11
$\begingroup$

You're an archaeologist working at the ruins of what appears to be an ancient, possibly advanced, civilization, and you've just been tasked to investigate a cave.

Well, it's not really a cave, per se - it's just one tunnel with no branches, and they believe it has marks and paintings scrawled on it. You bring a flashlight, a notebook, a phone (they didn't have the budget for cameras) and some other helpful resources like water.

As you keep venturing through the tunnel, you get a deep sense of foreboding. You move into a place where the walls are more cleanly cut; eventually, you get to the end of the tunnel where there has been some scrawling and chipping. As you reach to write it down into your notebook, a computerized voice starts up behind you.

WARNING - DISTURBANCE DETECTED. SECURITY SYSTEMS ACTIVATED.

You turn around in shock and fear, then you hear a massive shake. You look carefully ...to find that boulders have blocked off your escape path. Turning around again, you realize that you're trapped in a cave with no where to go. You try to turn on your phone, but the Wi-Fi connection is zero. Immediately, you assume this is a big prank of some kind; that Joe guy was always smirking at you when you were digging, and there definitely was equipment up at the surface.

"Okay, cool! You got me, guys!" You yell, but your voice echoes and it becomes clear this is not only serious, but very puzzling; ...how did this primitive 'civilization' ever learn how motion detectors worked; let alone how to use a voice effect to mimic English, if it was a civilization at all?

But there's no time for wandering and pondering when you're probably going to suffer some horrible fate. You turn around to find that a flat vertical slab of rock has appeared.

In this slab of rock is embedded a colorful keypad (you found that out when pressing one of the letters gave you a beep):

keypad

There is also a large grid of letters haphazardly stitched into a wall:

grid

What sequence of letters should you enter in the keypad?

Hints:

Look carefully. Very carefully.

Removed because, while I assure you there is grid deduction involved, a lot of steps are required to get to it.

Look at the letters. Oh, and also look at the colors in the keypad.

There are two parts to this puzzle that are pretty much independent from each other; Both answers combine to make the solution.

Compare the letters in the keypad to the letters in the board. Notice the differences. Oh, and you need to look at a 7x7 grid.

You are playing around with colors, considering manipulating their hex components when your phone beeps, and you're suddenly able to contact someone appearing to be very concerned about you. "You've been lost for a week. We found your location but it appears to be surrounded with some sort of ultra-steel so we can't get to you...Oh, there's a keypad and grid? Can you take a picture and send it to me? Huh... it looks like something's hidden in the keypad. There are multiple representations of letters faintly under the ones in the keypad. Scrubbing off the uppercase Y in the keypad reveals a lowercase y... yeah, I'd suggest using those letters instead for some of them. The w in the first row actually has one letter missing, by the way. You found little changes in color, so maaaybe since there's three changes, you could try a slitherlink or some sort? Oh, there's another part?...

...Attempting to match letters together doesn't feel right... ..The odd pixels. Comparing them with the background and doing something with the differences feels like the right way to go. Say, this is a 9x9 grid... And numbers on a 9x9 grid... Aha! The colors on the keypad. They could mean something - analyze their pixel colors, possibly their deviation from normal gray... 'prm2'? Maybe that could be a filter of some sort, I'm not sure. This is interesting-oh, sorry, I have an appointment to go to! See you later!"

$\endgroup$
2
  • $\begingroup$ hmm, i found a rot13(iregvpny erqurevat... qvq lbh zrna erq ureevat? be vf vg ab erq ureevat, orpnhfr vg vf abg pbeerpgyl fcryyrq?) $\endgroup$
    – Lezzup
    Jul 11, 2023 at 18:19
  • $\begingroup$ @Lezzup rot13(vg vaqrrq vf fhccbfrq gb or erq ureevat, ohg bar bs gur e'f orvat zvffvat pbhyq or vagrecergrq nf n pyhr...) $\endgroup$ Jul 11, 2023 at 18:22

1 Answer 1

3
$\begingroup$

Partial answer - data collection and a bit of progress? maybe?

I looked 'very closely'...

so closely, that the naked eye alone wasn't enough. Several of the colors that look the same are slightly different.

Here's a new image of the letter grid with those colors replaced by more clearly distinct ones.

letter grid
Slashed pixels were already that color in the original image. Key to the other colors:
image listing colors in the new image, and hexcodes they correspond to in the old image
There are also several 'one-off' colors. I haven't changed those:

list of hex codes and coordinates of those colors

Oh, and I should probably look closely at the keypad too.


image of keypad with colors changed

keypad color list
The border is all the color 101010, which I assigned to "dark yellow" in the other image. Presumably that should be taken as the 'default' version of the dark color in some way.

So, what messages can we get out of this?

In the wall grid: The hidden text along the bottom reads "comparison". (On the right, it reads "red her[r]ing" - not sure why it was included at all, in that case.)
Presumably, we'll be making some comparison between the new colors and the old colors - finding out exactly how they differ.

In the letter grid:
The letters AimSTVXZ are marked as a different color from the rest. This seems important, but it doesn't anagram to anything useful-looking.

The empty cells have a message, PRm2. Perhaps this could be short for "primes", or "prime 2", or something to that effect?

Let's make those comparisons.

I'm going to take the one-off cells and compare them to the 'foreground' in the rest of their cell.

table of compared color hex codes
They're all pretty similar!. In fact, the offset is in the range [-5,5] for each one.


This is where I'm stuck. There are lots of things I could potentially look at, but nothing jumps out at me. Here are some things I've tried:

  • looking for commonalities in characters that share a background color
  • looking for commonalities in characters that share a foreground color
  • looking for patterns in how background/foreground colors are assigned
  • interpreting the grid as a Sieve of Eratosthenes
  • reading all the letters in a certain color as a message
  • combining all the one-offs into a 5×3 grid (there are too many repeats and unused cells)
  • using the one-offs as 'instructions' to toggle pixels
  • looking for other ways to alter the characters in the grid to make them all letters (since so many of them are at most 2 pixels off from being 'valid' letters)
  • subtracting the hex codes of the keypad backgrounds from 255; using those as a message (they all end up in the printable ASCII range, with most of them being alphabetical!)

Here's a dump of all the transcription I've done, in case it's helpful for someone else.

one-off color:      r   g   b   very close to:      r   g   b   difference          R   C
R1C8 (Z)    0C0D10  12  13  16  dark yellow 101010  16  16  16  -4  -3  0   2   3
R1C9 (o)    2D2B2E  45  43  46  dark purple 2B2B2B  43  43  43  2   0   3   3   2
R2C2 (B)    0D0D0C  13  13  12  dark red    0D0D0D  13  13  13  0   0   -1  3   1
R3C6 (N)    151419  21  20  25  dark green  151414  21  20  20  0   0   5   2   1
R3C8 (2)    2D2E2F  45  46  47  dark purple 2B2B2B  43  43  43  2   3   4   3   2
R4C3 (n)    0F1011  15  16  17  dark red    0D0D0D  13  13  13  2   3   4   3   2
R4C8 ( ⸩ )  2D2E2E  45  46  46  dark purple 2B2B2B  43  43  43  2   3   3   1   1
R4C1 (=)    A3A7BF  163 167 191 lime green  A2A6BB  162 166 187 1   1   4   5   3
R5C5 (⌞)    0F0F10  15  15  16  dark red    0D0D0D  13  13  13  2   2   3   5   2
R5C9 (9)    2D2D2D  45  45  45  dark purple 2B2B2B  43  43  43  2   2   2   5   3
R7C1 (:)    0E0D0D  14  13  13  dark red    0D0D0D  13  13  13  1   0   0   2   1
R7C2 (ʌ)    0F0D12  15  13  18  dark red    0D0D0D  13  13  13  2   0   5   2   2
R7C6 (Y)    181717  24  23  23  dark green  151414  21  20  20  3   3   3   1   1
R7C9 (÷)    2C2D2D  44  45  45  dark purple 2B2B2B  43  43  43  1   2   2   5   2
R8C2 (P)    100D0D  16  13  13  dark red    0D0D0D  13  13  13  3   0   0   1   3
R8C7 (ᴀ)    151013  21  16  19  dark yellow 101010  16  16  16  5   0   3   2   2
R9C7 (h)₁   111011  17  16  17  dark yellow 101010  16  16  16  1   0   1   1   1
R9C7 (h)₂   111011  17  16  17  dark yellow 101010  16  16  16  1   0   1   5   3
dark red    0D0D0D
dark yellow 101010
dark green  151414
dark blue   151515
dark purple 2B2B2B
dark pink   363636
lime green  A2A6BB
    
pastel pink A6A6AA
pastel yellow   A8A8A6
pastel orange   A8A8A8
pastel cyan A9A8A8
pastel blue A9A8A9
pastel aqua AAA8AA
    
pastel red  A2A6BC
bright red  A3A6BC

dark purple ABABAB
bright yellow   F6FFFE
white   FFFFFF

A   B0A99A
b   84A88C
C   A2A6BC
d, m, P, T, X   AA9AAC
e, r    90B5AE
F   B09ABE
G   A9A184
h, S, V 91A89C
i, L    AFB590
J, n    9A92AD
K, U    91A9AC
O   ACA49C
q, Z    B29EA0
w   C6A28C
Y   A89C84
$\endgroup$
1
  • $\begingroup$ rot13(uzz, lbh ner pybfr... guvax nobhg jung tnzr vf cynlrq ba n 9k9 tevq :Q) Oh, and by the way, rot13(gurer ner gjb cnegf gb guvf chmmyr gung ner cerggl zhpu vaqrcraqrag sebz rnpu bgure; obgu nafjref pbzovar gb znxr gur fbyhgvba) $\endgroup$ Jul 12, 2023 at 12:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.