QT,

Why are you so resistant to accepting the solution? The FBI has and many numerous amateur cryptographers have. You seem intent on casting doubt on a verified solution.

Quicktrader, since you seem to not understand the reading pattern from available resources, I have tried to come up with a simple but complete "visual explanation" using doranchak’s Cipher Explorer. Short video is available here: https://streamable.com/5qh0tn

At first, 15th row is corrected according to the proposed solution. Then the deciphered text is highlighted letter by letter following the pattern.

You can play around with the demo yourself. Load doranchak’s Cipher Explorer at http://zodiackillerciphers.com/cipher-explorer/ and select Z340 Plaintext from the menu. Then open development console in your browser (by pressing F12), paste the code from https://pastebin.com/raw/4g3rLyG1 there and the highlighting sequence then begins.

The demo itself is not of much quality, because I just put it together really quickly, but I think it is sufficient for a simple explanation.

QT (and anyone else)-

There will NEVER be a simple 1:1 explanation of this cipher, because Zodiac did not create a simple 1:1 cipher. I quote from Nick Pelling’s excellent "Cipher Mysteries" blog:

* At the end of the top line, he wrote the words “LIFE IS” in his homophonic cipher (not transposed)
* At the end of the bottom line, he wrote the word “DEATH” in his homophonic cipher (also not transposed)
* The rest of the bottom two lines were written in his homophonic cipher, but with the letter-order of some of the words reversed (e.g. the Zodiac’s signature word “PARADICE” was written as “ECIDARAP”)
* The remainder of the cipher was split into two nine-line blocks, where the order of the letters was transposed using knight’s move reordering.

There was also, annoyingly, a single row of ciphered letters in the middle that were out by a single column, which skewed all the text passing through it. It’s fairly easy to see that once you’ve got everything else right, though.

So the thing was split into 3 uneven parts, of 9 lines, another 9 lines, and 2 lines. Transpositions were performed on these 3 parts, slightly differently in each part. The result of ALL THAT was then encoded homophonically.

Whether Zodiac’s "mistakes" in spelling and transposition were intentional, or accidental, is unknowable. The combination of all these elements is undoubtedly what made the whole thing so difficult to solve.

One cannot have a "simple" explanation of something that was created in a multi-stage complex manner.

Also, a disclaimer by me. I had nothing to do with this decryption, although I completely understand it. All credit to Dave and team.

QT (and anyone else)-

There will NEVER be a simple 1:1 explanation of this cipher, because Zodiac did not create a simple 1:1 cipher. I quote from Nick Pelling’s excellent "Cipher Mysteries" blog:

@Glurk – Are you joking? What’s the problem with writing CLEARTEXT letters above each HOMOPHONE?? Like the Harden’s did? Am I missing something?

@Chaucer – Because I have not SEEN it yet? If you fully trust FBI in checking the cipher, go ahead, I don’t have to do the same (prefer to use my own eyes/brain).

@Yxovec – THANKS a lot. You are the first one showing the approach of how the cipher shall actually be read (in detail). However, the illustration skips into letters at the beginning (1:1…..glurk). Fully ok (great illustration, btw). I will check now for myself if those (cleartext) letters actually do match the (original) cipher structure itself and how many errors occur over the whole cipher.

QT

The ZKDecrypto image that I posted previously shows the plaintext and the ciphertext.

This is AFTER the transpositions were applied. It’s a multiple-stage cipher construction, which obviously requires a multiple-stage decryption.

The transpositions are a precise, step-by-step mathematical process that can be repeated. Then that un-transposed result is decoded as a homophonic cipher. And in this case, every single character is used. No filler in this one at all.

Besides being confirmed by the FBI and every single knowledgeable cryptographer that I have seen comment on it, the plaintext references a then-current event, and sounds exactly like something Zodiac would write.

Nothing was added or taken away in the cipher. It was transposed, and homophonically deciphered. Not sure what else I can tell you.

The ZKDecrypto image that I posted previously shows the plaintext and the ciphertext.

This is AFTER the transpositions were applied. It’s a multiple-stage cipher construction, which obviously requires a multiple-stage decryption.

The transpositions are a precise, step-by-step mathematical process that can be repeated. Then that un-transposed result is decoded as a homophonic cipher. And in this case, every single character is used. No filler in this one at all.

Besides being confirmed by the FBI and every single knowledgeable cryptographer that I have seen comment on it, the plaintext references a then-current event, and sounds exactly like something Zodiac would write.

Nothing was added or taken away in the cipher. It was transposed, and homophonically deciphered. Not sure what else I can tell you.

"and sounds exactly like something Zodiac would write" helps as much as the FBI not having solved the cipher for more than 50 years.

All I asked for was to see cleartext vs. homophones plus the method how it shall be read..

It took until this morning until somebody provided this input. I had not the time to check but I for sure will not trust on ‘every single knowledgeable cryptographer‘. And yes, there are single cryptographers who have not commented on it.

ASA having looked into the SIMPLE input – now finally provided – which is ALWAYS required for any potential solution, I’ll either congratulate heartfully (doesn’t look bad for this one) or dismiss the approach – for myself. But please, EXCUSE that I’d like to first follow the idea of how it is assumed to be decrypted instead of believing the FBI or an anonymous crowd of congratulants or some text Z ‘could have said‘, thank you.

QT

Apparently the best methodology was just whatever required the least amount of tinkering. The FBI was willing to accept this even if it wasn’t accompanied by things like statistical significant null hypothesis tests or the ability to be falsified. Seems a lot of those high bar standards got scratched as long as whatever worked to get something coherent out of it all worked.

50 years of legit cipher methods failed.

If people here feel that some criticisms used to reject other cipher solves have been abandoned to allow this one in the door then you would probably be right. Once they did it they just had to accept that a lot of robust theories and methods don’t seem applicable here. So yeah feel miffed at how the goal posts got moved. However accept it’s probably the most likely solution we have.

The take home message from it isn’t the solved content. The take home message is there is no evidence the Zodiac serial killer understands anything more than how to muck up a cipher attempt so that it longer can meet the standards we have for legit ciphers.

I don’t think the FBI appears to have used alternative methods of validation at all. They just accepted it is what it is.

The point I’ve been trying to make here is that there was no "manipulation" at all. Not a single symbol was added, nor subtracted. Spelling corrections were made, but only after the solution was found.

The team figured out the transposition Z used, un-transposed it to its pre-transposition order, and ran it through a homophonic solver. That’s it. That’s all.

Completely and absolutely unlike starting with a message and trying to force it in there, this message was already in there, and was found.

I simply don’t understand how people cannot see that.

To put this another way. If anyone wishes to verify the solution for themselves, they can:

1) Take the 340 cipher (on a computer, or on paper)
2) Perform the mathematical transposition sequences as have been explained
3) Apply the key that was given

Steps 2 and 3 are interchangable. Do either one first.

And you get the cleartext. No random anagrams, no adding or subtracting letters, nothing. Just a pure step-by-step process. And with this cipher, it seems all 340 characters are decoded. Not even a bit of filler – which surprised me, in fact.

Max 5 errors?

Zodiac made way more than that in the 340. Spelling and encoding errors. Clear mistakes. How can you give a solution without acknowledging those (more than 5) errors?
Or, what does your request mean? Technically, the Z340 solution has no errors. It just fixes mistakes in the original code.

To put this another way. If anyone wishes to verify the solution for themselves, they can:

1) Take the 340 cipher (on a computer, or on paper)
2) Perform the mathematical transposition sequences as have been explained
3) Apply the key that was given

Steps 2 and 3 are interchangable. Do either one first.

And you get the cleartext. No random anagrams, no adding or subtracting letters, nothing. Just a pure step-by-step process. And with this cipher, it seems all 340 characters are decoded. Not even a bit of filler – which surprised me, in fact.

Couldn’t say it any better.

AZ Decrypt is running 600 000 permutations to see if can find something.

1:32 – Oranchak gets a gibberish message on one hit. He would normally ignore but he likes some words in it.

The pattern they are looking at for the Z340 is one down, two across and to keep going, even across to the left side from the right, so that you cover each symbol in that section. Then write out those hits as a new cipher.
They plugged in that rewritten cipher and got gibberish.

Oranchak used a crib grid to manually put in some of the sentences he liked from the hits at 1:32

This supplement produced at 3:22 brings up his input plus AZDecrpts processing of what he calls a really ‘clean’ message. That’s subjective as it has plenty of noise within it also but he likes ‘wasn’t on the TV show part’ of it because he can line that up with the Dunbar TV show which also featured reference to a gas chamber (one of the words Oranchak put in through crib grid).

So from there he has a key of 21 letters of the alphabet (4:46).

He uses this key on the remaining cipher and gets noise with some words he likes.

He skips the middle section for now because the key produces noise.

So he skips to the last two sentences of the 340 and uses the key there. He then rearranges it by selecting what looks like some straightforward words and reverses others to get the sentence at 6:10

Oranchak is addressing the middle section (7:55). He suggests a mistake for the middle part that is producing some noise. So they have to shift a line in the original Z408 in the middle.

It seems to me that even after they did that they still got a lot of noise in with some results. So they had to find another way and decided to omit "life is" in the top right hand corner from their reading rule of one down, two across (8:10).

8:23 has the result. It is as clean as they could get it for a garbled text. He then performs some ‘minor corrections’ which appears to be subjectively correcting what’s still garbled into some sort of functional real words you get the result at 8:27

They add in ‘new life’ to try and make more sense of it.

If you go back to 4:19 you will notice the end reads "SEND ME TO PAY ALL CE ALL THE"

Oranchak changes PAY ALL CE to PARADICE. So that part now reads "SEND ME TO PARADISE ALL THE"

At 9:04 they add the 3 parts together.

You get what you get there.

A rule with the least amount of fudging to build something legible from all of the cipher and not just some of it.

I don’t know how the FBI validated anything. My best guess is here is that they checked what Oranchak did and just went with his word on it being the best so far that won’t be beaten.

I don’t know if this presentation is even falsifiable. I don’t know if it can even meet criteria for developing some sort of statistical significance out of it. I don’t see how null hypothesis testing can be applied. So whatever academic journal will publish it will obviously have to do many things to explain why this won’t be meeting many of the demands expected of a true cipher solution.

IMO this solution is part Criminology using forensic psychology (subjective) and part cryptology (objective). Which is actually what is expected given the peer-review for "serial killers and codes" is practically non-existent or so small as to make the field mostly new original research.

If more people knew the bar for a code pass wasn’t set so high I am sure you would have had more people allowing themselves to manipulate stuff on top of whatever rule they found to get something readable. Since this got a pass, it seems they still can.

BDHOLLAND-

What you have described is the very messy process the team went through to obtain the solution. Solving ANY cipher is ALWAYS a tricky problem, with dead-ends, false leads, and etc.

But at the end of the process (of figuring out exactly what Zodiac did) you end up with a two-step solution. Apply a mathematical TRANSPOSITION and a homophonic SUBSTITUTION. And you get the cleartext. You can do it by hand and get the same result. That message is in there.

Now keep in mind, this solution is only a week old. I guarantee it will be written up in longer articles, research papers, and any cryptographer that takes an interest will go over it to check it themselves. In fact, here is a video of the author of Cryptool doing exactly that:

https://www.youtube.com/watch?v=hIrOftXgibg

He finds it to be correct, BTW. That’s Nils Kopal. Here’s his credentials:

https://scholar.google.de/citations?use … AAAJ&hl=de