Enigma Machine

[fibonatic]

I have wanted to build a Enigma machine in Minecraft for quite some time. At first I didn't know how I could do it, I had some ideas which involved huge piston tapes which would function as the rotors.
However not so long ago I got a much simpler solution, which involved binary representation of each letter and adding/subtracting to imitate the rotation of the rotors.
I recently also got a little bit into programming, so I made a simple emulator of this in JavaScript.
Currently I have a schematically setup in my head, however I still need to figure out a fast way of decoding/encoding 275 different outputs: 25 for each rotor (5 total) since the 26th letter would be zero (00000), 25 for the reflector and after that the signal would pass by the rotors again but in reversed order, so 25 * (5 + 1 + 5) = 275.

[icecoldjazz]

Take it a step further and introduce a mechanic that allows a letter to be encoded as itself. When encoding A, for example, A could never become A, it would always end up as B-Z. This was the fatal flaw that allowed the German code to be cracked at Bletchley Park.

[icecoldjazz]

As far as I'm aware, Turing cracked the code after he realized that a letter never became itself, even in the military version of the machine. The switchboard didn't make a difference apparently.

[fibonatic]

This is an informative video from Numberphile about the Enigma machine.
It is indeed correct that an encrypted letter can't be itself. However to use this to crack the code of one message (with no standard format) is very hard. Only if you know that certain words would appear in the message, it would be a lot easier to decode it.
But the mean reason why I want to build this is because of the aesthetics.

[newomaster]

Can't wait to see this when it's done
I don't think one has ever been made before

[Guy1234567890]

Yeah, very cool project idea!

[jmal116]

I thought of a similar idea (after watching the numberphile video), but my main problem is that I don't know what the rotors do the letters to code it.

[Thor23]

The rotors work like the switchboard on the front, excepting of course that the connections never change between each pair of letters. The signal goes into a rotor at one letter and comes out the other side at another. There are probably schematics to show how the letters in each rotor are connected, but you should be able to just wire them up randomly and achieve the same result.

[fibonatic]

There is a little bit more to it than what Thor23 said. It is correct that each rotor has some internal wiring which connects each of the 26 inputs to one of the 26 outputs. However the rotor can be rotated which would mean that each letter corresponding with a wire will be shifted both shifted on the input and on the output. After doing this three times (usually an Enigma machine holds 3 rotors) the signal will pass through a reflector which is symmetric, so if input A would output G, than input G would output A (this is needed to be able to decode a message with the same settings as the encoded settings). After this the signal is passed through the rotors again but in reversed order and direction.
And for actual wiring of the rotors and reflector I am using this source.
However I am not sure if I am implementing this correctly since I haven't been able to encode the message "AAAA" to "BDZGO"

[Thor23]

I don't think the encoding is a mistake on your part, I'm not getting it either. This page appears to have the same setup, and it encodes the first A as a U, so I don't think that wikipedia page is correct. Overall for the 5 A's I got UCFCK as the encrypted form, at least as far as the rotors are concerned.

[fibonatic]

One thing I have noticed which is different about my emulator is that I turn the rotors after I have pressed a letter, while in the actual machine the rotors first turn and then let a signal pass through all the wires.

Thanks Thor23, I have tested a few letters and they all returned the right letter, so it seems to work correctly. However for the 5 A's I got UBDZG (using rotor I, II and III, with I closest to the reflector and start position AAA).

By the way I also tried to find the settings which where used in the Numberphile video (since the start positions where given), however their machine also have a plug board.

[Nickster258]

That will be cooler than cool. I still think the name is more fun than the concept.

[Thor23]

Huh. I wonder, then, what sort of failure of understanding on my part is then preventing me from getting the correct answer. I don't suppose you could post the transformations for each letter as it goes through the rotors so I can see where I'm going wrong?

[Chibill]

This java program code may help you with this http://www.dreamincode.net/forums/topic/...a-machine/

[Xeomorpher]

>Java< I'm alright mate. <3

[Dornier]

Here is a good simulator: http://startpad.googlecode.com/hg/labs/j...a-sim.html

I have an enigma keyboard with plugboard, the rotor and ring controller and a base 26 adder on my RDF plot. 5 decoders have to be built, that's all.

[Klakkaros]

Hey guys Just seen the Enigma Machine discussion - I've been working one for a while and finally finished it the other day, please check it out!

http://www.youtube.com/watch?v=Isl1CfQUtWs

[Badidzetai]

This creation is Sooo badass. You're quite good at redstoning, man.
Anyway you'd rather classify this topic in Done section, no?