Online amateurs crack Nazi codes

Discussion in 'Top Secret' started by angie999, Mar 2, 2006.

  1. PsyWar.Org

    PsyWar.Org Archive monkey

    We tend to forget the power of the modern computer compared to the primitive
    equipment Bletchley Park had at their disposal during the war. I bet of they had
    the stuff we play about on these days the codes would have been decrypted in a
    flash
    David ._._.

    From what I understand Colossus performed rather well compared to a modern PC. Although a fairer comparison would be with a modern super-computer rather than with a general purpose desktop PC.

    But of course the computer has changed the face of encryption/decryption as well. I guess GCHQ has a much harder time of it these days in relation to code breaking. Anyone could grab a random data file off the internet and use it as a key for a code, like a one-time pad. No doubt though there are alternative benefits of the digital age for intelligence gathering.

    Lee
     
  2. geoff501

    geoff501 Achtung Feind hört mit

    From what I understand Colossus performed rather well compared to a modern PC.

    It performed well compared to anything else around at the time. The idea that Colossus could beat a PC was put about by someone using a PC simulator, only using those features available in Colossus (ie no storage etc) and was probably written in a slow high level language (BASIC or PASCAL).
    Colossus could process, at most, 25,000 characters per second, each of these operations was a series of hard wired boolean functions. I made a PC simulator some years ago, on a P200 (200MHz). A 'long run' to find the settings of a pair of wheels, was around 10 - 20 minutes on Colossus, on the PC a few seconds.

    Colossus was a brilliant piece of work and none of the designers got the full credit for their achievements.

    Anyone could grab a random data file off the internet and use it as a key for a code, like a one-time pad.
    How are you going to inform your secret buddy which random file you will be using today, if you are connected by an insecure channel?


    No doubt though there are alternative benefits of the digital age for intelligence gathering.

    Probably.
     
  3. Mark Hone

    Mark Hone Senior Member

    Geoff-I have received your stuff about Purple etc. Will have a closer look when I am back at school next week. Thanks again.
     
  4. PsyWar.Org

    PsyWar.Org Archive monkey

    It performed well compared to anything else around at the time. The idea that Colossus could beat a PC was put about by someone using a PC simulator, only using those features available in Colossus (ie no storage etc) and was probably written in a slow high level language (BASIC or PASCAL).
    Colossus could process, at most, 25,000 characters per second, each of these operations was a series of hard wired boolean functions. I made a PC simulator some years ago, on a P200 (200MHz). A 'long run' to find the settings of a pair of wheels, was around 10 - 20 minutes on Colossus, on the PC a few seconds.

    I was hoping you'd respond to that Geoff :) Thanks for setting the record straight.

    How are you going to inform your secret buddy which random file you will be using today, if you are connected by an insecure channel?


    Ah yes, I was thinking along the lines of a pre-arranged system. For example, send a message at 11.00 pm on a Thursday then use the master photo on the main news article on the BBC News front page. Or use an image sent to you in one of the chain-emails with pictures of amusing Chinese road signs or archive photos of the construction of the Empire State building, etc.

    But of course it is all worthless if someone is suspected of using such a code, all that is necessary is to intercept the coder's internet browsing history and it wouldn't take long to track down which image or data file they may have used as a code key.

    Anyway good to see this reported in the news today-
    Bletchley Park WWII archive to go online:
    BBC News - Bletchley Park WWII archive to go online

    Lee
     
  5. geoff501

    geoff501 Achtung Feind hört mit

  6. PsyWar.Org

    PsyWar.Org Archive monkey

  7. geoff501

    geoff501 Achtung Feind hört mit

    My German isn't quite up to it but looks intriguing and a clever idea.

    And Google refuses to find it to translate.
     
  8. geoff501

    geoff501 Achtung Feind hört mit

    A simplified example of RS44 (hope I remember, my crypto stuff is in storage)

    First a plain message:

    D Company will attack at dawn

    We add some word separators and make into groups:

    DXCOM PANYX WILLX ATTAC KXATX DAWNX

    We then take the stencil currently in use and select a start cell (I will select row ab, column cc). The receiver must have a copy of this stencil. These stencils are (or should be) changed frequently. The message is entered, starting at the chosen start cell, row by row:

    http://www.********.co.uk/WW2talk/rsplain.jpg



    The cipher message is then read out from the start cell, column by column. The start cell is in column 2, so each column is read out in numerical order starting from 2. If a column is blank, it is skipped. This gives the cipher text, which is transmitted, together with the start cell which is sent in plain to the receiver.

    ABCC DLTAP WWXCN MTNAD YAOTC XIXXA XKXLA

    At the receiver end, the stencil is first marked with the number of letters, (in this case 30) counting row by row, from the start cell. (the unused cells are penciled out - I've marked them in green here). The cipher message is then written into unmarked cells on the sheet column by column starting at the start cell. The plain text can then be read out row by row:

    http://www.********.co.uk/WW2talk/rscipher.jpg



    The actual key procedure was a little more complicated, a user selected 'offset' start value could be selected and if the start cell was a black cell, the cell to the right was used as the start. Plenty of opportunity for the cipher clerk to cock up! Stencils were made with random cells (always 10 per row) and the column numbers and cell co-ordinates changed with each stencil issue.
     
  9. PsyWar.Org

    PsyWar.Org Archive monkey

    A very nice explanation there Geoff. Is seems a lot like the transposition poem codes that SOE were using for a little while.
     

Share This Page