Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

The Turbo Entabulator: A 3D-printed Mechanical Computer

samzenpus posted about a year ago | from the because-you-can dept.

Technology 83

An anonymous reader writes "Have you ever been sitting there, quietly computing something and thinking to yourself, 'If only this process were somehow billions of times slower, less reliable, and involved lots of physical labor?' If so, the Turbo Entabulator is the machine you've been looking for! It's a (nearly-entirely) 3D-printed mechanical computer. With three single-digit counters for memory, it's driven by a hand-cranked, Jacquard-style punch card reader. You can even download the files and build your own."

Sorry! There are no comments related to the filter you selected.

Oh sure, but... (5, Funny)

millia (35740) | about a year ago | (#43964207)

it's not a turbo-encabulator.I've been wanting one of those for a long time.

Re:Oh sure, but... (0)

Anonymous Coward | about a year ago | (#43964283)

Yeah, but does it synchronize cardinal grameters?

Re:Oh sure, but... (1)

millia (35740) | about a year ago | (#43964333)

Yeah, I don't know how else you control side-fumbling.

Re:Oh sure, but... (0)

Anonymous Coward | about a year ago | (#43966953)

I take it the two sperthing bearings have to be in a direct line to the panametric fam? Last I heard, it had something to do with a logarithmic casing design.

Re:Oh sure, but... (2)

AliasMarlowe (1042386) | about a year ago | (#43964329)

I'm waiting for the print-it-yourself turbo Interociter! [wikipedia.org]
It will put all non-turbo interociters to shame...

Re:Oh sure, but... (1)

Medievalist (16032) | about a year ago | (#43966035)

Yeah, me too. I need one to fix my chronosynclastic infandibulator; the space-time interociter I got from Tom Servo is totally shot and every time I try to send a .jpg of the turbo subsystem in to metalunan support it just shows black fog. Cheap Gizmonic crap!

Re:Oh sure, but... (1)

sanman2 (928866) | about a year ago | (#43968107)

Dammit, I was hoping that somebody could print a simple Babbage machine

Yes. (4, Funny)

QilessQi (2044624) | about a year ago | (#43964225)

"Have you ever been sitting there, quietly computing something and thinking to yourself, 'If only this process were somehow billions of times slower, less reliable, and involved lots of physical labor?'"

Yes. And then I switch to a Windows box. Mission accomplished.

Re:Yes. (5, Insightful)

Mitchell314 (1576581) | about a year ago | (#43964417)

Every programmer at some point in their life wonders if they can make a computer out of a given thing in front of them.

Which given how many times computers and cpus have been made in minecraft and dwarf fortress, explains a lot.

Re:Yes. (1)

CastrTroy (595695) | about a year ago | (#43964831)

And by those standards, this project is even less impressive. I don't know if the creators of MineCraft ever intended for people to be able to create computers inside the game, but my guess would be no. A 3D printer on the other hand is designed to make custom mechanical parts. Building a mechanical computer using a 3D printer doesn't really required much of a leap from one step to the next. Take an existing design, and print out the parts. However, in a game like MineCraft, It would be quite a feat even figuring out what had to be done based on the limited number of blocks available, and you really couldn't even be sure it was possible until after you completed the design.

Re:Yes. (0)

Anonymous Coward | about a year ago | (#43965161)

And by those standards, this project is even less impressive. I don't know if the creators of MineCraft ever intended for people to be able to create computers inside the game, but my guess would be no. A 3D printer on the other hand is designed to make custom mechanical parts. Building a mechanical computer using a 3D printer doesn't really required much of a leap from one step to the next. Take an existing design, and print out the parts. However, in a game like MineCraft, It would be quite a feat even figuring out what had to be done based on the limited number of blocks available, and you really couldn't even be sure it was possible until after you completed the design.

Well, the creators definitely intended for people to make logic circuits, because that was the whole point of redstone torches (they invert the incoming signal). For anyone who knows how CPUs are made, it's a pretty simple connection in theory between logic gates and CPUs. The hard part is really just having the patience to construct the damn thing (mitigated somewhat if you use a world editor).

Re:Yes. (1)

Anonymous Coward | about a year ago | (#43966787)

... whilst playing in HardCore mode. Damn creepers keep exploding in the middle of the Level 2 cache! Half logic, half monster trap...

Re:Yes. (0)

Anonymous Coward | about a year ago | (#43965373)

Building a mechanical computer using a 3D printer doesn't really required much of a leap from one step to the next.

Let's see yours then.

Re:Yes. (-1)

Anonymous Coward | about a year ago | (#43964759)

This comment could have been from 1997 and is still funny. Tells you something about the state of Windows today.

Re:Yes. (2, Insightful)

Anonymous Coward | about a year ago | (#43964879)

No, it tells you that neckbeards are trapped in 1997.

Re:Yes. (1)

marcosdumay (620877) | about a year ago | (#43968949)

Old times... Windows had upgrades that actualy improved it by then.

Re:Yes. (1)

QilessQi (2044624) | about a year ago | (#43977597)

Oh, relax. I was going for the +1 Funny, and Windows is an easy punchline on Slashdot. :-)

Re:Yes. (1)

marcosdumay (620877) | about a year ago | (#43981337)

No problem, I got your joke.

I just can't see a troll like that AC starve :) It hurts my feelings!

Re:Yes. (2)

NoNonAlphaCharsHere (2201864) | about a year ago | (#43965187)

"Have you ever been sitting there, quietly computing something and thinking to yourself, 'If only this process were somehow billions of times slower, less reliable, and involved lots of physical labor?'"

No. I swore off on Perl years ago.

Re:Yes. (0)

Anonymous Coward | about a year ago | (#43967489)

Do you like it when men take shits on your face? It sounds like it.

Good for teaching ... (4, Informative)

gstoddart (321705) | about a year ago | (#43964231)

So nobody is likely to use this for actual work, but as a teaching aid, it definitely goes a long way. Explaining with a working physical device the principles of basic computing and Turing machine type things is pretty cool

And, if someone has done this, it's only a matter of time before we start getting some super awesome 3d printed Rube Goldberg type or Steampunk-type of devices.

Re:Good for teaching ... (4, Informative)

plover (150551) | about a year ago | (#43964625)

I was amused that he built it because his 3D printed Jacquard Loom was even less reliable.

3D plastic extrusion printing is fine for printing a pencil cup or a replacement game token, but a precision manufacturing process it is not. There's a reason machined parts have tight tolerances: without them, moving parts bind, jam, and break.

Re:Good for teaching ... (1)

spiritplumber (1944222) | about a year ago | (#43968175)

As the proud owner of a 3D printer whose parts I am slowly replacing with homemade ones (as a way to learn, mostly) I have to disagree. 3D printed parts do end up being a bit melty, but the difference between printing a gear and sanding it to tolerance, and carving one from scratch is huge.

Re:Good for teaching ... (1)

dj245 (732906) | about a year ago | (#43971203)

As the proud owner of a 3D printer whose parts I am slowly replacing with homemade ones (as a way to learn, mostly) I have to disagree. 3D printed parts do end up being a bit melty, but the difference between printing a gear and sanding it to tolerance, and carving one from scratch is huge.

I honestly can't figure out why you would do that when you could laser-cut, water-jet cut, or CNC* some wood or steel and be done in 1 step. If you're going to all the trouble of having an electronically-positioned tool, why not just hook it up to an electric motor and buy some milling cutters? Water-jet and laser cutting is outside the realm of most hobbyists, but homemade CNC machines are probably simpler than a 3d printer. You don't have to worry about adhesion, bead size, etc. I guess if you have a hammer everything looks like a nail.

*technically you should cut gears on a hobbing machine, but for hobbyist purposes a CNC should be fine.

Re:Good for teaching ... (1)

spiritplumber (1944222) | about a year ago | (#43977609)

For fun. I already have a CNC mill and I can easily borrow a laser cutter if I need to. I also understand that most people don't and want to come up with stuff for them.

Re:Good for teaching ... (1)

plover (150551) | about a year ago | (#44001803)

[ Disclaimer: I spent several years working in a precision grinding shop, grinding machined parts to very tight tolerances, measuring them with carefully calibrated tools, and achieving accuracy that lathes simply are not capable of achieving. So what I consider acceptable accuracy and what you consider accurate will likely vary more than slightly. ]

A 3D extrusion printer is not exactly a precision tool. There is a limit to the precision of the parts it can print. That's neither good nor bad, it's simply a fact of the design of the machine and the tolerance of the parts that it was built from, as well as the plastic extrusion process itself.

I fully understand that you can build a device with 3D printed parts, and it will function. But those parts are not precise, and devices built that rely on them will not operate with precision or accuracy. For most of the parts you're likely printing, that's not a requirement. For example, the parts that join rods together on a RepRap don't have to be accurate or precise, they just need to grip the rods securely. The person doing the assembly will align the rods accurately.

However, there are components in the movement chain where precision is important. If the 3D printed gear feeding plastic into the extrusion nozzle is not perfectly round, it can cause minute alterations to the feed speed of the filament, producing small differences in the thickness of the extruded plastic as a result. Building layer upon layer of the output will probably average out the differences rendering them invisible for most projects, but if you were to print a square wall with a perimeter that was a multiple of the pitch circumference of the extruder gear, you might get a wavy or bumpy wall as a result.

As more imprecise parts are added to the chain of parts controlling the motion of the printer, they will have an additive effect they on the accuracy of the printed results. I'm not saying you have a bad or crappy machine. Whether or not the printer is producing parts that meet your tolerance needs is up to you. Geckos, figurines, and weighted companion cubes don't have to be precise.

And yes, you can print an oversized part and grind it down to meet a tight tolerance, but that's not the result of a precision printer, that's the result of the process of a careful craftsman.

Re:Good for teaching ... (3, Interesting)

Bob the Super Hamste (1152367) | about a year ago | (#43964779)

The same can be said for any number of things. There is something to actually being able to see and touch something and being able to figure out how it works. This is why I spent the weekend with my 4 year old building a working automatic transmission with planetary and ring gears out of Legos. He became curious about how one worked after seeing a book about Lego machines at the library and made the connection between Lego gears shown in it and gears in a vehicle transmission. So he wanted me to build him one and show him how it works. Granted it doesn't have clutch packs (components are locked using shafts), torque converter, or a valve body (you move the shafts to lock components manually), but it does show how the gearing works. It has 3 forward speeds, neutral, park, and reverse, just like old standard 3 speed automatics did.

Re:Good for teaching ... (0)

Anonymous Coward | about a year ago | (#43968677)

Link to pics please, as a former lego enthusiast (way back when I was a kid and "expert builder" just came out). I would love to see this.

Clock of the Long Now (0)

Anonymous Coward | about a year ago | (#43965279)

In case anyone reading this hasn't heard of it, the 10,000 year clock being built by the Long Now Foundation will contain the coolest mechanical computer that I've heard of.

10,000 year clock [longnow.org]

Very slow computer. But unlike the one in the article, very reliable. And it was designed by Danny Hillis, who has designed supercomputers.

While I love cool things like the Turbo Entabulator, it is cool beyond words that that the computer in the 10,000 year clock has a serious purpose for being designed the way that it is.

people asking why bother (5, Insightful)

Guano_Jim (157555) | about a year ago | (#43964233)

We do these things not because they are easy, but because they are hard.

Hats off to the designer.

Re:people asking why bother (0)

Anonymous Coward | about a year ago | (#43964557)

The most /. worthy article in a long time!

Re:people asking why bother (1)

CastrTroy (595695) | about a year ago | (#43964777)

But this doesn't really seem particularly hard, just time consuming. People build mechanical computers all the time. Using a 3D printer to make the parts is a very obvious way of doing it. While I respect the person who built it just to further their own learning and to have a little fun, I would have to say that it doesn't really impress me, and isn't really that newsworthy.

Re:people asking why bother (4, Insightful)

lgw (121541) | about a year ago | (#43965057)

You can save some confusion and annoyance by just accepting that "on a 3D printer" is the latest "everything old is new again", just like "in the cloud", "on a handheld", "on the internet" and "on a computer" each were in turn. For each new engineering platform all the obvious stuff will be done for the first time on that platform, usually with some fairly minor cleverness involved. Easier just to accept the cycle of faux-new than to try to fight it.

Re:people asking why bother (1)

sensei moreh (868829) | about a year ago | (#43965615)

Kind of like writing HelloWorld! in umpeen different programming languages

Asimov's 4th Law (0)

Anonymous Coward | about a year ago | (#43964239)

A robot may not create another robot.

Quickly! (5, Funny)

fuzzyfuzzyfungus (1223518) | about a year ago | (#43964241)

Fetch my sturdiest manservant and the overclocking whip!

Re:Quickly! (4, Funny)

sconeu (64226) | about a year ago | (#43964287)

Yes sir, Mr. Burns. That would be Homer Simpson, sir.

Re:Quickly! (1)

oodaloop (1229816) | about a year ago | (#43964787)

Who?

Re:Quickly! (1)

digitrev (989335) | about a year ago | (#43965019)

One of your slobs in Sector 7G.

Re:Quickly! (2)

gooman (709147) | about a year ago | (#43965121)

"One of the mouth-breathers from Sector 7-G"

I guess it fails Turing even if prog'd correcly. (4, Funny)

Impy the Impiuos Imp (442658) | about a year ago | (#43964269)

Can you please not imagine a Beowulf cluster of these?

Re:I guess it fails Turing even if prog'd correcly (1)

EmagGeek (574360) | about a year ago | (#43964537)

Too late...

Re:I guess it fails Turing even if prog'd correcly (1)

loufoque (1400831) | about a year ago | (#43965795)

It doesn't appear to have memory, just 3 registers.

Holy Balls (0)

Anonymous Coward | about a year ago | (#43964285)

This line, "The book made it crystal-clear why people were maimed all the time in industrial accidents in the 1800s" is absolutely freaking hilarious!

Another step (2)

the eric conspiracy (20178) | about a year ago | (#43964289)

..towards the singularity.

Terrorism! Murka! 9/11! (3, Funny)

EmagGeek (574360) | about a year ago | (#43964413)

We have to put a stop to this dangerous movement of self-creation and innovation that is not under the protective regulation of government. We need common sense laws immediately that require all 3D printers to be registered, and licensure for their operators.

We cannot allow this threat to our national security to continue.

This Public Service Announcement brought to you by the Republican-Democrat Partnership Conference in association with The Foundation for Peace through Unity and Faith in Government.

Re:Terrorism! Murka! 9/11! (0)

RightSaidFred99 (874576) | about a year ago | (#43964899)

God, put a fucking sock in it. Anyone who uses any variation of 'Murka' should just be horsewhipped.

Well I can deduce. (1)

Virtucon (127420) | about a year ago | (#43964431)

1) No kids.. Dad has way too much time on his hands.
2) Wife is a Librarian and makes a good clamp when needed.
3) was bored one day and decided to make something with that $3000 printer the wife got him at XMAS.

This is cool but man are we going to be inundated with every novelty item that is 3d printed now? This is neat but the Lego Turing machine was IMO cooler [vimeo.com]

it's 3D printers all the way down (1)

Thud457 (234763) | about a year ago | (#43964553)

steam powered ^W^W via radio waves ^W^W^W with a computer ^W^W^W on the internet ^W^W^W 3D printed

Re:Well I can deduce. (1)

Bob the Super Hamste (1152367) | about a year ago | (#43964841)

Dammit now there is something else I will have to build out of Legos if my oldest son ever see it. I can't see the video (blocked at work) but if it something like the Babbage's Analytical Engine I fear for my life.

Re:Well I can deduce. (1)

Virtucon (127420) | about a year ago | (#43965063)

Worse yet, you'll need mindstorms

Re:Well I can deduce. (0)

Anonymous Coward | about a year ago | (#43965863)

The same guy also implemented a Cray 1 CPU in an FPGA, among other things. Unlike most [SD]INKs he's taking full advantage of his copious free time to do awesome.

Maybe this was what Linus was talking about (2)

intermodal (534361) | about a year ago | (#43964527)

when referring to the new kernel release which will involve more profanity use.

But can you imagine... (0)

Anonymous Coward | about a year ago | (#43964617)

A Beowulf cluster of these things?

Fp% g8aa (-1)

Anonymous Coward | about a year ago | (#43964659)

a child knows this mist&ake or

Does it use unilateral phase detractors? (1)

Alejux (2800513) | about a year ago | (#43964685)

to automatically synchronize the cardinal grammeters?

Jacquard loom (1)

bhlowe (1803290) | about a year ago | (#43964727)

Hey, semi-related.. If you type Jacquard loom [audible.com] into the search bar on audible.com (through 6/12) you can get a free copy of Jullian Assange Cypherpunks book.

Now make a water powered computer (0)

Anonymous Coward | about a year ago | (#43964737)

I'd quite like to see a water powered computer. (obviously the water would be inserted afterwards)

Re:Now make a water powered computer (2)

Mikkeles (698461) | about a year ago | (#43964901)

Do a web search for fluidics [google.com] .

Re:Now make a water powered computer (1)

Anne Thwacks (531696) | about a year ago | (#43964937)

If you will pay for it, I will be happy to build you a fluidic computer that uses water as its computational medium and executes the PDP8s instruction set (PDP8E if you cna afford it). I think a clock speed of 50-100 Hz is probably feasible for a Mk1, and, with considerably greater funds, several kHz might be possible eventually. (1 MHz is not impossible, theoretically - faster than a real PDP8s, but it is definitely beyond my skill level).

I can make it steam powered for about double the cost.

The MONIAC computer (0)

Anonymous Coward | about a year ago | (#43967223)

It was all done a long time ago....

The MONIAC [wikipedia.org] was an analogue computer that used "fluidics" instead of elecktrikery. A similar computer is discussed in the Discworld novel "Making Money" [wikipedia.org] , where it passes under the title of "The Glooper", for obvious reasons...

Great (0)

Anonymous Coward | about a year ago | (#43964893)

Now all we need is a jacquard-style 3d printer and the circle is complete

Mechanical calculators (4, Informative)

Clueless Moron (548336) | about a year ago | (#43964931)

I have one from around 1958, and it's quite satisfying to use. [photobucket.com]

They can multiply and divide as well as adding/subtracting. The above link shows the result of doing 355/113: 3.1415929 with a remainder of 23.

The top left is an accumulator, the top right is a counter, and the lower register is the number you want to add/subtract (entry register). So to do 355/113, the procedure is

  1. Pull all three 3 metal tabs on the sides to clear all registers
  2. Enter 355, press the rightmost red arrow button to shoot the entry register number all the way to the left
  3. Crank forward once. You now have 3550000000000 in the accumulator and "1" in the counter's leftmost position.
  4. Squeeze the two rightmost chrome handles together to clear both the counter and entry register back to 0
  5. Enter 113, press the rightmost red arrow button to shoot the number all the way to the left. You're done entering numbers at this point.
  6. Crank backwards to subtract from the accumulator until it is less than the entry register (takes three times). Don't worry if you overshoot; a bell will ring to indicate underflow and you just add it back. The counter now shows three in the leftmost position. The red dot indicates that it notes you started off subtracting, so it's counting backward cranks as +1 instead of -1.
  7. Press the right arrow to shift the entry register one position to the right
  8. Repeat the subtracting process, shifting right until you can't go any more right. You're done.

It sounds more complicated than it is, but really it's just long division. It takes about 20-30 seconds to do that division. That sucker works as well as the day it was built. I've looked inside; it's a mechanical marvel.

Oh yeah, those white slider tabs are for placing the decimal points where you want them

Re:Mechanical calculators (1)

lgw (121541) | about a year ago | (#43965123)

That's awesome. How much mechanical force is needed? I'd worry about catching my wrist on the right-hand tabs while cranking, unless the crank was surprisingly easy to turn.

Re:Mechanical calculators (1)

Clueless Moron (548336) | about a year ago | (#43966467)

Very little force is needed, and I've never gotten caught on those tabs. Actually, the force depends on how many numbers have to change: rolling over something like 999999 to 1000000 makes a noticeable difference in resistance. Really I should open it up and lube it.

I'm told you can still find these in remote villages in India and Africa and the like. They don't need electricity and are very reliable.

Re:Mechanical calculators (1)

tippe (1136385) | about a year ago | (#43965221)

I've always wanted a Curta [wikipedia.org] mechanical calculator conceived before WWII by an Austrian (but only mass-produced afterwards, I believe). It does more or less the same thing as the monstrosity that you linked to above (including multiplication and division), all in a hand-sized package. Now that's a mechanical marvel. I'd love to have one...

Re:Mechanical calculators (1)

interval1066 (668936) | about a year ago | (#43965571)

So you're essentially drooling for a slide rule. They can be had on e-bay for various prices.

Re:Mechanical calculators (1)

tippe (1136385) | about a year ago | (#43966183)

Yeah, I suppose it's more or less like a slide rule. Kind of like how a Rolex is essentially the same as a sundial (plus or minus several hundred precisely machined moving parts, assembled with amazing workmanship).

Re:Mechanical calculators (1)

Clueless Moron (548336) | about a year ago | (#43966803)

Slide rules give approximate answers. VERY approximate answers; their only advantage back in the day was that they were fast. These mechanical marvels give exact answers. Considering that when you divide it gives you a remainder that you can use to extend the answer to any arbitrary number of decimal places, they are in fact more accurate than a modern electronic calculator (apart from fancy ones like hp50g)

Anyway, why the negativity? Do you not appreciate well built complex machinery? My example was a Facit calculator, made in Sweden and extremely popular around the world. A similar marvel was the M209 cipher machine. [wikipedia.org] Even includes a printer, yet it fits in your pocket. I'd love to have one.

Re:Mechanical calculators (0)

Anonymous Coward | about a year ago | (#43967751)

With interpolation, you can often get six significant figures accurately with a slide rule...doesn't sound VERY approximate to me...

Re:Mechanical calculators (0)

Anonymous Coward | about a year ago | (#43973017)

A normal sized slide rule is good to 3 digits at best.

Re:Mechanical calculators (0)

Anonymous Coward | about a year ago | (#43965569)

You had a 3-d printer back in 1958? Wow!
Seriously, that is a nice adding machine and cool (thanks for sharing!), but you didn't design it, print the parts up yourself, and assemble it, right?

Re:Mechanical calculators (0)

Anonymous Coward | about a year ago | (#43967361)

Real men use a curta.

Would be useful in Revolution the TV show.. (1)

Danathar (267989) | about a year ago | (#43965311)

No electricity! (yes, the plot is insane) but this device would fit right into that world.

Turbo... (1)

interval1066 (668936) | about a year ago | (#43965409)

... cool. This upgrade is a long time coming. They base version was hell at rendering.

Dr. Doofenshmirtz (0)

Anonymous Coward | about a year ago | (#43965539)

The name of this device *must* be said in a Dr. Doofenshmirtz voice.

Curse you Perry the Platypus! You destroyed my Turbo Entabulator!

So... (1)

Progman3K (515744) | about a year ago | (#43965575)

Can 3D printers print 3D printers yet?

Re:So... (1)

ajlitt (19055) | about a year ago | (#43965893)

Sort of. [tantillus.org]

I have in my posession (1)

kilodelta (843627) | about a year ago | (#43965817)

The drawing for a four bit adder. I also have the components but need to assemble the circuit boards, etc. Be fun to have a computer where you can follow the action again.

Solve two problems at once! (1)

funwithBSD (245349) | about a year ago | (#43966025)

We will make the IRS and the NSA use these exclusively.

Here, have all our data. Let us know if you finish a search before the heat death of the universe.

fep!? (-1)

Anonymous Coward | about a year ago | (#43966747)

tHhe above is far Of OpenBSD. How

Quick question (0)

Anonymous Coward | about a year ago | (#43970547)

Does it Run minecraft?

Self-reproducing (1)

Ronin441 (89631) | about a year ago | (#43971103)

The RepRap project is an initiative to develop a 3D printer that can print most of its own components.

Until now, RepRap have been stymied by an inability to print any of the systems that control the printer. But, no longer! Simply print a mechanical computer to drive your 3D printer, and the goal of a self-reproducing device will be fulfilled!

Might be large, though.

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?