Zodiackillersite.com

Mr lowe wrote:your work is awesome doranchak ..Thanks for taking the time... trouble is for me I don't understand if what you put up is good bad or indifferent.

Yeah, I don't understand either. Can't really make a solid conclusion based on this.

Mr lowe wrote:doranchak can you run it through a solver in all directions to see if anything pops up when you get a moment.

Yes, I'm planning to generate many variations of your scheme to see if anything turns up.

doranchak wrote:

Mr lowe wrote:doranchak can you run it through a solver in all directions to see if anything pops up when you get a moment.

Yes, I'm planning to generate many variations of your scheme to see if anything turns up.

OK I generated over 2.5 million variations and scored them based on ngram counts, trigram counts, and presence of "mr lowe patterns". I retained the 20,000 that had the best scores. I'm running them through azdecrypt now. Should be done in about 2 days if all goes well. The downside is there are still many variations that aren't going through azdecrypt so there's always a chance of missing a diamond in the rough.

In the mean time...as the computer ticks over do you think any merit in me making a scytale but breaking the code into two halves then joining back together.

Something else I have noted when I make a physical scytale I wrap the first line `loop` under as `it seems more natural to make` rather than wrap over as most pictures show. not sure if it delivers a different outcome.

I think that is worth a try. If you describe your method I can try to generate many variations to throw at AZdecrypt for the next batch. No interesting results from AZdecrypt yet for the current batch. I think it is almost halfway done with the batch of ciphers I threw at it.

doranchak.. I take it from your non posting in this thread that we ended with a no show in the results from the solver..
That string is intriguing to me, something along the same lines as the pivots.
I will ponder some more on how they could be inter conncected.
cheers

1 19 37 44 11 56 8 60 31 18 35 7 49 30 59 40 63
55 20 5 40 23 38 18 34 20 42 37 51 58 19 20 29 16
46 3 57 22 16 5 14 32 5 19 32 39 19 12 30 1 22
16 61 7 10 28 22 31 21 37 8 8 26 33 19 10 17 6
6 24 26 23 52 25 55 4 22 39 45 9 4 13 2 20 19
30 50 10 28 13 5 36 6 17 17 15 19 53 34 43 48 55
36 27 62 34 5 19 6 16 46 36 51 17 47 7 24 23 51
43 15 33 44 31 49 3 23 13 31 41 50 41 19 41 11 29
40 16 15 19 19 9 27 13 3 54 29 38 7 25 21 19 53
21 40 5 23 15 5 19 16 11 3 21 38 8 51 51 40 47

"170" (top half 340) scy 21? columnar reformatted back to 17x10
Can you turn this into symbols so i can run it through crypto scope
the reasoning behind 21..it brought the period 19s in utilising my color formatted spreadsheet.
and below the bottom half

38 36 34 50 56 7 26 21 36 16 47 7 53 23 51 14 55
19 40 51 30 31 29 42 20 31 6 59 40 16 9 27 62 31
34 28 13 26 20 23 14 20 11 14 56 43 40 5 19 44 3
33 26 10 27 37 21 19 19 10 18 11 14 20 50 55 11 25
48 30 50 36 39 15 4 54 41 18 61 8 37 33 19 6 33
16 2 28 20 33 29 42 32 47 5 6 54 56 63 18 35 21
19 1 30 58 15 12 20 56 23 23 11 23 20 45 50 36 46
44 37 51 36 19 40 48 39 61 52 3 36 26 29 42 20 28
58 19 19 17 5 25 50 36 48 19 51 1 2 50 51 8 50
16 37 50 11 11 28 38 57 13 3 3 19 50 4 32 11 5

Mr lowe wrote:doranchak.. I take it from your non posting in this thread that we ended with a no show in the results from the solver..
That string is intriguing to me, something along the same lines as the pivots.
I will ponder some more on how they could be inter conncected.
cheers

It took a while to collect all the results in this spreadsheet: https://drive.google.com/drive/folders/ ... sp=sharing

But yeah, no strong results to speak of, apart from a handful that scored higher than average.

Each of the 20,000 ciphers in the spreadsheet is produced by applying the following decisions in sequence
1) Reverse the entire cipher text. Or leave it alone.
2) Rewrite it to some new grid width between 1 and 34.
3) Rotate the cipher 90 degress. Or not.
4) Flip the cipher horizontally. Or not.
5) Flip the cipher vertically. Or not.
6) Apply scytale, with a width between 1 and 34.
7) Rewrite it to some new grid width between 1 and 34.
8) Rotate the cipher 90 degress. Or not.
9) Flip the cipher horizontally. Or not.
10) Flip the cipher vertically. Or not.

Azdecrypt seems to produce higher scores for transformations that result in higher numbers of n-gram repeats. So, the longer "Mr Lowe" patterns don't appear among the top 400 results.

Take a look at the spreadsheet and let me know if you have any ideas. I'll try to look at your newer variation when I get a chance.

That string remains intriguing to me as well.

I have spent a few hours reading through those results. It will take me a few days to get through it all. Some little snippets pop up but not sure if they mean anything. I am going to look for sentence structures.
cheers

Yeah, me too. I will look through the results. Thanks for all of the work that you do, both of you.

The solver is very powerful, and can find a small handful of actual words in an otherwise imperfectly untransposed message. One with disconnected message chunks. I am wondering if the same word or words are showing up where the same symbols are grouped together, even if not in the same location of the untransposed message. Consider reverse transposing the solutions, and comparing them to find out if the same symbols frequently map to the same plaintext, or if the same words keep appearing in the same locations.

I am not sure how big of an undertaking it would be to repeat the experiment with a test message. Or maybe you already have. If you want a simple test message, with the answer provided in advance, let me know. One idea would be to make a simple transposition message, but with some filler. Like, say, positions 324-333 or whatever. My feeling is that a large portion of the message is exactly what we think it is, but there is a portion of filler or untransposed message, because 340 is not a multiple of 15 or 19. Or we could do whatever you want, at your leisure. Thanks.

Mr lowe wrote:1 19 37 44 11 56 8 60 31 18 35 7 49 30 59 40 63
55 20 5 40 23 38 18 34 20 42 37 51 58 19 20 29 16
46 3 57 22 16 5 14 32 5 19 32 39 19 12 30 1 22
16 61 7 10 28 22 31 21 37 8 8 26 33 19 10 17 6
6 24 26 23 52 25 55 4 22 39 45 9 4 13 2 20 19
30 50 10 28 13 5 36 6 17 17 15 19 53 34 43 48 55
36 27 62 34 5 19 6 16 46 36 51 17 47 7 24 23 51
43 15 33 44 31 49 3 23 13 31 41 50 41 19 41 11 29
40 16 15 19 19 9 27 13 3 54 29 38 7 25 21 19 53
21 40 5 23 15 5 19 16 11 3 21 38 8 51 51 40 47

"170" (top half 340) scy 21? columnar reformatted back to 17x10
Can you turn this into symbols so i can run it through crypto scope
the reasoning behind 21..it brought the period 19s in utilising my color formatted spreadsheet.
and below the bottom half

38 36 34 50 56 7 26 21 36 16 47 7 53 23 51 14 55
19 40 51 30 31 29 42 20 31 6 59 40 16 9 27 62 31
34 28 13 26 20 23 14 20 11 14 56 43 40 5 19 44 3
33 26 10 27 37 21 19 19 10 18 11 14 20 50 55 11 25
48 30 50 36 39 15 4 54 41 18 61 8 37 33 19 6 33
16 2 28 20 33 29 42 32 47 5 6 54 56 63 18 35 21
19 1 30 58 15 12 20 56 23 23 11 23 20 45 50 36 46
44 37 51 36 19 40 48 39 61 52 3 36 26 29 42 20 28
58 19 19 17 5 25 50 36 48 19 51 1 2 50 51 8 50
16 37 50 11 11 28 38 57 13 3 3 19 50 4 32 11 5

OK - here you go:

Code: Select all

H+M8|CV@KN:^j*Xz6
-BpzOUNyBSMF;+B<2
_Rq#2pTfp+fZ+1*H#
2b^k.#K(MVVW)+kdl
l%WO&D->#Z3P>LEB+
*5k.LpclddG+4y7t-
cYAyp+l2_cFd9^%OF
7G)8KjROLKJ5J+J|<
z2G++PYLR/<U^D(+4
(zpOGp+2|R(UVFFz9
Ucy5C^W(c29^4OFT-
+zF*K<SBKlXz2PYAK
y.LWBOTB|TC7zp+8R
)WkYM(++kN|TB5-|D
t*5cZG>/JNbVM)+l)
2E.B)<Sf9pl/C6N:(
+H*;G1BCOO|OB35c_
8MFc+ztZb&RcW<SB.
;++dpD5ct+FHE5FV5
2M5||.UqLRR+5>f|p