Breaking Enigma - Exploiting a Pole Position

[Music] [Music] right so we've done a on room 40 and one of the two patreon choices for Wednesday specials this month was the Enigma code and how it was broken so if you like cryptography and crypto analysis well we've come to the right place this week if you don't sorry but that's what they asked for so that's what they're going to get so before we get into how to break enigma it's probably a good idea to know what enigma is in the first place and the first person who just says it was a German code breaking machine I wish I could reach through the internet now I'm going to tell you some what how it works it is obviously an incredibly complex subject of which people have written multiple books just explaining how the machine works followed by lots of company complicated mathematics that show how you can break it and why it was so difficult to break but let's try and get this in a relatively condensed and understandable format at its heart the enigma machine was supposed to produce what's called a polyalphabetic substitution cipher but the messages that you plug into it that sounds very complicated and indeed the way it did it was very complicated but in very simple terms this takes a message and can be whatever message obviously and by application of a cipher key or code or what have you you end up with a message that you can then transmit that looks nothing like the original message and the phrase polyalphabetic substitution might give you an idea that this is basically substituting letters of the alphabet for other letters that were in your text to give what appears to be a bunch of nonsensical gibberish so this is different from a codebook such as was used in most of the German activities in World War one which we discussed in room 40 video where various orders and phrases were signified by batches of letters and occasionally numbers and it's also slightly different to some modern cryptography methods which try to hide transmissions or turn them into something that is not identifiable immediately as characters so trying to hide in background radiation or as static or something like that now instead the Enigma machine would produce a message that you could transmit in the clear and that anyone with the appropriate radio antenna could receive it's just that when you received it it didn't mean anything so how does this work let's look at a basic example so let's take a Pasic word like it hello now obviously there are 26 letters in the alphabet so we can convert each of those letters into a number corresponding to its place in the alphabet so we get 8 5 12 12 and 15 now we can just add a number to each of those let's say we add 5 so we get 13 10 17 17 and 20 and we can look up what those letters are in those and what not letters those numbers refer to and we get MJ qqt so we can now transmit MJ q QT which means nothing on the face of it but if somebody else knows that we are transmitting with a +5 substitution then they can just take away two take away five if once they've converted mjq cutie back into numbers and they will get the numbers that correspond to two hello out of it that's all very simple and obviously this including the numbers there's just an easy way of referencing how it's done you can just just substitute straight away now that's fine on the surface but it's very easy to crack this code apart from anything else there are certain letters that turn up very frequently in various languages and so if the message has any kind of significant extent then simply looking at the number of different letters and see which ones recur the most gives you an idea of what letter that most likely corresponds to in English for example the letter E is very popular so if you're reading a long message and you're using a plus five cipher and somebody sees L the letter J is recurring an awful lot that'll probably glue them in that that is therefore the letter E and from there it's very easy to break the rest of the code because you can work out what's the relationship in between J and E I'll write J is five letters further down the line than E what happens if I reverse all the other letters by five steps oh right I have a clear message great fantastic pretty easy to crack and at the end of the day you can do this with any of these single step substitutions because well you've got 25 other letters in the alphabet and it doesn't take very long to run all twenty-five possible permutations now you can make this more difficult by adding some kind of mathematical transform to your initial encryption key in the first case it was plus five but we could do something a bit more complex like say take the letter that corresponds to the number and then add say point seven five of that number to itself and see what we get there so again using our hello function eight five twelve twelve Dean becomes 14 8.75 so we round up or round down 21 21 and 26 point two five which then corresponds once you've rounded to the fractions 2nh you use ed now in that case the difference between the various numbers is different in each case so H has a difference of 6 e has a difference of 4 L has a difference of 9 and C's elegant same difference and o has a difference of 11 so this is going to be harder to spot because it's not the same reference all the time but as you guessed from L being uu you end up with the same number the same letter corresponding to the same letter just fire a different method so in this case if you see an H recurring in a long message you know the H will cut probably correspond to e which is a it's one of many clues many ways of breaking a substitution code like this and then you can work from there now you might not be able to immediately figure it out but someone with a fairly sharp brain can work out a fairly simple mathematical interference pattern like that but the common factor in both of these has been that we are simply increasing in this particular so we could decrease the numerical equivalent of the letters by a certain amount so that all going in one direction the numbers are all the other going to be higher obviously you can look back on yourself for hi about letter values like x y&z and vice versa going down but everything's going in the same direction you're not going to have a CH corresponding to a letter that's further down the alphabet and eek responding to a letter that's further back on the alphabet it's always going to be a steady progression as we said you might loop back on yourself for that is in fact still progression now where you can start throwing things into the mix making things a bit more interesting is by doing substitution that is not linear in that manner so you could substitute say D for P but you could substitute R for J obviously P is further down the alphabet but J is further up the alphabet relative just starting letters now that will confuse things a little bit more but again there's only so many combinations of the 26 letters of the alphabet that you can do so cracking the code for a good mathematician who's used to this especially with the clues that we mentioned before is a relatively simple matter certainly something that could be solved in military terms a lot faster than the information will become out-of-date so to make things more complicated and the solidedge a cipher we can move on to probability a probability is the well the chance of anything happening so let's say you flip a coin heads or tails well assuming it doesn't land on a side you have a one in two chance of getting a head or a tail but if you flip a coin twice in a row and you want to predict what that set of results is going to be obviously you can have heads heads heads tails tails heads or tails of tails so you've gone from having two to four I mean I think that's does that mean it's doubled well no cuz it's actually an exponential increase so you've gone from two to four but if you add in another layer of coin flips you now get eight but this isn't a case of the odds just doubling every time you go one step down it's a matter of squares cubes etc this is denoted by the notation ^ which is the small superscript number so in the case of a single coin flip it's 2 to the 1 which is 2 in the case of 2 coin flips and row it's 2 to the 2 or 2 squared which is 4 then 2 cubed or 2 to the 3 which is 8 then 2 to the 4 which is 16 plus 2 times 2 times 2 times 2 and so on and so on now if you take something that's got more than two possibilities so let's say you are rolling a normal six itíd dice well obviously rolling a dice once you have a one in six chance but if you roll the dice twice you've actually no got 36 possible combinations because it's six times six and the numbers got from there so if you have 26 possible combinations because you have 26 letters in the alphabet if you were to run your cipher through one set of substitutions and then run your number your cipher through another set of substitution so you let's say we take our transformations that we had before and we say run the word hello through the first one which was the plus five and then we run that result the M&J qqt through the second one which was the x 0.75 and add it to itself then we end up with vqc CI as an output now that's going to be much harder to deconstruct because you've got to there's no direct relation between vqc CI and hello you've got to deconstruct two separate layers of coding you've got to figure out what those are you so you've got to take it take off the 0.75 x itself and you've got to take off of the plus five and obviously the 0.75 plus itself is relating to the 13 10 17 17 20 that we got for the plus five which is nothing to do with the eight five 12 12 15 although there is a relationship there now that probably sounds very complicated doesn't it and this is the problem if you if you do to two sets of substitutions then you have 26 squared or 26 times 26 possibilities so assuming you've got a constant factor which is in this case we do have there's 676 possibilities okay well now let's try a transforming it through a third time well now we've got 17,576 possibilities and well you're thinking and now that's going to be a little bit harder to crack isn't if you've got 17,576 possible combinations to try that's gonna take a while especially in an era critically early before computers where you have to do this all by hand so how does this all relate to enigma well enigma worked by taking a series of rotors as its base point so each rotor had the 26 letters of the alphabet on it and the first rotor would be wired up such that when you pressed a button let's say a it would go through this rotor which had a wiring from one plate on one side to one plate on the other because the letter A will generate an electrical signal but that would be wired up to a point on the other side of the rotor that corresponded to a different letter so let's say a would transform into M but this wiring sequence was not sequential it was randomized so as we said before a could go to M but B could go to a and Zed might go to Q and all the other sorts of wonderful loveliness like that great so far and then there would be a second rotor and so you take let's say you're a has gone through to em and so it enter the second rotor as em and it would go through to the other side it'll come out as completely different letter because the second rotor which would be numbered would be wired up completely differently so whereas on the first rotor M might go through two t on the second rotor M might go through two d so you're a has now become a d and then the third rotor wired up differently still would do the same thing again and then on the uncertain ending mo machines the three rotor machines the signal would pass through what was called a reflector and it would then enter back into the third rotor coming in from the other direction at a completely different point so that's another transformation step and then it would go back through the rotors obviously because it's now entering the third rotor at a different stage exit a completely different stage and so on so it would actually go through seven transformation steps so that means that when trying to work out what how many possibilities there are just using that that's 26 to the seventh power and 26 to the seventh power for those of you are really cricket maths is eight billion 31 million 810 thousand one hundred and seventy six possible combinations so this is what getting into the realms where you can't humanly presume that any legitimately reasonable number of researchers is going to be able to try out all of those combinations in time to make that information that they might get out of it in any way useful at least by sheer random brute-force attack which basically means you try each one one after each the possible combination one after the other except of course we're talking about the naval enigma specifically although this will cover in it as a whole but the question was specific about naval enigma but anyway naval enigma in the Kriegsmarine a-- used four rotors which means you're going through four sets of rotors going one way then the reflector and then four sets of rotors coming back the other way which actually means at nine transformations which then translates to five point four to nine to the twelfth power which is trillion so you're talking about five point four trillion possible combinations but if you thought the numbers had stopped there oh no it gets even worse that's because big numbers are big numbers true but mathematical equations can help solve that I remember you said at this point in the early part for there are no computers true but you don't have to try every single possible combination you can try mathematical formulas that will give us certain relationships and that will help you to narrow down the options which means that a reasonable size team of humans can solve even such a complicated impossible encryption method in maybe a few hours maybe a few days but certainly enough to make orders like prepare to move out in three days worth intercepting and decoding well the Germans have thought about this so there were two further innovations to the Enigma machine in its base form just in in the actual box the first was the introduction of a stepper motor and this took the form of well a small lever so when you pressed the first key you'd end up with a letter and that'll be turned into another letter but then the rotor the first rotor remember that was setting up all these trillions of combinations would take a step around so whilst you could relatively easily for a given value of easy mathematically solve one set of equations to derive a message even something that's encoded within one of several trillion possibility possible ways by the time the operator hits the second key because the rotor has taken a step around you've got an entirely different set of possibilities so if we take our example of hello h will have been encoded one of several true trillion possible ways but then e will have been encoded in one of several trillion separate possible ways because we've just changed the sequence of the rotors in very mine because it goes backwards and forwards that's through the different rotor twice and obviously it's connecting two completely different letters because if our if we take go back to our previous example if our a is going through two M if M was previously going through two D but now our rotor has stepped one so M is actually going through two E and E is going to be connected to a completely different letter than what D was connected to and so our end result after just that first set of rotors can be completely different and then just to add insult to injury every so often either one full rotation of the first dial or possibly every half rotation of that dial or one or half rotations of the dial subsequent each other rotor would also change so over the series of a message that involved more than twenty six characters you would end up with at more than one possibly more than two rotors stepping round which means the number of possibilities just goes absolutely berserk but it's still all linear and can be solved within reason by a lot of people with the loss of equation boards but now you enter the plugboard for you see the plugboard is a section at the front and that's another completely random set of circumstances because at least with the rotors every single letter is being changed but with the plugboard this intercepts the signal between the key and the rotors and as you can see in this picture only some of them are wired up typically it was 10 sometimes it would be 15 sometimes another value but what this did as you can see is it transposes certain letters but it doesn't transpose all of them so even if you solve the trillions or quintillions of potential combinations that you'd get out of the rotors you're still going to end up with a bunch of absolute nonsense because half the letters have been transposed to completely random other letters and all of this assumes that the rotors are all starting at the same positions because each rotor obviously corresponds to the 26 letters of the alphabet they were all usefully labeled as you can see a through to z but they could always be different you could have different combinations so you have your first row so you could have your first row to set so that you type in the letter A and assuming it doesn't get switched by the plugboard the AE might not enter into the a on the first road trip might enter on to L and again on the second router it might not enter through the a socket there and my enter through G and then the next one it might enter through B and so on and so forth obviously 3 & 4 rotor variations so all of this is dramatically increasing the number of possible combinations and the difficulty of the math involved and thus might take making it take far longer to crack any given in a cipher added on to this is the fact that not only were the various machines using three or four rotors in the naval case obviously the four four rotors but there were more rotors available than that so for the army enigma units they actually had a choice of five rotors and for the kriegsmarine ii units they had a choice of eight rotors and you could choose any three or any four of those rotors for the for the future I'm just gonna refer to the Kriegsmarine one so you've got four rotors and you've got eight choices so now you can choose different rotor combinations as well so you could say right well today we're going to use Roach's one through two three and four in that order tomorrow we're going to use four two three and one the day after that we might use seven five two and three and so on and so forth so you're just making it exponentially harder at every step but enigma did have two serious built-in flaws one of which was that the various rotors were all fixed so all rotors marked one had exactly the same wiring all roads not 2 etc etc and this was necessary because obviously there were lots of different machines and you had to encode and decode on any given machine so if you could figure out what the rotor settings were for each rotor that massively reduced the number of possibilities that were available to you and secondly was the fact that because of the way we set up the Enigma machine could never code a letter back to itself now that meant that basically if you typed the letter a any number of letters might show up at the one letter that would never show up when you typed a would be a now you might think well surely that's a good thing because you might accidentally end up encoding the text exactly the same as it actually is well no because remember it's changing every time you press the key so you're always going to end up with a complete gobbledygook but the reason this is important is because remember those numbers we were talking about it actually massively reduces the number of combinations now there's a lot but that reduction is quite important so if we are looking at just going through the four rotors and back area militias to the nine transform so 26 to the power 9 which is if if the machine was able to rotate it back to any not letter including itself would give you just over five and a half trillion just under five and half trillion possibilities but since it can't there are only twenty-five possibilities in 25 to the nine is three point eight trillion now you might think well that's not especially comforting but it's not one and a half trillion possible combinations of which is a significant amount and obviously as complications get more and more those that slight change in numbers at the beginning a reduces the number of possible combinations at the end by significant numbers when it comes to trying to solve these things but B it's a known weakness in cryptanalysis terms because you know one thing that it cannot be so there's a whole bunch of equations that can try to solve for a substitution cipher which will swap any letter for any other letter including itself but by having a fixed point that you know whatever letter you've got it cannot possibly be that letter in the final decipherment is actually incredibly helpful for decoding these things there's a lot of very heavy math but honey but I think we've probably don't want to get into that I mean we're 25 minutes in and we haven't even gotten to the historical part of it we're still trying to explain how the machine works now of course the problem is once you've got this wonderfully in ciphered bit of text you can transmit it in the clear and it'll just be a complete jumble of gibberish how does the person at the other end know what's going on well the thing is because it's purely mechanical with the same starting settings any enigma machine that's using the same set of rotors will be able to decode it perfectly because remember the cipher is going through the rotors one way then back through the rotors the other way so it started is in finishing up at the point that it started out so if you take your gibberish text and you plug it into an enigma machine that started off at the same settings that the machine that's encoded in the first place was set to it will basically do a reverse loop through the entire machine and come out perfectly readable this is a good thing however how do you know what the other guy's settings were you can't very well transmit those settings in the clear which is what you'd need to do because of using put them inside fir'd and person doesn't know what the settings are it's useless if you transmit oh yeah by the way everyone today we're using wheel for then wheel to then wheel five then wheel one um that's not gonna help you with your encryption because everyone else who's listening is gonna go well thank you very much right now we know what what we what settings to use in our own machines if we have them and obviously you can capture any machines or as we'll come to in a minute you can build your own so what would happen is every enigma machine would come with a codebook and within this code book would be settings for every single day so every single German Enigma operator on a given network so obviously the army units the Luftwaffe units Kriegsmarine e units etc they'd all have their own little networks but anyway so let's say you're in Norway and you're waiting to receive orders from Kriegsmarine a high command and it is January the 5th 1942 so you will open your little book and you will look for the heading January the 5th 1942 and then you'll get your wheels out your rotors and you'll go right I need this wheel this wheel this wheel right but my four rotors or wheels in one after the other in the correct order and I'll then look at the ring settings all right ok I need to turn this one to see then I need to turn the next one to why then it takes and so on and so forth then it gets be doesn't it then you have to look at your plugboard you're right ok so today's settings we're going to move II to why we're going to move I to Q we're going to put K to you and we're going to put B to X and so on and so forth great fantastic you are now setup now assuming that nobody knows what the wiring settings in your rotors are actually are you have a machine that can configure a message in one of around three times ten to the 114 possible combinations which if you want to put another way is 100 de Killian as de Killian times the number of protons in the observable universe but if the enemy knows what the wire configurations on your rotors are it comes down a fair bit to a mere 100 sextillion possible combinations which is about 100 times of the number of grains of sand on all the beaches of the world put together other numbers that share this lofty order of magnitude are the lower end of the estimate of the number of stars in the observable universe and the number of atoms in one mole of a substance for those chemistry and physics nuts around otherwise known as a begad Rose constant the Germans also try to add further complication to the method by getting the rotor settings to be changed for each message the way they do this is you'd have your Enigma machine set up to the day setting but then the person who was sending you the message would choose a four-letter word or string of letters three letters obviously in the case of an army unit and they would encrypt this twice using the base settings so they might choose being the German the word eine ein e so that type I know eine into their Enigma machine and this would give them a eight digit string of letters and this would then be transmitted to the receiving station and the receiving station would put this string of letters into their settings and that would give them IANA ina so they would then both on both ends change their rotor settings to this particular combination so instead of whatever the daily setting on the rotor is you've now change it to this one that have been chosen purely for this message purely by that station adding a whole other layer of complexity to or had they you see this was actually an operational flaw we already discussed the built-in floors to the Enigma machine but what about the operational ones you might think well this is at a whole level of complexity to it that surely this is gonna make it more secure no actually because once someone on the other end of things that you don't want to be listening in on your messages has cottoned on to what you're doing with this opening statement if they know that encoding contains a repeated statement like ina ina then or if you wanted to say for for enjoyment of via via then they can look at the two blocks of four or in the army case two three two blocks or three and they know that there is a distinct mathematical relationship between those two because it's the same thing so okay so we know after four presses of the keys let's say using Ayane owner that E has become this letter in the first tart case and it's become this letter in the fifth case so for steps in we can actually work out a mathematical relationship and that can be the first step towards breaking a whole encryption so why have we just spent about half an hour talking about high concept mathematics for cryptography the interior of Enigma machines the various settings and a couple of sidetracks in two related subjects simply put to give some idea of just how colossal a task decoding this thing was bearing in mind that any decoding had to be done either by hand or if you had some kind of system to work out by you had to build a mechanical machine that would do some of those calculations for you but such a machine was useless unless you could work out how the Enigma machine was built in the first place which meant you had to first crack at least one of these ridiculously complex messages by hand now with all that in mind I'd like to introduce you to Mary Anne Wojcicki who is quite possibly one of the most brilliant people I've ever read about because he took one little bit of information this was before the kriegsmarine added the fourth rotor which was that the first three letters of a message were the same as the second three because that was the change of rotor signal rotor positions signal that we discussed earlier and based purely on that and those six character strings that were transmitted by various German stations he managed through sheer hard work and I have no idea what his IQ was to determine not only the cipher that those were sent in to decode them correctly from the mathematical equations that he derived from that cipher he managed to completely rebuild a three rotor Enigma machine from scratch and just to re-emphasize that he used six letters had a ton of math to reconstruct them world's most complex cryptography device at that point this being the early 1930s purely through the power of his mind if this man was not subsequently unable to walk at an even keel because of all the medals the Allied powers pinned to his chest I want to know why nonetheless having effectively rebuilt and reinvented the Enigma machine effectively from scratch that still didn't help figure out what the individual settings were for each day but a mind like this is not gonna be held back by such a minor thing as not knowing what one of several million possible combinations were from to start with so he invented the rather wonderfully named crypto logic bomb which was unfortunately not some kind of radio transmitted explosive the detonated aid Enigma machines at distance wonderful as that may have been instead this took the form of a number of replica Enigma machines all wired together and being tested repeatedly for different combinations of rotor and plug board settings to see if any of them would spit out something that resembled a readable text using the various transmitted six-letter and obviously later eight letter encryption keys and then once you had that you were pretty much on your way to such solving the actual cipher key for the day to give you some idea about how efficient this was by 1938 they could reconstruct any key that the Germans had for any given day in about two hours which unless you were transmitting a signal that basically read attack in the next half hour meant that German transmissions were now basically not encrypted unfortunately it was in 1938 that the Germans introduced rotors four and five remember we said they had three in the army units and initially it was just those three and then they introduced four and five in the Kriegsmarine orcs head now had four and we're introducing another eight and this increased the number of possible combinations to well beyond that which the Polish machines could counter in a reasonable time period due to the fact that the poles have managed to work out what the internal wiring of each rotor was they managed to produce these this was the product of another Polish cryptologist Henrik the Gauss key and each of these sheets represented the known starting position of each rotor and by superimposing various combinations of these one on top of the other using the initial encryption key you would eventually end up with only a single aperture showing as a possible valid combination and this would tell you that you had successfully decoded that letter and from that they could work out what the order of the rotors were what the settings of each of those rotors were etc etc and thus they could work out the entire cipher key this worked for a couple of years until the Germans changed how they encrypted those message keys in the first place but it was good for a couple a couple of years as we said by the beginning of the war the British had made some headway into cracking certain up parts of enigma but nowhere near as far as the poles had and credit where credit is due it must be noted that the poles efforts had been helped greatly at the beginning of the 1930s by the fact that the French had managed to secure a number of documents related to enigma manuals on how to use it pages of keys and suchlike and this had helped greatly in reduce keys initial mathematical constructs to solve the mysteries over the initial 1930s Enigma machines on the creeks arena side of things the Germans were still also using encoded books like they had in World War one so that meant even if you decoded the transmission you'd still have blocks of text that didn't actually mean anything unless you had the codebook to look up what those are various code blocks meant but the army communications were be easier to in the late 1930s the pols shared much of what they'd learned with the British and the French who quite bluntly were completely astonished at the complexity of the efforts that they'd gone through and just how far they'd gotten to give some idea of the level of difference the pols as we said earlier take a couple of hours to crack any random daily German transmission whereas the next best effort on the British side was occasionally being able to read parts of a message if you happened to know Italian the Germans were helping somewhat by being rather lazy with a number of their encryption keys which was of each you might remember from World War one as well so when they were transmitting those initial changes in setting well it would be very difficult to think of well going about the kriegsmarine a standard a four letter word that's different all the time so quite often you would have the 1930s opponent of having your password set as password or one-two-three-four which would be the key pitch being translated would just be a a a a or b b b b and so if you encode be eight times to get your two sets of B's it becomes vastly easier to solve back what that encryption key is and therefore all that encryption setting is and therefore reverse engineer the rest of the settings whilst the gems would stop transmitting this duplicate encoded cipher around 1940 thus making many of the aforementioned methods much more difficult if not impossible to use the German invasions did have one small benefit which was that by invading Poland they forced the Polish cryptanalysis team to flee to France and then by invading France they forced both of the French and polish crypt analysis teams to flee to Britain which now meant that you had British French and polish cryptanalysis teams all in the same location all able to work on the same thing despite the German changes to their security and settings there was one major flaw in a lot of their messages which although it wasn't quite as egregious as the earlier flaws and therefore took longer to solve for did allow for a number of messages to still be broken and that was simple human laziness basically almost every message although not all and began with the letters a n and X and the reason for this was that each rotor had 26 settings corresponding to 26 letters on the alphabet but this didn't leave any space for numbers or punctuation or spaces and so the job Germans had chosen to substitute the letter X for a space and the letters aan and other German for to and you can see how of great many military messages might read to whoever the messages for and this obviously enjoyment translated to an and then X for the space so knowing that almost all the messages they received after they sent through the key settings would start with an incited version of a and X meant the deciphering was still possible albeit slower and less frequently there were other flaws that would allow the breaking of certain enigma cipher keys on certain days such as the fact that a message might be transmitted by enigma say from Berlin to increase Moreno Western HQ but it might then be relayed out to something else say a port or u-boat or something like that that didn't have an enigma machine in a lower level cipher which was much easier to break if that message could be broken and they knew from their radio transmissions that the message had gone from Berlin to cruise marina western HQ and then out to another destination they knew that the text of the message was the same and if they had a copy of that message they could then have a starting point with which to break that particular enigma cipher heat for the day which obviously would then allow you to read all the other messages that have been sent using that same key that particular day there were numerous other operational flaws and mistakes that were made but I won't go into all of them right now although it must be noted that one particular new rotor wheel that was added was sold very quickly when some poor old German signals man was turned told to send a dummy message and he just decided to send all our holes because he was bored presumably and this is where that factor of the Enigma machine not being able to code back to the same key came back to bite them because when they received the message I will intercept a message I should say over at Bletchley Park they know just why in a minute there's this massive bunch there's not a single letter L amongst them hang on a minute and very quickly they were able to solve that encryption which then allowed them to work out the wiring of that new rotor and of course where that mention it's a good time to mention at Bletchley Park say Bletchley Park was a location in England where all the decoding was now taking place with the exception of some cryptanalysis team members who for various reasons usually having being that they'd operated in Vichy France for a while so they weren't necessarily sure if they've been compromised were working elsewhere at Bletchley Park they began to work on new methods of solving the Enigma in ciphering and one of these was an enhanced version of the Polish Cryptologic bomb as you can see here and this was basically a series of multiple identical copies of the Enigma rotor wheels all running round together and with this they would take an educated guess as to a portion of the message using various commonly known items that were including in many messages such the aforementioned anx and this would then be compared by the machine with the ciphertext that had been delivered to it and since a number of the capabilities of enigma were known as we said such as the fact it couldn't code back onto the same letter these rotors could all work their way around until they found a combination of rotor settings that didn't produce text that contradicted these known capabilities the necessarily mean it was the right setting but it meant it was a setting that may possibly work so you couldn't note that down and then off it would go again so this would eliminate all the chap settings all the ones that were theoretically possible but wouldn't actually work with that particular text the syph are still accelerating the process of decoding of course while we're on the subject of great minds it's worth mentioning that of course at Bletchley Park was Alan Turing genius mathematician himself this was where simply having more money and Industry helped greatly whereas the polish cryptanalysis team had only been able to produce six machines to try and decode enigma rotor settings the British using said industrial advantage as well as obviously polish and French as well as their own expertise were able to produce several hundred of these machines which individually were larger than the Polish ones anyway and these could be manned by a team of several hundred operators who would effectively use these known weaknesses in German transmission discipline to try and brute force the remaining possible rotor settings meanwhile out on the front lines various German code books were being recovered and other allied efforts were factoring in it to the fight against enigma encoded transmissions specifically the fact that the Allies had invented a way to track u-boats via their radio transmissions actually helped decipher enigma messages believe it or not and the way this worked was that because the u-boats now knew that to transmit or receive significant amounts of information took a long time and aircraft with a radio direction finder would be very likely to lock on to them come in and depth charge them they were put under a system whereby their messages had to be very short but to be very short and to communicate large amounts of information had to use a set of stock code phrases much the same way as how a naval signal flag can communicate a lot as long as you have a flag book that tells you what they mean of course this meant that once a u-boat had been captured with its codebook intact you then had a limited set of phrases and code groups that the transmissions could actually be which again reduced the complexity of what you had to solve for of course not all signals used of that particular code book but there was enough traffic floating back and forth using the codebook that this could be used to crack the Enigma cipher key for the day which could then be used to crack all the other longer messages whilst also pushing their prodigious talents to use in working out some of the starting points from which to try and decrypt a cipher key during the polls and others were also working on inventing newer larger and faster mechanical rotor comparison mechanisms the successors to the Polish Cryptologic bombs which would help decipher the rotor settings faster still now Bletchley Park is frequently known for inventing what's quite possibly the world's first computer Colossus but Colossus does not feature in the story of deciphering enigma because that was developed to decipher an entirely different set of German transmission nets which used a completely different encryption device in 1942 of course America entered into the fight and if Britain had money and industrial might well in excess of what Poland had been able to bring to bear on the problem adding the Americans into the situation only made things worse for the Germans and much better for the Allies with a fresh pool of resources and brainpower to draw on Turing and the Polish inventors cooperated with some of the finest American minds to produce even faster larger and more capable mechanical analysis devices with the bombs the Germans were continuously suspicious of the Allies success in Andrey routing convoys to avoid wolf pack's and knowing what the sizes the wolf packs of u-boats were but many of the most suspicious Vence fortunately seemed to time in with other thing for example in early 1942 there was a major upgrade to the signal security and level of complexity in german Kriegsmarine enigma systems and for a while this massively reduced the output of Bletchley Park also known as ultra intelligence and obviously more more ships started to go down now the Germans might have picked up on this coincidence except for the fact that around the same time they were ordered to attack targets off the American East Coast which were easy targets for a number of other reasons and so the overall increased effectiveness of the u-boats was masked by this particular campaign off the east coast and so they didn't quite cotton on to the relationship between increased numbers of kills and the increased transmission security likewise during a period when Enigma traffic was being read quite easily and the number of u-boats being surprised and killed by Allied aircraft was going up dramatically once again the Germans became suspicious that their Enigma codes were being read however this coincided with the recovery of a crashed Allied war mother happened to have a cavity magnetron still on board and the remains of a radio direction-finding system which then seemed to confirm to the Germans that in fact these kills had been made by aircraft equipped with radio direction finders however whilst the number of various enigma networks as well as the level of security the complexity of the dial settings etc was being increased steadily throughout the war the simple fact of the matter was that by about 1942 going into 1943 the Allies had combined the best of the Polish French British and American cryptanalysis teams as well as applying the considerable industrial and financial leverage of both Britain and the u.s. towards producing these mechanical Cryptologic bombs all based of course as we said before on the Polish model and of course with the Enigma machine being an electromechanical device with a finite number of settings the Allies could simply keep producing more and larger Cryptologic bombs which were of course off themselves massively complex machines and so sooner or later the sheer number of possible rotor combinations and other settings that the varicella are decoding machines could try in a relatively short space of time would simply exceed the number of possible permutations that an Enigma machine was capable of and this in turn meant that during the course of the last two to three years of the war the amount of time required to decode any given transmission dropped dramatically in mid 1943 a decent Enigma code combination it could take two days to decipher whereas by 1945 it was taking maybe ten minutes on some of the easily solvable ones and even the most complex late-war devices with multiple settings and different methods of code security could still be read in a day or two of course using all this information was another matter but it's estimated by most authors that the simple act of being able to read break and read certain enigma messages obviously an increasing frequency as the war went on shortened the war considerably estimates between very between one and three years and in some cases especially with some of the earlier messages possibly even changed the course of large sections and campaigns of the war entirely especially around the u-boats a campaign so there you have it a short history of roughly how enigma was broken some of the key players in it and hopefully a bit more of an understanding as to how an Ingo worked in the first place which is that previously serves to underline just a massive accomplishment this was by the various cryptologists in the Allied cause and if you particularly enjoy the nice warm feeling of your brain melting out of its ears in a desperate attempt to flee from what you're trying to do to it I will include a link in the description to a copy of the cryptographic history of the work on the German naval enigma which was one of a number of sources but probably the main source that was used in the making of this video hope haven't confuse you all too much and tune back in in a few days for some slightly easier to explain warships that's it for this video thanks for watching if you have a comment or suggestion for a ship to review let us know in the comments below don't forget to comment on the pinned post for drydock questions

Channel: Drachinifel

Views: 179,916

Rating: 4.9032016 out of 5

Keywords: wows, world of warships, Enigma, Kriegsmarine, Luftwaffe, codebreaking, Bletchley Park, Turing, Alan Turing, Collossus, Bombe, Rejewski

Id: rekaj8bBWiU

Length: 53min 56sec (3236 seconds)

Published: Wed Dec 11 2019