19

The text is:

ntmilEmondt enae
Leala ed hThtrc 
smve till pg   n
tut ane una olas
hrertwil incma t
ieeenak ioo anTl
es otah foer del
smeormresr  nmhi
m  f ugrpsmci io
 thmmoT tir luay
beoa   a nfsoner
t oris sihhsvlad
sb fengtfned ewr
 irnssortaa htam
ervegs  hntgdn i
 senly  ia eoftn
bito itofshirude
 trsecolucfia ra
 phef fid tyaene
thowuoyffi w  ro
nde sr pnfct ta 
 T f  et e a lir
 fhiout ioilimos
utortru  tn  c t
nce  raero  ls n
osenn  bo  Ee ml
Elt  stigsimtet

It comes with the following riddle, which is part of the intended form of solving the puzzle.

Where Air meets fire
and land marries sea
head south and east
Therein lies the key

I'm pretty sure it's a transposition cypher, based on letter frequencies, but beyond that, I have no idea.

I know that the end result plaintext is a spell in 5th edition D&D, which means it's (probably) in the format:

[Title]

[Ordinal] level [spell school]

[Description]

The [spell school] is Transmutation, and I know that the spell level is a high one, so the [Ordinal] is probably "Sixth" or higher ("Ninth" being the highest possible). This was learnt from in character knowledge. There's also a well above zero chance that the first word of the [Title] is "Elminster's" or "Elminster".

Does anyone know any good analytical techniques for solving this sort of transposition cypher, some possible cyphers it could be, or perhaps , way of solving it through the "intended" method using the riddle?

Thanks

Someone asked a question as an answer, and then deleted it (obviously, using an answer to ask a clarifying question is not the way to go about things). The question was something along the lines of 'is it a spell from the D&D 5e PHB?'

The answer was no, probably not. It's a spell for 5e, but this is the third of three puzzles, and the first two were homebrew spells, so I suspect the third will be too. The previous two's names both started with "Elminster's", which is why I suspected this one might as well.

I used some in-character knowledge to get a clue. Apparently "fire" refers to heating the scroll, and "air" refers to the blank spaces in the text. When the physical scroll was heated* the following map appeared. The blue dots weren't actually there, I've just added them to make clear where the blank spaces are. For reference, the map is of the Moonshae Isles, in Faerûn (part of the Forgotten Realms D&D campaign setting).

Apologies for for the fact that some information was required that would have been impossible for anyone here to obtain on their own. It never occurred to me that such information would have been part of the puzzle.

* unfortunately, the invisible ink had faded by the time I tried this, so it didn't work. The above image is of a separate copy of the scroll, with the text superimposed over it digitally.

Jim Cullen
  • 293
  • 1
  • 6
  • Is the space behind each line intentional? as in, was it in the file you got? – DrunkWolf Dec 16 '15 at 09:09
  • Actually, never mind. The spaces aren't there in the source. It's an oddity with SE's rendering of code blocks. – Jim Cullen Dec 16 '15 at 09:13
  • yes i noticed that just now – DrunkWolf Dec 16 '15 at 09:13
  • you can try the tools at http://rumkin.com/tools/cipher/ for some clues – CaldasGSM Dec 16 '15 at 10:36
  • 3
    The letters in the first four lines can spell "ninth level Transmutation spell". – f'' Dec 16 '15 at 10:55
  • Yeah @CaldasGSM, I've used them as much as I can, but unfortunately I haven't been able to find anything that helps. – Jim Cullen Dec 16 '15 at 11:50
  • 2
    the first 2 line of the riddle can refer to para-elementals. smoke (air and fire),ooze (earth and water) – CaldasGSM Dec 16 '15 at 12:44
  • May be useful: http://puzzling.stackexchange.com/questions/5905/what-characteristics-of-a-ciphertext-can-be-indicators-of-a-particular-cipher – A E Dec 16 '15 at 14:10
  • @AE unfortunately, most of the techniques shown there seem to be for determining types of poly-alphabetic or substitution cyphers. A frequency analysis of the above text reveals that it is almost certainly a transposition cypher: all the letters used represent themselves, just in the wrong order. – Jim Cullen Dec 16 '15 at 14:45
  • I think somewhat more than just transposition is involved. For example, the chiper does not contain the capital A. – wythagoras Dec 22 '15 at 21:01
  • Why would it need to contain a capital A? – Jim Cullen Dec 29 '15 at 09:16
  • Based on this map could the place "Where Air meets fire and land marries sea" be Ash Lake? – Patrick Roberts Jan 18 '16 at 10:08
  • Which things in that image appeared and which ones were there beforehand? (black shapes, reddish outlines, small grid squares, 5x5 squares) – f'' Jan 25 '16 at 06:30
  • The original was nothing but a grid of text as transcribed above. The grid was not there, and is only shown on this version as a guideline — it's not part of the puzzle. I'm waiting on word regarding the reddish outlines and the green shading. The only thing that is definitely supposed to have shown up is the dark black of the islands. – Jim Cullen Jan 25 '16 at 07:05
  • "Air meets Fire" sounds like smoke or embers, and "Land marries sea" sounds like coastline. – ASCIIThenANSI Jan 28 '16 at 21:30
  • @ASCIIThenANSI At least part of "air meets fire" has already been solved. Fire referred to physically heating the page (see the picture for what that resulted in — a map), and air supposedly refers to the blank spaces in the text. My best guess at the moment for "air meets fire" is places where there is both a blank space and land on the map. I agree, coastline seems most likely for land marries sea. – Jim Cullen Jan 29 '16 at 02:49

1 Answers1

6

I haven't fully worked it out, but the spell appears to be:

Elminster's Elemental Embodiment, Level Eight, Transmutation

And the solution's technique appears to be related to:

Sliding rows and columns "South" and "East" (wrapping around letters that go off the end).

For example, the following image shows a partially corrected (by trial and error) solution:

partial solution

In it, I've just:

shifted columns south by 1, 1, 0, 0, 2, 1, 1, 3, 3, 1, 2, 2, 3, 1, 1, 0 cells respectively. You can see fairly clearly the first three or so lines, as well as the last several ("This effect lasts for up to a minute, or until concentration is lost").

I don't have more time to work on this, so I'll just dump some final notes here, for others to pick up from:

1. The errors in my transcription clearly seem to correspond in some way to the positions of the islands. I.e. it gets more "corrupted" towards the centre where there are more islands.
2. However, there's still other issues, such as the missing "e" in "Elemental", and "t" in "Embodiment".
2. You clearly need to shift some rows East before (and/or after/during) shifting columns South. For example if you shift the 4th row East by 1 space first, then shift the columns as shown above, you can also read most of lines 4-7: "The caster and up to three willing allies...".
3. It's possible that some of the shifting requires shifting only partial rows/columns...
4. The amount you shift columns South by, initially seems to correlate with the number of blank spaces that appear on island squares in that given column, but it falls apart further along...

Alconja
  • 37,250
  • 16
  • 116
  • 164
  • Wow, thank you so much for this! Though I will say, your output doesn't quite match what I get when I use the same array to shift by. I get this. Even better, though, when I change the penultimate number from the array, I get this. How did you arrive at that array, by the way? My guess at first was that it's a count of the number of times in that column that "air" meets "fire", but that doesn't hold up after the first few columns. – Jim Cullen Jan 29 '16 at 09:33
  • Since I can't seem to edit the comment, I'll add a new one. I accidentally missed the last section of your answer when writing the previous one. My guess, right now (haven't had the chance to properly try it out yet), is that perhaps the number of eastward shifts relates to the number of coastlines on that row. And perhaps when that is combined with the correct number of blank spaces over land, that this might produce the result. – Jim Cullen Jan 29 '16 at 10:02
  • 1
    @JimCullen - it's entirely possible I stuffed up writing out that array, as I hadn't been keeping proper track of things as I experimented and manually rewrote it after the fact. In terms of how I arrived at it originally: based on your hints/observations, I noticed the letters Elminster partially appearing in the first and last rows, I also noticed the word Level woven through the lines starting at row 2, so just started shifting things to see what I could find. Then having shifted the columns noticed how some rows were only slightly off and could be corrected by shifting then too (eg row 4). – Alconja Jan 29 '16 at 10:24
  • 1
    And yes, I agree that the islands/coast lines must indicate how/when things are shifted as that's the only information remaining that's not actually been used yet. – Alconja Jan 29 '16 at 10:25
  • 1
    Messing around with this some more: The caster and up to three willing allies take on an elemental form. This form is random, though [some?] may be more... – f'' Jan 29 '16 at 19:16
  • Places that the text has problems might have something to do with the letter "c" on land (land marries sea), but I'm not seeing a convincing pattern. – f'' Jan 29 '16 at 20:21
  • @f'' interesting, in what way did you mess around with it to get that bit of text? – Jim Cullen Jan 30 '16 at 10:19
  • 1
    @JimCullen I shifted rows 4, 7, 8, and 10 one space to the east, and row 11 two spaces to the east, before using Alconja's shifts (but column 15 is 2 south, not 1). Some letters were out of place, but the text was still readable up to that point. – f'' Jan 30 '16 at 18:25