Reverse engineering electronics

Question

I haven't found this question anywhere else so I don't think it's a duplicate. So here it goes, some background information of the world.

Suppose there is a civilization with access to magic which has taken the place of traditional technological advancements.

Things such as fast transportation, aid in farming, etc are all enhanced by magicians and the majority of resources are poured into further developing and advancing magic (which for all effects and purposes is basically another branch of physics since magic in this world exists thanks to the properties of certain elements and not, well magically).

While knowledge of electricity exists it is mostly used as a tool for battle and not for transferring and storing energy or using it to power machines.

For this purpose another form of energy is used, let's call it aether, which is more efficient. The idea of computers has not been well realized yet. (If I had to draw a comparison I'd say technology is at late 19th century levels more or less)

If that civilization was to stumble upon a computer, like the one we are using today (circa 2017 home computer) but it wasn't in a usable state, could they

• Recognize what its purpose may have been?
• Understand the concepts and technology that make it work?
• Reverse engineer it to further advance their own technology and how long would a process like that take?

Would any of this change if the computer was usable instead of broken? (I'm not asking whether they have motivation to study the computer or any reason to, I'm asking if they could do it)

Answer 1

Finding a dead computer would give little to no hint to its usage (unless their "magical concoctions" happen to be very similar, physically, but that would be a loooooong shot!).

Finding a working one would be even more "magical", since modern computers depend on a lot of "external environment", including, but not limited to, a power grid, not to mention the "other end" of any needed connection.

If you, by mystical chance, manage to lay your hands on a fully working device then next step would be to understand how to operate it and that is another big stumbling block, unless you get a brand-new computer which powers up in a very detailed dumb-proof tutorial guiding the user unto the intricacies of operating on "folders" to start a "browser" or an even more mystical "productivity suite" (would they know what "productivity" means?).

Even if all these steps are somehow performed (and I think it may be material enough for a long novel) understanding the working may prove an impossible challenge (depending on specifics of your magical setup).

Remember Clarke's third law; corollary is such a box will be studied as a "magical artifact" and, unless your magic is based on a good deal of Science (as you seem to hint), I see no way a sorcerer can divine enough Quantum Mechanics to understand a transistor, let alone understand what are the billions of them embedded into the CPU supposed to do.

You would need some kind of Rosetta Stone to bridge the "techno-magic" gap (perhaps in the form of a digital Encyclopaedia contained in the device itself or a paper-based one found nearby).

Answer 2

You, a gentleman from the 1800s, discover a strange flat silver box, with a hinge at the front (or so it appears from the image of a fruit on the lid).

Under the lid, where you might expect to see a tray of chocolates, you see a tray of letters and symbols, randomly arranged. Perhaps you recognize the alphabet, but if so, it is scrambled, and placed in the middle of the tray, facing away from the hinge.

The letters appear fixed in place - trying to remove one with your fingers, it pops down as if on springs. They could be mechanical buttons, but none, at first glance, appear functional. If you have experience with typewriters, you might intuitively get the concept of keys; otherwise, you will at least understand the concept of buttons and an alphabet.

Below the buttons is a faint recess, such as one might rest a small lithographic print or business card into, but unfortunately there is none there: it is empty.

At the sides are various holes of various shapes, five on one side and three on the other. One looks like a coin-slot, the others are squarish and go in to various depths, some having golden metal springs within.

On further investigation, one button, the one on the top-right, appears to be a toggle, making an illuminated image of a beach appear and disappear from the inside of the lid.

When the beach is shown, so too is a small rectangle in the middle of that picture, containing the text "enter password".

Most of the buttons at that point add a dot to the box, the exceptions being:

• "delete" removes one of the dots;
• "enter return" makes the rectangle shake from side to side;
• "caps lock" makes a light on that key light up or go dark;
• "fn", "control", "shift", "alt option", "command" and the keys marked "F1" through "F12" do nothing;
• "esc" at the top left, and the strange symbol at the top right both turn everything dark again;
• Some arrow buttons move a blinking line left and right in the little rectangle.

You have, functionally speaking, plumbed the depths of this device. You take it and show it to the greatest minds in the nation: none discovers more about the picture box. Some propose that perhaps placing a lithograph into the tray below the buttons and pressing the correct combination of buttons will make the image of the beach change to that of the lithograph, but nobody makes it work. In playing with this, they discover that messing with this moves a picture of an arrow around the screen, but it appears to do nothing (though tapping it when over the rectangle moves the blinking line).

Someone pushes a coin into the coin slot, but it just falls out again and appears to do nothing.

Sometimes, it blows a friendly warmth of air at you.

Eventually, the image grows dimmer and stops showing up.

Someone suggests dismantling it to find how it worked. The lid appears to be a single piece, but the tray appears to be riveted together, so you drill out the rivets.

Clearly, it's a machine. An electrical machine. You remember it giving off a friendly warm, so the function of the fan is obvious (warming the user). But the rest?

Below the fan, probably a heating plate.

The big items - probably one stores the electricity, and another stores the image of the beach. The remainder... complex electrical circuits. That's OK, you understand electricity. The wires are glued down to a board, but it's still obvious enough.

So many very, very tiny components, but you can quickly identify the simplest ones - the capacitors, resistors, and diodes. You discover that the light in the "caps lock" key operates as a diode, and by providing a voltage to it yourself, slowly increasing, establish that it took about 0.7v to illuminate.

You discover that the coin slot was not a coin slot, but rather probably for inserting some complex plug, as were all the other indents around the side. One was clearly for an auxiliary power source, with thick tracks leading from it.

For the devices you do not understand, you measure the resistance and capacitance, and then plot the V-I curves between each of them. In doing so, you intuit the behavior of the standalone transistors, the power regulator, the clock crystal.

Thousands of interested minds join you in your quest to reverse engineer the "seascape device". The effort makes the decoding of the enigma machine look trivial. Entire warehouses quickly fill with written observations. Your understanding of electricity proceeds in leaps and bounds. The science and engineering of electronics becomes a thing, and solid-state devices are developed.

Between you, you work out that the board has multiple layers; that clock signals are used to tell the little black boxes to process their inputs; that the boxes have inputs and outputs, and those are made of high voltages and low ones.

The other stuff, though... it leaves you baffled. The best you can come up with, for these strange black boxes, is that some of the pins that come out of them are almost certainly for providing power.

X-rays and perhaps destructive microscopic examination confirm that the black boxes are protective shells around impossibly-fine wires which connect to small fragments of crystal etched with even-more-impossibly-fine details.

You know that SOME sequence of inputs - high voltages and low voltages - to the various chips, gives some sequence of outputs. It's not always deterministic, but sometimes it is.

Assuming you found an entire warehouse of these devices, what more can you find out, even by destroying them? You can map out SOME of the commands of the CPU, once you realize it's not a heating element because there are too many lines going towards it, and figure out it's the heart of the thing. You can identify that it's reading memory, or sending commands to the hard drive controller area.

Ultimately, though, your tools and your analytical machines get better and better, and you manage to crack it, with scanning electron microscopes and such, to decypher every logic gate and transistor, and describe the behavior of the thing completely.

And you look at each other and say "in creating systems which could analyze this one, we've been forced to develop far better systems ourselves. This is no longer magical or even impressive, it's just a laptop. An outdated one."

Answer 3

No, it's a dead loss

A magical society on discovering a computer will most likely spend their time trying to work out the correct sequence of words to make it go, as they would with a magical artifact, and not realise that it needs power.

Unless magic

The only way they're going to understand this machine is if their magic allows identify and repair. The chips are all black boxes, the other components are other coloured boxes and that's even if they understand that it's a complex object in a protective case not just a thing whole and complete.*

Magic usually works by command words not manual switches. Are they going to understand the concept of a power switch? Even if they have electricity available, will they understand the mystic runes "230V 50Hz AC"? Without such understanding they won't even get it to power up.

Do they understand the concept of the device in question? This is critical to them being able to identify it. They have to have something to compare it to, whether magic mirrors or otherwise, the concept of having something that does at least part of what a computer does must exist.

_{* Which is pretty hard to tell with the machine I'm using to write this}

Answer 4

This question can be boiled down to: could a nineteenth century technology level society, with or without magic, understand how a lump of electronic machinery works?

The simple answer is no. They lack any equivalent machines to recognize it is a computer. Even if their society possessed Babbage computing engines, as those were mechanical and not electronic, there is no way that the appearance of the computer would tell them its function.

They lack the microscopy to determine the structure of solid state integrated circuitry. They might recognize the wires and wiring in the device, so it is possible they could realize it was an electrical device. This does depend on whether they use electricity and conduct it via wires.

The main trouble is a non-functioning computer (circa 2017) will look to people in nineteenth century level society that it could be a very sophisticated work of art.

Unless the magicians in this world possess a form of magical divination, and as suggested by @Separatrix in his answer, which enables them to plumb its purpose and function. However, their divination would have to extend an understanding of the electricity supply needed to run and power the computer. Of course, if their powers of divination were that good, then reverse engineering computers would be as easy as pie.

Then in all likelihood, they create computers powered by their aether. This would save on building power stations and all the other industries attend on the manufacture of the advanced electronics necessary to make computers. This is the other unspoken problem with reverse engineering a piece of technology nearly a century ahead in terms of technical development. It isn't simply a matter of reverse engineering one type of device, it is also a matter of reverse engineering all the industries, services, trades and professions to make all the parts and components, materials and services to make it work. This means building an entire sector of the economy that doesn't exist in their world.

In summary, the simple answer is no, but so is the complex answer where the magicians have the powers of divination to reveal its actual purpose and function. This is not simple reverse engineering, it then becomes completely rebuilding their world to accommodate a new technology.

Answer 5

Unless magic allows understanding of the science used to create something, then they will be unable to understand it.

If they do get it running (eg it's a tablet and has battery power) they will probably assume it's magic - the "technology" they know.

Can you reverse engineer a computer by looking at it? I sure can't. I think anyone would be at a loss if ordered to "duplicate that black box that produces non-magically powered pictures on the screen" even in the modern age. Even with magic allowing them to see what is happening inside, the tiny feature size on modern computers, and the use of so physics far in advance of the time would render it useless.

I imagine if you took a current laptop back to the 1990's, the would recognise it was technology, and that it was simply a more advanced version of what they had. But there is no way they could replicate it (14nm silicon manufacturing technology? Eh, nope, in the 90's they were three orders of magnitude off). I doubt they'ed even understand it properly. Todays CPU pipelining and architectural designs are so far advanced that even after I studied them for a year, a modern CPU is still mostly a box of "magic".

---

If a faster than light space ship jumped to Earth, and allowed humans to look at it's engines - but only with human tools (eg scanning electron microscopes) we would end up, after months of study with:

1. No idea how it worked
2. No manufacturing ability to duplicate it

Same situation with a latptop 100 years ago, let alone middle ages.

---

However, you can bootstrap a civilisation to any given level of technology by providing sufficient education methods within only a few generations. The difference between current technology and that of 1000 years ago is simply that today's engineers were taught on material 1000 years newer. There is nothing that prevents setting up a modern-day technical university in the middle ages - if you could somehow provide it with electricity, and all the supplies (food, chemicals, metals etc.) that it requires as part of tuition.

Answer 6

I've vetted and edited a lot of books. Assuming your magicians could reverse engineer a computer without having electricity as the basis of their energy infrastructure is what I call a "technology dichotomy." You'd be surprised how many stories are written involving time travel without having first (e.g.) created the wheel --- basically because time travel was simply the process the author used to justify telling the story they actually wanted to tell (but that's another issue).

Simple answer: no. No understanding of electricity. No understanding of printed circuit board manufacturing or silicon fabrication. Being an electrical engineer myself, they'd not even realize that printed circuit boards can have many (sometimes a lot) of layers. Chips are even worse. Transistors aren't discrete items on a silicon chip, they're often mushed together depending on the doping configuration to improve manufacturing or operational efficiency. And that's assuming you had a microscope capable of even seeing the doping surface, much less the cross-section. (And I haven't even started talking about magnetic hard drives, operating systems, BIOS, or what an "icon" is...).

Our magicians would be at a complete loss, not having even a simple multimeter to start testing things. It would be no different than asking someone completely trained in all the latest steam technology to work it all out in a room lit by natural gas. They couldn't.

EDIT: After reading and thinking more about this, let me explain that the body of knowledge needed to reverse engineer a modern computer is MASSIVE. I've designed chips that went into space, and yet I couldn't begin to reverse engineer even the smallest fraction of what would be needed. Hyper-threaded multi-core CPUs? Encryption and compression technologies? Wireless mice? Electromagnetics! Think about Windows 10 and trying to figure out how it worked with no computer background at all. It's actually humbling to realize just how much knowledge has been accrued so we can play MMORPGs.

If today's modern technology were likened to the top of a mountain and nobody existed who understood those lofty peaks, literally the entire mountain would need to be understood to climb it again (and the cost, quadrillions of dollars world-wide over the last 75 years...). The task is so large that there is no fundamental difference between reverse engineering the computer and re-discovery of all that knowledge. It would take that long.

An an interesting Logan's-Run-esque question is, what would happen if our Earth were suddenly devoid of the people who understood those things? Let's say just three generations of them? Could our technology-dependent society survive the time required to re-aquire all that knowledge? Probably not.

This is fun. Cheers!

Answer 7

The point at which this becomes a nonstarter is that the computer isn't an a usable state, the only reason we understand as many archaeological finds as we do is that we can draw parallels from those finds to objects we still use. They may still see some useful advances in material science from dismantling individual components and get some interesting ideas from things like capacitors if they assume this is an advanced but broken Aether based device, since a number of modern electronic components are actually kind of intuitive in design when seen in detail, for my money anyway.

A working computer may allow them to glen some information as to function from which they may be able to extrapolate something useful about the how side of things but dead electronics are remarkably uninformative.

Answer 8

While I'm agreeing with the other answers, I think Ihave a nice metaphor do help understand why they couldn't figure it out.

You get a map of a city, complete with street names. What can you tell about the city, you could make educated guesses about population density, but really you can't infer that much about the city.

Now imagine you've got a 10,000 foot view of the same city starting 6am on an average Monday morning, you get to see how the population flows, you can identify residential areas, industrial areas, business areas. You see kids flocking to schools, then back again. Study the city for weeks and you'll identify markets, recreational spaces, places of worship, transit hubs etc.

Examining a piece of dead tech tells you very little about it, studying a piece of running tech tells you much much more about it.

That said, if you start poking around a running motherboard to see how it works, you'll quickly and irrevocably break it.

As has been mentioned, early 19th century folks would find reverse engineering integrated circuits an almost impossible task, and they were familiar with basic electrical circuits (such as light switches). Your fictional magic users wouldn't have a hope (unless they can magically divine it's components and their operation at microscopic levels.

Computers are horrendously complicated machines, they're so ubiquitous we often forget that.

Answer 9

What if some of the concepts were present already?

Charles Babbage is credited with inventing the first computer long before the technology existed to make an electronic computer. He died in 1871 still working on his general purpose computing engine, but his work popularized the ideas of computing and formed the foundation of knowledge necessary to understand (and ultimately create) electronic computers. It's quite possible that similar thinking would have been going on in the magical world as well.

Magicians would clearly need to understand energy flow. The next step towards a computer is thinking of energy as holding state and representing information and then the idea of combining different flows logically to create and store desired information states. These are concepts you could easily imagine being present in a magical universe if they were important for advanced magic, or were being worked on in theoretical circles. The third leap of insight and the one that takes us to the idea of a magical computer is: what if I could chain a bunch of these systems together and do advanced calculations, store arbitrary information etc.

If your goal is a plausible way to import computation to your magical world, pick some of these concepts and add them to your world.

It would still be hard to reverse engineer a computer though. One of the barriers would be how advanced and compact it is. Reverse engineering a simple logic circuit like this adder would be trivial. Pump some magical energy in one end and watch the effects, but the adder(s) in your computer have been shrunk to the size of a human hair and is sandwiched in layers of silicon. It's hard to conceptualize things at that scale.

It would be far easier to reverse-engineer something out of a computer history museum, a child's toy (walking, talking robot?) or something designed to teach the fundamentals of computing. A college notebook containing an undergraduate's computer architecture homework would be sufficient to kickstart a computing revolution in a magical world. A simple computer like a Raspberry Pi or an arduino might do (It certainly would if it also included the book).

Answer 10

I'll touch specifically on the electronics side of things, as many of the answers focused on the running and using the software itself.

Regarding the electronics that make it work, reverse engineering them would be a nearly impossible task. The only reason we can even understand a modern processor design is because we have the previous version of it to design on. We use software to understand and model the processor designs and make sense of the billions and billions of transistors that make up the circuit.

Trying to reverse engineer that with our currently technology is a monumental task. Trying to do it without would be essentially impossible.

Something like a 1980s machine, where the design is far simpler and a single person could comprehend the entire system would be a more achievable goal, though many of the issues mentioned above would still exist.