Ten Weirdest Types of Computers

An anonymous reader writes to mention that New Scientist has a quick round-up of what they consider to be the ten weirdest types of computers. The list includes everything from quantum computers, to slime molds, to pails of water. "Perhaps the most unlikely place to see computing power is in the ripples in a tank of water. Using a ripple tank and an overhead camera, Chrisantha Fernando and Sampsa Sojakka at the University of Sussex, used wave patterns to make a type of logic gate called an "exclusive OR gate", or XOR gate."

It ain't ...

[Anonymous Coward]

a computer if I can't get pr0n.

Re:

[Anonymous Coward]

OK editors, forget this is /.? Do you really think you have to explain the words behind XOR?

Re:

[uglyduckling]

Well... uh... imagine this piece of paper is the universe, and this pencil is a quantum singularity in n-th dimensional space. No, better still, imagaine that this salt cellar is the universe and this napkin is a small black hole - uh, no, that doesn't work... in fact, *thunk*.

Re:

[lattyware]

Now we just have a majority of opinionated, non-technical idiots and children who once installed Linux screaming about technology they know absolutely nothing about.

And those guys who whine about how shit things are, and yet still manage to post.

Re:

[cyphercell]

I knew that once. Don't use it much anymore.

What about the weirdest computer of all?

[MaDMvD]

The brain.

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[Anonymous Coward]

except, of course, for "Krang" in Teenage Mutant Ninja Turtles, the historical characters in Futurama, etc...

Re:What about the weirdest computer of all?

[Naughty Bob]

You'd be interested to know that rat's brain cells have already [newscientist.com] (the linked article is from 2004) been harnessed to fly a virtual F-22.

The singularity, as the man said, is near.

Re:

[Daimanta]

The singularity, as the man said, is near.

And I still don't have my flying car!

Re:

[truthsearch]

No, but apparently you do now have your flying rats... oh, wait...

Re:

[The Great Pretender]

FTA - "Ferdinando Mussa-Ivaldi of Northwestern University in Chicago has shown how a few brain cells from a lamprey, a primitive eel-like vertebrate can be used to control a robot."

I would like to be the first to welcome our robot controlling, primitive eel-like vertebrate Overlords and welcome them to harness the power of our captive rat brain cells for virtual war.

Re:

[Snowmit]

I however would like to NOT welcome the flesh eating [seanbaby.com] primitive eel-like vertebrate Overlords.

Seriously scientists, what's wrong with you?

Re:

[geekoid]

Singularity is a myth perpetuated by people with no imagination.

Re:

[Naughty Bob]

Singularity is a myth perpetuated by people with no imagination.

So when, mighty imaginer, will computing power plateau?

Re:

[kvezach]

So when, mighty imaginer, will computing power plateau?

Who knows, but it's not important. You can't make a fire build a house, no matter how powerful fuel you're pouring on it. In other words, it's going to take more than just brute force to scale up intelligence.

Re:

[Naughty Bob]

That's pure conjecture- But anyhow, not only do we have no understanding of what, exactly, constitutes human intelligence/consciousness etc, but to describe the ongoing attempts to approximate it as 'brute force' tells me you are equating the clock-cycle race with progress in the area (or you have a rather glamorized impression of brutes....).

Re:

[kvezach]

You talk about "computing power" as if it plateauing would prevent the singularity, and Kurzweil et al. make diagrams like these [lifeboat.com] that pretty much link Moore's law and the Singularity together.

As for brute force not leading to scalable intelligence, just take a quick look at tree search. That's exponential; it's just that, as an example, the game tree for chess is narrow enough that computers can beat grandmasters. And it's not just chess. Many other puzzles, when generalized, are NP-complete, and many ga

Re:

[hcdejong]

That's only true for limited values of "fly", though. The task was to keep the aircraft straight and level in variable crosswinds. This can be accomplished with a simple feedback loop. Things like landing are much more complex.

Re:

[knarfling]

Actually, check out Anne McCaffrey's Brainships series. Although the first one, "The Ship Who Sang" was written in 1969, several others like "The Ship Who Searched", and "The City Who Fought" were written in the early 1990's. The last one that I know of, "The City and the Ship" came out in 2004.

I realize that the only one written only by Anne McCaffrey was the original, "The Ship Who Sang", and the others were co-written by other authors. (Usually that means written by other authors using McCaffrey's univ

Re:What about the weirdest computer of all?

[c6gunner]

In his early Known Space stories collected in Tales of Known Space [amazon.com] Larry Niven forsaw a future 1975 (ha) where the brains of people managled in car accidents are integrated into spacecraft for guidance, allowing them to continue contributing to society even if their bodies are gone.

Some idiot who can't even handle a car ends up causing a massive accident, and Niven wants to let him drive a spaceship? Great idea! What could possibly go wrong?

On the bright side, I hear collisions at relativistic velocities are rather painless....

Re:

[Whiteox]

http://en.wikipedia.org/wiki/Donovan's_Brain [wikipedia.org] was the earliest piece I read ages ago.
The article also explores the development of the theme.

A reference to the title can also be found at the end of the Larry Niven short story Becalmed in Hell, in which the character Eric, who lives as a brain and spinal cord on life-support, and works as the directly-connected controller of a NASA exploratory vessel, signs a telegram "DONOVAN'S BRAIN", either as a joke of his own or because Niven surnamed the character "Donovan" in homage.

Re:

[The_Dougster]

In his early Known Space stories collected in Tales of Known Space [amazon.com] Larry Niven forsaw a future 1975 (ha) where the brains of people managled in car accidents are integrated into spacecraft for guidance, allowing them to continue contributing to society even if their bodies are gone. This entire idea of "brain in a jar" science fiction seems to have faded out with the 1970s.

Well, maybe jars. Personally, ever since I invented that harmonic portal to the other dimension where I learned how to e

Re:

[syukton]

An analog asymmetric multiprocessing system complete with random-access memory and a variable-speed bus. Truly, quite weird.

Re:

[syukton]

The strange dreams people have would seem more random than associative, although they may indeed be associative on some level. Although the single best example I can think of is simple creativity. One's ability to create new things, come up with new ideas, look at things from a different perspective, those seem like random-ish processes.

Re:

[Reziac]

Worse -- MY brain!

[looking around the room]

Or possibly my computer, which has a mind of its own. ;)

Re:

[ppc_digger]

I'm pretty sure Pinky is weirder.

Re:

[mapkinase]

Brain is quite lousy computer.

Re:

[slimjim8094]

The brain is Turing-complete, as in it can emulate a Turing machine.

So does that make people computers by definition?

Re:

[geekboy642]

Hi, oh Anonymous Coward. You may have forgotten something, seeing as you've been posting on Slashdot since before CmdrTaco registered, but "Anonymous Coward" isn't exactly a "low uid". In point of fact, I'd say you rank at just about a 12-digit UID.

Go troll Digg, that's a good place for idiots who have nothing better to do than spew pointless venom.

Re:What about the weirdest computer of all?

[c6gunner]

So I gather you disagree, from your semi-intelligible response? Did we (collective humanity) or did we not create the computer you are reading this on? Did we not bring about the technological advances that are stated in the very article you are replying (hardly) to? If so, then are we not the ultimate computer? Equipped w/ the highest resolution video, audio, CPU/logic, etc?

Nope. We also created the nuclear bomb, but we're not the ultimate explosion.

Your logic is faulty because there is no rule which states that extremely complex systems have to be created by even more complex systems. This is the same logical fallacy which creationists often advance in order to "prove" the existence of God: the idea that because humans are complex, there must be an even more complex being which created us. In reality, it is quite possible for complex systems to be created as a product of random chance, or natural selection.

As for humans being equipped with "the highest resolution video, audio, CPU/logic, etc", that's just plain silly. Computers can detect and display video at resolutions (and in light spectrums) which are undetectable by the human eye. They can detect and produce sounds which would be inaudible to us. And when it comes to raw number-crunching ability....well, don't be silly. I'd like to see you sit down and brute-force an NT LM hash in your head. Hell, I'll be generous and let you use a pen and paper!

BTW, the guy you were responding to was clearly making a joke. Lighten up.

Re:

[edumacator]

I think the underlying principle the original post was trying to make remains valid. And c6gunner, I think you made the leap in inference when you think he's saying complex things have to come from other complex things. I think he was implying that the human wetworks are pretty impressive to have dreamed up taking ones and zeros and making them into such marvelous things. You have to give him that right.

And to the resolution point. I think you're thinking too literally. Show me a computer that can discern

Re:What about the weirdest computer of all?

[c6gunner]

I think he was implying that the human wetworks are pretty impressive to have dreamed up taking ones and zeros and making them into such marvelous things

The thing is, the human mind doesn't really do that anyway. Who do you know that programs in binary? And can you imagine a human being creating Windows Vista from scratch, in binary?

Human accomplishments, much like evolutionary progression, are cumulative. Our brains are basically simulation programs - they take data, feed it through a series of filters and rules, and then act on the output. However, since we're able to learn, each successive generation gets a different set of rules and filters, allowing us to work out new problems without first having to go back to basics. As such, it's wrong to say that the human mind created modern computers - rather, the human SPECIES created modern computers. There's a huge difference there. All of our accomplishments owe as much to natural selection and the passage of time as they do to the complexity of the human brain.

But yes, I'd agree that the human brain is an amazingly complex piece of machinery, which is impressively adaptive. If that's what he meant, then we are in 100% agreement.

And to the resolution point. I think you're thinking too literally. Show me a computer that can discern a great masterpiece from a technically proficient work.

Show me a human that can.

Even if we ignore the fact that judging masterpieces has nothing to do with resolution, your argument still makes no sense because the judgement of "masterpieces" is subjective. Show a Picasso to an African tribesman, and he'll probably use it for kindling. On the other hand, the artwork of his own people will doubtless hold great value to him, while being nearly worthless to the average westerner.

Re:

[Raenex]

Show a Picasso to an African tribesman, and he'll probably use it for kindling.

So would I, except for the other people who think it's worth lots of money.

Re:

[edumacator]

But yes, I'd agree that the human brain is an amazingly complex piece of machinery, which is impressively adaptive. If that's what he meant, then we are in 100% agreement.

So we are primarily on the same page. I'm going to hold on to that. :)

Even if we ignore the fact that judging masterpieces has nothing to do with resolution

Forgive me, I'm an English teacher, so I take liberties sometimes with language. I was using the broader definition - analysis into clear-cut components.

Show a Picasso to an Africa

Re:

[c6gunner]

I would argue that the subjectiveness of our evaluation is indicative of the complexity of the human mind.

There is certainly some truth to that. The ability of our mind to adapt to different environments is certainly one of it's more important aspects. However, our ability to make subjective judgements isn't exactly unique - you could easily "teach" a computer to subjectively critique art. The only caveat is that the computer's sense of "fine art" would be completely different from our own, and would c

Re:

[RespekMyAthorati]

Human accomplishments, much like evolutionary progression, are cumulative. Our brains are basically simulation programs - they take data, feed it through a series of filters and rules, and then act on the output. However, since we're able to learn, each successive generation gets a different set of rules and filters, allowing us to work out new problems without first having to go back to basics. As such, it's wrong to say that the human mind created modern computers - rather, the human SPECIES created mo

Re:

[CustomDesigned]

Your logic is faulty because there is no rule which states that extremely complex systems have to be created by even more complex systems. This is the same logical fallacy which creationists often advance in order to "prove" the existence of God.

You are correct when talking strictly about complexity. Intelligent Design people, however, are talking about "chosen" or "specified" complexity. For instance, the number Boggle arrangements is its complexity, but when a person selects a particular arrangement (as opposed to tossing the cubes), that is specified complexity. If an arrangement forms an English sentence, then your judgment of the likelyhood that this was a random roll or selected by an English speaking person would revolve around the Tota

Re:

[jd]

The philosophical problem comes with detecting design by non-human intelligence, possibly not even part of this universe. There are a ton of presuppositions as to how to recognize a message vs noise.

Yes and no. You cannot distinguish any signal from noise, but you CAN distinguish certain categories of signal from noise. For example, virtually nothing emits much of a signal at the frequency popularly known as "the water-hole". It's (almost) certainly not a maximum for anything. Natural sources are also gen

Re:

[CustomDesigned]

we only have to know that nature neither thinks nor organizes.

But nature apparently does organize on occasion, because we are here. (Or else something or someone else did the organizing.) So you are looking for signals that are different from most signals in the same way that we are different from most of the products of nature.

Re:

[c6gunner]

This is, of course, so incomplete as to be bullshit. Thermodynamics only requires entropy to increase globally, or in closed systems. The Earth is not a closed system. Therefore any statement that entropy must increase on Earth because of the laws of thermodynamics is simply false. They easily permit locally decreasing entropy at the cost of increased global entropy. Want proof? Clean your room.

Why the hell are you posting as anonymous? The simplicity and accuracy of that paragraph combined make for a tr

Re:

[c6gunner]

Your sig says, "I don't respond to Anonymous Cowards." Yet you just responded to one.

Funny - every time I respond to an AC, someone feels the need to point this out.

My sig is just a warning: if you WANT a response from me, don't post anonymously. It's not a promise, a moto, or a written contract, it's a simple suggestion. On the other hand, if I feel that a certain AC post deserves a response I will respond to it, but such cases are so rare that I can still count them on the fingers of one hand.

Yo

Re:

[CustomDesigned]

Congratulations! That straw man is history! If I had said anything about entropy increasing on Earth, I'd sure be embarrassed. Now I will say something: over the long term, entropy *will* increase on Earth according to thermodynamics - it is an inexorable as 2 + 2. The sun can only hold it at bay for a few billion years.

You can be as proud as you want of Human accomplishments, but it is all destined for the cosmic heat death and ultimately meaningless - no matter how many meta-universes you propose to

Re:

[Aetuneo]

Wait, there are 90 Megapixel sensors for computers now? Why did no one tell me about this?

Re:

[c6gunner]

Wait, there are 90 Megapixel sensors for computers now? Why did no one tell me about this?

Not for your home PC, no. Check the resolution on spy-sats, though, or on the Hubble (230+ megapixels).

Re:Random chance cannot create complex systems

[jd]

Give me just one example of a complex system that was created by chance.

Earth. Jupiter. Saturn. Alpha Geminorum. The Andromeda galaxy. The United States tax system.

Re:

[BungaDunga]

The Game of Life (and cellular automata in general) can produce very complex behavior from a very small set of rules. If I run through a set of random live/die rules (one cell dies if it's surrounded by 4, not surrounded by four, surrounded by 3 or 5, etc) I'll eventually come up with one that does something interesting. It could be totally automated and random, but leaving each cellular automata to run and then checking on it and looking for complexity will produce some patterns that /look/ "designed" but

Re:

[jd]

Certainly. Cellular automata were used by Von Neumann to prove that simple rules were sufficient to produce a "Von Neumann machine" - a machine capable of replicating itself - and even a "universal constructor" - a machine capable of constructing any pattern at all from a blueprint. I am not positive, but I think a "Von Neuman machine" pattern is known for Conway's Game of Life.

Re:

[tsjaikdus]

>> You're saying things like a watch can be created by accident?

No, making a watch involves a designer.

However, the designer can be created by accident.

Re:

[Killjoy_NL]

Leaving a rock on another rock and watching it for a full day can be done by accident (sundial)

Re:

[dj_tla]

Actually, a watch can be created through random chance if you think about what watches are made of: gears, ratchets, hands and springs. Randomly put them together in various configurations, keep the good ones (through a natural selection-like process) and randomly mutate those, over and over and over. Eventually you'll get clocks.

Video that illustrates the process: http://www.youtube.com/watch?v=mcAq9bmCeR0 [youtube.com]

Re:

[c6gunner]

I've never heard anyone believe in your claim that God had to be more complex just because we are complex.

Then you've never argued with a creationist. EVERY argument about God eventually breaks down to the idea that a complex system such as a human being cannot be created by chance or natural selection, therefore "God".

A higher power does not imply greater complexity.

Yes, it does. If it's not more complex, then it's not a higher power. If you fail to understand simple symantics like that, I don't see

Re:

[c6gunner]

Okay, but that doesn't say anything about requiring God to be more complex than the creation and thus breaking some known or unknown rule.

*sigh*

I can't believe I'm explaining the creationist argument to a religious person :p I get the feeling I'd be better off just agreeing with you: "Yep, you're right, 'God' is real simple. He wears a safety helmet and rides the short-bus to school".

But anyway, based on what you already stated I'll respond with this: the human brain is very complex but speaking i

Re:

[jd]

So I gather you disagree, from your semi-intelligible response? Did we (collective humanity) or did we not create the computer you are reading this on? Did we not bring about the technological advances that are stated in the very article you are replying (hardly) to? If so, then are we not the ultimate computer? Equipped w/ the highest resolution video, audio, CPU/logic, etc?

Alan Turing was rather abhored by the collective humanity at the time, due to his orientation, so I'm not sure collective humanity h

Wetware

[Rassleholic]

The one I find most facinating is MONIAC [wikipedia.org]. A cookie to whoever gets it to run linux.

No Conway's Life?

[pclminion]

Conway's Life is Turing complete. I guess, to a computer scientist, it's not really surprising that an automaton could be Turing complete, but it's still pretty damn awesome to think that little cells replicating on the screen are capable of carrying out any arbitrary computation -- as well as self-reproduction.

I wonder, with a large enough simulation, if self-reproducing, intelligent entities could evolve out of just a few simple rules (and it's really only one rule, if you code it a certain way).

Re:

[frovingslosh]

The article missed a lot, but certainly a serious candidate would be the Wireworld Computer, a cellular automaton computer that actually (slowly) computes prime numbers and displays them, done by implementing a digital computer as a cellular automaton. This is an amazing computer, only one op code, and you can watch the data as it flows through the computer, including the stack of 64 registers (a few unused in this program).

Re:No Conway's Life?

[frovingslosh]

The article missed a lot, but certainly a serious candidate would be the Wireworld Computer [quinapalus.com], a cellular automaton computer that actually (slowly) computes prime numbers and displays them, done by implementing a digital computer as a cellular automaton. This is an amazing computer, only one op code, and you can watch the data as it flows through the computer, including the stack of 64 registers (a few unused in this program).

Sorry, due to a typo the link was lost in the previous post.

Why are these weird?

[The Ancients]

weird |wÉÉ(TM)d|

adjective

suggesting something supernatural; uncanny : the weird crying of a seal.

â informal very strange; bizarre : a weird coincidence | all sorts of weird and wonderful characters.

I don't really see them as 'weird' as such - different, and fascinating, and many seem to point a potential way forward for computing. I don't see why we should refer to technology moving forward as 'weird'.

Re:

[Itninja]

Yeah. The word weird comes from the olde tyme term 'wayward', meaning something unearthly or unnatural. These computers are maybe unusual, but certainly not weird.

Re:

[tirerim]

I don't think that's right. 'Weird' comes from Old English 'wyrd', meaning 'destiny', or more specifically the power to control destiny, as in 'the weird sisters'. That grew into the 'unearthly' or 'unnatural' meanings, which then evolved to the modern meanings, but I don't think it ever had a form of 'wayward'.

Regardless, though, the word's etymology doens't have much bearing on what it means today, which certainly includes just 'strange' or 'unusual'. There are plenty of more extreme examples out there

Re:Why are these weird?

[value_added]

weird |wÉÉ(TM)d|

Weird is trademarked?

I'm in trouble.

Re:

[Dr. Cody]

Don't try to out-weird me, three-eyes...

Pneumatic computer

[Thelasko]

When I worked in manufacturing I would occasionally rig up some logic circuits using a series of pneumatic valves. If only a few conditions had to be met (like don't open door if bucket raised) it was cheaper and easier than installing a PLC.

Personal favourites

[ozamosi]

My personal favorites are computers built in Game of Life [rendell-attic.org] and a model railroad [rendell-attic.org].

Re:

[ozamosi]

Crap!

The railroad link is supposed to be http://www.monochrom.at/turingtrainterminal/abstract_eng.htm [monochrom.at]

K'nex Computing.

[KinkoBlast]

You know, someone on Youtube showed off logic gates in K'nex. But it was only and, or, and not.

Has anyone figured out how to do an xor in k'nex without horrible permutations along the lines of (in scheme, since it's easy for me to think in today)

(define (xor a b) (and (not (and a b)) (or a b)))

?

Re:

[Actually, I do RTFA]

NAND and/or NOR gates

Or XNOR...

Re:

[Ceriel Nosforit]

All of logic is compromised of AND and OR alone. You use these to build NOT. Or you could use NOT and OR to build AND, or NOT and AND to build or.

Really, AND and OR can be described with truth tables which you can take to be arbitrarily defined: We just design them that way because it allows us to do a whole bunch of other things.

Add quantifiers to this and and you get all of math. ALL of it. Basically, the sum of human knowledge, and potential for knowledge, can be summed up as the truth tables for AND an

Re:

[Actually, I do RTFA]

All of logic is compromised of AND and OR alone.

Well, the point is that, yes, using only AND and NOT gates you can build anything. Using only OR and NOT gates you can build anything. You cannot use AND and OR gates to build anything, because you cannot build a NOT gate. However, given only NAND gates or only NOR gate, or only XNOR gates, you can build anything.

Because it is easier to create one type of gate, most chip manufacturers create either NAND or NOR gate chips, and wire them up however (XNOR is

The one at Unseen University

[TheWoozle]

Hex [wikipedia.org]

Some better examples

[Animats]

Some better examples:

• The Great Brass Brain [jhu.edu], an analog tide predictor. It was built in 1910, and used until 1966, for regular tide predictions.
• The Bay Model [army.mil], a working 1.5 acre model of water flow in San Francisco Bay. Built in 1956, in use until 2000. (You can still visit, but it's not used as a research tool any more.)
• SCEPTRON [aip.org], a mechanical filter bank of quartz fibres which could record and play spectra onto photographic film. This was trainable as a speech recognition system. Early 1960s.
• The Iconarama. [ed-thelen.org], the USAF's Etch-A-Sketch. This was one of the first large screen displays, basically a plotter/slide projector combo. It could write, but not erase selectively, so units were used in pairs, allowing a redraw by the unit not projecting, then a lamp switch. 1950s.

Re:

[Intron]

About 40 years ago one of the railroads built an analog computer for controlling switches and retarders in a gravity classification yard (hump yard). It used ball bearings rolling down tracks to model the rolling cars. Google didn't find anything online about it, so I don't have a link.

The Game of Life is Turing Complete

[still_sick]

http://en.wikipedia.org/wiki/Conway's_Game_of_Life [wikipedia.org]

It is possible for gliders to interact with other objects in interesting ways. For example, if two gliders are shot at a block in just the right way, the block will move closer to the source of the gliders. If three gliders are shot in just the right way, the block will move farther away. This "sliding block memory" can be used to simulate a counter. It is possible to construct logic gates such as AND, OR and NOT using gliders. It is possible to build a pattern that acts like a finite state machine connected to two counters. This has the same computational power as a universal Turing machine, so the Game of Life is as powerful as any computer with unlimited memory: it is Turing complete. Furthermore, a pattern can contain a collection of guns that combine to construct new objects, including copies of the original pattern. A "universal constructor" can be built which contains a Turing complete computer, and which can build many types of complex objects, including more copies of itself.[4]

homebrew purely optical computer

[smellsofbikes]

They're all impressed by using waves for building logic circuits.
Want to build your own cheap, brilliantly visual set of logic gates to show kids how digital computing works? Nightlights. Each one is a NOT gate. You put two close to a third's sensor and you have a NOR. Put them some distance away with some blocking material around them (this is fussy) and you can get a NAND. A little bit of thinking and combinatorial logic and you can build anything else from those. I've built stacked, carrying half-adders this way, and it's pretty cool to watch small binary numbers get added.
Two nightlights, each with its bulb by the other's sensor, are a flip-flop. Now you have memory.
For extra credit, you can build a ring oscillator by putting an odd number of nightlights in a ring, so each is seeing the next one's sensor, and use that to clock your half-adders and flipflops.
If I had a lot of money and time, it'd be fun to see how far this could be extended (before I had to start hiring kids as tube runners to keep the whole works going.)

Re:

[harry666t]

Argh, damn mod points, never around when you need 'em. Imaginary +1, Interesting from me, sir.

Re:

[smellsofbikes]

Well, don't mod me interesting -- go build one! Seriously. Nightlights cost like 2/$4 at BigLots! and similar places. The main problem is hooking them all up without potentially electrocuting yourself or going bankrupt buying cheap extension cords. (and for the ring oscillator use the same type of nightlights throughout because if you have different types, they'll often have different delays and sometimes it'll screw up the oscillation and you'll lose pulses.)

Re:

[smellsofbikes]

It's *really* messy. I've done a couple with extension cords, which is the safest way, and I've done a couple with just hookup wire run through the holes in the prongs and wrapped around, when I was in a hurry. The nicest setup I've seen, a guy had done a big random set of nightlights each one taped to a piece of wood, and all those nailed down to a plywood bed, using extension cords. It was just a big mass of them with chaotic patterns flickering across it, but it got me to thinking about how I could tu

Weirdest storage.

[argent]

OK, let's go back a ways and look at the weirdest storage systems.

Mercury delay lines are a good one. Delay lines in general, actually. I recall readong once about a free-space delay line using a laser beam between Earth and a retroreflector on the moon.

CRT storage tubes are another.

What about stochastic computers ? Robust, cheap...

[franois-do]

Stochastic computers represented any value between 0 and 1 (both included) by a probability. A set of random bits were sent according to that probability.

Multiplication, always a problem with analog computers at the time, was very simply, quickly and cheaply done by an AND chip (one of the inputs had to be decorrelated of the other by a delay line to avoid parasitic correlations). The addition was a little more tricky, but getting (p1+p2)/2 could be achived with just three basic circuits, if I remember well. Of course you had to remember that the value was scaled, well, exactly the same king of caution you had to observe with analog synthetizers at the very same time.

Details here for whoever is interested... and knows somebody reading French ;-)

http://fr.wikipedia.org/wiki/Calculateur_stochastique [wikipedia.org] The complexity of keeping trace of scaling, decorrelations and the like could be taken away by monitoring them from an associated PC, now that I am thinking about it. Try it ! You will like it ;-)

Puzzle computers

[Bob Hearn]

Conway's Life was mentioned, but that is still a deterministic computer.

Many puzzles have been shown to effectively be nondeterministic computers. E.g., you can make a sliding-block puzzle that is solvable if and only if a given traditional computation succeeds.

Science News story:

http://www.sciencenews.org/articles/20020817/bob10.asp [sciencenews.org]

Personal plug:

Games, Puzzles, and Computation [mit.edu]

They mentioned water computer but...

[Yold]

they didn't mention MONIAC, which is the coolest analogue computers IMHO. http://en.wikipedia.org/wiki/MONIAC_Computer [wikipedia.org]

ah Man, is that what pratchet was talking about

[geekoid]

in Making Money? Gee, I guess I'll 'have' to reread it, shucks~

How about an old one?

[thatskinnyguy]

What about the Antikythera mechanism [wikipedia.org]?

More Weirdness

[Purity Of Essence]

In A.K.Dewdney's Scientific American column (and subsequent books) he documents many unusual mechanical computing devices that solve a range of computationally expensive problems. In a chaptered entitled Analog Gadgets in the book The Armchair Universe he describes several mechanical computing devices that solve a number of many computationally expensive problems (with some caveats):

* a spaghetti powered sorting machine
* computing a convex hull using a board, nails and a rubber band
* finding the shortest path joining two nodes of a graph network using brass rings and string
* finding the minimum Steiner-tree for any number of nodes using pegs sandwiched between parallel sheets of plastic dipped in a soup solution
* a prime calculator using a pair of lasers and parallel mirrors

In the next chapter, Gadgets Revisited, he presents:

* a way to compute the best-fit trend of a graph using a board, nails, rubber bands, and a rod
* finding the longest path through a network of nodes using segments of string knotted together
* computing the forth power of a number based on the principle of elasticity and the deflection of a bar of aluminum
* or the third power of a number by using the same principle applied to a weight placed on the bar
* light refraction computed with soap film suspended between stepped surfaces
* optimal position for a refinery using a board with holes, string, a brass ring, and weights proportional to the cost of transportation for each source of raw material
* number averaging using interconnected graduated glass cylinders
* cubic polynomial solver using a water tank, a balance beam, two scalepans, and a variety of solids to represent terms of the equation: a cone for x, a paraboloid for x and cylinder for cx, and a sphere for d

Re:

[Purity Of Essence]

I hit Submit accidentally.

That should have read:

cubic polynomial solver using a water tank, a balance beam, two scalepans, and a variety of solids to represent terms of the equation: a cone for x^3, a paraboloid for x^2 and cylinder for cx, and a sphere for d

In the Tinkertoy Computer, Dewdney covers the well known Tic-Tac-Toe playing Tinkertoy computer built by MIT, as well as a fanciful computer based on ropes and pulleys featuring an inverter, an OR gate, an AND gate, a multiplexer, a flip-flop, and an ad

Marble adding machine

[tvelocity]

This article reminds me of a very interesting video on youtube about a marble adding machine [youtube.com]. It is constructed out of wood, and the creator also has made a video explaining how it works [youtube.com], in case anyone would like to build one on his own.

ripples!

[dwater]

For those half blind speed readers among you, it's ripples in a tank of water...

Domino Digital Logic

[jone_stone]

This article makes me think, of course, of my experiments in domino digital logic [pinkandaint.com]

Why does everyone screw this up?

[CrazedWalrus]

That would make possible things that are unfeasible with today's computers â" such as rapidly factoring large prime numbers to crack cryptographic keys.

Thanks Bill Gates. That really would be a neat trick.

ions in an electric field

[Jeff1946]

Ion paths in an electric field were determined by streching a rubber sheet between walls whose edge heights were proportional the electric voltage on those edges. The rubber sheet would obey Laplace's eqn just like electric fields do. If you roll balls down the rubber sheet they will follow the similar paths to ions in the electric field. Conducting solutions can also be used to for a similar purpose for systems that only vary in two dimentions, like a cylindrical lens. Here a small electric probe can b

XOR gate

[Jurily]

used wave patterns to make a type of logic gate called an "exclusive OR gate", or XOR gate."

Why the explanation? Are /. readers braindead nowadays? What kind of "news for nerds" needs an explanation of what a XOR is?

No dominoes?

[MrNougat]

No domino computer? Really?

http://www.pinkandaint.com/oldhome/comp/dominoes/ [pinkandaint.com]
http://en.wikipedia.org/wiki/Domino_computer [wikipedia.org]

Slime mold?

[ibbie]

Now I feel bad for eating all of those poor things in Nethack.

Re:

[good soldier svejk]

You ate Lord Running Clam! [philipkdickfans.com]

Mmmm rather lame article...

[Lisandro]

These are all (interesting) variation of basic logic gates implemented without electronic components. I was expecting something in the vein of the transputer [wikipedia.org]...

eMachines?

[morari]

Oh wait, those aren't weird so much as lame... and probably can't actually be counted as computers, per say.

Not useful

[ajs318]

An exclusive-OR gate is not useful by itself, because there are an equal number of ways of getting a one out of it as a zero.

Any logic function can be built up with just NAND, just NOR, or NOT and either (AND or OR). There's also an odd logic function, BUN (= BUt Not; output is 1 when A=1 and B=0, 0 otherwise) which is sufficient (you can make it into NOT by tying A to logic 1, AND by inverting B or NOR by inverting A). These properties, though, depend on asymmetry in the truth table -- and the EOR fu

Nominating #11 - Ice Ice Baby

[DynaSoar]

Protonic processing using doped ice.

In water ice, it is protons, not electrons, that move under voltage. Use pipes instead of wires, fill them with water and freeze. For gates etc. dope the water to give it differential response to voltage.

As with some of the others, there's no good reason to do this other than its neatism.

Re:

[NonSequor]

I wasn't impressed with that one either. You could do it by setting up two ripple sources so that the wave peaks will be out of phase when they reach a certain point. If both or neither ripple source is turned on, the water will be calm at that point, but if only one is turned on the camera will see ripples at the point.