it keeps me coming back to watch every time I see a new video.
I’d love to hear your take on the modern weapons the PRC keeps acquiring that are really just reverse engineered designs and tech from both the US and Russia/Soviet Union. Do you think their home grown capabilities are truly providing cutting edge stuff like stealth and carriers and space rockets, or are they providing the cutting edge of the designs they stole 10, 15 or 20 years ago. Just an idea.
Have a great day! Cheers from St. Louis, MIssouri USA
My father worked on one of the Apollo projects engineering part of the comms. He mentioned how unbelievable it was even then how fast the chips were shrinking in size, fast enough that finished designs would need adjustment because the size was reduced during the design phase.
Too much reliability lol. Love your humour.
Thanks for the video
Cheers
Wow, that 3 computer 'voting' system at 5:00 has shown me, once again, that I underestimate the nerdiness of Hideaki Anno & staff displayed in Gainax's works. Who would have guessed that art imitates life. I can only imagine that ended up in Eva because they couldn't cram it into Top o Nerae! and held onto the idea. I wish nerdy space sci-fi was still in vogue.
That kind of system was also used in the Space Shuttle; there were four computers running the same software, and if one computer disagreed with the others, it would be voted out. The Shuttle also had a fifth computer, same hardware but completely different software, as an emergency backup in case something went really wrong with the primary software or as a last resort if the other computers had failed.
I learned that the Saturn V rocket's "voting" system worked like this: if input 1 and input 2 agreed, use input 1. If they disagreed, use input 3. No need to compare all three inputs, just compare two. Logically, it's the same as comparing three and taking the two that match. If input 3 is used and it's bad (does not match either 1 or 2), the flight is lost anyway.
This channel is my new favorite... amazing an in depth content.
Great documentary work. I was 6 years old when Yuri Gagarin astonished the world with his achievement and as so I've been through all the "spatial race" times and advances. To me, it's still hard to believe how dedicated were all those who made it posible and where electronics has gone. Greetings from Paraguay, just on the opposite side of your place in the world
Personally, I think you covering tech and industry on a global scale, not just Asia, would be great. You explain things well and make it interesting. You'd have a lot more content for videos too.
Man, I love your vids. Found you randomly a week or two ago, been watching steadily since. Fascinating subjects.
As a stateside viewer, please keep making them. I really enjoy them and have learned so many tidbits from so many of your videos.
Also, the way I see it, the only people on this earth that can yell at you about the content of your videos are your Mother, and Grandmothers and that is it.
The rest can go pound sand.
Keep bringing us the great content!
One of your best videos to date, enjoyed every minute of it.
Excellent research as usual. One of my favourite channels.
Great Episode, Thanks a lot!
Love your videos and content research, don't apologize for making such a good content
"It get's me yelled at". That's actually sad that you feel that way. Me as a true follower, i like whatever content your are producing since it's always interesting and well spoken/written. If i were you, i'd try actually to extend the scope to actually whatever you personally find interesting, not just asia-related things. This one was really nice! Much appreciated, as always! Keep up the good work champ!
Very nice video! Also i find your tone of speach very professional
Thank You!!
＼( °□° )／
there, I yelled at you
Fairchild was an amazing story. Noyce was such a legend.
Rarely have I enjoyed a good video so much, this site is a luxury.
Yeah it's funny how many people don't know that the integrated circuit came about in large part due to the Space Race and the need to compactify the hilariously huge electronic components of that day into something that would fit in a reasonable sized rocket.
@@asgdhgsfhrfgfd1170 I do think you're inverting the problem. All of the practical researches are funded by private companies in search for ROI. The more fundamental sciences won't create any profit (at least not tangible and patents can easily expire before the manufacturing process matures enough for commercializing a product) so public founding is the way to fulfill this ungrateful task.
The thing is, space research is an in between that brings new challenges allowing techs or process to be created or allowing them to improve and evolve faster.
@@asgdhgsfhrfgfd1170 we got a lot of interesting technology out of the Space Race; integrated circuits becoming more widespread was just one of them,I know for a fact that a lot of things like velcro and especially a lot of stuff like solar panels miniaturization and all of that stuff and especially GPS came out of the space race.
Vacuum tubes still wins the beauty contest with silicon.
No need to apologize, I like your videos
As an ex Titan II launch officer I never knew the design choices of the guidance, but knew it was way better than the USSRs. Now I know the why. Well done!
"...and rode that rocket all the way to the moon." Excellent!
the funny thing was that had nasa done the apollo program 10 years later and they'd have had microprocessors and RAM, the AGC was very advanced but in its own ways a technological dead end as there were technologies on it that wouldn't be replicated for decades such as many multi layer pcbs, I can't recomend enough CuriousMarc's AGC rebuild video series
Amazing video once again
First time you make so many funny parts in a video, well done 👍 did enjoy it a lot, also laughed 😁
Another great video 👏🏻👏🏻 One observation tho: at the beginning the AGC displayed is actually the Display and Keyboard Assembly or DSKY. The AGC is partially shown at the bottom left of the frame.
Just discovered the channel and glad am I that I did :) In my view this is very close to public television level stuff, the info and presentation especially. Wanted to comment on a thing as people always do in the comment section @ 6:04 I disagree with the generalisation that tubes "suck" they "suck" for that applications, no doubt! but they can be reliable and perform very well in other applications. Tube reliability is the result of the circuit design and not a inherent problem of tubes. Tubes, transistors, integrated circuits, all have their quirks and sensitivities and if you design the application respecting the datasheet recomended parameters all of them can be reliable. I own both tube and solid state lab equipment and can atest that both types can work reliably for many decades. It has been my experience that tube faliure is very very rare in good quality equipment.. cheap equipment is another story and the tubes in them have a hard life usualy, old budget tube based tv-s probably the worst. End of rant :)
such a great video !
Thanks for showing a PCB shot of the HP35 :) well done recapping the story of the AGC. There is much more if you research the creation of the AGC software when there was no term for software engineering - Truly groundbreaking
6:04 Thanks for saying that. At the risk of making some audio enthusiasts cry...
To the moon ! And BACK. Definitely an important detail. Voyager space crafts are still active a great achievement.
The best channel on Semi
Love this 😍
Fascinating! I had always though integrated circuits were invented in the early '70s,.
Thank you for this interesting bit of history!
Great Stuff! Thank you.
Thank you. I love your videos.
GREAT STUFF
Wow great stuff thanks!
Love the closing comment. ...and I agree.
The channel CuriousMar has lots of Apollo electronics vids. A few years ago his team got the AGC going again and most recently, the radio system. And for good measure, see the vid
Ultimate Saturn V Launch with Enhanced Sound on the channel Starship Trooper
Funny, just yesterday I was binging on just this subject. You Tuber CuriousMarc got a Type one running that a guy bought at a salvage electronic junk yard. Only up to 6th episode and I think over 26 of them. Very in depth look see. Way over my head but I bet Asianometry would get it. Check him out if american stuff is ok for you. .
Thanks!
My complement for this excellent video. All of your video’s are. Please consider a research/video about: Carbon based chip / nano tubes
Nice transistor boards 😻
Americanometry
The photo at 17:55 gives dark dots illusion :D
If nobody ever yells at you about your video's you probably have nothing to say. You have plenty to say, and I like that. Top quality content with a lot of background and some informed opinion.
Nothing wrong with that, not at all.
Keep up the good work👍
Moon Machines has an episode on this subject. There is also a YT series on the restoration of one.
Somebody out on the interboobs did a detailed analyses of the the AGC. At one point during one of the moon landings shit hit the fan and the AGC basically worked like a boss and saved the day. The AGC was waaaay ahead of its time.
Ok, did anybody else crack up when he said: bringing him back safely, that part matters? 1:05
Seriously I love your humor! @asianometry
Although most people nowadays consider the Apollo Guidance Computer to be ancient technology, it still possesses a most elegant system architecture that rivals any modern processor. (see en.wikipedia.org/wiki/Apollo_Guidance_Computer)
Upgrading the architecture (steam-punk-style) would be to replace the core rope memory with SSD (USB) and throw a few ARM processors into the logic modules. The simplicity of the system leads to emergent scalability. Of course, it all has to be radiation-hardened for the journey to space, but that technology is mature today.
Excellent, well researched & totally agree with your conclusion. IMHO, the Minuteman II program was a far more significant factor in the development of ICs than Apollo. Another was the "calculator wars" of the mid-late 60s, which provided the "economy of scale" Fairchild were hoping for when they decided to sell ICs below cost-price. One small, nit-pick: the IBM 608 was not developed until 1955 & was released in '57 (not '53 as you state).
Some fans can lead your channel down a dead road.
Americans will yell at anything for no particular reason, don't take it personally LOL.
The fact you keep uploading content so often, which such detail level, shows an enormous research level behind every video, it amazes me that you still have time for working, traveling, eating and sleeping!
quote "I try to not make videos about America because it gets me yelled at"
as an Ozzie, I have to say that I laughed out loud at your words !
"America", whatever one's political bias/stance/creed, IS, in the same sense that "air" or "distance" IS, so we cannot ignore it, anymore than we can ignore gravity or sunshine.
please keep up the great work, your videos are exceptionally good.
A backup system really does though at least give a lot of peace of mind.
Jack Kilby actually made his integrated circuit from germanium, as he had trouble getting a hold of the silicon he wanted to use instead.
6:04 nice
Somebody got an AGC working again a few years ago. There's a video about it on UA-cam
Excellent story. Now tell them the story of the calculator, (Casio? I forget) that brought in the Central Processing Unit as a single $30 integrated circuit. That was the springboard for Intel.
Every guitarist absolutely disagrees with the statement "vacuum tubes suck" ;)
16:15 "Did he just enter the atmosphere?"
Since your narration is so professional and serious, these hidden Eastereggs make for even better comedy gold! I love it
God dang it Leroy
Amazing to think that these early computers had less computational power than the NES that came out a decade later.
You're audience includes a lot of Electrical Engineers. When has anyone had a conversation with EEs and not gotten yelled at?
A tidbit about the Minuteman guidance system, Jim Williams of some fame in the EE community, cut his teeth repurposing Minuteman guidance computers for lab use at MIT when he was a technician there.
Also to keep the yelling going, you might have mentioned Jean Hoerni along with Fairchild and the planar process...
Great! More like this please. How did the civilian market for ICs compare in size to the demand created by the military and NASA?
At that time? The civilian IC market basically didn't exist. NASA built the first computer using ICs.
0:15 - That's not the AGC! That's just the keyboard and display for it!
Where do you get information for these?
Small nit: Jack kilby and Robert Noyce were determined to be co-inventors of the IC patent. Although it is true that Kilby was first to announce, the lack of interconnect was considered crucial to the invention.
This is all arround my favorite video till now from you, but If you get yelled at for videos concerning USA and things..., its totally ok not to produce them! To not get annoyed, or worse eventually doubt yourself. Keep releasing what you like, and don´t feed the trolls :)
Fascinating stuff, thank you. A question: how did this fit in with the ‘hand-woven’ hardware, i.e. the computer made up of wires passing through or around metal rings; was this limited to the Lunar Module?
That would be for the memory. Core memory (RAM), Core rope memory (ROM). The rope memory was quite dense for the time. The AGC had about 14K of rope memory for the main program and lookup tables and about 4K of Core memory for it's ram. The Core memory was non-volatile meaning it would keeps it's data during power failure, and while it could be written to and read from, reading the data destroys the data so there would always have to be a write back after a read. Rope memory could not be written to after manufacture. The data was literally woven into the memory as what determined a 1 or 0 was dependent on how that wire was woven. One single wire could also be woven through multiple cores so that one wire could 'hold' multiple bits. Those bits would be selected via another set of wire that were also woven through said cores.
the 'hand-woven' hardware inside the AGC, was called core rope memory. it was the memory of the AGC that stored the programs. these were part of the computer in the same way your ssd is part of your pc (following this analogy the IC's where the CPU).
That's just the (magnetic) core memory and core rope ROM. Integrated circuits once they started establishing themselves quickly killed them off.
@@SianaGearz Specifically it was the MOSFET transistor that enabled VLSI silicon based MOSFET ICs that enabled it, consigning CRM to history
no both the lunar module and the csm used an AGC which had rope and core memory. the lunar module even had a simplified backup computer to launch and orient the spacecraft if the AGC died for some reason
Of course, the proximity fuse of ww2 fame used special vacuum tubes and was designed to survive the shock of being shot from a gun.
Well, I guess it wouldn't have been a disintegrated circuit that took us to the moon
8:41 Ah the 2N696 - I remember it well. And the 2N1131, its PNP brother.
Hi Jon, are you doing something on the 7nm Chinese chip?
Yep, there wasn't much demand for miniaturization beyond discrete transistors in the consumer electronics of the time. Look at the large amount empty space available inside a radio or TV of that era. The drive for ICs at that time came from military and civilian space efforts to reduce the size and total mass of electronics.
Both vacuum tubes and transistors require wiring them up. Every component, even ICs, require some wiring up to other components, but the point you were trying to make got lost in the shuffle because the main problem with transistors (and tubes) is that to do computing, you need lots of these devices and the wiring can introduce problems the IC doesn't have.
Also, when you say 4100 ICs (and later 5,000), I find it hard to believe there are thousands of chips. I suspect you mean the number of transistors in the ICs.
The AGC used chips that contained two NOR gates each, so yeah, the Block 1 AGC used 4100 chips.
I am a SWE and have no professional hardware experience, but I do love it as a hobby, I have been designing tons of PCBs over the years. One thing I have always been curious about is why most ICs (like ST ARM chips, Atmel, the ESP32, etc) all require external capacitors and resistors for things like decoupling, pullups/pull downs, etc.
Why is this, couldn't they design the silicon wafer to incorporate these components? Is it impossible for them to design a semi-powerful ARM chip for example without these external passive components? Is it because the external passive components are expected to handle higher power levels/temperatures than the silicon could endure, or is it impractical to build something like a 22uF capacitor into a wafer design?
@8:05 - was the caption meant to be Autonetics, rather than Automatics?
‘But vacuum tubes suck…’
*Triggered guitarists and audio geeks have entered the chat* 😂
In fairness, tubes got a lot better towards the end of their development. Subminiature tubes ran at lower voltages and were much stronger and less subject to shock. They were used in ballistic missile guidance systems for quite some time due to their resilience to EMR versus ICs, apparently.
That IC was not used on Explorer 18. The quarter is about 20 years too new.
NASA and the ICBM programs were the "killer app" of their day.
Whoever is yelling at you They’ll be OK proceed with whatever interests you. This is your UA-cam channel not theirs
Great video.
I will say that my understanding is those first orders did make Fairchild financially solvent temporarily and convince investors of the viability of the business.
I walked by the picture of the traitorous 8, and the stainless flying F in the final slide countless times. It’s kind of sad that Fairchild is no longer in business, although their first commercial sized fab is still producing IC’s
Maybe a story for another video is video game development when Fairchild competed with Atari releasing the Fairchild entertainment system FES.
5:04 is showing an SMD moulded Package. These came in the 90s.. Should be rather through hole metal housing.
OK. Later on everything looks authentic :-)
“Pilot Gordon Cooper…like a boss”….that’s the 2010s version. The 1970s version is “Pilot Gordon Cooper…Tightened that bad sucker inside the runway like a mother. Sheee”.😂
Great video, subbed and I love your channel, but I've got to tell you that "solder" rhymes with "fodder" (the l is silent).
Not in British English it doesn't 😊
@@cncshrops Good to know!
I don't think transistors needing manual work was the reason they switched to using ICs. Making a PCB is much much easier than making a Silicon Die and has much better yield. Also, IC packaging needed manual work to connect the bond wires to the pads. The main reason everybody prefers ICs is the transistor density (area), speed, and power. Smaller interconnects mean less capacitance. Less capacitance means higher speed and better power efficiency. Add all that to having a physically smaller circuit.
TIL NASA came up with the MAGI system...