Jarlve, could you give me a hint? There is no distinctive spike, except maybe at period 6 reading right left top bottom. Could you tell me the period and transcription direction, and if there are nulls at the end of the message? Thanks.
Period 1, left-to-right, top-to-bottom. No hint toward the chunk size.
Here’s something else. I was looking at bigram repeats map. Images that show where the periodical repeats are and visually found a large repeating structure in the 408 at period 10. Here is a compiled image with in the middle the actual bigram repeat map at period 10. On the left and the right side the structure is offset on eachother to show the amount of repeats involved. If not a coincidence then how does this work?
Period 1?
You were describing a way to detect Largo’s columnar rearrangement scheme. I thought it would be interesting if we could come up with some statistic to do just that so I made such a cipher to see if we could. No extra periodical transposition or direction was applied. The message existed from left-to-right, top-to-bottom and then I chose a chunk size and randomized the order. Can someone figure out the chunk size? If so, then it would be interesting to see what it would yield on the 340.
Plaintext: AVADAKEDAVRA Chunk size 4, rearrangement order "3421" Plaintext: 1234 1234 1234 AVAD AKED AVRA Ciphertext: 4312 4312 4312 DAAV DEAK ARAV
I have tried to solve your cipher but had no success. I had tried chunk sizes 2, 3, 4, 5, 6, 7 and 8 (a total amount of 46232 permutations). Then I read the last posts again and recognized that you are talking about columnar transpositions. Just to clarify I want to make sure that we are talking about the same thing and so I have created a sheet which describes what chunk based transposition means to me. May you have a look and tell me if we are talking about the same thing?
Jarlve, maybe I have found some snippets:
icantdursintoa chinaetothewas mindstathehasf undermssalarec atragicallmesa theatechrongod seeidealthatas onstoletrapols ledourniinproh ecanortsmuchei thenistarehass igointerinside lceanumphastha rdforalromasco ntimeapondermi lrenounttherai neagliondersso npriothatsents ashairostsegin gandaustricere
Do you see some parts of your plaintext? Some columns are still missing, but I am trying…
Yes, we are talking about the same thing.
I was just wondering if anyone could figure out the chunksize somehow. I suggest not trying to solve it. You found no snippets, it matches the original plaintext by less than 1%. Nice looking sheet by the way, do you have a background in graphics? The use of complementary colors is a dead give away.
Here is an easy chunk based transposition cipher which you can try to solve.
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
This could be fun, can anyone find the chunksize I used?
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
Don’t reveal the answer yet. I am working on it. I think that there are at least 10 columns. Columns 5 and 10 should, when put back together originally, should sit next to each other. But I don’t know where. And columns 4 and 9, when put back together, probably sit next to each other. But I don’t know where.
Columns 1 and 5, when put back together, should sit next to each other also. So I think that columns 1, 5 and 10 should all sit next to each other.
hi guys.. ChunkSize .. word of the month, love it… I have been reading and trying to comprehend how it all fits together always thinking that its the 70s he’s an amateur and its from a book… I really like the concept of Chunks as it makes sense. I have been over scytale like a rash and have not had any great returns except its concept fits and suits the period scheme 19s real well.. chunksize might bring that back into play. I always thought it real messy a huge long string of single line writing wrapped around a stick.. Much easier to write it out and then cut it out into say five rows of four …chunksize… easier to handle and it will still change the code considerably when unravelled. also may be the explanation of a pivot.. anyways ChunkSize has given me a new path to follow and I will run some excel spreadsheets up and drop them onto jarlvies new programme. but time is not kind at the moment, im busy helping Mrs Clause get ready for our big day..
Cheers all.
Nice looking sheet by the way, do you have a background in graphics? The use of complementary colors is a dead give away.
Thank you. But I do not have a background in graphics. I use "Numbers" from the Apple iWork office suite which offers nice built in default colors.
Here is an easy chunk based transposition cipher which you can try to solve.
Thank you for the second cipher. I have found the solution by brute forcing a lot of permutations. Maybe I have an idea how to solve chunk based ciphers with large chunk sizes without brute forcing. More on that later when I am sure that my idea could work.
Solution: (Scroll down if you want to see it)
Chunk size 7 Order: 5, 6, 2, 4, 1, 3, 0 AZdecrypt 0.992 (Practical Cryptography 5-grams) Normal, Index of coincidence Score: 24644.59 Ioc: 0.06565388 Entropy: 4.078139 Chi-square: 44.98024 Characters: 336 Letters: 22 comfortableamonga llofthemladehalso notesinthesameint erviewobstinacyan dtenacitynotbeing afraidtogetembroi ledincontroversyt hatsverymuchaturk ishtraditionthats partofmycharacter tooicanbeverystub bornthatsprobably beenbeneficialfor thedevelopmentoff ullylogichedescri beshimselfasaname ricanmathematical lyorientedelectri calengineerofiran iandescentbor Multiplicity: 0.1755952 Characters: 336 Symbols: 59 .&LX?R0'ID*KL)(M! DDHA2C=LVNT@CBD+& >HS:+Y<WC=+!L*Y60 :,8O@5?I+9Y(BJ4N> 12@(K/O046&9I:Y>P !XZ'OTW)U*2:LI7&Y D=1O>-)(SRH8@,+40 CN9+F:Z4L3.C!2;7E Y+C0RN%Y0OH<SCB9+ [!,0&AL4/CBZNJ9*7 0H)O-K6I@F=R4+0GI I?,(9C!2+"Z&INID4 I*=>I:<@XO.YKDA)Z WC:Q=8*DH$L@6S?XX 3V#4D&MYJC*T:+/7O I=+CYL+@DA'+N6!L: RO-'>L!0C*LN9O-BD D4),Y:<2@Q=D=JWZO .KD@>PY(=:RHXO7K< Y'>T*+.=6SI?,
Mr lowe:
I am glad that you like the concept of chunk based transposition. Sometimes I turn off my computer and just take pen and paper to see the things from Zodiacs perspective. His first cipher was broken very fast and I am sure he was upset. I also think he read in the newspapers how the Hardens solved it (repeating bigrams and so on). What could he have done to prevent the people to break the new cipher by using the same technique? Of course … Vigenere, Bifid, Plairfair or whatever would make the cipher much more secure. But there are many more and even simpler techniques which can do the trick. So I came up with the idea of chunk based transposition. You said that you think that this could be an explanation for the pivots. Can you explain that in more detail? Sounds interesting!
Well done Largo. I hope I’m not distracting you too much from your other work.
I’ve been looking at the encoding again and ran all combinations of removing 2 rows from the cipher and scored the encoding. The highest peak occured while removing rows 12 and 14 (pivot rows). Then ran all combinations of removing 2 columns from the cipher. The highest peak occured while removing column 8 and 13 (pivot columns). Could anyone confirm this find?
In this plot, for the left square, the x-axis is the first row removed and the y-axis is the second row removed (stacked on eachother). For the right square, it is the same but with columns. Note the even/uneven bias for removal of rows.
Here is how it looks when formatting the 340 in a 10 by 34 grid.
It finds rows 20 and 24, exactly the positions of the horizontal pivots. Wow!
HER>pl^VPk |1LTG2dNp+ B(#O%DWY.< *Kf)By:cM+ UZGW()L#zH JSpp7^l8*V 3pO++RK2_9 M+ztjd|5FP +&4k/p8R^F lO-*dCkF>2 D(#5+Kq%;2 UcXGV.zL|( G2Jfj#O+_N Yz+@L9d<M+ b+ZR2FBcyA 64K-zlUV+^ J+Op7<FBy- U+R/5tE|DY BpbTMKO2<c lRJ|*5T4M. <--- +&BFz69Sy# +N|5FBc(;8 RlGFN^f524 b.cV4t++yB <--- X1*:49CE>V UZ5-+|c.3z BK(Op^.fMq G2RcT+L16C <+FlWB|)L+ +)WCzWcPOS HT/()p|Fkd W<7tB_YOB* -Cc>MDHNpk SzZO8A|K;+
I’ve been looking at the encoding again and ran all combinations of removing 2 rows from the cipher and scored the encoding. The highest peak occured while removing rows 12 and 14 (pivot rows). Then ran all combinations of removing 2 columns from the cipher. The highest peak occured while removing column 8 and 13 (pivot columns). Could anyone confirm this find?
What did you score them with? AZdecrypt?
Very interesting finding. I wonder if you would see the same effect if you randomly rearranged the symbols in the selected rows or columns instead of deleting them.
I’ve been looking at the encoding again and ran all combinations of removing 2 rows from the cipher and scored the encoding. The highest peak occured while removing rows 12 and 14 (pivot rows). Then ran all combinations of removing 2 columns from the cipher. The highest peak occured while removing column 8 and 13 (pivot columns). Could anyone confirm this find?
What did you score them with? AZdecrypt?
As stated, the test is scoring the encoding. Which is my 2 symbol cycles measurement and another thing added.
Very interesting finding. I wonder if you would see the same effect if you randomly rearranged the symbols in the selected rows or columns instead of deleting them.
Good suggestion, here are the results. The color brightness is the encoding score. Each images is made up of small 10 by 10 pixels subsquares, each sub subsquare are 100 different randomizations of the same intersection.
Image 1 (rows by rows):
X-axis are the randomized rows positions 1 to 20. Y-axis are the randomized rows positions 1 to 20.
Row 14 gives good returns.
Image 2 (columns by columns):
X-axis are the randomized columns positions 1 to 17. Y-axis are the randomized columns positions 1 to 17.
Column 13 very clearly shows good returns. The diagonal shows increased brightness because randomization of 1 column (overlapped) on average is better than scores of 2 randomized columns.
Image 3 (columns by rows):
X-axis are the randomized columns positions 1 to 17. Y-axis are the randomized rows positions 1 to 20.
Column 13 pops out again and there is no sign of interaction.
This loosely seems to confirm my test for row and column removal a few posts above although there were was not much return from row 12 and column 8. Perhaps the disturbance does not affect the entirety of the row and/or column.
