The Legend of Satoshi Nakamato, FINAL STEP PUBLISHED.... 4.87 BTC GRAND PRIZE!

[feedo]

Do you still count the flame? They carry information to look at the image where the details matter.

You still do not follow me, you do not trust the rabbit, still many of you can not see me in the picture.

Why?

In your world it is hard to understand that nothing is the same?

Because in the picture it is not an ordinary world.

WONDERLAND

[sini]

freenode ##1flamen6

[neogabe]

freenode ##1flamen6

Perhaps it's better to keep the channel private or at least not posting the name here. It could become full of trolls.

[neogabe]

If anyone has the ability to get this scripted out faster than me, I would really appreciate it if you would share your code with me.

Maybe you could modify this code... https://bitcointalk.org/index.php?topic=766000.msg28303552#msg28303552, you have an input data example too.

[kn0w0n3]

There are 8 unique combination of border/fill/height for each of the flames on the border (10 if you count the orange ones, but they look like noise to me).

• Red,Green,Tall
• Red,Green,Short
• Red,Purple,Tall
• Red,Purple,Short
• Yellow,Green,Tall
• Yellow,Green,Short
• Yellow,Purple,Tall
• Yellow,Purple,Short

I think there is a real possibility that this a representation of octal, but not direct octal > text. I think it might be encoded a couple of levels... Something like octal > hex -or- binary > Morse > etc.

There are 40,320 possible ways to assign 0-8 to each of the color/height combinations. Also, not knowing the direction to read the information makes it more of a shot in the dark. So manually checking this might be unreasonable, and even scripting this is going to produce quite a bit of information to sift through. It would be nice if there were some clues available that might narrow the possibilities. I will try to script this out when I have some time. I think it might be best to grab a small sample rather than the whole thing.

If anyone has the ability to get this scripted out faster than me, I would really appreciate it if you would share your code with me.

I can modify my code to do this, but I don't know what you are going to do with the output, After you've represented the 152 flames as octal numbers and converted that to binary you'll have 456 bits. I'm not sure how you would convert that to morse, seeing as morse is trinary.., but let's say you could. We still don't know starting position, so we could try all 456 possibilities combined with your 40,320 possible ways of assigning octal numbers is 18,385,920 outputs. Wanna read the inner and outer tracks both directions? 73,543,680 possible combinations. Don't know whether to start with inner or outer flames? 147,087,360 possible outputs. Forgetting that maybe all 8 tracks were painted out of order and seeing as, in reality, you will also need to brute force the spaces, your output is going to be astronomical..., so what I'm saying is, unless you can think of some ways to narrow this down, I don't think it's worth the effort. I mean you could compare the output to a dictionary, but that doesn't do you any good if she/he/they didn't hide a message.

[Pan Troglodytes]

There are 8 unique combination of border/fill/height for each of the flames on the border (10 if you count the orange ones, but they look like noise to me).

• Red,Green,Tall
• Red,Green,Short
• Red,Purple,Tall
• Red,Purple,Short
• Yellow,Green,Tall
• Yellow,Green,Short
• Yellow,Purple,Tall
• Yellow,Purple,Short

I think there is a real possibility that this a representation of octal, but not direct octal > text. I think it might be encoded a couple of levels... Something like octal > hex -or- binary > Morse > etc.

There are 40,320 possible ways to assign 0-8 to each of the color/height combinations. Also, not knowing the direction to read the information makes it more of a shot in the dark. So manually checking this might be unreasonable, and even scripting this is going to produce quite a bit of information to sift through. It would be nice if there were some clues available that might narrow the possibilities. I will try to script this out when I have some time. I think it might be best to grab a small sample rather than the whole thing.

If anyone has the ability to get this scripted out faster than me, I would really appreciate it if you would share your code with me.

dude, you forgot to factor in "blob" which is the fatness of inside of the flame and comes in two varieties (fat and thin). It makes for another bit for each flame and makes the problem you are posting intractable.

Also, the rule (at least for the previous puzzle) was: no brute forcing. But of course, I am desperate too and some little brute forcing never hurt anybody

[captainnoob]

your output is going to be astronomical..., so what I'm saying is, unless you can think of some ways to narrow this down, I don't think it's worth the effort.

Yeah, I started doing more of the math last night and it just seems unreasonable to decipher or decode any information in this way. I wanted to lean towards finding some Morse code after looking at the ribbons (.--./.-. stood out as WR). But that could have easily been coincidental.

I think I am grasping at straws as a result over-thinking. I read through a description of a previous puzzle and it pushed me in the direction of looking for pure data or a bitstream somewhere. Now I am leaning towards the "finding pictures in clouds" approach.

I am very entertained by this puzzle, but I am pretty sure I am spinning my wheels. I would really love to brainstorm in real-time if anyone has a discord or anything set up.

[captainnoob]

dude, you forgot to factor in "blob" which is the fatness of inside of the flame and comes in two varieties (fat and thin). It makes for another bit for each flame and makes the problem you are posting intractable.

Also, the rule (at least for the previous puzzle) was: no brute forcing. But of course, I am desperate too and some little brute forcing never hurt anybody

I did think that there might be something to the thickness, which led me to consider a possible code 128 mini-key barcode being embedded in some way. After failing to generate any functional barcode from this, I gave in to the notion that the brush strokes look like they were applied pretty quickly (or hastily for lack of a better term). When I watch my wife paint, she gets pretty into it sometimes and can get going pretty fast. It creates quite a bit of irregularity and similar looking brush strokes. So to me it seems more reasonable to chock some details, like this one, up to artistic flow.

I am probably wrong about why I am wrong. I have not come up with anything that I have solid confidence in.

In regards to the brute forcing, I agree after giving it more thought. If there was a significantly smaller number of possibilities and I had more confidence in the idea, though, I would not consider it brute forcing.

[neogabe]

your output is going to be astronomical..., so what I'm saying is, unless you can think of some ways to narrow this down, I don't think it's worth the effort.

Yeah, I started doing more of the math last night and it just seems unreasonable to decipher or decode any information in this way. I wanted to lean towards finding some Morse code after looking at the ribbons (.--./.-. stood out as WR). But that could have easily been coincidental.

I think I am grasping at straws as a result over-thinking. I read through a description of a previous puzzle and it pushed me in the direction of looking for pure data or a bitstream somewhere. Now I am leaning towards the "finding pictures in clouds" approach.

I am very entertained by this puzzle, but I am pretty sure I am spinning my wheels. I would really love to brainstorm in real-time if anyone has a discord or anything set up.

You have an IRC channel up in this page.

[crax0r]

Troll BS aside, here is some realnew research:

First of all, I think there are several "keys" involved here.

Code:

011010 - Ribbons. Ribbons are cut and ordered, perhaps they indicate something will have to be cut and shuffled around.
011    - Bishop melting points. Melting points "melt away", perhaps indicating 011 will have to be removed.
11110  - Phoenix spikes. Connected via branch with the Dove or into a left/top leaf. Might indicate Dove connection or data insertion at leaf place.
1000   - Dove's "tails". Might indicate data insertion between the leaves or connection with 11110 bits from Phoenix. 

Some legend:

Code:

H: Height flames bits, with 011 even bits REMOVED ("melted away"), not XORed out
I: Inner color bits
O: Outer color bits

Analysis:

Step #1 - Group 3 streams line by line, separate by 7, 6, and the remaining bits, notice the "011010" Ribbon "key" is after 7th bit in each stream:

Code:

H:  1011000 011010 111111011110000100011111111110101011011000011110000110010101010
I:  0111001 011010 1000011110001001000001001101111010010111111001100100011100111010101100001101111110001011011101000000000100111100000010110100100101101010111
O:  1110101 011010 0100011010011111111010001000111000011110101000111100011101000000010100011000010110011010110010011001110000111001111010001010011100010001101

Step #2 - Mark 1st and 7th bit:

Code:

H:  (1)01100(0) 011010 111111011110000100011111111110101011011000011110000110010101010
I:  (0)11100(1) 011010 1000011110001001000001001101111010010111111001100100011100111010101100001101111110001011011101000000000100111100000010110100100101101010111
O:  (1)11010(1) 011010 0100011010011111111010001000111000011110101000111100011101000000010100011000010110011010110010011001110000111001111010001010011100010001101

Step #3 - Re-order the streams so 1st bit matches 011 pattern (wild guess):

Code:

I:  (0)11100(1) 011010 1000011110001001000001001101111010010111111001100100011100111010101100001101111110001011011101000000000100111100000010110100100101101010111
O:  (1)11010(1) 011010 0100011010011111111010001000111000011110101000111100011101000000010100011000010110011010110010011001110000111001111010001010011100010001101
H:  (1)01100(0) 011010 111111011110000100011111111110101011011000011110000110010101010

Step #4 - Merge first 7 bits of each stream into one pattern:

Code:

Codec: (0)11100(1) + (1)11010(1) + (1)01100(0) => merge "connector" bits =>  (0)11100(1)11010(1)01100(0) => 0111001110101011000

Step #5 - Find that pattern:

Code:

                                                                           [found here###############]
I:  (0)11100(1) 011010 1000011110001001000001001101111010010111111001100100(0)11100(1)11010(1)01100(0)01101111110001011011101000000000100111100000010110100100101101010111
O:  (1)11010(1) 011010 0100011010011111111010001000111000011110101000111100011101000000010100011000010110011010110010011001110000111001111010001010011100010001101
H:  (1)01100(0) 011010 111111011110000100011111111110101011011000011110000110010101010

Mind blown? There's more...

Step #6 - Extend the pattern following the same logic, to see if there is anything extra there:

Code:

What if we extend it a little bit...
I:  (0)11100(1) 011010 1000011110001001000001001101111010010111111001100100(0)11100(1)11010(1)01100(0)01101(1)11110(0)01011011101000000000100111100000010110100100101101010111
O:  (1)11010(1) 011010 0100011010011111111010001000111000011110101000111100011101000000010100011000010110011010110010011001110000111001111010001010011100010001101
H:  (1)01100(0) 011010 111111011110000100011111111110101011011000011110000110010101010

Ribbon key discovered: (0)xxxxx(1)xxxxx(1)xxxxx(0)xxxxx(1)xxxxx(0)xxxxx

Even more? There seems to be a reversed version of the key too: 010110:
I:  (0)11100(1) 011010 1000(0)11110(0)01001(0)00001(0)01101(1)11010(0)10111(1)11001(1)00100(0)11100(1)11010(1)01100(0)01101(1)11110(0)01011011101000000000100111100000010110100100101101010111
making it 010110+011010, mirror? But with different data inside, like in Wonderland?

Step #7 - Guesswork#1:

Code:

(0)11100(1) 011010 10000111...
(1)11010(1) 011010 01000110...
(1)01100(0) 011010 11111101...
(0)01101(1) missing 
(1)11110(0) missing 
(0)01011(0) missing

Step #8 - Guesswork#2:

Code:

(0)01101(1) search for 0011011 -- found but without 011010 next to it...
(1)11110(0) search for 1111100 -- found but without 011010 next to it...
(0)01011(0) search for 0010110 -- found, WITH 011010 next to it!!! On the "O" stream:
O:  (1)11010(1) 011010 010001101001111111101000100011100001111010100011110001110100000001010001100(0)01011(0)011010  110010011001110000111001111010001010011100010001101

Step #9 - Guesswork#3:

Code:

(0)11100(1) 011010 10000111...
(1)11010(1) 011010 01000110...
(1)01100(0) 011010 11111101...
(0)01101(1) missing 
(1)11110(0) missing 
(0)01011(0) 011010 11001001...

Now, last 3 steps I marked as "guesswork" as these might be purely coincidental. Chances for re-occurrence of the 3 patterns I was looking for, followed by 011010 in a random stream of (152-13) bits is around 0.2%, so coincidence is not probable but is not "impossible".

Step #10 - Internal patterns

Code:

There are two interesting internal patterns:
1) 1000011110 - exists in all 3 streams. Could be very likely a combination of Dove's tail bits and Phoenix spikes (1000 + 0? + 11110).
2) 0111010000000  - exists in the longer streams. Interestingly, if you remove the "codec line", then this pattern occurs at the same location in the "left-over" bits. Extra note, the "10000000" is 80 in hex, and happens twice, 2nd location could be interpreted as "00000001", making both re-occurrences useful for WIF key.

I:  (0)11100(1) 011010 1000011110  001001000001001101111010010111111001100100 (0) 11100 (1) 11010 (1) 01100 (0) 01101 (1) 11110 (0) 01011 0111010000000 0010011110000001011010010 0101101010111
                       ----------                                                                                                         -------------
O1: (1)11010(1) 011010 010001101001111111101000100011 1000011110 101000111100 0111010000000 1010001100
                                                      ----------              -------------
H:  (1)01100(0) 011010 11111101111000010001111111111010101101 1000011110 000110010101010
                                                              ----------
O2: (0)01011(0) 011010 110010011001110000111001111010001010011100010001101



With "codec" removed:

I:  (0)11100(1) 011010 1000011110  001001000001001101111010010111111001100100 0111010000000 0010011110000001011010010 0101101010111
                       ----------                                             -------------
O1: (1)11010(1) 011010 010001101001111111101000100011 1000011110 101000111100 0111010000000 1010001100
                                                      ----------              -------------
H:  (1)01100(0) 011010 11111101111000010001111111111010101101 1000011110 000110010101010
                                                              ----------
O2: (0)01011(0) 011010 110010011001110000111001111010001010011100010001101

Now, the remaining data can be cut/shuffled around, codec part can be removed or the first 7bits removed instead (cause in real data there should be no duplicates). All of that makes SH1TTONS of possibilities of getting 304, 296 or 256bits. In fact, about 5-60 billion.

Let's face it, the authors underestimated difficulty of making a hard & solvable puzzle. Like OnTheMF said once, it's hard to make a puzzle that is complicated&solvable, but easy to make a complicated&unsolvable one. This is the latter case I'm afraid.

The authors simply did not consider the other "view points" on the data they presented us with. To them, it might look like a straight forward case A->B->C, while in reality its more like a tree, or better, a forest, where each tree has million branches. Each one logarithmically increasing possible&probable ways of solving it. IMHO, The Authors should come forward and give some solid clue or at least "burn" the bad leads. Until then, it's a waste of time & I'm out.

[RealOnTheMF]

Let's face it, the authors underestimated difficulty of making a hard & solvable puzzle. Like OnTheMF said once, it's hard to make a puzzle that is complicated&solvable, but easy to make a complicated&unsolvable one. This is the latter case I'm afraid.

Yes, I really think that's the case. It's actually very difficult to design a puzzle that is challenging yet solvable. One has to take great care to ensure the solution is achievable without the benefit of having the solution to begin with.

With mistakes being made in previous puzzles, I think it's clear that the author is not very rigorous in their design. Further, at the time this puzzle was made it was worth a lot less, so there was much less incentive to do it properly. The fact that the prize money was once worth $100K USD is also a good indication that this may be unsolvable. Look at all the crazy solutions to other ARG's that are worth only a couple hundred bucks. You could imagine those solvers, as well as many more would be attracted to the larger prize of this puzzle. Lots of very brilliant people have worked on this and still nobody has even a confirmed first step.

From the authors perspective I'm sure it seems simple, but to those without the benefit of knowing the solution, it might not be solvable. This is basically the definition of hindsight bias.

[Ailcab]

Definitely! Well as for me it looks like simple to see this puzzle but somehow it feels its difficult to solve it. Oh, maybe im not good at looking into this art yet this is impressive.
 

Let's face it, the authors underestimated difficulty of making a hard & solvable puzzle. Like OnTheMF said once, it's hard to make a puzzle that is complicated&solvable, but easy to make a complicated&unsolvable one. This is the latter case I'm afraid.

Yes, I really think that's the case. It's actually very difficult to design a puzzle that is challenging yet solvable. One has to take great care to ensure the solution is achievable without the benefit of having the solution to begin with.

With mistakes being made in previous puzzles, I think it's clear that the author is not very rigorous in their design. Further, at the time this puzzle was made it was worth a lot less, so there was much less incentive to do it properly. The fact that the prize money was once worth $100K USD is also a good indication that this may be unsolvable. Look at all the crazy solutions to other ARG's that are worth only a couple hundred bucks. You could imagine those solvers, as well as many more would be attracted to the larger prize of this puzzle. Lots of very brilliant people have worked on this and still nobody has even a confirmed first step.

From the authors perspective I'm sure it seems simple, but to those without the benefit of knowing the solution, it might not be solvable. This is basically the definition of hindsight bias.

[Pan Troglodytes]

Troll BS aside, here is some real research(...)

well, what can I say? If your point was to show that if you mess with the data long enough then you will always find something and it is not possible to tell the real leads from garbage, then you are partly right, but ...

dude, you forgot to factor in blob

yes, it is my last comment to other member

But seriously, if you factored in blob, you would immediately see that your "regularities" fail because for blob the sequence starting at the 8th bit is 000110 and thus it invalidates the first step of your analysis.

[crax0r]

Troll BS aside, here is some real research(...)

well, what can I say? If your point was to show that if you mess with the data long enough then you will always find something and it is not possible to tell the real leads from garbage, then you are partly right, but ...

dude, you forgot to factor in blob

yes, it is my last comment to other member

But seriously, if you factored in blob, you would immediately see that your "regularities" fail because for blob the sequence starting at the 8th bit is 000110 and thus it invalidates the first step of your analysis.

What blob? The sequences that I pasted are 76+152+152 bits, there are no other flame-bits after removal of 011 pattern...

[RealOnTheMF]

Troll BS aside, here is some real research(...)

well, what can I say? If your point was to show that if you mess with the data long enough then you will always find something and it is not possible to tell the real leads from garbage, then you are partly right, but ...

dude, you forgot to factor in blob

yes, it is my last comment to other member

But seriously, if you factored in blob, you would immediately see that your "regularities" fail because for blob the sequence starting at the 8th bit is 000110 and thus it invalidates the first step of your analysis.

There's absolutely no evidence that the width of the inner flame is intentional and that it is a real data track. Even if it were valid data, it doesn't mean the above patterns are invalid. If you read crax0r's analysis, he found that these patterns don't necessarily start at the beginning of the data tracks, so that could still be consistent with the "blob" data. (Even though I'm on the fence as to weather or not that's real data.)

[crax0r]

I agree, the skinny/fat "blob" track does not follow the format AND seems to have a lot lower entropy, so it could be a text or an "AND" mask (given a lot of '111' occurrences). In the analysis above I ignore that track as it's not a solid/reliable data source.

Skinny/Fat Pros:

• Paint-style-wise, some flames seem to have inner color intentionally "painted over" to achieve "skinny" look
• 0/1 ratio seems to be 50%/50%

Skinny/Fat Cons:

• There are around 14 places that are not 100% readable, some users read those as skinny, some as fat - this doesnt happen w/colors or heights.
• They do not follow the format of other tracks
• They have lower entropy (less "random" data), they might be a  text track.

Moreover, the pattern as described in the analysis, where first 7 bits of 3 tracks build a pattern that later fully appears in one of the tracks, has only 0,03% chance of happening at random. Now, on top of that add the fact that following 6 bits in each track are the same AND they are actual 011010 ribbon pattern AND the fact that "connector" bits also are (0)(1)(1)(0)(1)(0), you get a chance of around 0,00004%. I don't want to bother calculating a chance of same thing but with 4th piece added (like in the "guesswork" steps).

We didn't "just look" at the bits, we carefully calculated chances of those things to happen, and they are crazily low, and yet, the pattern/logic is visible with naked eye without any computation at all.

[sini]

There are several places that are not 100% readable, some users read those as skinny, some as fat - this doesnt happen w/colors or heights.

This is enough to discredit the entire idea.

[rezpkt]

Let's face it, the authors underestimated difficulty of making a hard & solvable puzzle. Like OnTheMF said once, it's hard to make a puzzle that is complicated&solvable, but easy to make a complicated&unsolvable one. This is the latter case I'm afraid.

Yes, I really think that's the case. It's actually very difficult to design a puzzle that is challenging yet solvable. One has to take great care to ensure the solution is achievable without the benefit of having the solution to begin with.

With mistakes being made in previous puzzles, I think it's clear that the author is not very rigorous in their design. Further, at the time this puzzle was made it was worth a lot less, so there was much less incentive to do it properly. The fact that the prize money was once worth $100K USD is also a good indication that this may be unsolvable. Look at all the crazy solutions to other ARG's that are worth only a couple hundred bucks. You could imagine those solvers, as well as many more would be attracted to the larger prize of this puzzle. Lots of very brilliant people have worked on this and still nobody has even a confirmed first step.

From the authors perspective I'm sure it seems simple, but to those without the benefit of knowing the solution, it might not be solvable. This is basically the definition of hindsight bias.

Sorry, have to disagree here. Just because you guys are salty you’ve spent loads of time on the flames and found nothing doesn’t give you the right to diss the authors here. It will go to the person who is able to solve it, due to their way of thinking.

[RealOnTheMF]

Sorry, have to disagree here. Just because you guys are salty you’ve spent loads of time on the flames and found nothing doesn’t give you the right to diss the authors here. It will go to the person who is able to solve it, due to their way of thinking.

Welcome to the forum. Nice of you to make an account just to post that. Nobody needs a "right" to make a comment on a public forum. If you read my post, my conclusion didn't stem from my own frustration with the puzzle, but from a reasoned list of arguments. You're certainly entitled to disagree with that reasoning, that is your prerogative.

[xMossx]

There are around 14 places that are not 100% readable, some users read those as skinny, some as fat - this doesnt happen w/colors or heights.

They seemed ambiguous to me at first too. I found that looking at how she painted the typical fat/skinny flames, you'll notice qualities in the more ambiguous ones that make them easy to categorise.