3
$\begingroup$

Previous ungood creation here

Attribution to: Me :)

Also sorry for all of the rot13("zrzrfcrnx") per se. I'm just trying to include a bit of a "humor" element into my ciphers. :)


Goal: Same my first puzzle, which is to uncover the secret (final) message. Each part of the puzzle (should) depend on the previous part (except for Part 1)


Part 1

17 5 c c 7 f f 4 a f 2 d 1 e 1 7 9 e 7 14 f 4 5 3 12 19 10 14 14 8 9 13 6 9 12 13 14 10 1 12 14 9 17 1 13 e 14 13 15 12 5 9 6 14 8 9 13 6 9 12 13 14 10 1 12 14 17 1 13 7 f 9 e 7 14 f 2 5 4 9 6 6 9 3 15 c 14 3 f d 10 1 12 5 4 14 f d 19 6 9 12 13 14 10 15 1a 1a c 5 2 15 14 7 f f 4 a f 2 1 c 14 8 f 15 7 8 d 1 b 5 13 15 12 5 14 f 15 13 5 1 16 9 7 9 e 5 12 5 3 9 10 8 5 12 14 f 4 5 3 f 4 5 14 8 5 e 5 18 14 10 1 12 14 e f 14 5 14 8 1 14 14 8 5 b 5 19 9 13 5 1 13 19 14 f 4 5 4 15 3 5

Part 2

Aooxxfv,x biw'vih krbzom kyti!v Il fhbhwwfac aomjwc keyyp ibpwamsgi biwf xos onwww esg rsw,l qbbfi,h sqyr kfmp, ab jllux turxes. Drbqccw, dab rcx oqiy M yswb mc xwh Wtcttag mc fvhuvi turxesv? M pycr, mt'k lhh omny K aes wtxf jslhi xs rwjxovi wbgq! Xhaq pov nxmv alef G pov mqnq gvesrbbj EUAu sj mq mpb! Hrfifmrg krnti yvcpk xhw zbhzmvy qtirsrbcq "BRL" yew dwdbblxhfa qc fst!m Vrahpgv,s I lfbbnl L'pgi oehrq mryd fqrkd wlhijl,h aqg wodtxw vspyo sxhwpt dxdcfgw!p

Part 3

Hzss3?hzss3?hzss1?Kw~k8l?rj|w?vromzllviz1?Hwvsz?V?xjzll?kw~k?kwvl?|pjs{?w~iz?}zzq?z~lvsf?{z{j|z{?}f?}mjkz?ypm|z3?V8r?xs~{?fpj?{z|v{z{?7rplk?svtzsf6?kp?lkv|t?~mpjq{?yjssf3?zloz|v~ssf?vy?fpj?h~qk?kw~k?xmzzq?|wz|tr~mt3?w~?w~1?Wpozyjssf?hz?{pq8k?w~iz?~qf3?fpj?tqph3?vqlk~q|zl?py?vk?}zvqx?lozz{m~q?svtz?h~l?kwz?~ss2omp}~}sz?|~lz?vq?rf?yvmlk?|vowzm?7hwv|w?w~l?lprzwph?xpkkzq?-?joipkzl?lvq|z?V?oplkz{?vk6?fpj8mz?{zyvqvkzsf?qpk?xzkkvqx?yjmkwzm3?pm?~mz?fpj1?Hwf?{pq8k?fpj?ujlk?MZYZMZQ\Z?O~mk?-?vy?fpj8mz?lkj|t1

Part 4

Niqp, psh uivqjh ks hphvvxxrrq ylrx N arw egjvvwmek gq tiiyxp inumcc. Imu M rxii qjrkmbp xyey M rpfq yjii xf pvmi Gsqcsmhu Whyfvvw? Nnwf, hnh psh mrfa ylrx vp sihjv ks uqpu eqp karpxp wnb cigviiw fru xrp rlqgiiw, lqy'ii wihyvtiu xt lrzr c wzb gc jmk uulewi? Klvu qzkmx si ugpgjzp zr bthvv ys jsyxi klj prwg reix, blzgu kw ymihvr hphvv ssk sag, flx yaf pnaiiw!

Part 5

14ea 1488 14ea 14f1 38 148f 1490 14f5 3f 34 14f5 1490 40 1486 148d 148f 38 1490 1490 40 14e9 38


Hint 1

Sometimes you just have to take everything quite literally.

Hint 2

Part 5 can only possibly be brute forced in its entirety through pure guesswork, however its decoding method can be deduced logically, and does contain the final message in its entirety.


A perfect answer would include:

Each part (decrypted)

How each part was decrypted

The final answer

$\endgroup$
1
  • 2
    $\begingroup$ Having read through the answer and worked on this a bit myself, this is infinitely better than your first attempt, great job at taking the feedback on board and making a good quality puzzle! $\endgroup$ Commented Oct 12, 2023 at 18:37

1 Answer 1

5
$\begingroup$

To begin, I enjoyed this puzzle much more than the previous one. (I worked out the previous one, but never posted an answer because I couldn't determine what the final answer was supposed to be.)

This is a series of ciphers, with each solution providing hints on how to solve the next one.

For the first cipher, I immediately noticed that

It contains the digits 0-9 and the letters A-F, which makes me think it is probably hexadecimal (base 16).

Converting the hexadecimal to decimal gives
23 5 12 12 7 15 15 4 10 15 2 13 1 14 1 7 9 14 7 20 15 4 5 3 18 25 16 20 20 8 9 19 6 9 18 19 20 16 1 18 20 9 23 1 19 14 20 19 21 18 5 9 6 20 8 9 19 6 9 18 19 20 16 1 18 20 23 1 19 7 15 9 14 7 20 15 2 5 4 9 6 6 9 3 21 12 20 3 15 13 16 1 18 5 4 20 15 13 25 6 9 18 19 20 16 21 26 26 12 5 2 21 20 7 15 15 4 10 15 2 1 12 20 8 15 21 7 8 13 1 11 5 19 21 18 5 20 15 21 19 5 1 22 9 7 9 14 5 18 5 3 9 16 8 5 18 20 15 4 5 3 15 4 5 20 8 5 14 5 24 20 16 1 18 20 14 15 20 5 20 8 1 20 20 8 5 11 5 25 9 19 5 1 19 25 20 15 4 5 4 21 3 5
which is immediately recognizable as A1Z26 (numbers represent the letter at the same position in the alphabet).

Substituting letters for numbers, we get
WELLGOODJOBMANAGINGTODECRYPTTHISFIRSTPARTIWASNTSUREIFTHISFIRSTPARTWASGOINGTOBEDIFFICULTCOMPAREDTOMYFIRSTPUZZLEBUTGOODJOBALTHOUGHMAKESURETOUSEAVIGINERECIPHERTODECODETHENEXTPARTNOTETHATTHEKEYISEASYTODEDUCE
or, with spaces/punctuation
Well, good job managing to decrypt this first part! I wasn't sure if this first part was going to be difficult compared to my first puzzle, but good job! Although make sure to use a Vigenere cipher to decode the next part. Note that the key is easy to deduce.

Using the clue from the first part, we use

a Vigenere cipher to decrypt the second part. Scouring the post for a likely key, I immediately settled on REDUX in the title. This turned out to be incorrect. Then I tried EASYTODEDUCE, which turned out to be correct.
The decrypted Vigenere with a key of EASYTODEDUCE is
Wowzers,t you'reh stilli here!r It honestlyy wouldo haven expectede youb toa quits bye now,t since,e youn know, my first puzzle. Anyways, did you know I used to use Cryptii to create puzzles? I mean, it's not like I was ever going to release them! This was just when I was into creating ARGs of my own! Encoding stuff using the bitwise operation "XOR" was definitely my fav!t However,o I thinkh I'vee keptx youa longd enough,e goc solvei somem othera puzzles!l

At first, I thought there must be some mistake, because although this is readable, there are a lot of extraneous letters. Then I noticed that half of the plaintext had no extraneous letters, and I wondered if they were there on purpose.

I extracted the extraneous letters, and ended up with

thirty one base ten to hexadecimal

For the third cipher, I used the directions in the second solution, and decrypted using a

bitwise XOR, with the byte 1F (which is 31 base 10, in hexadecimal).

This gave the solution

Well, well, well. That's much impressive. While I guess that this could have been easily deduced by brute force, I'm glad you decided (most likely) to stick around fully, especially if you want that green checkmark, ha ha. Hopefully we don't have any, you know, instances of it being speedran like was the all-probable case in my first cipher (which has somehow gotten 2 upvotes since I posted it) you're definitely not getting further, or are you. Why don't you just REFERENCE Part 2 if you're stuck.

Moving on to the fourth cipher, I followed instructions again, and used a

Vigenere cipher (the same as Part 2), with a key of REFERENCE

This provided the following solution:

Well, you seemed to understand what I was referring to pretty easily. Did I ever mention that I also used to like Polybius Squares? Also, did you know that in order to hold all twenty six letters and ten numbers, you're required to have a six by six square? This might be helpful in order to solve the last part, which is hidden under not one, but two layers!

The last cipher proved to be the trickiest, but by following the directions in the fourth solution (and a bit of fiddling), I managed to work it out.

First, I noted that, like the first cipher,

This appeared to be hexadecimal.
Converting to decimal, I ended up with the string
5354 5256 5354 5361 56 5263 5264 5365 63 52 5365 5264 64 5254 5261 5263 56 5264 5264 64 5353 56

This looks very much like the coordinates to a Polybius Square, as mentioned in solution #4, but, also as mentioned, the square is 6x6.

I started out with

the following square:

A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 1 2 3 4
5 6 7 8 9 0

but decrypting gave me

mostly digits, but some Zs as well.

Then I realized that

this was probably using the gimmick of combining two letters (usually I and J) that the 5x5 Polybius square uses (in order to fit 26 letters in a 5x5 grid). So I shifted all the digits one to the left:

 A B C D E F
 G H I K L M
 N O P Q R S
 T U V W X Y
 Z 1 2 3 4 5
 6 7 8 9 0  

That gave me the string

23 15 23 26 5 18 19 20 8 1 20 19 9 13 16 18 5 19 19 9 22 5
which, as everyone who reads A1Z26 fluently can easily see, says
WOWZERS THAT'S IMPRESSIVE

$\endgroup$
4
  • 1
    $\begingroup$ Good job, I was on the last polybius square :P Just a little correction, key for part 2 wasn't redux $\endgroup$ Commented Oct 12, 2023 at 18:35
  • 1
    $\begingroup$ WOWZERS. That's a lot of layers. $\endgroup$
    – Someone
    Commented Oct 12, 2023 at 18:36
  • 1
    $\begingroup$ @BeastlyGerbil Right. I tried REDUX first, and forgot that that wasn't correct (I went through it all before I began writing it up). $\endgroup$ Commented Oct 12, 2023 at 18:39
  • $\begingroup$ @GentlePurpleRain That is 100% correct! Good job! :) $\endgroup$
    – CrSb0001
    Commented Oct 12, 2023 at 18:43

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