unforum

[Infernux]

just found this, but im having trouble compiling and running it. its a solitaire brute force solver.

http://www.galad.com/programm/solitair/sdown.htm

[jaikaiman]

I don't really know if this is correct, but I had a look at the link downloaded the file, and had a quick look. Its a bunch of text files, so I scanned down the download page a bit and found a link for The solitaire game...well, the board game anyway. Is this relevant?

Heres the second link
Solitaire Brute Force[/url]
P.s. I hope you are good at reading German lol

[kian]

Food for thought

[GinGenie]

Trying something different here - alternative ways of using the deck of cards

The message on the cards says 'I spread the cards out .... it took me some time...', so perhaps he was playing clock patience with them.

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

The (almost) interesting thing about this game is that if the deck is in the same order at the start and the cards are laid out in the same way, then each time the result of the game is the same. Instead of placing the cards next to the correct number, if you just keep them in a pile then this could generate a key.

More to the point, if you shuffle an untouched deck perfectly seven times (using out shuffles I think, apologies if its the wrong way round), then lay out the cards one at a time starting at 1 oclock, then 50 cards are used before all the kings are found.

Taking that the 108 letter string might be divided into groups of two, as in 54, then -2 at each end also makes 50.

Thats as far as I got, no joy so far in actually decoding anything, but it seemed like a good a reason as any to say hello, still here and all that!

[Infernux]

The part that annoys me about the card is the part of the Ciphertext "LUL" This is annoying because in Cryptonomicon, and even in Bruce's solitaire page on his website, he says that any blank spaces should be filled in with X's. This means that there shouldn't be any chunks that say LUL, but that whatever LUL was in plaintext should have had XX added onto the end, so you would get an ending chunk of 5 letters.

Think this might have to do with anything?

[Mindez]

Padding it with Xs isn't important. It's just a standard that some/most people use. It is acceptable to have a 3-letter block at the end.

And because it isn't at the end, it leads me to believe that we don't use the heat-fading letters, so that it will be at the end.

[James Siegesmund]

[dmatos]

After reading the article on the solitaire cipher it seems to me that the encryption keyword needs to be more complex than the actual message. I have been trying using the "disappearing letters" as the message and the long string as the keyword and have seen some semi-words starting to show up.

Has anyone else looked at it from this standpoint and had any luck? I don't have time to look into it any further today, 2 1/2 year old screaming and a Thrashers game tonight. Please let me know if something comes of it!

From http://www.schneier.com/solitaire.html


Remember, though, that there are only about 1.4 bits of randomness per character in standard English. You're going to want at least an 64-character passphrase to make this secure; I recommend at least 80 characters, just in case. Sorry; you just can't get good security with a shorter key.

[jaikaiman]

Im not sure if this has been posted before so trout me if it has.
The following is the deck order of a brand new pack of unshuffled motor brand casino cards.

Joker 1 = 53, Joker 2 = 54.
AD - KD, AC - KC, AH - KH, and AS - KS.

The first joker is in full colour with the second joker being in black and white. Both jokers have 3 stars on them (from what I could see).
Both jokers are holding three cards in there left hand QH, KH, AH.
Both jokers are dropping 5 cards from there right hand (in order top to bottom) 5D, 4D, 3D, 2D, AD.

There are no markings on the back of any cards. Although the cards can come with either Red or Blue backs (Hot Cold?).
The back of the card box has 3 stars on it, and also the number 88.
The Motor brand serial number is Poker 8418 and the part number is CSYJ-8418.

I hope this helps...maybe it will jog some inkling of thought somewhere...oh please god please.

[Infernux]

Personally, i think the pack of cards was only meant to say that the cipher had to do with a deck of cards. I think people are going into too much detail about the pack of cards.

[RedZed333]

Re: Food for thought

[Mecha Angel]

Re: Food for thought

[jaikaiman]

I actually bought a pack of these cards so that I could study them more closely (in the flesh) plus i fancied a game of poker...anyway, the compass needle is double ended and therefore points in both directions. Ho hum. I thought it was interesting though, that if you draw a line through the compass needle it splits the card in half from corner to corner exactly. However, I would probably agree with Infernux that we might be clutching at straws in our frustration.

[Marnok.com]

Okay here's what I see.

This solitaire cypher business needs 3 things - a keyword, a deck order, and a message. Am I right?

We have 3 things - vanishing text before, 108 in the middle, vanishing text after.

To me, it seems we have key before, 54 card deck order in the 108, and possibly message after. Key and message could be swapped. Message can be short because it is just a name.

This would imply that the key is either those 10 characters directly - Or that it is solvable in some way (not solitaire) to reveal the key - and the deck order is similarly solvable without solitaire. Only the final message component requires the solitaire method.

The 108 character section is probably 2-character pairs like KC / 4H / 7S listing the cards themselves. Depending on whether suit or denomination is first, there are maximum 6 possible plaintext characters for suit (if each joker has it's own suit data) and up to 15 posible plaintext characters for denomination (assuming each joker has it's own character here, which it may not)

Now, this is where my experience fails me (if it hasn't already).

Does knowing the number of plaintext characters help us decipher the card order? Also knowing that a particular letter appears 13 times, and that their are 4 sets like this... seems like it ought to be a clue.
Again knowing that each number appears precisely 4 times in the plaintext, plus 4 each of JKQA and a character for ten (T?), ought to help a bit.

If the "star" comment refers to an asterisk, perhaps that unique plaintext creates a unique identifiable ciphertext? Especially since it is likely to be accompanied by a "J" or equivalent, which won't be common in the plaintext.

Would knowing the likely letter in each alternate character aid us? Assume the perplexian card decks use suits with the same letters. We need to be sure if it is Hearts Spades Clubs Diamonds... though of course if we just identify 4 different suits and get the denominations right, without knowing for sure which suit is which, that's a very finite number of deck combinations to try. The only reason I can think getting a direct translation is important is because the key might be encrypted in the same way.

Does this sound reasonable and sane? Any of it?

[Mindez]