Sign in to follow this  
buckysrevenge

New stuff on Tumblr!

Recommended Posts

SGGCSNTSSSFEESRWOTOSOYGL

LESMDETNOEDRTSOOOTSTGERO

AFFESAEEENTETEONEYRTGDTY

EDNEESYEYTGEDNRLRIIETEST

RSYEGUERRE.OMMM.MC...EGR

RW.RFHWWE.HHMRMCHHHFWW..

.A.M...IIE.SSNEEEEIARNN.

..HHLWHKHNMLHRNKSVRLRHWV

DMLBDVRTHFBTBEERHERVDHLD

HVDTGSYMN.OOO...N...SNSR

.NNNNONNEAAAAOII.TW...TT

TTTTWWTTTTTTP..GG.FFLW..

EHVHNTTT.H.DD..RW.IALLUA

ZZB.E.OIPB..BEAANRI...RR

UO.OE..OEEOOUAOOOO.OIIOI

IIIAOA.OAAAATTSDGSTTTL..

.LNFSSII.PW..CL.IFEVFFFP

LN.YH.MCCRAU....EEELEAEO

TTTGUOUUEE..EOOOORUKAAEY

ESIILAINWIIUSNN.L..EOOAU

..SOUIAASASUEIUNSI..$...

.O.YE.IAAA.....X...OFTF.

GGSBAJOOBPPAAOOOAAE....L

........OEMLMBAH..ANZZUU

I don't even know where to start with this one.

The full stops aren't consistently spaced with a substitution cipher, but there's probably a key hidden somewhere.

Share this post


Link to post
Share on other sites

It's a 24x24 layout. Here's maybe a better representation of that, but I don't know if it helps.

puzzle.png

puzzle.png.701ddbf791c0879a71b8a1f5d51b9

Share this post


Link to post
Share on other sites

It looks like a transposition cipher, but again that would need a key.

The thing is, there are entire lines that have no spaces in them going horizontally, so even if it wasn't a 24 column cipher there would be super long words. It might be worth rotating the grid 90 degrees and going by row instead.

Share this post


Link to post
Share on other sites

I don't really have a clue about how to solve these kinds of puzzles, but one thing I noticed when browsing the new Tumblr is that the combination "fi" is always replaced by a "dot". Could that helped in finding a key?

Share this post


Link to post
Share on other sites
It looks like a transposition cipher, but again that would need a key.

The thing is, there are entire lines that have no spaces in them going horizontally, so even if it wasn't a 24 column cipher there would be super long words. It might be worth rotating the grid 90 degrees and going by row instead.

I think you're right about the transposition cypher - and yeah I wonder if its been transcribed in columns and then switched to a different divisor. The first three lines having zero spaces certainly implies that its not meant to be read that way and vertically the dots generally only small clusters rather than tons at a time.

Potential number of columns in the solution are the divisors of 576 which are 1 | 2 | 3 | 4 | 6 | 8 | 9 | 12 | 16 | 18 | 24 | 32 | 36 | 48 | 64 | 72 | 96 | 144 | 192 | 288 | 576

(sidenote this post was making the same point as Pyradox as he was editing his post and now I've edited mine like 5 times, getting a balance between keeping trains of thought clean and just splurging random unsorted thoughts over the thread for posts at a time is tricky!).

Share this post


Link to post
Share on other sites

No number of columns results in anything that looks like English or more subsitution cypher-y vertically (there's always double-spaces and/or triple letters), I think we're on the right track but its something a bit more complicated.

Also to confirm - there's almost certainly not a substitution cypher involved - the letter frequencies are in the vicinity of the English average.

e : 56

o : 46

t : 41

a : 37

s : 31

n : 29

i : 28

r : 27

h : 21

l : 20

u : 16

g : 15

w : 15

f : 15

m : 14

d : 13

y : 10

b : 9

v : 7

p : 6

c : 6

z : 4

k : 3

x : 1

j : 1

q : 0

Share this post


Link to post
Share on other sites

Yeah, I think that's definitely the case.

What interests me is the $ sign, but lack of numbers. If this is all written out in plain text then it wouldn't make sense to have it there, but if it's possible to have numbers then why wouldn't there by any?

It might be a clue as to the language being used.

Share this post


Link to post
Share on other sites
Yeah, I think that's definitely the case.

What interests me is the $ sign, but lack of numbers. If this is all written out in plain text then it wouldn't make sense to have it there, but if it's possible to have numbers then why wouldn't there by any?

It might be a clue as to the language being used.

Yeah I've been wondering about it - I'm silly and my first thought was that it was the name of a variable and that maybe the message was about some cool thing we can mess with in the game when it comes out; until I realised that was making the bizzare assumption that the game has been rewritten in Perl or something :P.

It's possible it is against a number but that number is written in words and the dollar sign was just a shortening to make the message the right length.

Share this post


Link to post
Share on other sites
I don't really have a clue about how to solve these kinds of puzzles, but one thing I noticed when browsing the new Tumblr is that the combination "fi" is always replaced by a "dot". Could that helped in finding a key?

What's weird is the "-" is visibly there, but it's "fi" in the source (and if you copy and paste it, it pastes as "fi"); must be some css hoodoo I guess. The only words where the fi are replaced are first, final, and finding (the only appearances of that letter combination other than "Fine" in Double Fine, which is not changed).

Share this post


Link to post
Share on other sites

Re: "fi," it sounds like this is merely a mis-handling of ligatures, certain combinations of letters (including "fi") that are sometimes rendered with a special character in the font. If the browser is trying to render "fi" as a ligature but the font is missing that character, it'll appear as a dot.

http://en.wikipedia.org/wiki/Typographic_ligature

http://aestheticallyloyal.com/public/optimize-legibility/

https://medium.com/typography/baeff3613070

So probably unrelated to the puzzle.

(I don't see a dot for fi in Chrome on a Mac. The font spec appears to be 14px/22px share_tech_monoregular, "Lucida Console", Monaco, monospace, sans-serif. share_tech_monoregular: http://www.google.com/fonts/specimen/Share+Tech+Mono )

Share this post


Link to post
Share on other sites

This reminds me of the scrambled tome in Fez that I (to my knowledge) was the first one to solve...

If the letters are showing up as the same frequency of English language then they're jumbled in some specific way. But the nature of the jumbling here is weird. It doesn't look like a simple scrambling algorithm has been used that can be easily reverse engineered, like the FEZ book. The block of text has too many distinctive features. For example:

There are a LOT of repeated letters - e.g. 6 Ts in a row. If this was being scrambled in a normal way, getting 6 of the same letter in a row would be very rare, but here it happens twice in quick succession, on the letter T:

.NNNNONNEAAAAOII.TW...TT

TTTTWWTTTTTTP..GG.FFLW..

And even in that small excerpt there are several other repeated letters.

So we have an interesting paradox - letter use that suggests scrambled english text seem to rule out simple substitution ciphers or something like a Vigenere cipher, but patterns in the text that make it appear unlikely that it was scrambled in a way that can be reduced to a simple de-scrambling algorithm. If it is scrambled, it has to be something more subtle than that.

Share this post


Link to post
Share on other sites

It had a solution? Nice.

Anyway, it's often the case with a transposition cipher that the columns are re-arranged according to a certain key - so 12345 may become 51423 for example. The problem is, we would need a 24 (or however many columns there really are) digit numeric key with no repeated numbers, and it's not apparent where we would find one of those.

The pictures all appear to be pictures rather than hidden .zip files, and the "Introducing hack n slash post" on the tumblr doesn't really have any obvious 24 letter phrases that don't have any repeated letters that we could convert into a column order.

That being said, I don't think it's possible that that's all there is to it.

If we assume that the line with the $ sign is a number, not a variable, the only one we can make out of the letters on that line is "ONE". In which case the letters above and below the "$ONE" don't match up to any sort of discernible English - you've got combinations like IOOU, DBRB, FYX. and stuff. This would suggest that both the rows and the columns would need rearranging before we can get anything discernible, but there's no suggestion of how to do that, or in what order.

I also tried rotating the grid 90 degrees and creating "$TWO" out of the letters there, which is the only possible number you can make from the letters in that column. It went equally well as there's a .S.A and a .T.U underneath it.

So either the $ sign is a variable and not a number, or there's another layer to the puzzle that we don't have yet. Maybe both.

Share this post


Link to post
Share on other sites

Having no obvious key from the tumblr page, I thought it *might* be a grille cipher, but the only potential grille we might have is the QR code from the previous puzzle and it's only 21x21, leaving three columns and rows out. Hrm...

Share this post


Link to post
Share on other sites

I still think it's more subtle than that. If it was just rearranged letters via some simple algorithm, it wouldn't contain so many repeated letters. And if it were a grille cipher, there'd be no reason for the letters to correspond so closely to what would ordinarily be expected, unless they deliberately did that, which is POSSIBLE, but I doubt.

I think the letters are rearranged but I haven't got the slightest idea what method would get them to be rearranged into a state that is so... well, orderly. It's not completely orderly, but it TOO orderly.

This is the original:

SGGCSNTSSSFEESRWOTOSOYGL

LESMDETNOEDRTSOOOTSTGERO

AFFESAEEENTETEONEYRTGDTY

EDNEESYEYTGEDNRLRIIETEST

RSYEGUERRE.OMMM.MC...EGR

RW.RFHWWE.HHMRMCHHHFWW..

.A.M...IIE.SSNEEEEIARNN.

..HHLWHKHNMLHRNKSVRLRHWV

DMLBDVRTHFBTBEERHERVDHLD

HVDTGSYMN.OOO...N...SNSR

.NNNNONNEAAAAOII.TW...TT

TTTTWWTTTTTTP..GG.FFLW..

EHVHNTTT.H.DD..RW.IALLUA

ZZB.E.OIPB..BEAANRI...RR

UO.OE..OEEOOUAOOOO.OIIOI

IIIAOA.OAAAATTSDGSTTTL..

.LNFSSII.PW..CL.IFEVFFFP

LN.YH.MCCRAU....EEELEAEO

TTTGUOUUEE..EOOOORUKAAEY

ESIILAINWIIUSNN.L..EOOAU

..SOUIAASASUEIUNSI..$...

.O.YE.IAAA.....X...OFTF.

GGSBAJOOBPPAAOOOAAE....L

........OEMLMBAH..ANZZUU 

And this is what the same letters, scrambled by an algorithm look like:

N.VGITOAOSSAILP.S.AN.WGE

ALS.ESGSLA....H.ONLSAAV.

BO.ESTCNTTES.FYTUOHVDFO.

YEIHMNT.RD..DDOIHRU.TNTE

YG.GSAE.USIEUEY..SHOASIA

.R.OUNUNTFFLIUTINWSUTMGA

W.HRU.O.LTLTEIDEV...OORU

RIBDEE.AZIEOHRNLOPSRTBTC

.TBOVAS.IK...OOB.NJAGTH.

R.OAAR..RPLNEAAIUIEIO.OM

.AWT.E.HIZNWDABSF.ABEUEY

TPWNREA.SSN.TO.NESOA..OT

$ATEOAEWFHOTIAOFNRTENGEH

AHH..OMNF.TAGOEOEOMHIFHE

ODTWLIIOAH.VHENFA..ERNUD

LPF.SAGRIHENTYTCLSEME.R.

M.TOOTO.REA.GFTTYWIT..AD

UEROMNFPSIEYLC.ES.AEIE.I

M.ZKFIR.SCEEH.SGNW.TY.HT

ZEOOMN..A..HAAER..ST..

AVNG..OME.OOOT..ORN..T.G

EOW.OEL.IDTERYT.LTLIG.ER

R..DKCEWMERRESEBN.SWS.RL

FEW.U.MBNL.MWL..UEESELO.

Note how repeated letters are relatively rare, and the periods are all over the place. We're miiissing sommmeeething

Share this post


Link to post
Share on other sites

I was. That was an interesting read because the heart pieces are where I gave up on Fez. I've finished everything else, decoded the languages and everything, but those things stumped me.

Anyway, I checked the glyph room concept art and it just says "The Five Boxing Wizards Jump Quickly", just like the final image (only spelled correctly unlike last time). And that's 31 characters so I'm not sure if there's a way to make a key out of that. I've also read a few methods of using transposition cipher keys with repeated letters, but without a 24 letter key (if that's what's required) I'm not sure that'll help much.

I'm also pretty sure there's no sequential series of letters that can make a number after the $ sign, assuming again that it's not a variable, which would imply some sort of vertical shift instead of just horizontal.

Share this post


Link to post
Share on other sites

At first I thought it's likely to be code, just because of the number of periods. There are too many of them for that to represent ends of sentences, but there could conceivably be that many in a block of code.

But then you wouldn't expect the only symbol to be $, and the frequency analysis would likely not match so closely. So maybe instead . stands for space, or something else.

Share this post


Link to post
Share on other sites

Yeah, I've been kind of assuming that they represented spaces for that reason, but that also runs into the problem of the top three lines having no spaces.

Share this post


Link to post
Share on other sites

maybe the web address is some clue?

80219964612/sggcsntsssfeesrwotosoygl-lesmdetnoedrtsoootstgero-a

note the dash before the last a, not present in the text.

Share this post


Link to post
Share on other sites
Yeah, I've been kind of assuming that they represented spaces for that reason, but that also runs into the problem of the top three lines having no spaces.

I don't think that's a problem. I think it's fairly clear that the whole text has been scrambled, if it's been scrambled, so if the spaces are represented by . then the dots could be anywhere at all.

The real puzzle is what scrambling algorithm could be used that would clump letters and spaces together in the way that we see them in the text.

Here's one thought. It could be scrambled along these lines, though perhaps not exactly:

Take the first letter of the first word, then the first of the second word, etc, until you reach the end.

Then take the second letter of the first word, etc, until the end.

Whenever you reach a space, put . instead, and then the next time it goes around, skip that word entirely.

The reason I like this one is that it makes partial sense of the way that so many (but not all) letters are clumped together, and the . for spaces, for the following reasons:

- You would expect it to be a while before any spaces appeared, because the first time around it would just be the first letter of every word. Indeed, that's the case, if . is space.

- You'd expect spaces to become more common as the text goes through, which it also does.

- Because vowels often appear in the same place in words, you'd expect to find long strings of them, which we do: e.g. UO.OE..OEEOOUAOOOO.OIIOI

IIIAOA.OAAAA

- Similarly for consonants.

I haven't been able to find an approach like this that works, so far, but I wouldn't be surprised if this is something like where we need to be looking.

Share this post


Link to post
Share on other sites

Oh yeah I didn't include the dots in my frequencies above - there's 107 of them or 17.9% of the text which is pretty consistent with the frequency of spaces in English (at least I believe that's about 17%ish, don't have a source to quote for that on hand).

I think KestrelPi is on the right tack here, there's something a bit more different from a simple transposition cipher going on here. Though I also think the space distribution being more like English reading downwards rather than across is potentially significant.

To satisfy my feeling that straightforward substitution won't cut it I hacked up a quick script and left it running overnight to see if I could get anything from changing the width of the grid and moving the rows around to produce something readable column-wise and didn't get anything that looked even close to English. Of course even with hyper-aggressive pruning logic, combinatorial explosion meant that I could only thoroughly test between 1x576 and 24x24, but frankly if the correct configuration is 32x18+ it would have to have a key hidden away somewhere to qualify as anything other than cruel and unusual punishment. :P

Share this post


Link to post
Share on other sites

Perhaps the $ represents the start or end of the sequence and is there to give us a starting point of some kind. Something else to think about.

Share this post


Link to post
Share on other sites

Maybe the fez tome approach would work here too. Can anyone think of a distinctive word that has cropped up in the puzzles so far that might be in here? Preferably a reasonably long word, with at least one rarer letter in it.

Share this post


Link to post
Share on other sites
maybe the web address is some clue?

80219964612/sggcsntsssfeesrwotosoygl-lesmdetnoedrtsoootstgero-a

note the dash before the last a, not present in the text.

I'm pretty sure you can't customize tumblr urls, so it's probably just using the - to represent a line break. I don't think that's part of the puzzle, and I'm not sure anything else on the tumblr is, otherwise I would definitely have expected the glyphs on the concept art to have even a hint of a new message.

Anyway, the only recurring, distinctive words I can think of are either common lettered or short - Incantation, Glyphs, Wizards etc. We could also try looking for Vertical and Horizontal like in the AF video - that would seem to have some relevance here.

Share this post


Link to post
Share on other sites
Oh yeah I didn't include the dots in my frequencies above - there's 107 of them or 17.9% of the text which is pretty consistent with the frequency of spaces in English (at least I believe that's about 17%ish, don't have a source to quote for that on hand).

I think KestrelPi is on the right tack here, there's something a bit more different from a simple transposition cipher going on here. Though I also think the space distribution being more like English reading downwards rather than across is potentially significant.

To satisfy my feeling that straightforward substitution won't cut it I hacked up a quick script and left it running overnight to see if I could get anything from changing the width of the grid and moving the rows around to produce something readable column-wise and didn't get anything that looked even close to English. Of course even with hyper-aggressive pruning logic, combinatorial explosion meant that I could only thoroughly test between 1x576 and 24x24, but frankly if the correct configuration is 32x18+ it would have to have a key hidden away somewhere to qualify as anything other than cruel and unusual punishment. :P

Hmm, if there's 107 spaces, there's 108 words. This might be useful info.

EDIT: hold that thought, I'm onto something, maybe!

There are actually 103 dots, implying 104 words. If we assume that $ is another non-space character separating 2 words, that brings the estimate up to 105

It's rather a coincidence - or possibly not - that there are 106 characters before the first . appears, if the block of text is read left to right:

SGGCSNTSSSFEESRWOTOSOYGL

LESMDETNOEDRTSOOOTSTGERO

AFFESAEEENTETEONEYRTGDTY

EDNEESYEYTGEDNRLRIIETEST

RSYEGUERRE.

I think this STRONGLY suggests that the scrambling of the letters is linked to positions of letters within the words.

But what position in the words are these first 100ish letters? Probably not the first. You'd expect a few more Ws, way more As, it doesn't fit with the letter frequencies for first letters in words.

I'm not sure, but here might be a clue. About 50% of the words in the english language end in E, T, D or S.

SGGCSNTSSSFEESRWOTOSOYGL

LESMDETNOEDRTSOOOTSTGERO

AFFESAEEENTETEONEYRTGDTY

EDNEESYEYTGEDNRLRIIETEST

RSYEGUERRE.

55 of these letters are E, T, D and S. I think this first block of text is the last letter in all the words.

To back this up, here are the most common letters to end words, with the most common first:

E, S, T, D, N, R, Y, F, L, O, G, H, A, K, M, P, U, W

And here is the same for the above short block:

E, S, T, O, R, G, N, Y, D, L, F, A, I, C, M, U, W

A pretty close match, especially for such a short block of text. And I suspect the I crept in there because the I's are just the WORD 'I', which isn't counted in an analysis of last letters.

This might be a useful resource to unpick the rest of this: http://scottbryce.com/cryptograms/stats.htm

Share this post


Link to post
Share on other sites

That would help to explain the weird distribution - meaning we just need to figure out what significance the "packs" of letters that follow them are. A lot of them have too much repetition to be the remaining set of letters in the word, but there is a noticeably lower percentage of ETDS as you go further down.

Share this post


Link to post
Share on other sites
That would help to explain the weird distribution - meaning we just need to figure out what significance the "packs" of letters that follow them are. A lot of them have too much repetition to be the remaining set of letters in the word, but there is a noticeably lower percentage of ETDS as you go further down.

Well, I think that's to do with the fact that there's much more likely to be certain letters at certain points in a word.

For example, in the above sentence if I just took the second letter of each word (or a space for 1 letter words) I'd end up with:

E.HHOOIHAHHUOIOEEETEON.O

That looks an awful lot like the sort of strings we're seeing in the ciphertext, right? More repetitions than you'd expect if it was simply jumbled text, but not easily predictable.

The task here is made harder in that if we're to assume that they're counting from the LAST letter backwards, because all the words are different lengths, the first letters of all the words are going to be variously spread throughout the ciphertext, so that the first letters of short words will be near the start and the first letters of long words would be at the end. I suspect it's a little more subtle than that...

... but at this point I'd be amazed if it wasn't true that:

* The . represents a space

* The cipher is to do with positions of a letter within a word

* The first 100 or so letters in the ciphertext are the last letters in all the words of the plaintext

Share this post


Link to post
Share on other sites

More supporting evidence. The most common 3rd letters in an english word are:

ESARNI

Because of the problem above, we'd expect those ESARNIs to be spread about around the text, if we're counting from the end backwards, but there are still a lot of 3 or 4 letter words in the english language so we should be able to find a block of text in our cipertext which has a lot of ESARNI in it. Here's one possibility from the ciphertext:

SSNEEEEIARNN.

..HHLWHKHNMLHRNKSVRLRHWV

DMLBDVRTHFBTBEERHERVDHLD

HVDTGSYMN.OOO...N...SNSR

.NNNNONNEAAAAOII

I feel like we're close.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this