Dungeon Master Forum

(part 5)

C. Specific attempt to decrypt a single word

a. Steps for verification II

a. Find all english 9-letter words (names and artificial words necessarily excluded – if this was one, then it was not decryptable by this method, but the likelihood of an artificial word with 4 identical letters in it seems narrow anyway)
b. From the resulting group take all those who have exactly one letter appearing exactly four times.
c. From the resulting group take all those who do have exactly one other letter that appears exactly twice.
d. From the resulting group take all those where the letters appearing four times and twice are in exactly the right places (“aquantana” has a’s in the places 1-4-7-9 and n’2 in 5 and 8)
e. If any of the (hopefully few but –more hopefully- more than one) result words seem useable try to work on from the decrypted word towards the rest of the phrase.

Please note: I hope my reasoning can be followed by logic or deduction. If however I have made an error or missed something, please let me know, my mail address can be found at the end of the document. If you are interested in detailed lists of the steps described above, please contact me, I will gladly send them to you.

(End of part 5)

(part 6)

b. Results of those steps II

a. Searching my way through the Internet I found a place that offers lists of x-lettered words in english (http://www.xato.net/files.htm), which also contained a list of 9-letter words. I chose to take those as a basis since the list seemed long enough (24873 words) to justify a try.
b. From the given list I collected 204 words containing no more than one letter exactly four times and used it as a base for the next step. (Side note: 13 letters do not appear four times in any english word, those are I, h, g, u, w, b, y, k, v, x, j, q and z)
c. From the given list I collected 96 words containing no more than one letter exactly twice and used it as a base for the next step.
d. From the given list I collected (only) 1 word having the letters appearing four times and twice in exactly the right places (1-4-7-9 and 5-8 respectively). This word was “subsenses”.
e. The result did not seem to justify further decryption of the phrase (After exchanging the letters “a”, “q”, “u”, “n” and “t” for “s”, “u”, “b”, “e” and “n” it now read “gryeix jerebm ubey ki skebow redeim b os dey wefes eeocsre subsenses” which does is not only highly unlikely but impossible because auf the single “b” in the middle, which should be an “a” or an “i”. Side note: The only two words I found that have one letter 4 times, one letter twice, three letters once, and an “a” or “i” on third place (not minding the correct sequence) were “scissions” and “scissures” – both unlikely candidates for a DM scroll.

(end of part 6)

(part 7)

D. Analysis of results

Following the main assumptions (assuming my logic has no hole) yielded no useable results, therefore it follows that at least one of the assumptions made in the hypothesis was wrong.

E. Further assumptions for future decrypting

1. The phrase makes sense (this can neither easily be proven nor rejected – at least I can see no obvious way) I still guess, that this assumption was correct, but there is a reasonable possibility of a simple programming error. Speaking against that is the highly suspicious arrangement of the words, that are no words but sound too close to words for an accident (jernum, skebow, rednim, enocarn, aquantana)

2. English seems the most likely choice – french and german also appeared as languages in DM, but english was always the main language (also for the programmers). I therefore also think that this assumption was correct.

3. Each letter stands for another one (and only one); no letter stands for two different letters when (de-)crypting. That’s already a tricky one. I still think it is true each letter is unique, but I am not sure anymore, that the letters are scrambled, so I hazard the guess that maybe this assumption was wrong.

Following the note after the three tables (where I tried to find out if any letters “behaved” weird in the crypted phrase according to the percentages of the appearance of letters in english I found that from the 21 different letters (only h, l, p, v and z are missing) only 6 really are outside their usual percentage – I, s and t are underrepresented and x, j and q appear too often. But on the other hand this is a short phrase (157 letters) and there may be “special” (artificial) words in it. I am therefore inclined to believe that this is not a scrambling puzzle but a kind of anagram – the word length probably still being correct overall (though not necessarily in the same order, although this also is likely – what other reason could there be ?!), but the letters given are exactly those of the “true” phrase – only shuffled more ore less randomly. Side note: In none of the other scrolls ever a “q” appears – according to this theory we would have two here. This worries me a bit, but it’s just a hypothesis anyway. Feel free to correct me.

(end of part 7)

(part 8)

4. The words have the same length, no matter if crypted or decrypted also seems like a problem to me after this attempt. Perhaps this is true (but then the letters may or may not be scrambled and/or may be randomly distributed through the phrase leaving only the word length as an indicator), but I cannot rule out the possibility that some kind of forming rectangles and then reading the meaning column after column (or alike methods) might lead to a result. So to be on the safe side I also have to assume that this assumption was wrong.

F. Conclusions

I think that my logic holds for this try, so I find it most likely to assume, that what we have here is some kind of anagram (maybe with scrambled letters as well), where only the length (and possibly the order) of the words serve as indicators.

You will probably argue that Christophe already made these basic assumptions in his outline (anagram, word length, order of words) but I think it is valuable to have underlined that by trying another way and failing – there now exists a proof that scrambling alone (based on “regular” english words) does not give any satisfying results.

As a last side note I would like to point out the funny fact, that there actually is an english word where two letters appear four times: “seeresses”…

I very much enjoyed this attempt, thanks to anyone who read this far – I hope future attempts will find this a useful basis. If you have any comments, feedback or corrections to make or want to share new angles for another decryption try you are very welcome to contact me via the forum or directly by email: picturesinthedark@hotmail.com.

GP, 2003-01-09 (was longer than I thought! ;))

(end of part 8 and whole text)

Dear Cristophe,
the answer to your question about the scroll is that as the atari version came out first, this scroll was placed so that the magic map may later be programmed easily into the amiga version.This was revealed in a magazine called amiga format, probably before you were born, but i cannot remember the exact issue, however, rest assured, my memory is fine. I hope this helps you out.

This sounds strange : why the magic map would have been easier to program because the scroll was there? I am not sure this is the true reason for that scroll.
Is our beloved small boy webmaster that young?? :)

What rot. The magic map was put in the Amiga version to help the player out. This was because FTL got a lot of negative feedback from Atari users that CSB was too hard!

I agree with Sphenx: putting a scroll in the dungeon would certainly not help developers in preparing the arrival for a magic map.
The only fact is that the magic map replaced that scroll.

Anyway, that does not help at all in understanding the meaning of the scroll... unless the hidden message is something like "This useless cryptic scroll will be later replaced by a magic map" ;-)

And, no, I'm not THAT young.

Indeed most DM-players on this site aren't that young, oh great one and only ancient dungeon-master-veteran dermott cassidy. Who the f u c k you think you are?

huh? i could swear that scroll isnt magic map in the amiga version anyway, just some scroll telling you how to cast a magic map spell..........(actually i am very positive)

Magazines aren't the best guide (the propogation of the MON GOR SAR cheat shows that!) - Gambit has done a heck of a lot of research into the game versions, if he says no, then it's 99.999999999999999% no : )

lmao@ lunevar (in a nice way) - everyone is usually so polite in these forums and in general, it's fun when someone loses it once in a while : )

I wonder if any of the DM developers can be contacted about this question =)

I must say I'm playing CSB at the moment (Atari version using CSBWin), and its insanely difficult! I keep going round and round in circles!

I remember playing it on the amiga, and I remember the magic map, although how much use it was I do not remember...

Still, I'm sure I'll hack it eventually. I peeked at the walkthrough a few times and it was completely useless, because I'd already done some bits in a compeletly different order, and the maps were all wrong and it just drove me insane.

Beowuuf: Well, certainly I was not very polite, but it just unnerves me when someone thinks he is the only one around who knows that there was a time before windows. I had my first Computer when many years ago, long before there was a C64 and, well, that's it: As Christophe said:
"And, no, I'm not THAT young"

Simon: Why don't you use the maps on this site: They aren't wrong, you won't get any better anywhere.

Back to the main theme of this thread. Anyone still willing to help and try and fulfill the quest?

maybe there is a diffenrent key for any word?
these keys could be somehow connected to each other, like following some pattern.
this scroll drives me mad^^

welcome to the club... I just hope there is some sense after all - otherwise we're wasting valuable energy...

How could such a key word (or the first one if the rest follows) be found?

I cannot figure any english word for "aquantana" for example - if the theory is that each word stands for itself!

Even if there is any sense at all, it will be difficult to find as long as we don't even know whether it is based on English or Elvish or anything else.

I guess you just have to assume english for trying if there are no other clues - if it's elvish we can not discover that and I guess that would make the riddle worth less - so I'm hoping we have a "standard" language here, even if it's spanish ;)

Picture: lets say here, internally, that with probability of 99 % it is without sense. But I think, that on 1 % it means somethink. Look:
1) The goups btw. words can be shaken. I mean: we´ve got the word "Grynixjernumqueykiskebowrednimuosdeywefnaenocarnaquantana", and we must find 1. which letter is which and 2. breaks btw. words. Hard enough, or impossible, huh? Come back to the sure think that "e is the most common in english" (try e = n or a or e) and then .... ??? don´t know. (kind of dissertation thesis, isn´t it?)
2) Look to library - there can be some books about problems like that (I remember E. A. Poe wrote a story about solving similar (much more easy) problem, he used "e is the most common", although I think it´s not our problem, we can find another examples in library). What about to give this problem to university (some departments for English), it could be very interesting problem for them?
3) Christophe, are you sure that the sentence sounds so? Isn´t it "Grynix, jernum.." (with comma after grynix?). It´s important, because then "Grynix" can be a name or st. like....like "Greetings". Then, "jernum" could be st. like "journey", "ki" "key" etc. It doesn´t make sense enough, but just think about it. I´m not very skillful in English (you see), so find someone who is!

No, there is no comma after grynix.

By using a web tool for crosswords (http://www.anagram-engine.com/cgi-bin/a ... search.cgi), I tried to find out what word may be hidden behind "aquantana".
If we suppose that each letter is replaced by another one (always the same), here are the only english words of 9 letters with identical letters in positions 1, 4, 7 and 9 (as in aquantana):

alcantara
azcarraga
ellesmere
elsewhere

None of these words has the same letter in positions 5 and 8, so we can conclude that either:

the hidden word is not an english one (however it may be a name)
the message code does not use simple letter permutations

Huh, I found, it´s not so easy..
What about "seismisms"? (s 1,4,7,9; m 5,8)
This one was a joke, but I´ll search for more..

Uhh, I can´t find anything more than "subsenses" (see PicturesIn.., but it´s impossible, because u = b) and "seismisms". It´s without sense, I know, but interesting is, that then u = i, which gives some sense in this sentence. And more - Uau!: Then "enocarn" could be just "embosom" or "imbosom" and we´ve got more letters... hmmm, it´s very imbecilly, I think, with so math we can build a whole non-sense sentence... logically (mathematically) correct, but it will be more ununderstendable then the original one.
We can go on with these speculations (search through web for more vocabularies), but it´s wasting of time, I think. It´s totally wrong direction. What makes me a little worry: there are no ..uh.. double-letters (bloody language, this English..) in our sentence. I mean - heLL, tOO, bEEr, sEE, leTTer, woRRy etc. I´m not master in English (you see), but I think, these words are relatively common, in English.

Uhh, I see, what a fool I´m. In seismisms, there is "i" in 3 and 6... umf, I was tired a bit. I think I´ll have to stop drinking and smoking joints, or then I won´t be able to choose a character in the beginning of DM..

Hi Christophe & Jenda,

I alos found the "elsewhere" (aquantana) but ruled it out since 5th and 8th letter are not identical. My current guess is that the word lengths are a hint, but the letters may just be shaken over the whole text - even the words might have another position.

Maybe it would be an idea to ask someone who works with languages - I sent an email to someone I found on the internet specialising in anagrams - after an initial answer I received no more emails :(

The really hard parts of this riddle are the facts that a) we do not know if there is any chance of solving it, b) we have no real clue about the language being used (supposing a is true) and c) we have no clue about the method at all.

So I'm kind of hesitating to put in too much time without further clues - my spare time is limited. Anyway, I'll have a look at it from time to time...

Keep smoking that stuff, maybe you'll get an idea, Jenda ;)

Regards

PitD

I suppose you have to smoke something more weird to solve this...

Havg a point there Lunever... but no giving up yet, there were not enough hard tries so far for a real solution. We will see if further attempts will give better results...

PitD

PItD:You mean you will start to smoke weird things when pondering over the scroll?

Do you guys know, that you are the first ones, who make me smoking grass? (if this knew my mum..)
There is the last way how to solve it in an "exact way" (matematically) - i mean, download some anagram solving program somewhere, then run our sentence through it (I think there is more than 1.0e+500 possible solutions, which will write as this software) and then make a algorythm, using some dictionary on it (from MS Word etc.), which will choose sentences with more than e. g. 90 % of English words. And from this sentences (may be hundreds of them) choose the relevant one. Easy, isn´t it? Two problems: (1) Who will write the programm, which will choose sentences with english words from the anagram sollution? (2) Will be NASA able to lend us their best computers for several weeks (I don´t think my 900 MHz Athlon will be able make it faster then in 900 years)?
So I think these kinds of solutions are dead. So: We can´t solve it matematically/statistically, but logically. Not quantitative, but qualitative approach..hmmm, where is my grass?
PS Sorry for my English..have a little headache after this (damn good Czech) beer..

No Lunever, I will not do that, never have done ;)

@Jenda: quite agree with you - but I cannot find any lever where to start at the moment since the most likely word to be decrypted ("aquantana") did not work out for me and Christophe as well - so how to start is the problem.