Hey all,
Just something I decided to do in response to: http://zodiackillerciphers.com/wiki/ind … tle=Pivots
In the large picture I personally didn’t spot anything special between the two ciphers, although ofcourse the pivots are. Maybe this is another reminder that the 340 appears to be homophonic, schemed horizontally.
Yellow squares indicate repeats at any distance, orange adjacent. If wanted I could easily summon up images like these with repeats at specific distances, just let me know.
Repeats per grid line:
340 horizontal: https://www.dropbox.com/s/cvnp7xljfaui4 … h.jpg?dl=0
340 vertical: https://www.dropbox.com/s/c7d7q8282zi00 … v.jpg?dl=0
340 diagonal left: https://www.dropbox.com/s/qpi4d2s0okq44 … l.jpg?dl=0
340 diagonal right: https://www.dropbox.com/s/twtdavieho7fe … r.jpg?dl=0 EDIT
340 all directions: https://www.dropbox.com/s/bzpwv3xd2cd89 … r.jpg?dl=0 EDIT
408 horizontal: https://www.dropbox.com/s/0j0gbgihy9kw2 … h.jpg?dl=0
408 vertical: https://www.dropbox.com/s/eugwqdeidagwg … v.jpg?dl=0
408 diagonal left: https://www.dropbox.com/s/d4gj0824owk76 … l.jpg?dl=0
408 diagonal right: https://www.dropbox.com/s/faxc04e72nl46 … r.jpg?dl=0 EDIT
408 all directions: https://www.dropbox.com/s/uttkwd49b1vyf … r.jpg?dl=0 EDIT
I’m planning to add some more images later.
I’m sorry, there was a small error in my code not showing some repeats on the diagonal images. I marked the ones with EDIT in the original post.
I am thinking of building some sort of database of homophonic substitution ciphers to compare the 340 to in all sorts of tests. Symbol counts should be in the 54-72 range and if the cipher length is greater than 340 only a 340 part will be used.
If you happen to have such samples (I already have the samples that come with ZKDecrypto) please don’t hesistate to share them!
8 ciphers were tested, this is probably not enough to come to any conclusion. But I do want to note a few things. The pivots (which I call orientated "diagonal left") seem to fit nicely into the grand scheme, averages. And a possible tendency for the ciphers to exhibit a higher rate of diagonal left repeats than diagonal right repeats. I guess if true then this must be a plaintext thing, possibly local to english, and a possible indication of in which direction the plaintext was written.
How I counted the repeats in the following data: if a line has 2 symbols that are equal this is counted as 1 repeat, if a line has 3 symbols that are equal this is counted as 2 repeats, etc.
8 ciphers tested:
average symbol count: 62.875
total horizontal repeats: 206 (25.75 average)
total vertical repeats: 396 (49.5 average)
total diagonal left repeats: 282 (35.25 average)
total diagonal right repeats: 252 (31.5 average)
total repeats: 1136 (142 average)
————————————-
340.txt
————————————-
symbols: 63
total horizontal repeats: 18
total vertical repeats: 56
total diagonal left repeats: 35
total diagonal right repeats: 27
total repeats: 136
408.txt (top 340 part tested)
————————————-
symbols: 54
total horizontal repeats: 21
total vertical repeats: 45
total diagonal left repeats: 40
total diagonal right repeats: 30
total repeats: 136
340.ccactus.example.txt
————————————-
symbols: 63
total horizontal repeats: 24
total vertical repeats: 47
total diagonal left repeats: 37
total diagonal right repeats: 26
total repeats: 134
340.gardi.example.txt
————————————-
symbols: 65
total horizontal repeats: 37
total vertical repeats: 46
total diagonal left repeats: 42
total diagonal right repeats: 45
total repeats: 170
340.mikec.example.txt
————————————-
symbols: 63
total horizontal repeats: 39
total vertical repeats: 58
total diagonal left repeats: 31
total diagonal right repeats: 33
total repeats: 161
340.mikee.example.txt
————————————-
symbols: 61
total horizontal repeats: 26
total vertical repeats: 48
total diagonal left repeats: 33
total diagonal right repeats: 38
total repeats: 145
340.tonyb.example.txt
————————————-
symbols: 63
total horizontal repeats: 25
total vertical repeats: 57
total diagonal left repeats: 33
total diagonal right repeats: 25
total repeats: 140
340.tonyb2.example.txt
————————————-
symbols: 71
total horizontal repeats: 16
total vertical repeats: 39
total diagonal left repeats: 31
total diagonal right repeats: 28
total repeats: 114
I am thinking of building some sort of database of homophonic substitution ciphers to compare the 340 to in all sorts of tests. Symbol counts should be in the 54-72 range and if the cipher length is greater than 340 only a 340 part will be used.
If you happen to have such samples (I already have the samples that come with ZKDecrypto) please don’t hesitate to share them!
I could probably generate these all day long using a program, but would it be of help? I don’t know, but I don’t think it would. All of the example ciphers included with ZKD were made "by hand" (as far as I know) and were meant to confuse or stump the program, but it manages to solve them anyways…
If a line has 2 symbols that are equal this is counted as 1 repeat, if a line has 3 symbols that are equal this is counted as 2 repeats, etc.
When you say "a line" I assume you mean a line of 17 symbols, as the ciphers were written, correct? But is this important? When ZKD loads and solves a homophonic cipher, it deals with the entire cipher as one single line, one single cipher.
It really could be lines of 5, 10, 20, 30, 40, etc…, but the display of the program, or how the cipher was written really has no bearing on its solution, does it?
To be clear, I mean no disrespect to your work at all, I’m glad to see it, but I’m not sure if this is very useful. I think that ZKD can solve 340 ciphers with 54-72 symbols all day long. That multiplicity falls in the range of .1588 to .2117 which are easily solvable:
——————————–
I don’t believe in monsters.
Hey glurk,
I’m looking to build a "database" with a large variety. I wrote a program that generates homophonic and I will include a few of those with different settings and I’m happy to include some of other programs, just not a lot.
I’m not sure myself if this work will be useful, but I’m happy to do it. The main idea is to compare the 340 vs a wide variety of other ciphers.
Thanks for the info.
Hey glurk,
About the ciphers that come with ZKDecrypto. You said they were made by hand and to confuse, do you still remember the specific confusion scheme for each cipher?
Thanks
Hey glurk,
About the ciphers that come with ZKDecrypto. You said they were made by hand and to confuse, do you still remember the specific confusion scheme for each cipher?
Thanks
I have all of the keys to all of the included ciphers, yes. But they were not all provided with the program itself. I will send them to you, I still have your email address, I think.
I don’t remember all of the exact details of all of them, but I know one had no vowel "E" and another was a ‘nomenclator’ where some symbols represented entire words, etc.
-glurk
——————————–
I don’t believe in monsters.
