unforum

[hairysocks]

To refer back to my last post and SteveC's reply, what I am suggesting is that it is indeed RC5, but the decrypted data is not just plain text. There are 176 bytes in the data so if the decrypted data is 176 and the actual message that makes the answer is found by taking every 5th character from this decrypted data then it may not be easy to automate the spotting of the correct answer.

e.g. say the resulting decrypted data is this (in character and byte values):

'H' 00 01 09 13 'E' 1F 0E 01 00 'L' 08 02 05 I4 'L' 08 05 0A 17 'O'

then the message is 'HELLO'

Such a decrypted data file will not be detected as the right one if the automated process of checking the decryptions from rccrypt is looking for a file with a long string of readable text.

[SteveC]

RE: Help with RC5

at this stage, just suggestions on decoding mechanics, but basically without any encryption knowledge there's not a lot anyone could do yet...? If we can get a widescale decrypt effort going then hopefully more, but until then...

RE: every fifth character, every 13th word, every 12th night....

My opinion is that if they were to do something liket that it would effectively make the code unbreakable. Hell, most situations in the world use this system to make the code unbreakable as it is, we're having to assume that there's a weakness we can exploit!

[bertyb]

Hi,

I am in the same position as doomsdayred with very little knowledge on this type of thing, but:

I actually have access to 4 desktops and 3 laptops currently and would have no problem in setting some of them running if a client could be created - that could run non-stop but would need assistance with this.

I am here if I can help....but to the guys cracking on now - best of luck. Have had confirmation from MC tho that this card does need a collaborative effort as opposed to shuffled which doesnt....they wouldnt say much about Riemann though - but none of that belongs on this thread so I will shut up now.

[Guin]

If collaborative effort is what we need then I will certainly put in my 2 peneth. I have access to 5pcs at home and if need be over weekends 15 and a sever.

Let me know what I can do

[SteveC]

[bertyb]

Here is the wording I got.

The 3 cards in question were this one, shuffled and Riemann.

[Uhtoff]

Asking possibly a very stupid question...

At the start of the original 13th Labour thread it was mentioned that we needed an iv in order to decrypt the code. Did we not in the end, or did we decide exactly what it should be? Mention of it seemed to disappear fairly rapidly in the thread and as I don't even seem to be able to work out how to compile rccrypt I can't check for myself!

[BrianEnigma]

[babycakesjase]

Hey all,

I've just started recently in my quest for the Receda Cube, and am impressed by the daunting level of progress you guys have managed to make. (That was my token n00b disclaimer).

I actually managed to get this card yesterday, and started reading your threads - the idea of running rccrypt on it makes a lot of sense (at least to my limited crypto/math mind); but a couple of suggestions (and tell me if I'm stating the obvious.

It seems that any key to this is likely to be a word or name, so maybe if we start by trying to force it through, but limiting the input keys to words/names/common text in PXC and add something to try all leet-speak variables (I'm not good at coding myself, but I'm sure that shouldn't be too hard to factor in), we could reduce the number of combinations to evaluate hopefully without losing our chance of getting the solution.

And secondly, to help evaluate the text, would it help if we came up with a list of words to look for in the "decrypted" text to act as flags for a potentially correct solution. Obviously strings like T H E are very common, but could we come up with something more unique, tied to PXC perhaps, that would lessen the chances of looking at nonsense?

Please tell me if I'm off base with either of these ideas. I'm also happy to contribute some computers to the effort - particularly if over the weekend - I can probably add a full small office environment (six or so PCs dedicated to the task).

Cheers all,

Jase

[SteveC]

Jase,

We have an even better trump up our sleeves for identifying a correct solution. If we expect the result to be readable text it will fall entirely in the ASCII range of characters (62 out of 256). We don't really need to go any further than that.

As for the key, yes, I think guessing keywords has to be at least tried. That's why I put that webpage up that facilitates that for you. Of course, due to some quirk of something, that webpage isn't working, but I'm sure I'll solve that before I die.. Sometime..

Now, must PM Bri and see if we can get moving on this..

[bertyb]

Will anybody be able to access the website once it is working?

[SteveC]

Sure, as they were before it broke

It's http://www.kallisti.co.uk/pxc/ but currently gives out errors rather than results.

[CoolCats712]

Okay, I just picked up a copy of Cracking DES. I've read through most of it, but haven't found any particularly noteworthy key possibilities, either in plaintext or hex. However it does have full chip schematics and source code for slapping together your own cipher-breaking powerhouse. Anyone feel like putting one of these babies together? (Kidding/?)

I still think the five cows is pointing to this book, maybe not for a specific key, but rather for a specific way to brute-force it. Most of the book is public domain, so there's no problem scanning and posting relevant sections. Take a look at the contents (http://www.amazon.com/gp/reader/1565925203/ref=sib_dp_pt/102-6015221-6047307#reader-link)
and let me know if anything looks promising. Hopefully this will spark something in someone more knowledgeable than me (which isn't saying much).

[VJBeam]

Hello,

Found this site which I think shows most of the Cracking DES book (looks that way anyway, maybe CoolCats712 can confirm?)

Hope its of some use, I have read some of it but as CoolCats712 said no real signs of a key.

http://cryptome.org/cracking-des.htm

[CoolCats712]

Hey! Perfect! It looks like the links to chapters 5-8 are broken. Those are all source code and diagrams, though. If anyone really wants to see them, I can post.