[Jabba]

it was mentioned earlier in this thread or the other (13th labour thread) by someone (sorry i forget who). the book is Cracking Des : Secrets of Encryption Research, Wiretap Politics & Chip Design by Electronic Frontier Foundation, John Gilmore (Editor)
found here in the recommended reading page.

[sackofpotatoes]

What book are you talking about, Jabba? distributed.net seems to be down...

[Guin]

Jabba wrote:

the book that received the only 5 cow rating from ditributed.net it that a book that covers the topic of the encryption method used in this card? if it does, then the cows on the card hmay have only served as a hint to get us as far as that book and the explaination of how to approch solving it.

Yep seems a reasonable idea.

[Jabba]

the book that received the only 5 cow rating from ditributed.net it that a book that covers the topic of the encryption method used in this card? if it does, then the cows on the card hmay have only served as a hint to get us as far as that book and the explaination of how to approch solving it.

[arnezami]

Seej wrote:

That depends. If MC calculated their figure based upon an assumption that we would check all possible keys then they're probably closer to the mark. However, Chimera and Guin have made the fairly reasonable assumption that of all the possible ASCII characters that the key could contain, it almost certainly contains exclusively alphanumeric ones, allowing us to skip lots and lots of possible keys.

Yes. But this (reasonable) assumption is largly based on the idea MC "wouldn't be that cruel". The card nor the hints so far give any indication which (specific) characters to exclude. In fact this quote from Michael Smith seems to be indicating how many different characters MC has used for the password. And it doesn't appear to be just alphanumerical.

I wonder if they expected many more computers/participants...

Seej wrote:

If we get to 100% and still haven't found an answer then we just go on to try all the other possibilities. Well, except for FF FF FF FF FF FF FF FF as I can now exclusively reveal that it doesn't work

Hasn't going through the entire 256^8 search space already been deemed impractical?

[Seej]

That depends. If MC calculated their figure based upon an assumption that we would check all possible keys then they're probably closer to the mark. However, Chimera and Guin have made the fairly reasonable assumption that of all the possible ASCII characters that the key could contain, it almost certainly contains exclusively alphanumeric ones, allowing us to skip lots and lots of possible keys.

If we get to 100% and still haven't found an answer then we just go on to try all the other possibilities. Well, except for FF FF FF FF FF FF FF FF as I can now exclusively reveal that it doesn't work

[arnezami]

Quote:

"I think about 98 percent of the puzzles that we've released on the cards have been solved," he said. But there is an exception. "One is a very complex piece of cryptography that we've calculated would take about 30,000 computers running in tandem several months to solve. They haven't managed to get that one yet."

Those are quite disturbing numbers actually. Since we "only" have roughly 1000 computers. Meaning it would take at least 2 months x 30 = 60 months with our current resources.

[Guin]

So far the 13thlabour client has completed all numerical and a, b and c keys. we are now progressing steadily through the d's

Based on comments made by Mind Candy such as "you have risen to the challenge" and that "you should start this sooner rather than later" and particularly like the interview Michael Smith gave MTV

http://www.mtv.com/news/articles/1528282/20060410/index.jhtml?headlines=true

Quote:

"I think about 98 percent of the puzzles that we've released on the cards have been solved," he said. But there is an exception. "One is a very complex piece of cryptography that we've calculated would take about 30,000 computers running in tandem several months to solve. They haven't managed to get that one yet."

a distributed client is most likely the correct way to go - although it could be possible to guess the key

[TheDarkMoomin]

nomadcelt wrote:

Ideas: - 64/12/8 - could this be a reference to a book? Page/Chapter/Paragraph?

This has been discussed before - with the books considered being The Labours of Hercules and a book about DES found from the distributed.net site, the only one with a rating of 5 cows. This book doesn't have actual page numbers unfortunately.

I did have a thought about the Labours of hercules book. I would imagine this has been published in numerous editions and formats by different publishers, so it is more than possible that the page and line numbers will vary between different copies. For this to work we would need the right edition.

nomadcelt wrote:

- 251 - is the card number, itself, a clue? Clearly 256 is more common for all things geeky but still.......

Haven't read anything suggesting that aspect yet - can't think of a link yet though.

I also agree that the brute force approach is unlikely to be what MC are looking for, and when its cracked we may well be kicking ourselves. That said, my PC is still running keys in the background, and I don't think its a bad route to persue in the mean time.

Has anyone else had any ideas?

[nomadcelt]

Ideas:

- 64/12/8 - could this be a reference to a book? Page/Chapter/Paragraph?
- 251 - is the card number, itself, a clue? Clearly 256 is more common for all things geeky but still.......
- have those in perplexcity academy been of any use? There is an entire cryptographic department.

[nomadcelt]

I'm only relatively new to this but I think far too much effort is being put in to this. I think there are some rather perverse people at Mind Candy currently having a big laugh at our expense. When, IF, it is cracked I rather suspect that we will be told of a far simpler solution.

I'm new to this and am going to start working on my own, from the ground up....

[groovygirl20]

go on http://www.connected.ltd.uk/pxc/ to guess what you think the key is...

i dont know what it means by \....

and if its a "rejected" key then the word is followed by "is no good" hence my sad post above

[TheDarkMoomin]

groovygirl20 wrote:

none of the bold words work but Quote: The decoded file for key 'Melatron' vaguely looks like a \ hmm... a \ might work....

Cheers for trying them, they were a bit of a stab in the dark tbh.

Out of interest, what sort ouf output do you get from a 'wrong' answer? And how do you mean it vaguely looks like a \?

Sorry, still haven't got to grips with the testing aspect of it!

[groovygirl20]

none of the bold words work but

Quote:

The decoded file for key 'Melatron' vaguely looks like a \

hmm... a \ might work....

[TheDarkMoomin]

Ifurita wrote:

you can go to 13thlabour.tk, download the client, and start procesing work units along with the rest of us monkies

Cheers, I can't at work, but I'll have a go at home tonight - couldn't keep track of which progam I was supposed to be using!!