unforum

[hairysocks]

I think the idea of a collaboration on cracking this card is a good one. Its also what is suggested by MC (I can't remember where I read that, but it was on a reply to someone's question regarding this card, or Riemann or Shuffled).

But there seems to be an assumption that the data to decrypt is obvious - the bytes as shown on the card; and the form the decrypted file will be is also obvious - a text message.

These assumptions must be questioned.

I have mentioned in the other #251 discussion that the data may not be in left-right, top-bottom as shown on the card. It may be in any one of 8 possible orders. All variations are in the attached ZIP file so you can try them. The file that matches the order shown on the card is ppc_lrtb - "lrtb" means "left to right, then top to bottom", which is the way we read English. Others, such as Arabic readers, would find "rltb" more intuitive. Anyway, the card itself has no indication which way to read the data, so we should try all possibilities.

The second assumption to question, namely that the decrypted version of the data will be a bit of readable text is also not necessarily valid. It could be an image file - BMP, JPEG, PNG, GIF being the most obvious to check for.

Even if the answer is plain text, it doesn't mean the whole file is plain text - there may be some non-printable characters at the beginning of the decrypted file that would cause a simple check for plain text to fail.

So, anyone writing a cracking program to solve this card should consider these points. I hope I haven't shattered your dreams of breaking this encryption quickly!

[Guin]

knocked this up. Just needs to be linkified. If its any use I could knock up a simple web page with an explanation and download details etc. ANy ideas?



Okay so I used the Assault on precinct 13 as a theme

[hexDa3m0n]

[ambskunk]

No one seems to have directly emailed MC asking any questions about the implementation of the RC5 on this card so I emailed MC asking if the cypher is in accordance with the RSA competition rules (http://www.rsasecurity.com/rsalabs/node.asp?id=2101). At least an answer to this may clear up once and for all questions about character ranges and if the answer starts with the ol' "The unknown message is:". I haven't received a response but will let you know if I do.

[chimera245]

Progress Report

[hairysocks]

My reason for suggesting the decryption could give an image file is because even with a very small image file it could be possible for it to refer to a longer message - think about the Barcode Recipe card and the Perplex City postcard, and the data grid barcodes on them. I assume most people have managed to decrypt them? Anyway, without giving too much away they both point to web sites.

I've written a Java app that take the decrypted file from rccrypt and checks for strings of text and tries to open the file as an image. In some cases the data is structured as a monochrome BMP and the image produced is certainly large enough for more than one data grid barcodes. I have some examples at work, I'll post some when I get there.

chimera245 said

[hairysocks]

Here is an example of a decrypted file being in a BMP file format. In this case the rccrypt tool was used with the -p option, the "password" was 8eLkarEH (which is herakles, backwards with 8 in place of s and letters H, E and L upper case) and the data on the card being used in the "English" way of reading it - i.e. left to right, top to bottom.

The image in the file has been enlarged to 400% of its original size so you can see the individual pixels. I know the image here doesn't look like a solution (!) but it shows the potential - you could have a barcode, or even a piece of text displayed in this many pixels.

[hexDa3m0n]

OK.
So it may be a picture then.....am attaching a list of the words that I have tried, if anyone has an urge to help check through these, I am thinking about using yousendit, or talking my friends into hosting them for me.....

[BrianEnigma]

[hairysocks]

[Juxta]

Well. I've never worked with cryptography. Ever. However, I can spot ass-hattery at 200 paces. Unnecessary.

As for Bri's point, if you'd care to read it, it makes rather a lot of sense...we're discussing the decryption of a PxC puzzle card here. Whilst Mind Candy/the CRT have created it to be a stern test (and seemingly one which they are looking for us to apply significant man/CPU hours to) they aren't attempting to create something which can *never* be deciphered. As was suggested, we can either take the approach of covering every available angle and possibility which will take a LONG time, or we make some educated and calculated decisions to narrow the search down. Now, back to the experts...play nice eh kids?

J

[Amorya]

[BrianEnigma]

[Jakeo]

Re: Progress Report

[c1023]