Cereal

[gfoot]

Wow, that took a lot of guesswork. Is the bitstream intentionally so corrupted? There are many extra zero digits; it's pretty hard to sift through the possible answers, and I had to actually send quite a few to the server because it wasn't really clear if they'd be right or not.

I'd also be interested to know what the official full message is - I got "the answer is? rice cold", but just after "is" there's something odd in the bitstream that doesn't make sense to me - punctuation perhaps? Unless I decoded it wrongly, the word separator is not an ascii space, so who knows what the punctuation means.

[gfoot]

Oh, I get it now, heh. I somehow managed to interpret it without actually realising what the encoding really was.

Having looked the answer up in wikipedia, it all makes sense now.

[cyberwoozle]

I've never heard of the Golomb-Rice-Code before, but i was totally fixated of the idea, that it must be some kind of serial (RS232) transmission (because of the unison serial and cereal). So i tried all kinds of Start-/Stop-/Parity bit variations and during these tests i found the first characters. And then the rest wasn't that hard. But i did get "rice code" and not "rice cold" !?

[bok]

Congratulations!
Indeed, the message is encoded using a Rice Code. When I made the puzzle, I wasn't sure how difficult it would be to solve. It took a while for anyone to crack it, it I guess it was a bit harder than I planned.
It's interesting to see that one of you got to the answer without any knowledge of Rice Codes.

[MalNivek]

Nice challenge!

Finally solved it when I re-visited here after some months..

As everyone is saying it is about serial signal and we have a prime number, so at the beginning I though it is some asynchronous signal, and I thought: what if it spells out "The answer is...." as in the other puzzles? But that only gives me "rise s?te" because half of the code for "c" and "s" are the same. Finally get the answer by looking at the bit pattern in details.

[gandhi]

Thank god there was "the answer is..." encoded in the message

[trofi]

As others i tried hard to decode it as RS-232 and got "code" in one of modes:

Code: Select all

("d",7,"B",1,"p",1,"E",2,"j",1,"BE","id"): |}|AC]csrx_'G,~~~~`code

Which means:

Code: Select all

data bits: 2
start bit: 1
parity bit: 1
stop bits: 2
junk bit: 1 (stream start)
big-endian
non-inverted

But all the auxiliary bits are garbage: parity does not match, start/stop bits are random, etc.
The presence of ASCII symbols suggested these are unscambled char bits.

Characters are mostly lowercase, start with '110'. That allowed me to
roughly split stream into chars.

My final table:

Code: Select all

--           t+(d)         h             e       `            a             n               s+(c)
enc = "11111 1100100 1111  1101000 1111  1100101 1100000 1111 1100001 1111  1101110 11111   1100011 " ++
--           w+(g)         e             r+(b)   `            i             s+(c)           u
      "11111 1100111 1111  1100101 11111 1100010 1100000 1111 1101001 11111 1100011         1110101 " ++
--           `             r+(b)         i             c               e       `
      "    0 1100000 11111 1100010  1111 1101001 1111  1100011 1111    1100101 1100000 "++
--           c             o             d                    e
      "1111  1100011 1111  1101111 1111  1100100         1111 1100101"

Looked like 7-bit ASCII encoding with prefix '1111' that does not change meaning of a char and '11111' which adds +16 to ASCII value.

I had no idea what Rice Code is Very nice challenge!