Week 8 - Puzzle 10 Discussion

[Mel-o-rama]

As requested - this thread is for discussing Puzzle 10 (so as to not spoiler this for people who haven't attempting it yet).

[Jonas Grumby]

I want to see the puzzle!

[Ultimate Anyone?]

As requested, the infamous Puzzle 10:

Spoiler:

No instructions.

KMJVSDDMNMRHDBKZRBSM

UKGIVYMK
HLLHJMEHJCYP
DUKMCHISUBHYYBHTOI
DDNJHIGERJRURPNFCVVDAADBGZAREEAGHTDBORM
ENSBAUNWUWZTDNFZA
WZRFCFZGOGZYHNFC
TCDDTBLZNYMTTWCHDZGMEMH
MUMFRJRDARA
THMRIESKCNWAVRRZ
MABWGOLT
HBNLWPLLFAKKN
HKUSDNDFRRFJN
ABMDSBVJMVFBD
NPAHBFLMGDMUKICGVSVFLPAILFPSEPVKVKIGPA
MWJZKFACYWZLZSGKJSGK
AZSOKSM
RMHMUCMHDSMHFWOCKCHO
IFZPMSABGFHAERPHPI
BMCIUSPRKZ
ZRFGNSZKDBZKUCPNJNKP
DBFITTNSF
LBZGFMBZHCZAIHNB
JZVHCHOXPMVDBVLMRF
ZPKTFMCFCAJRCUHHVMT
SBNIBRTBZUBZEK
ZPFLMCTBUG
RJECPTCGPLCFP
MADNLWKJDFNM

Link to end of the game

[Mel-o-rama]

Hmmm - I'm already seeing patterns I didn't see before. Interesting.

[SirVLCIV]

DUKMCHISUBHYYBHTOI

DUKe, MiCHigan, ISU, Brigham Young....

[Mel-o-rama]

I see a lot of double letters. No triple letters. Also, there are quite a few palindrome sequences like CKC, VKV, GOG.

Also, I just noticed that every row has at least one roman numeral in there.

[SirVLCIV]

Originally Posted by Mel-o-rama

I see a lot of double letters. No triple letters. Also, there are quite a few palindrome sequences like CKC, VKV, GOG.

Also, I just noticed that every row has at least one roman numeral in there.

I noticed when analyzing for pure cryptogram that there where VERY few two letter combos that repeated (like a normal 'th').

[Mel-o-rama]

Good observation - that's strong evidence that a simple "cryptogram" is not the case.

Here are some other random observations - none of which reveal an awesome pattern - but who knows - maybe you guys will see something:

Double letters: DD, LL, YY, DD, VV, AA, EE, DD, TT, RR, LL, KK, RR, TT, HH

Palidromes: MNM, HLLH, RJR, RUR, DAAD, WUW, FCF, ZGOGZ, MEM, MUM, RJR, ARA, DND, FRRF, VSV, VKV, KVK, ZLZ, MHM, CKC, PHP, NJN, HCH, CFC

Roman numeral letters in each row:

MVDDMMDM

IVM / LLMC / DMCII / DDICVVDDDM / D / CC / CDDLMCDMM / MMD / MICV / ML / LLL / DD / MDVMVD / LMDMICVVLILVVI / MCL / M / MMCMDMCC / IMI / MCI / DC / DI / LMCI / VCXMVDLM / MCCCVM / I / LMC / CCLC / MDLDM

I may have missed something - so if you think you see a pattern, but not quite - check to see if I missed something.

Also, M is the most frequent letter, but I think X only shows up once.

[Abstract Actuary]

Some other things I noticed/considered:

-I think there are going to be multiple steps (decode, . . . )
-There is a line of "instructions" or "title" then 28 words.
-The instructions line is 20 letters.
-The biggest line is 39 letters, there is one other very large one, 3 20s, multiple 18s, mutiple 16s.
-There are a lot of Z's, but a normal distribution of Q's and X's. It makes acronyms less likely.
-Too many different double letters for a straight cryptogram to be likely.
-Row 17, pattern "MH" repeated 3 times.
-The double letters at the start is not a good sign for straight cryptograms.
-SSH likes AO history - I looked into first posts on the AO, AOTH thread, AOTH2 thread, SSH's friends, SSH's first posts, SSH's first threads, Poster names.
-I looked into translating to morse code, then graphing? then making some sense out of it
-Translate to Braille?
-Change to numbers based on A=1 or phone pad (but then what?)
-Could be cryptogram, acronym, letter shift (subset of cryptogram).
-I think it is one of the above, but not directly. How about something like shifting the first letter 1 (forwards or backwards) and the second letter 2, etc. etc.
-Could be some sort of elimination technique (place letters into a crossword and word search away letters, cross out pairs (somehow?), etc.) and eventually you are left with something legible.

[Mel-o-rama]

Shot in the dark - a variation on the phone pad idea: take each letter and replace it either with itself or another letter on the same keypad?

AZSOKSM could be: CYROKRO

Hmmmm - maybe not.