mcb
Junior Member
Posts: 68
|
Post by mcb on Mar 6, 2019 19:39:29 GMT -5
Any progress?
|
|
|
Post by fennster on Mar 7, 2019 20:51:59 GMT -5
Nothing on my end. I haven't really tried recently but I want to see progress. I love my solve and what I saw when I was BOTG. Hoping that this has some sort of confirmation.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 8, 2019 15:50:34 GMT -5
Nothing on my end. I haven't really tried recently but I want to see progress. I love my solve and what I saw when I was BOTG. Hoping that this has some sort of confirmation. Prob should ask a 12 year old. But depending on how smart the child is it could be a 7 year old or it might have to be a 16 year old. Come children get it - some do not it just varies.
|
|
|
Post by zaphod73491 on Mar 23, 2019 0:44:30 GMT -5
A report on my latest (failed) approaches. With the impending reveal from Jenny of the keyphrase length, I asked myself: "What would you do with this information if you had it now?" Simple: if it was less than 80 letters, that would mean some portion of the key is repeated. That provides leverage into cracking the Vigenere. So I wrote software that allowed me to explore all possibilities from 20-80 letters, looking for a "peak" vs. keyphrase length. (It's definitely more than 20 letters long, otherwise my code would have solved the cipher immediately.) The only minor peaks were at 25- and 50-letter lengths, so I'd say there's a small chance the keyphrase length is 50 letters (with 25 as a harmonic). A keyphrase that long relative to the message does not provide enough leverage to crack the cipher. That leaves crib-dragging of appropriate phrases from Jenny's book.
I've tried over a hundred Fenn-relevant phrases from the book, and none yielded a set of letters that had decent English letter frequencies. I didn't just check full phrases--I checked any 20-letter subset of each phrase against any position in the cipher text. And 25-letter segments, and 30-letters: more than enough that a match would stand out like a sore thumb. Nada.
Clearly the keyphrase is not amongst the 100+ I typed into my database. So I'm going to have to type in more of the book (of course skipping chapters that have nothing to do with Fenn or the Chase).
|
|
|
Post by fennster on Mar 26, 2019 14:33:59 GMT -5
A report on my latest (failed) approaches. With the impending reveal from Jenny of the keyphrase length, I asked myself: "What would you do with this information if you had it now?" Simple: if it was less than 80 letters, that would mean some portion of the key is repeated. That provides leverage into cracking the Vigenere. So I wrote software that allowed me to explore all possibilities from 20-80 letters, looking for a "peak" vs. keyphrase length. (It's definitely more than 20 letters long, otherwise my code would have solved the cipher immediately.) The only minor peaks were at 25- and 50-letter lengths, so I'd say there's a small chance the keyphrase length is 50 letters (with 25 as a harmonic). A keyphrase that long relative to the message does not provide enough leverage to crack the cipher. That leaves crib-dragging of appropriate phrases from Jenny's book. I've tried over a hundred Fenn-relevant phrases from the book, and none yielded a set of letters that had decent English letter frequencies. I didn't just check full phrases--I checked any 20-letter subset of each phrase against any position in the cipher text. And 25-letter segments, and 30-letters: more than enough that a match would stand out like a sore thumb. Nada. Clearly the keyphrase is not amongst the 100+ I typed into my database. So I'm going to have to type in more of the book (of course skipping chapters that have nothing to do with Fenn or the Chase). Holy cow! My hat is off to you. That is more than I could imagine. The end of the month can't come soon enough. I'll be happy when we can put this one to bed. With any luck it will confirm even just a little bit my solve.
|
|
|
Post by fennster on Mar 29, 2019 10:24:37 GMT -5
Tick tock. Counting the hours.
|
|
mcb
Junior Member
Posts: 68
|
Post by mcb on Mar 30, 2019 7:03:14 GMT -5
Tick tock. Contando como horas. Minutes ... Espero que valha alguma coisa.
|
|
|
Post by crm114 on Mar 30, 2019 10:50:20 GMT -5
Hi,
Completely new here, and just starting on this, but I coded it up in a spreadsheet to play with. Awaiting the key length hint. The transposition at the end is both a blessing and a curse. It seems like the puzzle would likely be set up to jump back and forth, meaning there are some hints to both the key and the message, as opposed to brute force key trial. The transposition spreads letters out, so you can guess at words in either, if you have a chunk of the other piece. So I like the reverse engineer idea someone had way back on page 12 (edit:Ironwill, P. 12) and coded functions to go forward and back in steps, allowing guesses at letters by hand.
Anyway, a couple of ideas spring up I haven't seen posted: - the message is likely signed "f" or "ff." Certainly not much help, but it's a start. - "treasure" is hidden in every puzzle? Is it hidden in the Vigenere decryption step? If so, this would generate hints to both key and message. - plotting the transposition number vs it's index yields some patterns that catch the eye and might be a good place to start. The transposition doesn't look totally random to my eye. Why not random? The puzzle was constructed in reverse, most likely. Start with f's message, transpose, maybe part by hand (hiding treasure?) then randomly, then Vigenere encrypt, etc.
|
|
|
Post by zaphod73491 on Mar 30, 2019 12:32:32 GMT -5
Hi crm114: good thoughts. There is a decent chance the message ends with a single letter f since that would imply that either the 5th letter of the keystream is a T, or the 48th key letter is an A, depending on whether Jenny's 80 codes are addresses or indices. (Jenny's 5th code is 80, and the 80th code is 48.) She could have done it either way. The only guidance we have on that is the magic square for ordering the puzzle pieces, but no guarantee that she maintained consistency.
|
|
|
Post by zaphod73491 on Mar 30, 2019 12:50:52 GMT -5
I ~have~ tried crib-dragging hundreds of Jenny/Fenn-relevant words across the ciphertext, looking for decent "hits". If TREASURE is in the keyphrase, then one of the more likely positions for it is the 12th through 19th letters since this yields plaintext letters of GVEEIEDH (and if it's TREASURES plural, then it yields GVEEIEDHN). However, if TREASURES starts at the 12th letter of the keyphrase, then the keyphrase is over 60 letters long because TREASURES doesn't produce another good "hit" in a later position.
|
|
|
Post by crm114 on Mar 30, 2019 13:13:54 GMT -5
Thanks, zaphod
Maybe I am confused. I thought Jenny confirmed the process (on page 13 or so, I think) as:
Step 3: Vigenere decrypt the 80 character string "SVRWYULOKKOZMIEAYULFUTIITYWBLBHKAVCAZUAUMWXCLLQFRMJMYPJLSVLCUSOKLLICTXBXACUHRBVG" using the key phrase.
Step 4: Descramble the result of Step 3 using the 80 numbers "41 10 73 22 80 28 76 6 32 53 61 19 1 36 56 23 65 40 13 67 29 2 21 45 39 3 79 51 49 24 27 11 17 69 4 71 50 5 30 57 25 7 60 35 12 20 8 70 52 26 62 42 14 54 31 9 33 72 55 37 15 58 75 66 43 74 68 59 16 46 34 18 77 63 44 64 78 38 47 48"
I know about the other thread where the Descramble is applied to the "SVRWY..." before Vigenere decryption as an alternative, but it seems like Jenny has discounted that. Is that what you are talking about? Maybe the inverse of the descramble is used in Step 4?
Edit:nevermind, your post was clear upon reread.
|
|
|
Post by crm114 on Mar 30, 2019 13:22:22 GMT -5
Zap,
Good info. So I was thinking of exploring (crib dragging I guess you call it - total noob here) not with "treasure" in keyphrase or message, but in the Step 3 result - thus hidden treasure? Treasures are usually found at the end.... Have you looked at that?
|
|
|
Post by zaphod73491 on Mar 30, 2019 13:34:28 GMT -5
Crm114: ah, so you're wondering if TREASURE (or some other likely Fenn word) is in HIS message rather than Jenny's keyphrase. To be honest, I don't think I've tried approaching the problem from that angle since I figured I was much more likely to guess Jenny's keyphrase (or words in it) than words in Forrest's. Only slightly more difficult to investigate, but perhaps if TREASURE or TREASURES *is* in there, then it will be obvious which is the correct descrambling method.
|
|
|
Post by crm114 on Mar 30, 2019 13:46:19 GMT -5
Well, I am looking for "treasure" or other possible words in 3 places: key phrase, the ff message, and the result of Step 3. The latter one is a twofer if it exists - you get letters in both the message and the keyphrase that way. I wouldn't be surprised if treasure exists in all 3.
|
|
|
Post by zaphod73491 on Mar 30, 2019 14:41:37 GMT -5
According to GeneticBlend, apparently not. She said there were no confirmers in the scrambled result of step 3, other than after doing the final descramble. Too bad -- Jenny could easily have done this (e.g. have an obvious word pop up at the beginning of the output of the Vigenere decryption).
So anyway, I've tried looking for TREASURE in Forrest's message to see what the corresponding Jenny phrase letters would be. In descramble method 1, the best hit is at position 58: ESVEDAUT. In descramble method 2, the best hits are at position 66 (ROEVNGEN) or 39 (ADOMTART). If I expand the crib to "THETREASURE", the best locations are:
Method 1, position 14: HEESDIITBDH Method 2, position 43: SNEEECLHHAG
Unfortunately, not enough proof to decide which descrambling method is the correct one, or even if TREASURE is actually in Forrest's message.
|
|