Top 5 Weaknesses of an Enigma | Bletchley Park

I have   a prototype that worked like this. That made it possible to encode letters in a much larger number of ways (an astronomical number of ways) using the board. But the problem was that it was prone to human error. You need 52 sockets, you need twice as many sockets, they are arranged in four rows and you need 26 cables. Each socket needed a plug, so you couldn't leave any plugs out, you had to plug one plug in the top two rows into one in the bottom two rows. If you got confused and connected two top plugs or two bottom plugs together, then the machine would not work.

If you plugged in an odd number of sockets, then the machine didn't work, and all the wires had to be long enough to extend across the socket board, so there were a huge amount of wires and a huge amount of tangled wires.  It must have been an absolute nightmare to use. The designers opted for a simpler layout: they had 26 sockets with these double connectors and a maximum of 13 cables. It's much easier to use, but it increases the complexity of the machine considerably, not so much, but that's not really the problem here. The number of settings was further limited by the regulation that 10 cables should always be used, now the maximum number of permutations you can achieve with this type of plugboard is actually achieved with 11 cables, and it would be even better to use a number variable number of cables, say any number between 6 and 12.

But they always use 10, and that limits the number of configurations you can get from this component. If the socket board wasn't being used to its full potential, that wasn't as big a problem as the fact that you swap letters in pairs and that means the socket board doesn't obfuscate the action of the rotors as well as it should. This is a little unintuitive, but the board swaps letters before the current flows through the rotors and then after the current leaves the rotors it changes everything again, and that means that with certain techniques you can deduce a lot about what the rotors are doing. ignoring the plug completely.

At the end of the process, you only have to determine the configuration of the socket board, something that was actually not difficult for an expert cryptanalyst to do by hand using his knowledge of the German language, as long as he knew the configuration of all the other components of the machine . - all other parts of that day's encryption key. This board reciprocity was also exploited by Gordon Welchman in his contribution to the Bombe Machine, the key-finding machine used to crack Enigma, which he called the "diagonal board." That was the component that made the Bombe machine a practical solution for cracking Enigma.

All that means is that a non-reciprocal socket board with single-ended sockets, while it would have been a nightmare for traders, would have made the cryptanalyst's job much more difficult. The last major weakness of the machine has to do with these rotor rotation notches. What they do is they are a feature that produces a characteristic pitch of the rotors, which is fundamental to the effectiveness of the Enigma Cipher, the fact that every time you type a letter, a different cipher is applied.  The notches are on each ring of the rotors and you can move these rings and that allows the user to compensate for the internal wiring of the rotor, compared to what is shown in the numbers and letters around the rotor, which is shown in the little windows in the machine.

Setting the ringtone is essential for a secure messaging procedure.   It means that you can share information about the initial position of the rotors for a particular message by sending what is called a flag before your message. The indicator has two parts, one is encrypted and the other is unencrypted, so if you receive a message and you have your Enigma set up the right way, according to the daily key, you can use the unencrypted part of the indicator to decrypt the encryption. part that reveals how to configure your rotors to decipher the rest of the message.   For an attacker who does not know the ring configuration, it is impossible to do so.

However, there is still some information that an attacker can deduce from the indicator, because the rotor ring bears the notch that allows the rotors to move. This always happens when the same letters are displayed in the windows. The BP cryptoanalyst put together a helpful memoir for this, 'Royal Flags Wave Kings Above'. Thus, rotor one always spins when the letter 'R' is displayed in the window, rotor two spins when the letter 'F' appears, and so on. This is a classic example of a feature designed to add complexity that actually weakens a cipher by giving the attacker, in this case, a way to differentiate between different rotors.

That is the basis of several crucial techniques for breaking Enigma, including Cillis and Banburismus, which we will talk about in another video.   But both techniques are based on the fact that if two indicators can be deduced from different messages, either because they are not random, due to human error, or due to a procedural deficiency, if two indicators are found that are close to each other in terms of the rotor positions, you can tell if a rotor has spun between those two positions, and if you know that, through a process of elimination you can discard rotors that would have spun if they had been in use, which is why there are these ways based on the variable position of the notches to know which rotors are on the machine on a given day, at least the rotors in the right and middle positions.  Now, from about the middle of the war, the German Army Air Force changed the order of its rotors, not once a day, but three times a day.

This actually helps Bletchley Park, because they use the same three rotors but they circulate them, so if you have the order 1, 2, 3, it becomes 3, 1, 2 and then later in the day, becomes 2, 3, 1. Before this, the British could never identify the left rotor because it never makes another rotor move, and now they have the opportunity to identify that rotor because, at least part of the day, it is now in the right position of the machine. The easiest way to improve this part of the system would be to place the notch in the same position on each rotor. The downside to this would be that some information would be readily available in the unencrypted portion of the indicator.   So, let's say you set all the notches for the rotations to occur at the letter 'A', an attacker would know whenever there was an 'A' in the message indicator, that a rotation of the rotor would immediately follow.   That information could be exploitable, so in my opinion even better would be to go back to the old design, where the notch was not on the ring, but on the wiring itself.

There will be no relationship between the position of the ring and what is shown in the windows and notch and therefore there will be no way to identify the rotors from the message indicators. On the other hand, that would allow your cryptanalyst to know the exact position of the rotor wiring in the machine if he detected a rotation of the rotor by some other method. But all methods historically used to identify rotors in use were based on the notches being on the letter ring.   Placing the notch in the rotor core could have resulted in other

weaknesses

I hadn't thought of, please let us know if you think of any.

That was a serious weakness, a necessary feature that really helped the Bletchley Park Codebreakers. That is all for now! In a future video we'll talk about some of the human errors made by the German operators that made the Bletchley Park code-breaking transmissions possible, but we hope you enjoyed this video. Goodbye for now!