Loading...
  OR  Zero-K Name:    Password:   

Marvin encryption scheme

13 posts, 613 views
Post comment
Filter:    Player:  
sort



CHrankAdminDeinFreund
8 months ago
(edited 8 months ago)

Marvin Encryption Scheme



Story


quote:

[14:59] Marvin1: haz aisr uadu zraiar
[14:59] Orfailius: glaives could be more 🗡
[14:59] Orfailius: 🗡 🗡 🗡 🗡 🗡 🗡 🗡 🗡 🗡 🗡
[15:00] Marvin: io ua aiatdf osfpsufpdu osuf iaz pdns
[15:00] Anarchid: dagger is an extremely fast alpha raider
[15:00] Marvin: kaz
[15:00] DeinFreund: not even rot13
[15:00] Anarchid: oh look marvin is back
[15:00] Orfailius: who?
[15:00] Marvin: ,audtd kommst du in ein level
[15:00] Marvin: gazdr
[15:01] Anarchid: anyway; dagger is an ultrafast alpha raider; its strenghts are speed and alpha, its weaknesses those of the raiders: low hp, low range, aoe
[15:01] DeinFreund: what is that cipher
[15:01] Anarchid: among raiders it's special by being fast and alpha, so nerfing hp doesn't reduce its uniqueness
[15:02] DeinFreund: i think we should nerf Marvin's cipher
[15:02] Marvin: im HATDODDADZAIDTSAWSZSDGFD
[15:02] Anarchid: it's not a cipher, look at the letter distribution
[15:02] Anarchid: "sdgfd"
[15:02] Marvin: R3
[15:03] Anarchid: repetitive letters, and repetitive letters placed closely on the keyboard
[15:03] DeinFreund: not on neo
[15:03] Marvin: jazsits uatdud
[15:03] DeinFreund: not on dvorak
[15:03] Anarchid: it's like trying to get a random number and getting 31321233123
[15:03] Marvin: hazf
[15:03] Marvin1: nahdoazg
[15:03] Anarchid: because your idea of getting a random number is mashing on a keyboard
[15:03] Anarchid: and keyboard is not random
[15:03] Marvin: ok marvin1
[15:03] Marvin1: haha
[15:03] Marvin: ):
[15:03] Anarchid: he's mashing numbers on a qwerty
[15:03] Anarchid: *buttons
[15:04] Marvin: ok haidctd komm,laudf.q
[15:04] DeinFreund: it could be an interesting cipher
[15:04] Marvin: !
[15:04] Marvin: oaf
[15:04] DeinFreund: whose characters are distributed by qwertz mashing distribution
[15:04] Marvin: Anarchid komm
[15:05] DeinFreund: maybe characters are encoded by the geometric distance between the letters
[15:05] Anarchid: oh qwertz
[15:05] Marvin: hdzagsdiats jatdjatd
[15:05] Anarchid: yes that explains why so many z
[15:05] DeinFreund: with highly frequent characters having low ids
[15:05] DeinFreund: i need to implement that cipher now
[15:05] Marvin: jerted
[15:05] DeinFreund: it'd fool Anarchid
[15:05] DeinFreund: then it'd fool everyone
[15:05] Marvin: i
[15:05] Marvin1: ksgdiazdtdudt
[15:06] Marvin: ?
[15:06] Marvin: <
[15:07] Anarchid: how do you compute geometric distances?
[15:07] Marvin: kfzfe


Hint


String lines[] = new String[]{
    "qwertzuiop",
    "asdfghjkl",
    "yxcvbnm"
};
int distanceVerticalPerLine = 4;
int distanceHorizontalPerColumn = 1;
int splitCharacterDistance = 8; 
//split distances > this into multiple characters because keyboard too small
String mostFrequentLetters = "etaoinsrhldcumfpgwybvkxjqz";
//distances for letters, characters not in this list remain plaintext


quote:
[15:07] Anarchid: how do you compute geometric distances?



Example


now we can finally speak in marvin

ahlbtr krtt afhb ldjnvrtlkrvn kvaadumk ie cfstlaqtf

same meaning:
fbxwhg yqww xbmh wfroudszumut duffszsy bf ietkvrfku


FIFDHDMMVM


fitwataavnmdecvnbgsea fb assfysxalbb awrjecmjz

gfiikz qqygh jjeznngkvcnsnnszr kvcduifg jcxhjfifg otd sg yvygzscedvfituii fiotzss eecewhwtgge uh jeeeuf tuzwxhss nmbhxhhwzifiiih

gmvyn snmm duikvcew xhdd gygkrtt ru myahkrejkrkh vmdfifrr tlll baarrwe jzwiuufifs goj hppp hjfifajhhc
+4 / -0

FIrank[ffc]killer
8 months ago
....wat
+3 / -0



CHrankAdminDeinFreund
8 months ago
(edited 8 months ago)

FIrank[ffc]killer
adadgddgf sfhkhkhkhhfdgdadgjj ljljljkhhfsfhkhfhkkhfhfsfsfsaa dadadgjlkhhfsas das dgdgfsfhfsfsfhh hfsfsfsfsfhkkjgjgfhh khkhkhgddadggjljgg jgdadasfgdgdaddd addgjhkhffggddgjgjll

dfhkhffhffhfss fssfhkk hfhkhkkhkhhkhhkhgdgdgjljgg
splitCharacterDistance = 2;


afawtzititijffafawdxcc fbfedhjiplitzewadrdxaxc y vdrgfjhuplkgkghupuuu plhupozopllilpllpijfjk gfbxx ysett fjilhdffthhg faxsggvdvddvgnjnbxde

lkupilkupp lkmkk ozesseffasycngvysa
    int distanceVerticalPerLine = 2;
    int distanceHorizontalPerColumn = 1;
    int splitCharacterDistance = 3;

+0 / -0

GBrankTheEloIsALie
8 months ago
(edited 8 months ago)

Deciphering is probably a bit harder...

Also, if your goal is to have a cipher that looks like keyboard mashing (i.e. consecutive encrypted letters are close on the keyboard), aren't you doing it the wrong way around? Your cipher transforms actual keyboard mash into "likely" English letters, not likely English letters into keyboard mash.

It should be possible to design a (also easily reversable) cipher that does that because typical written English is around 1.0-1.2 bits per letter, which even the highly redundant nature of a qweqrtasdga string should be able to encode.

(PS: https://xkcd.com/1530/ )
+0 / -0



CHrankAdminDeinFreund
8 months ago
(edited 8 months ago)

GBrankTheEloIsALie
it's very easy to decipher just start lagmcfiff vb tljfikaccjkvh kdut ltuphs znmcjkre ggg

I don't think you've completely understood what it actually does. To be taken into account your critic has to be written in Marvin :P
+0 / -0



CHrankAdminDeinFreund
8 months ago
(edited 8 months ago)

It comes with built-in message signing. Assuming a long enough text, every encoding of the same text is unique. This means except for random chance, nobody can create texts with your "signature" without having your private key. Decoding is deterministic and removes the added entropy.
+2 / -0


USrankAdminSprung
8 months ago
[14:59] Marvin1: haz aisr uadu zraiar

A beautiful example of Classical Marvin by a native speaker. You can tell it's not Vulgar Marvin because it does not parse using contemporary grammar.
+3 / -0

IMrankCovertMagic
8 months ago
From the thread's title, I expected this to be yuuioti, but it wasn't.

+0 / -0


ROrankSigero
8 months ago

+1 / -1

USrank[I]burp
8 months ago
(edited 8 months ago)

I'd suggest we write in ROT26 "encryption scheme" from now on, it's twice as secure as ROT13 and easier for me to decypher, since my brain is trained very well in it.
+3 / -1


ILrankAdminhokomoko
8 months ago
USrank[I]burp don't use the antiquated ROT26, it's been proven to have some serious weaknesses lately. I suggest using more modern encryption schemes such as ROT52 and ROT104.
+5 / -1

LVrankSenaven
8 months ago
Are he even real human?
+1 / -0

ESrankElTorero
8 months ago
(edited 8 months ago)

quote:
I'd suggest we write in ROT26 "encryption scheme" from now on, it's twice as secure as ROT13 and easier for me to decypher, since my brain is trained very well in it.

Had to use a web in order to decypher this message...
PS:
if anyone wants it, here you are

quote:
I'd suggest we write in ROT26 "encryption scheme" from now on, it's twice as secure as ROT13 and easier for me to decypher, since my brain is trained very well in it.

+2 / -0