That looks really cool. So when the queue of operations is set up, the hillclimber will try various values for the input arguments for each of the operations? Or do you determine those ahead of time when setting up the queue?

Thanks.

This is the current design philosophy, but would like to change the wording of transposition to manipulation (which encompasses anything):

A problem we currently have with the 340 transposition hypothesis is that it’s hard to test specific transposition hypotheses/schemes, especially for those not trained in programming. The idea is that we can define a transposition scheme by its operations. So if you have a particular transposition hypothesis for the 340 you just select the necessary operations, throw a couple of test ciphers at it, if these solve proceed on to the 340. This way we will be able to rule out specific transpositions schemes more easily.

For the first version(s) the search mode will be random. I haven’t thought about the hillclimber yet and want to add multiple search modes, random, brute force, hillclimbing, etc.

There is a changeable minimum and maximum operation stack size (so that you can fix it or make it variable). The operation queue is a list that the search can use, in any valid order (unless the order is locked, see my previous post for that). For instance, the Reverse operation is not allowed to stack. The arguments are random and not predetermined, although there is a specific range of valid inputs for each operation. An example, with the Skytale operation it is checked if gcd(period,cipher_length)=1, etc.

I want to allow setting up custom ranges and values for the arguments but not yet for the first version(s). The first version(s) also will only be able to use dimensions native to the cipher length. If the cipher length is prime and you have selected a bunch of (x,y) operations then it will look like this: Manipulation operation(s): No (valid) operation available, No (valid) operation available, No (valid) operation available.

In time I’d like to add stuff different from transposition. Removal of characters, expansions of symbols, anything you can think of.

Everything I said is subject to change, whatever works best!

AZdecrypt 0.992: https://drive.google.com/open?id=0B5r0r … HJuYmU3Z2s

Added,

– Various statistics to the output files.
– Lots of structural improvements to the program.
– Ciphers up to 10.000 characters and 5000 symbols can now be processed correctly.
– The option to save or not to save log files (can get pretty big after a few days of processing).
– Cribbing functionality for numeric ciphers only (works on top of everything) see the included 5.txt for an example.
– The option to normalize scores by IoC or entropy, the latter seems to work better when the cipher is more randomized.
– New solver mode External and option Multiplicity weight, try with 1.3 to normalize the 340 between different cipher lengths.

– The first iteration of the Manipulation solver, can be found next to Start processing, it can solve a variety of transposition ciphers. Supported manipulation operations: Diagonal, Flip, Mirror, Offset, Offset rows, Offset columns, Period, Period rows, Period columns, Reverse, Rotate, Skytale. Click on the operations to add them to the operation queue, in the queue you can remove operations by clicking on them or lock them with the "l" key. Or also increase/decrease the solver’s tendency to select these operations by changing the ratio with the "+" and "-" keys. To activate the Manipulation solver you must click Start processing through the Manipulation solver menu, it will override any other Solver modes.

If you have any questions or problems let me know.

AZdecrypt 0.992 (Practical Cryptography 5-grams) Progressive, Index of coincidence

Score: 15906.63 Ioc: 0.07962867 Entropy: 3.852247 Chi-square: 114.9212 Characters: 340 Letters: 19

herceanbigivethem
helltoorseattoyeh
lousalsohetheolhs
feeanadobieollrye
seallifminailargu
edrnaaofomrgacest
onlycruessahbtlei
thesefoolshallsee
mtalslorealsomery
flasblnsloeatalof
slrunieisalestayo
etsarsiontratlala
leefoolhinalstudr
ahahnenerstsbrill
olavourerecbsonfl
istillytoenteache
rstlevertlaaelihe
llherlesiofhtuthe
iagmetailsaolofrs
cashhegfloodmiyul

Multiplicity: 0.1411765 Characters: 340 Symbols: 48

h e r c e a n b i g i v e t h e m
h e l l t o o 24 25 26 a 28 29 30 31 e 33
l 34 35 36 37 l 38 39 h 26 t 33 e o l h s
42 e e 43 n a 44 30 b 45 e o l l r 31 e
s e 37 l l 48 f m i 50 51 i l 52 53 g 54
e 44 r n 51 a o 55 30 m 56 g 51 c e 25 t
o 50 l 31 57 24 58 e 38 36 59 h b 28 l e i
t h e s e f o o l s h a l l s e e
m 29 37 l 61 l 39 r e 51 l 36 34 62 63 53 31
55 l a 38 b l n s l o e 43 29 51 l 34 55
38 l r 54 50 48 e i 25 a l e 61 t 37 31 o
e 29 36 a r s i 30 50 t 53 37 28 l 52 l 51
l 63 e 42 34 o l h i 50 51 l 36 t 58 44 r
a h 51 h n e 50 e 53 61 28 36 b 53 48 l l
34 l 59 v 30 35 53 e 56 e c b 38 39 50 55 l
i 36 28 45 l l 31 t o e n 28 e 37 57 h e
r 36 t l e v 63 56 29 l 51 a 26 l i 33 e
l l 33 26 56 l 26 36 i o 42 h t 54 t 33 e
i 51 g m 26 29 43 48 l s a o l 30 55 56 36
c 37 25 h h e g 42 l 39 o 44 62 i 31 58 l

Important bug fixes for the manipulation solver: https://drive.google.com/open?id=0B5r0r … HJuYmU3Z2s

And with some last minute changes I included a bug which would prevent the program from solving any transposition, because the cipher string was being reused all the time. Fixed some smaller bugs also and included the option to change the arguments. When an operation is in the queue hover over it and press the key of the argument you’d like to change. You could for instance change Flip(x,y) to Flip(17,20). If the program detects that arguments are unvalid then the "No (valid) operation" message will be generated.

x = dimension x
y = dimension y
t = (transpose=1), (untranspose=2)
a = the period or offset or other typical argument (1 to 4 for Rotate, 1 to 8 for Diagonal)

I deleted a post of mine in: viewtopic.php?f=81&t=3063 It was not so positive and I’m getting aggravated over nothing.

But in short, I’m leaving the forum and quitting my cipher work. It has been enough, best wishes to everyone!

I am sorry to hear you are leaving – I hope your absence is not permanent!

Be well, and I hope our paths will cross again.

I deleted a post of mine in: http://zodiackillersite.com/viewtopic.php?f=81&t=3063 It was not so positive and I’m getting aggravated over nothing.

But in short, I’m leaving the forum and quitting my cipher work. It has been enough, best wishes to everyone!

Jarlve I truly hope you just take a break. Don’t let that writer get you down. Writers write for effect and it was unfortunate that he chose to portray you in a manner that you are uncomfortable with. I do not think anyone here thinks less of you in any way because of that article. Whether or not it is solving ciphers or trying to catch the Zodiac Killer in another way we are all subject to the attitudes of those that think our endeavors are less than worthy.

I think it is important to embrace your passion, it’s what will give you satisfaction in the end.

sorry to hear youre leaving jarlve. the work you guys are doing is 90% of what keeps me interested in the zodiac case. i feel like you are sooooo close to an "ah ha" moment. i hope youll take a bit of time for yourself and then maybe return to continue your fascinating work.

sorry to hear youre leaving jarlve. the work you guys are doing is 90% of what keeps me interested in the zodiac case. i feel like you are sooooo close to an "ah ha" moment. i hope youll take a bit of time for yourself and then maybe return to continue your fascinating work.

Totally agree masootz ..the combination of smokie doranchak and jarlve seems to work brilliantly together. Throw in daikon and glurk for special comments and you have a real class act.
total respect to you jarlve

Jarlve-

I certainly hope that my comment here: http://zodiackillersite.com/viewtopic.p … 063#p48080 didn’t upset you.
Having been at this for a rather long time, I’ve found that keeping a sense of humor about the whole matter keeps one sane. It doesn’t have to be all-or-nothing.

Take a break, and please come back. Please. You are needed.

-glurk

Jarlve-

I certainly hope that my comment here: viewtopic.php?f=81&t=3063#p48080 didn’t upset you.

Oh come on. Who doesn’t like The Eagles?

But seriously, just take a little time off. Clear your head. You’ll be drawn back eventually.

I do not think anyone here thinks less of you in any way because of that article.

The Kernel piece? I might have missed something but it seemed to me the writer’s intention was to pay tribute to the cipher guys – it certainly didn’t come across as a negative portrayal to me.

Anyway, I can only echo what others have said here – and hope Jarlve will be back: Taking a break is not a bad idea, though – this case tends to wear one down.

It’s a pitty that you have left the forum Jarlve. I’ve read most of your postings since a couple of months and learned a lot from you. You gave me new ideas and motivation to keep working on the cipher.
I hope that you are just taking a break and come back. Sometimes it is good to keep back from things and start again some weeks later.

I really like your great tool "AZDecrypt" and it is one of the most important tools for everyone in this forum. It would be a pitty if the development on this tool stops. If you really decide to quit your work on the cipher it would be great if you can put the source code online. I am very familiar with C++, Python and a couple of other languages but unfortunately not with FreeBasic. But I would like to give it a try and implement a command line functionality for it. This would make me able to pass all my deciphering attempts directly to your fantastic tool and archive some sort of automation.

Best regards and all the best for you!
Largo