Zodiackillersite.com

marie wrote:So I wondered what is being used for the alphabet, allowing how many substitutions per letter.

My solver, AZdecrypt is totally free in this regard, whatever fits best. Btw, welcome to the cipher subsection of the forum!

Jarlve wrote:My solver, AZdecrypt is totally free in this regard, whatever fits best. Btw, welcome to the cipher subsection of the forum!

Thanks, I'm not the worlds best programmer, but have dabbled as I have worked in genetics and the newest datasets required I learn a bit.

I am assuming by your comment that it is totally free, that the program somewhat randomly iterates through various combinations until it finds a high scoring match to a known "dictionary."

I saw the link to the counts page, very helpful, and I see partly where I may have had miscounts, partly my bad hand writing notations. But I did not include the gibberish at the bottom, which may have affected it, and there is also a case where E is substituted to an S (and my notation confused my count), so yes I did do corrections that way, too.

I should also say I agree the FBI may be wrong, however when I play with the oranchak web toy, lines 4, 10, 14, 17, 18 quickly become gibberish no matter what you type elsewhere. I do think it is a substitution cipher, possibly with a different line order or pattern, but I also hesitate to think its too complicated as I doubt Z was too much of an expert on the subject.

Right now I can't play with your program as I have a Mac and easily run executables, though I have used workarounds in the past. Just out of curiosity, has anyone tried running the cipher through without those 5 lines?

-m

Hey marie

The solvers use n-gram data, which are the frequencies of any combination of n letters appearing in English text, they are compiled from many books.

That some rows are filler is a possibility I guess, though I feel maybe somewhat unlikely and it's what the FBI have been hinting at. I'm planning to re-run a test removing all combinations of up to 6 rows and another test with up to 5 columns, the reason why I won't go further than 6 and 5 is multiplicity (it's somewhat of a difficulty rating for our solvers and is simply the number of unique symbols divided by the number of total symbols/characters: 63/340 = 0.185 multiplicity).

Tried your suggestion but nothing shows up, I verified that a cipher with the same multiplicity and 5 lines removed was easily recoverable by the solver.

Code: Select all

limrtimewithstren
etscountbeaadeonf
caphisworeofsulla
heislyintrowsoria
timmaindentiarebo
ursofthewherealst
oreandunshealsose
ndisasomeachanbro
wsmaryitbactation
edhimatertriasoca
lbecausetrachohim
acehepretirewords
thallcoontmanifre
tainedbychancedth
ribleticlonintohs

HER>ÐÌ^VPËI²LTG±Ä
NÐ+B¢·OºDWy•<»KÆ£
BŸ„ÃM+uZGW¢£L·¤HJ
¸¼M+¤ÔÊÄIµFP+ˆ³Ë/
оR^FÌO-»ÄCËF>±D¢
·µ+KѺƒ±uÃXGV•¤LI
¢G±JÆÊ·O+¸Ny¤+¹L¼
Ä<M+Â+ZR±FBßA°³K
u+R/µÔEIDyBÐÂTMKO
±<ÃÌRJI»µT³M•+ˆBF
¤°¼SŸ·+NIµFB⃾R
ŸBX²»„³¼CE>VuZµ-+
IÕ´¤BK¢OÐ^•ÆMÑG±
IFËÄW<½ÔB¸yOB»-CÃ
>MDHNÐËS¤ZO¾AIKƒ+


You may seem words appearing but it's a low scoring solution and it's very probable that not much is there.

A different line order has been investigated on another forum I believe, basicly the solver doesn't care much what row order the message is in.

Thanks Jarlve!

I was just going with one of the FBI brilliant ideas to hopefully eliminate it. If I have any of MY own brilliant ideas, I'll let you know, but also set up my computer to run executables.

-m

One thing I noticed is the word "because" came out, which I thought would be in the cipher somewhere as Z used it three times in the first cipher, and he likes to make excuses. I haven't and won't have time till Sunday to see if it leads anywhere, but its kinda intriguing. -m