Postcard Code: Difference between revisions

From Perplex City Wiki
Jump to navigationJump to search
m (→‎Theories: typo)
Line 89: Line 89:
2215 45 484 848 4651 00503 215
2215 45 484 848 4651 00503 215


These perfectly correspond to the number string under the Semacode on the postcard.  By entering http://www.soogees.com/221545484848465100503215.htm we were directed to another puzzle, another GIF.
These perfectly correspond to the number string under the Semacode on the postcard.  By entering http://s1.soogees.com/221545484848465100503215.htm we were directed to another puzzle, another GIF.


However, [http://soogees.com/elohtibbar.htm the puzzle trail has now been completed] and its author, Ehsan, had this to say about the link:
However, [http://soogees.com/elohtibbar.htm the puzzle trail has now been completed] and its author, Ehsan, had this to say about the link:


'''''Trivia:''''' ''The number is an unsolved string from the upcoming ARG, Perplex City. Even though soogees is not affiliated with PC, you are encouraged to play it anyway, because it seems to be very interesting indeed.''
'''''Trivia:''''' ''The number is an unsolved string from the upcoming ARG, Perplex City. Even though soogees is not affiliated with PC, you are encouraged to play it anyway, because it seems to be very interesting indeed.''


== Results ==
== Results ==


* Nothing conclusive... but interesting.
* Nothing conclusive... but interesting.

Revision as of 16:02, 24 August 2005

The code of 221545484848465100503215 is beneath the Semacode on the Perplex City Postcard, does this mean something? This page is going to be used as a workpad for the puzzle.


Images

Postcard Semacode


Theories

  • We've tried using numerical translations into Kanji and Hebrew. No luck, though "2210" gives us "SEEK."
  • Various ROTing, and conversion to ASCII then ROTing proved...nothing but gibberish.
  • Equally, running modulus calculations on the number to bring it into number spaces for ASCII etc. AND converting to ASCII as base 10 and base 16 AND ROTing the results gives yet more gibberish.
  • Ditto multiplying the number by 2210 (the number next to the semacode stamp on the card) and trying the above operations.
  • Tried translating to EBCDIC: still gibberish.
  • Too many digits to be an ISBN or a latitude/longitude. Though could be two (22:15:45 48:48:48 and 46:51:00 50:32:15, see below for further speculation on that!)
  • Tried mapping numbers to characters in the text of the postcard (i.e. first character in Sente's message is 00, second character is 01 etc.) trying various starting points and different combinations of including/ignoring spaces, hyphens, line-breaks etc. Results were meaningless so ROTed them but still no obvious message.
  • Also tried mapping numbers to the letter on the www.perplexcity.com home page.
  • Also tried mapping numbers to other various strings on the postcard as well as other sources on the www.perplexcity.com site, with different offsets, base 9, 10, etc. For all ~5000 results see: Base 10 and Base 9
  • Shish tried converting to music using 'beep' command in Linux and Tanner tried mobile phone ringtones but no obvious answers or recognisable tunes were found. Tanner has now broken out into doing Postcard Code Remixes as MP3s, available from his site.
  • Frequency analysis of the numbers was similarly unsuccessful. Assuming the string was one-digit numbers gave far too much repetition of characters (for example the 484848 would probably be tetete or etetet) for a meaningful message. Assuming the string was 2 (or more) digit numbers prevented frequency analysis simply because there wasn't enough data to work with (frequency analysis works best over a very large amount of encrpyted data).
  • Currently trying various ways of converting number to binary (all possible methods of writing number in binary have been listed on unfiction forum by Seej) then converting that binary to morse code, i.e. 0 = dot, 1 = dash (or vice versa) and interpreting that series of dots and dashes (actually more complicated than it sounds as morse code characters don't always have the same number of dots and dashes so there's no clear way to tell when one character has ended and another has begun).
  • Binary number strings from above method also all converted to ASCII but non-alphabetic characters were output.
  • It's not a prime number.
  • We've tried indexing in every way to every available text (ie, 31 could be 3rd sentence, 1st word; 3rd paragraph, 1st letter, etc.)
  • It's not a phone number or a credit card.
  • graphing it in various ways hasn't revealed anything.
  • Doesn't appear to be related to a deck of cards because of repeating sequences.
  • Assuming code is a series of 2 digit numbers:
    • Reasons to believe this:
      1. "484848" jumps out as a pattern
      2. the numbers are clearly clustered and not very random
      3. all 12 numbers range from 00-51
    • Since numbers range 00-51 we've tried several simple translations using 2 alphabets (ex. A=00, B=01,...Z=25,A=26,...Z=51).
      1. In some cases "484848" translates to "www", but the rest doesn't look like a URL (also tried tinyurl)
      2. In other cases, it translates into a set of letters which can be scrambled to come up with "cube" but the rest is undecipherable. Note: letter on website refers to "XXX Cube" where XXX is blacked out. It looks like about 6 or 7 letters in length, so it could be a proper name.
    • Translating the number in ASCII using Base64 results in "WPtwwwuzAygP" (again resulting in the appearance of "www")
  • It has been suggested that the number may simply be a key for a cipher such as MD5 or RSA but there is no way to verify this until further encrypted messages are received on which the number can be used as a key. This is unlikely to happen until the game starts properly.
  • Tried converting the number into different numeric strings and converting to hex, to get unicode and Adobe Postscript characters. Provided Gibberish thus far, even after ROT'ing.
  • Tried converting the number to an IPv4 address a few different ways. There are probably more. (Link)
  • Tried plotting pairs of numbers as (x,y) coordinates, i.e. the first coordinate would be (22,15). Then joined up coordinates in order plotted, but it didn't give any meaningful shape.
  • If 484848 means www, no web extensions can be found through any rots, both forwards and backwards. This also indicates that even if the code is a continously changing code, it isn't a web address.
  • Could be a continously changing code every pattern of letters.
  • None of the below means anything to us yet:
    • 22 15 45 48 48 48 46 51 00 50 32 15
    • 221 545 484 848 465 100 503 215 = html co-ords? 221, 545, 484, 848, 465, 100, 503, 215
    • 2215 4548 4848 4651 0050 3215
    • 221545 484848 465100 503215
    • 22154548 48484651 00503215
  • Tried converting numbers to letters in a 52-piece alphabet, and rotting the letters with code 2210, for example, 22 becomes W and is changed to ROT 2, 15 is P and is changed to ROT 2, 45 is T and is changed to ROT 1, 48 is W and isn't changed at all. No solutions.
  • Tried adding each number cumulatively to the previous number/sum of numbers, then converting into letters. All ROTs produce nothing. However, only done using single digit numbers.
  • It's not a UPS, FedEx or TNT parcel tracking number.
  • Spankit realised that the number multiplied by e (2.718...) gives a fairly accurate (0.0014%) estimate of avogadro's number (which is the number of molecules in a mole of any substance). Current spec includes "e-mole" and that this was just a silly joke.
  • Attmempted typing the number into a note pad program, copying a picture of this, then putting that into a graphic editing program. Tried flipping numbers around and upside-down to little avail- However, the last four digits, 3215, when flipped vertically and horizontally, seems like "Size"
  • Realized that the all the numbers the code are also numbers which are used in "leet speak." Trouble is, is that the number two can stand for either 'R' or 'Z', and the numbers eight and six are both sometimes used to stand for 'B.' The last four digits like this spell out "Eris." Also attempted reading it backwards, with "sire," the reverse of "Eris," being the only real word to show up. However, "Sireo" and and "Sasir" seem familiar, somehow...

Latitude/Longitude

Just for completeness and due to the luck of hitting the Crozet Islands almost exactly here is an analysis of the number as a set of coords. There are four possible translations of the two possible locations, I'll list them below with links to satellite imagery of each. Each refers to counting the coords as North or South and East or West.

22:15:45 48:48:48

N/E: Saudi Arabia in the Khali Desert somewhere

N/W: Mid-Atlantic - can be ignored IMHO.

S/E: Off the coast of Madagascar - once again, can be ignored?

S/W: Southern Brazil

46:51:00 50:32:15

N/E: West Kazakhstan on the Caspian Sea coast - Just into the water - not land, can be ignored again?

N/W: Off the coast of Newfoundland - and ignored again?

S/E: Crozet Islands in the extreme southern Indian Ocean

S/W: South Atlantic off the coast of South America - ignored again.

The Soogees Link

For those who don't know, Soogees was created as a fun puzzle, quoted as "a little diversion". We traversed our way through the puzzles, finally reaching the TheEnd.htm puzzle. After modifying the colour table in photoshop, we got this string of numbers:

2215 45 484 848 4651 00503 215

These perfectly correspond to the number string under the Semacode on the postcard. By entering http://s1.soogees.com/221545484848465100503215.htm we were directed to another puzzle, another GIF.

However, the puzzle trail has now been completed and its author, Ehsan, had this to say about the link:

Trivia: The number is an unsolved string from the upcoming ARG, Perplex City. Even though soogees is not affiliated with PC, you are encouraged to play it anyway, because it seems to be very interesting indeed.

Results

  • Nothing conclusive... but interesting.