unforum

[booba]

Gotta love your tenacity, tripitka!
Other stuff from grngecko.com:

"The encryption step performed by a Caesar cipher is often incorporated as part of more complex schemes."

http://www.grngecko.com/code/ceasarshift.html

and
http://www.grngecko.com/code/cipherclassics.html

Good luck!

[SuperSlug]

Gargoyle Page

[Skipjack]

Boy, I don't remember the solution to the gargoyle... what came before it?

I'm stuck at the mummy-looking guy with some sort of symbol cipher. I just haven't had the attention span to sit down and do a frequency analysis on it.

[EDIT] Actually, poking through this one, I realize I suck at frequency analysis.. Is there a utility you guys are using to do this?

Also, I've counted more than the usual 26 characters used in this one. (???) So, perhaps I'm underestimating this one. I don't recall anyone else mentioning it... so it must be an easy one that everyone is blowing past except me.

[EDIT...again] ARGH! I found the key for the cipher in google image search! However, the page that contained the picture is no longer there. And.. the only thing I've got to work with is the tiny thumbnail that google offers. Ah well... it's better than nothing, I suppose. God bless Google.

[SuperSlug]

I tried to send you a PM but I guess you didn't get it.

You are at the infamous circle cipher. The reference you need is a pdf document, page 24.
www.nku.edu/~christensen/section%205%20symbols.pdf

Good luck! (There are lots of hints many pages back in this thread.)


I've advanced to the blocks page where there is lots to ponder.

[tipsila]

I sent him a PM on the 8th, too. Before his last edit, so he must not have gotten it.

[tripitaka]

After the amount of time I wasted earlier on looking for binary strings (based on the block letters being either red or black) I know its likely I'm on another deadend track with homophonics but, for the record, I'll pass on what I have so far.

Using the Z378 key as a starting point, I established the following partial key and mapped it sequentially to two consecutive plain text alphabets, making the dangerous assumption that the slugs do nothing but reset the instances of each letter.

Partial Key:

AGM?DW?L???????E??????I?EH?T????????????????????????

Reading from the left hand side of the blocks puzzle this key results in a readout which begins as follows:

"I am delighted we..."

The huge gaps in my partial key mean that I cannot read further until I asign some more key letters. This probably sounds more promising than it is in reality - actually I'm really not sure I have the tools or the patience to find out!

[silentdream]

It sounds like more than anyone's gotten so far. I thought about trying that then come to realize I have no idea how. Being that I have quite a bit of free time on my hands, (along with my friend who's been working on this with me getting ungrounded), if you'd be willing to point me in the right direction I could work on it some and report back with my findings?

[tripitaka]

Cheers Silentdream. To answer your question, the approach requires that the first time a letter appears it maps according to the first half of the key:

A stays as A
B changes to G
C changes to M
etc according to the first 26 letters of the key

The second appearance of the letter will map to a different letter based on the second half of the key (eg B goes to T according to my partial key).

And a slug is presumed to do nothing except to act as a reset so that the next instance of a letter is assigned to the first half of the key (which does seem a pretty simplistic assumption).

BUT please don't feel encouraged to waste a load of time since there are good reasons to be skeptical:

I mentioned the above key is loosely based on the Z378 (which is a pretty obscure cipher to start with as well as having been used here in reverse to how it was designed)... anyway I additionally cut out the first 4 letters of the Z378 to make it neatly fit two alphabets AND then eventually changed half of the assigned letters in my partial key from those in the truncated Z378 version in order to improve the garbled readout when applied to the blocks puzzle letters. In other words, this almost boils down to developing a 52 letter homophonic key by hand which seems uncharacteristically difficult compared to the earlier levels in Torment.

Still, I guess homophonic encryption remains a possibility and if, by sheer good luck, the message really does begin with "I am delighted we..." then building the rest of the key should be a manageable task BUT I don't yet have a high level of confidence that these opening words can be relied upon since quite a bit of "encouragement" was needed to make them appear!

[silentdream]

considering I got freaked out by a movie, it'll do me some good to keep my mind occupied.

Let me make sure I've got this right. With a hypothetical key of left (I know that the key needs to be much longer but this is just the basic principle I'm looking for). baab would be written elft?

edited for silly mistakes.

[tripitaka]

that is correct (for a 2 letter alphabet)

[silentdream]

I see what you mean here. From my work I'd get something that look like "far and few". Which sounds something like what I got the other day when I was just messing around with taking some weird idea I had, running it through a vigenere cipher then ceasar. I'd get something like jggfhadffgthe. Exciting as it was that I got english, i'm sure it meant i was miles off

[tripitaka]

Exactly silentdream! When I next have some time I'll muck about with Cryptool and see if automation can help this process.
[Edit: I've had a look at Cryptool and unless I'm missing something it doesn't seem to allow you to manually set the homophones for individual characters... time for me to give up on this I feel]

I am faintly encouraged by the homophone possibility but suspect that trying to solve for a homophonic encryption by hand will just be an enormous waste of time.

[silentdream]

More than likely tripitaka. But worth a shot if you have the time and no better leads. Something about the blocks being turned bothers me. I mean if it was of no importance then why not have them in a general direction? Or some kind of pattern like certain colors a certain way. I had an idea for the only relevant blocks were the ones facing the same way as their preceding bunny, but came up with nothing when I tried various ciphers. Then again a number of things could throw that off.

Or this is greengecko tormenting me. Fitting just as well.

[tripitaka]

To prove how slippery these homophones are:

Using the coding system I described above it is straightforward to write a key which translates the blocks up until the second slug as follows:

"iamdelightedwemeetagainwecannowgetonandmoveinto"

that is -
I am delighted we meet again We can now get on and move into...

Actually the coding constraints for the second half of this sequence are very loose due to few repeating letters (for example the string "get on and move into" could be replaced with "go on with more inte"..[resting?] but, in any case, the wheels start to fall off at the 2nd slug (3rd column) since the frequency of coming across already allocated letters increases rapidly - which constrains the code much more tightly than in the first lines. Of course its possible the rotation of the slug in question changes something else so it doesn't necessarily prove that the above translation is wrong, but you know what I mean.

[Skipjack]