unforum

[locqust]

I think we established that the 108 characters is a bit too long for a name so perhaps a sentance instead.
Has anybody thought that perhaps it could be a website address? With random letters in the string like you find in the URL's of some huge sites.

I seem to remember that some1 emailed Garnet and asked if Sente had anything to do with the card and he said something like "indeed he did tell me what the code says and it will lead you to the answer."
A website URL makes sense as the answer. A secret site that tells us who made the virus.

Oh update on my theory, Ive been taking each line as two messages encrypted with the same key. Now with that it is possible (apparently) to decode each message by subtracting them from each other without using the key. No luck so far on it, basically cos I dont really know how to do it as Im not a cryptologist.
If anyone is interested in trying as well (a cryptologist would be nice! I know a couple but they are too busy to help really.) look on the bruce scheiner page towards the end when he talks about security of the cipher etc. That gives you the Kloos.

[anon]

I made up a spreadsheet that can attack two messages encrypted with the same key... I used it on the two sets of ten heat letters. I will try it on the two lines.

[locqust]

Cool, also, if you can, try with the heat letters on a selection of normal letters. Picking ten off each line.

[UKver2.0]

I don't understand why people care what the standard deck order for some given set of cards is. I don't see how knowing the order of this Motor deck was supposed to be something so great*. And here's why I don't get it...

If you go back to the beginning of this thread (first couple pages) you'll find a program that can handle 1,000,000 deck orders at a time with as many passphrases as you want along with the added bonus of processing with the jokers swapped, the deck reversed, and the suits swapped (24 possibilities).

Now if you create a 1-52 deck (meaning sequential) and place the jokers at the beginning (slots one and two - ie. 53,54,1-52), then another with one joker at the beginning and one in the number three slot (53,1,54,2-52) followed by one with one joker at the beginning and one in the number four slot (53,1,2,54,3-52) etc. until you get to one joker at the beginning and one at the end (53,1-52,54) - then if you continue to make further deck orderings by moving the joker at the beginning of the deck toward the back until you have 1-54 you will have created just over 100 deck orderings. (Save them as your layout file.)

Load up the little program and check the "swap jokers" box, the "swap suits" box, and the "reverse order" box. Now anytime you run it you will be using every possible new-out-of-the-box deck ordering with the jokers in every possible location. So if the Motor deck had been S-H-C-D or H-S-D-C or whatever, you'd have been covered. If the deck was A-K or K-A, you'd have been covered too. Are we supposed to use the face-up order or the face-down order? Covered! Was the star joker moved? Covered! As far as new-out-of-the-box orderings are concerned, no matter what deck you use, you are covered.

We've had this tool since the beginning. My first attempts were made with this exact setup. I'm sure all the people that were trying to figure this out eight months ago used the same idea. So we've eliminated all these new-out-of-the-box orderings long ago. (Well, I admit our passphrase lists are probably incomplete, but any time I come up with a new one I run the program again and I instantly eliminate that new-out-of-box order/new passphrase combo.)

If people are sure that a new out-of-the-box deck order was used then load up the program with the 105 deck layout file and get busy on those passphrases/cipher texts. You're almost there!!!

As for me, I'm reserved to the horrible task of making the 108 = a deck order. I think it is the only logical path to victory.

NOTE: If you use the layout I've attached, please double check that it is correct before using. "so-and-so's" transcription error of the 128 characters really screwed me over and I'd hate to screw someone that way. No offense to so-and-so cause somebody who had the card should have checked it. Not your fault.

* I've never known a new-out-of-box order to be completely jumbled. Suits aren't always in the same order and some decks go A-K while others go K-A, but that is the extent of variability that I am aware of. Using the program with this layout file and having the appropriate boxes checked wouldn't cover a completely jumbled deck. So yeah, if the Motor deck turned out to be the one odd-ball deck that was shipped out in some weird order I could see the interest in knowing the order. Once it was revealed that it was just a variation of suits/A-K/joker placement all interest in the actual order should have died.

[Guin]

Saturday morning thought

[oliverkeers13]

UKver2.0: The importance of the deck order is that, if we don't know it, and we aren't sure that it's in the right order (regardless of starting place, which would make it a lot easier if we did know), we have (54!)-(53!) combinations available. That assumes that we're using a deck of 54 cards, which we don't know, there may also be 53, or 52 cards, so the actual amount of possible orders are: ((54!)-(53!))+((53!)-(52!))+((52!-51!), a huge number.

[echidna]

[dps05]

OK -

following up on my post a few days ago I used the full 128 letters as 'ciphertext' and ran them through the solitaire cipher programme (http://www.cif.rochester.edu/~la002k/solitaire.htm#fields)
using "IFENTROPYWINS... ...YOUCOLD" as the passkey

to get:
XCQACTNADHTQBLKTDDUFUVDLGUJPMNPFGYMIFJJLKQJANZGSHGAQZAILROJHZDTLYELSONOHEYPUHXWCUENIHKFQXNFGZIXFUVQVDSJAJZZNZIVXVDBGOHUNUXDLXJJZ

then using Guin's idea (if I understood it properly) removed the heat-sensitive letters (first and last 10) of the 'cleartext' to arrive at:

TQBLKTDDUFUVDLGUJPMNPFGYMIFJJLKQJANZGSHGAQZAILROJHZDTLYELSONOHEYPUHXWCUENIHKFQXNFGZIXFUVQVDSJAJZZNZIVXVDBGOH

I'm just running frequency analysis and other basic substitution cipher cracking things on both the 108 'deciphered' letters and then on the 128 if I don't get anything promising. Will post back later...

Also did anyone find any alternative 128-bit encryption techniques? Although I have to say I am right behind this being a solitaire cipher.

[UKver2.0]

[locqust]

I think what ollie meant was there are 2.3 x 10 ⁷¹ possible combinations of deck orders not just 54

[Guin]

couple of thoughts that I have had re possible reason for using this particular deck (and in relation to finding a pass phrase / word that decodes the 108 characters route)

significant texts are:

MADE IN CHINA
MOTOR
PLASTIC COATED (its on it its valid!)
88
CSYJ-8418
MOTOR BRAND HIGH QUALITY PLAYING CARDS
MOTOR POKER 8418

this is the only other information that Garnet has been provided. could it be relevant?

A day of cleaning out an attic for your mother gives you time to think about stuff. there are several possible alternatives to how we solve this and, until one is proven correct, they are all valid.

im off to complete a proofing for Riemann tonight though

[oliverkeers13]

[UKver2.0]

[UKver2.0]

[oliverkeers13]

Are you certain that it would be 54!? I can't really believe that it would, as it would mean that it takes into account the possibility that there is only one card.