Puzzle #4 - Color Code

Have you guys noticed that the puple lines consist of 6 pixles in width each, and that they got green values too?

I tried to pick out some values:

purple line 1: 40 20 0 0 20 39
purple line 2: 46 23 0 0 21 42
purple line 3: 40 20 0 0 19 37
purple line 4: 38 19 0 0 20 41
purple line 5: 41 20 0 0 20 41

I think they look a bit "odd", since there is always two zero's in the middle, and different numbers on each side. Just a brush-thing from making the picture, or something to use for us

[edit on 10-8-2008 by rihred0]

Sry, double post

[edit on 10-8-2008 by rihred0]

Originally posted by AshleyD
Oh. Just thought of something. Is the number 16 significant in any way? Or any other number significant in any way?

Edit: NM. Just noticed it wouldn't matter for what I was thinking of.

[edit on 8/10/2008 by AshleyD]

16 pages and no answer.
The image has 16*16 = 256 different colors.
But none of this probably matters..

Originally posted by rihred0
Have you guys noticed that the puple lines consist of 6 pixles in width each, and that they got green values too?

Yes, and only on the even lines, as Ian mentioned earlier.

This was discussed earlier in the thread with a possible solution being artifacts of the color translation. But, why would those artifacts occur only on even lines? Doesn't make any sense. They must have been intentionally set up that way.

Here is the color data from the image, in text CSV format (which you should be able to open in any spreadsheet program):

www.4shared.com...

There's five files:

The red/blue channel color values from the entire 282x59 image

The green channel color values from the entire 282x59 image

The clut index numbers from the entire 282x59 image (probably not useful)

The red/blue channel color values from the 47x29 solid color blocks on odd scanlines

The green channel color values from the 47x29 solid color blocks on odd scanlines

The first three tables (the 282x59 ones) are stored pivoted, so the Y-axis is across, in columns, and the X-axis is down, in rows. This is so the spreadsheet isn't too wide.

Enjoy!


Edit to mention: These color values are taken from the 282x59 pixel rectangle with its origin at (259,10) in the original image 8uhnbsgt62bs.gif. Everything outside that rectangle in the original image is pure black (color=000000).


[edit on 10-8-2008 by Ian McLean]

hey guys is it possible.

the text hidden within this puzzle is not spelled grammatically correct but is phonetically correct.

like if i wanna say, "look where you can see the end of time"

luk wheer u kan c the nd of tiem

This is so damned hard...

I fell like screaming....

Hope to add something useful soon.

I keep running into this


nfacme]dd�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf�fh]^n


Ian the files are great
I am going through all of em, using line 16... who knows out of all the lines this right above was in line 16 .. so. it was in green odd file


this is all of green odd in dec translation

942294044�290469�692224�0929042�424294994�44009

aZUW`ZSZX�VdSZ^d�^^WVUX�S`VaSZS�ZWZTd\``Z�
Z\SSaqiceph_gf�du_jmu�mnedcf�_pdq_ha�iejbulppi�il__qphbdog^fe
�cu_ilu�lmdcbe�^ocp^g`�hdiaujooh�hj^_pohbdng^fe�cs_ilu�lldcbe
�^nco^g`�hdiasjnnh�hj^_onfacme]dd�br^gjs�jkcbad�]mbn]e_
�fcg`rhmmf�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf
�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf�fh]^nnfacme]dd
�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf�fh]^nnfacme]dd�br^gjs�jkcbad
�]mbn]e_�fcg`rhmmf�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_
�fcg`rhmmf�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf
�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf�fh]^nnfacme]dd
�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf�fh]^nnfacme]dd�br^gjs�jkcbad
�]mbn]e_
�fcg`rhmmf�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf
�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf�fh]^nnfacme]dd
�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf�fh]^nnfacme]dd�br^gjs�jkcbad
�]mbn]e_�fcg`rhmmf�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_
�fcg`rhmmf�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf
�fh]^nnfacme]dd�br^gjs�jkcbad�]mbn]e_�fcg`rhmmf�fh]^nngacnf]ed
�br^hks�klcbad�]nbn]f_�gch`rinng�gi]^nphbdog^fe�cu_ilu�lmdcbe
�^ocp^g`�hdiaujooh�hj^_pphceoh_gf�du_ilu�lmedcf�_odp_ha
�heibujooh�hj__pb\WXbZTZZ�XgT\`h�`aXXWZ�TbXbTZT�\X\Vg^bb\
�\^TTb1.,,1.*,,�,3*.13�11,,,,�*1,1*.,�.,.,3.11.�..**1






[edit on 10-8-2008 by ragster]

I was dreaming of this puzzle last night, I think my subconscious is taunting me. It knows the answer but won't tell me. I may not post a lot but do know that I'll be in front of my computer all day today trying to crack this purple #*%$@!.

reply to post by ragster


Doesn't the new clue say to focus on the colours? I know the information in the GIF file defines the colours, but i dont believe they want us to look into the bits and bytes (or any derivation) of the colours to be honest.

On the verticals, the third almost black line, where the values are 18 17 16 16 18 17 15 17 16 and so on... the 15's are all of the ] ^ ' _
When I take those out I end up with this:
nfacmedd brgjs jkcbad mbne fcgrhmmf fhn

Not sure if that's anything.


[edit on 10-8-2008 by oregongal]

I have had a scary thought...

What if we have had the correct cipher text since the first few pages, but have just not been able to decipher it yet! Perhaps the green values mapped to letters is correct, but the decryption programs we are using have a dictionary without the words in the solution. Its not impossible to think that the words in the solution may well be common in the 'conspiracy world' but not common in the dictionary used.

Im not entirely sure how these tools work, so i may be wrong. But i got a feeling it could turn out to be one of those horrible 'DOH' moments.

Has anyone thought that there could be a hiden file in the .gif? It's possible and I've retrieved files from pics before but for the life of me can't remember how.

lifehacker.com... rivacy/geek-to-live--hide-data-in-files-with-easy-steganography-tools-230915.php

It's nice to know I'm not the only one crazy enough to still be tearing my hair out over this.

Ian -- thank you for the files

I think we can count on there being some sort of spelling mistakes, or phonetic spelling, or acronyms, to throw off the crypto programs. Otherwise we would have broken the code before we broke that website

What really has me convinced that we're looking at a changing-offset code is that next to last word (driftwood, liverpool, disagreed, etc... The possibilities are so limited in a monoalphabetic cipher that I'm sure we would have gotten it by now.

I've been thinking maybe the filename is a keyword for the offsets -- I think I'll look at that angle this evening. I'm also interested that the green values are all the same in lines 9 and 49 (those are the only two, not counting the middle block).

Hopefully tonight will be the night. I don't know how much more I can stand of these purple bars.

edit to respond to Gareth's post which appeared while I was writing:

I'm almost completely convinced that we've had the right ciphertext all along. I think it's possible that one of the brightest lines, or lines 9 and 49, contain the right values instead, but there's so much overlap in all of those that if we had found the right way to break them any would have provided enough info (like when Ian broke abnvetnosecret or whatever). All along, the codes have been designed to be breakable by logic rather than by brute force programs.

[edit on 10-8-2008 by americandingbat]

reply to post by oregongal


yea this looks good

i reversed it

an in decryption it seems like it has potential

NHF FMMHRGCF ENBM DABCKJ SJGRB DDEMCAFN

NHF FMMHRGCF ENBM DABCKJ SJGRB DDEMCAFN

to this

FUL LDDUNIWL RFKD AOKWEH THINK AARDWOLF

to this


THINK AARDWOLF

to this

Aardwolf (CIA report)
From Wikipedia, the free encyclopedia
Jump to: navigation, search

Aardwolf is a nickname used for a CIA station chief's formal assessment of conditions in their country.

Two aardwolves are described in State of War: The Secret History of the CIA and the Bush Administration by James Risen (2006). These aardwolves were sent to CIA's headquarters by the Iraq Station Chief in Baghdad in August and November, 2003. Both frankly detailed the worsening situation in Iraq. His assessment was at odds with the assessment of those in Washington, D.C.

to searching ats for something along these lines


I have no idea if it is anything but hey

its worth it

peace out guys

reply to post by americandingbat

I've also tried the original middle-row green text:

RJEGQIAHH FVBKNW NOGFEH AQFRAIC JGKDVLQQJ JLABR

with a polyalphabet (Vigenere) cipher, through a cycle length of 5, with different modes of resetting the cycle between words, ignoring spaces, etc. I got vaguely excited for about two seconds when I got the weak hit:

YSVOHPJYP MESSED EWXMNY RXOIIZJ AOBKECYHQ ATRIA
(decode key=7,9,17,8,17)

but no 'luck' other than that.

I don't think it's gonna be that tough, number-crunching wise; I think the solution, if it's along those lines, may have to do with using one row of colors as polyalphabet offsets against another row.

Another approach: how about 'sorting' the columns in each 'group', perhaps by the color values in the upper-rows, then translating the middle-row text as a cryptogram? Caesar sequentially shifts his colors...

Originally posted by americandingbat
It's nice to know I'm not the only one crazy enough to still be tearing my hair out over this.

...I'm also interested that the green values are all the same in lines 9 and 49 (those are the only two, not counting the middle block).

Yes I noticed that too, and spent some time playing with that as a cryptogram, also adding rows 1 and 57 (which seem like they might be offset) to it; no luck.

I'm currently obsessing on row 52 (62 in the original image), due to the mention of '62' in the filename. It's an even row, but it's almost entirely made up of solid 6x1 blocks, except for a single stray pixel, located at (443,62) in the original image.

Before i go to sleep, a new bit of information ive been considering that may be helpful to you...

The image is unusually wide, 800 pixels. Also the image is 80 pixels high. Also the filename begins with 8 and ends with 62 which is 6+2=8. Thats a lot of references to 8. So we got 800, 80, 8. Coincidence?

Not sure how 8 could fit into anything yet tho. Off to sleep now, enjoy

reply to post by GarethAyres


Also, 62 in hex is the letter 'b' and 62 precedes 'b'.