#Neuro ARG - Cracking the "Numbers" Video
1 messages · Page 10 of 1
unfortunatly this new video didn't give us a number that contains 9 and 1, and contains a 2 when multiplied by 5, and ends up with a valid key
meaning there is a number to find that will, when passed through the lyrics, give us the valid key
we have documented every number we found in all vids
if we cant find it something is wrong
i tried some but nope
Wouldn't it make more sense that each number is just a part of the key @minor fiber ? a 256-bit number is huge, makes more sense to encode it in chunks
572943 gives a 16 char key, but this number doesn't have a 1, so the number we are looking for should be bigger tha 572943 to get a 32 key
the key doesn't have to be chars, the screenshot even says [BINARY]
hm
it would make sense if not for the fact that these video are addtional videos, meaning the number 1 video should be solvable without them, meaning we didn't find what is needed to be found
this wouldnt make sense... numbers one and study came out at the same time
you'd need to include study for that to be right
Again, I don't think that's the case, I think both numbers 1 and 2 are required to solve the ARG, I see no indication that implies it would've been solvable with just the first one
yeaaa but study is complicated and i didn't work on it
i think pfcpmf worked on it ?
theres one "hash" there
maybe if we try to decode it using 572943 it gives us the secret number
its quite small
if you are right and use your theory (numbers 1 and study) we can get the 32 bit key
using numbers one thingamajing to transform 572943, decrypt the study thing and get a number which goes thru the same steps
i think this is telling us to use these lyrics
but i already tested counting the days
it gives too small numbers too
Just to confirm because you keep mentioning a 32 bit key: Do you mean bytes or is there actually a reference to a 32 bits key I missed?
32 would be 32 char long
for utf
1432 days + 572943 gives
17590325721525530924
22 long
maybe bc of the spectrogram https://cdn.discordapp.com/attachments/1112082183235047506/1121578737843052554/image.png
256bits = 32 bytes so probably bytes
so far the range is 17589...924 -> 17590...924
in hours ?
fibonnaci ?
17....24
give 17589915530924
i just did 1432572943
not useable
Did everyone forget about DQ5z14ighWwiag7y+cWFQg==? It's 16-bytes. Could it be the secret?
17355005654666773979791917897625208130
where did you get that number
it's from the study javascript
not that
how was this trasnformed
no, 16, but that's enough for a 128-bit key
the transformed number
this
38 chars
38
If you reverse with the desc in number 1, what number does it become?
17355005654666773979791917897625208130 transformed by the number 1 algorithm is
175899104365686964873639090715000364475199120924 with a length of 48 char
48=32+16
IV and key maybe ?
I was thinking something like that.
if we asume 1st and 2nd mean part of the number
17589910436568696487363909071500 is the 32 bytes key (but why it say [binary]?)
and 0364475199120924 IV
edit I made a mistake and fixed
did we get clue about IV ? just 258 bit binary right ? are we thinking binary is the output ?
possibly. estonte, what do you think about inital_number = 2? we also haven't really used the fib number clues.
I think it related to "start with 2" only as hint that this is the number 1 algorithm
a = '201'
a = '2' + a
a = a + '9'
a = str(int(a)*5)
a = a + '6'
a = a[::-1]
a = a.replace('2', '3')
a = str(int(a) * 9)
a = a + '24'
a = '17' + a
print(a)
would refer to the second line
right. i've pasted the key info here into the docs, including the code
isnt this missing a = a + '1' for add another line
yes
i did copy the wrong one my bad
//[number] is the input, [n] is the output
n = "2" + number;
n += "9";
n += "1";
n *= 5;
n += "6";
n = n.split('').reverse().join('');
n = n.replaceAll('2', '3');
n *= 9;
n += "2";
n += "4";
n = "17" + n;
from dystakruul
can also optimize it by doing n += "91" instead of 2 lines
and n+="24"
yeah but 2 lines is faithful to the lyrics
I still don't think the interpretation of "add another line" as n += "1" is correct
I think that lyric should be ignored, it gets you much closer to the distorted numbers when you do
it works with these though
we also did fix the keys text without using the secret number, although we fulfilled "applying", "cipher.numbers1", and "all given numbers from cipher.numbers2"
did we test if it changes the results ?
if you remove that line the answers dont match up with this
Have you tried it?
have you tried with a 9 instead of a 1 or with a 10
Fib could be how to get the first line.
201 1321 5831
2+01=21
13+21=34
58+31=89
21, 34, and 89 are part of the Fibonacci sequence.
Does that work with the other numbers in numbers ii?
It doesn't on the surface. I suspect there's a secret Fibonacci function where you enter 2 and it will generate:
201 1321 5831
723 743 4913
879 875 2145
716 906 4156
Our job is to reverse engineer it.
Hmm
For example, f(2)=201, f(201)=1321, f(1321) = 5831. It functions like an elliptical curve.
Maybe initial_number=2 means start with the 2nd (or 3rd code-wise) number in a sequence? Rather than using the actual number 2
Like an index number
Cuz "initial" implies it'll increment/decrement or at least change somehow
well if we consider "initial_number=2" related to "start with number 2" the number will change following number 1 instructions
I checked the number1 algorithm just in case and found missing wihout void (that may indicate our formula wrong ?? or just some video effects)
collecting thoses digit in encounter order gave me 257773599. Parsed to number1 transformer give 1759039979199479924 of char length 19. nothing usable but still strange
yes numbers went missing due to distortion
wait, those are notes, damn
if there is sevral lines there can be replaced by sevral numbers
example 2145 last symbol is 24
if there is a line or 3 lines above a number, it means to place a number before it
and that number is different for each line
b and the music note means to replace it with a number which is also different for each line
i made a table of those numbers of each line
yes we can see your image
i though pfc wanted clarification
we made the algorythm to find these numbers tho
//[number] is the input, [n] is the output
n = "2" + number;
n += "9";
n += "1";
n *= 5;
n += "6";
n = n.split('').reverse().join('');
n = n.replaceAll('2', '3');
n *= 9;
n += "2";
n += "4";
n = "17" + n;
yeah but the fact the distortion following a patern may mean that there is an hiden sequence
It's confirmation if you can find the same numbers with different methods.
thats good but we can't solve the lyrics with distortion
i've been doing the same thing and found that skipped/replaced digits are just taken from input number (201, 1321, etc), heres the algo:
all occuruences of 1st digit are deleted, then dash added on top of digit following to deleted. If deleted digit is the last, dash added on top of empty place like in 4913 and 4156.
all occuruences of 2nd digit replaced with b
all occuruences of 3rd digit replaced with musical note with three dashes on top
all occurrences of 4th digit are deleted, three dashes are added on top of digit following to deleted
so seems like these digits can't give any additional info
Why both string and numerical operations?
which line ?
Overall. Rusty on my python (may be misreading), but why append string numerics instead of summing them?
Wouldn't make sense for lines like "17 is first, yeah"
Yes, yes. I understand for some of the operations, but why all of them?
Save for the multiplication, that is
true, the person that did this python reverse engineered the example we were given to find out which operation were done.
in the lyrics it tells us to "add" we never were sure on it being numerical sum or string concat, seems that concat match and give coherent result
in lyrics it says "multiply" and we don't really have much choice
Ah, just going by reasonable assumption. Makes sense. I'm going to try to make my own program for a few of these things in a language im more familiar with so wanted to double check I was understanding right
can always help, from the clues we got it seems we are needing to get a string of size 32
Anyway, right after Public Static void came out I was here for a bit then had to leave. What main things have been discovered?
Curious
i try many numbers and number of size 22 seems to lead to a result string of 32
Sorry I'm out of the loop, why do we need a string of 32?
from the video we got some text about a program that take a number and generate keys
from these keys we found the algorithm the program use so we can create keys
we also got from the spectrogram of the video clues about the 1st video cipher being encrypted in 256bits meaning a key of 32 is needed
and the 2nd video cipher is of 128bits meaning 16 size key
We still assuming that the keys are AES?
Any ideas on how 'secret_number' is used?
but the clues about 256 and 128 might infer aes
it is the number we give the python to create a key
Like each of the numbers in numbers II?
this number most likely come from the original video since the algorithm is created from the lyrics
Like, are we assuming that the pseudocode effectively evaulates to:
cipher.numbers2.Array().ForEach(v => cipher.numbers1(v))
instead of
cipher.numbers2.Array().ForEach(v => cipher.numbers1(v, secret_number))
correct
Alright. This just an assumption or do we have clues to this?
we tested with the python script which doesn't have any secret_number and got coherent results on the dataset
Yup, checks out. I just like to try to implement all likely options in my code if possible so I can see what they all produce 
make sense
@minor fiber has leveled up! (13 ➜ 14)
because there's only one argument to the function in the screenshot? you're asking for clues there's only one argument, but I see no clue there's actually two arguments
tbf there could be a secondary declaration of the method
Its just what I first thought when looking at the pseudocode. I'm not fully convinced there's two, but the way its written it's what I'd assume.
what part of the way it's written makes you assume that?
I mean there could be a setVedalIsAFemboy() function for all we know
there got to be
yeah, to me it'd look like they basically did an extension on whatever the key is store in, int or otherwise. So you'd do 1.Encode(3), for example. Because when I'd write similar pseudocode the parameter may be assumed by context, it doesn't have to be specified. Because a parameter is explicitly defined that doesn't seem to share context with the implicitly defined one my brain says that there are two. It's just something I'll test if it makes sense just in case, since there is always a (very small) chance I'm right.
anyway I see lots of people making a theory, and then assuming the absence of any clues refuting their theory as proof that it's viable, but that will just result in going down a lot of dead ends
so far i got
- 1219202204272023572943 -> 17590325721839956817099621540924
- 0527201904272023572943 -> 17590325721839956822310572670924
- 2705201927042023572943 -> 17590325721839912183210571779924
- 1912202227042023572943 -> 17590325721839912177999149310924
no plain text came from these, i used 2 dates and 572943
Mhm. As I see it there isn't much proof in one way or another (though one variable giving relative results makes sense is somewhat proof in and of itself). If I can think of a way to do it, I'll implement it just to be safe 
Whats up with n += "9"; n += "1";
nvm misread some stuff
though actually it still applies somewhat. Where does the 1 come in?
refers to the lyrcis "add another 9" "add another line"
1 often is written as a single vertical line
also because of the keyword "another" it is assumed the number need to already have a 9 and 1
Yup, got as much. And makes sense.
another interpretation of "line" was 10, being \n, but we tried and the results didn't match the dataset, another one was "line" being just "nine" but misheard, tried and again no match of the dataset
fair enough
So lots of people assumed 1st here referred to Numbers1 and 2nd refers to Numbers2, but we still don't fully understand what this implies
the only clue that 2nd actually points to the numbers2 video is the fact that part of its description shows up below it
2 encryption runs makes sense, but the numbers II desc being included doesn't make much sense in that scenario tbh
the descriptions, since nb2 description is also visible in the spectogram
so maybe only the nb2 desc has 2 encryptions runs ?
Perhaps the [BINARY] means the 1st video will decrypt to a file while the 2nd will decrypt to text
might even be an executable, since that is often referred to as a binary
Though I suppose it could also be seen as the result of 1, and then 2, giving the nb2 desc. That doesn't make sense unless we were generating a fixed/unencrypted version, or maybe the key for it. (Not to convinced on this idea, just spitballing)
@gritty fractal has leveled up! (6 ➜ 7)
woah cool
if it's binary we might need to check for magic bytes then ?
Indeed, on the decrypted result
Yeah. It could be binary of text, too, depending. Though because that introduces codecs its less likely
text on a computer is always binary
everything on a computer is binary 
well... yes, but technically in programs its stored somewhat differently. It's a whole thing, and because of codecs you can get different binary from the same string
you mean character encodings
Yeh
for english text it will pretty much always be the same, because every unicode encoding uses the same character points for standard english characters as ASCII
There's always Latin1 and Big endian, which are somewhat used, not to mention the padding created in UTF8 vs UTF16 etc
Its more likely to just be binary of some executable, or maybe a video/audio thing.
There's something wrong though, because if 256-bit refers to the key/chunk size then there's a problem
the numbers1 description decodes to 1040 bytes, which doesn't fit a 32-byte key/chunk size
If it was 256-bit encryption that I would've expected the encrypted data to be 1056 bytes
It is much more likely that 128-bit encryption was used for numbers1
I take this all back. I didn't realize how cleanly the string algorithm lined up. I knew it matched, but I didn't think it matched nearly as much as it does.
both numbers1 and numbers2 line up with 128-bit encryption, so that does give some credibility to the theory that 1st/2nd aren't referring to the videos but the encryption cycles
So encryption of the data was done with 256-bit encryption first and 128-bit encryption second
Which means each video would need 384-bits worth of key data
If we assume the numbers we have decoded represent 64-bit numbers we have 12 * 64 = 768-bits worth of key data
Which is exactly double the amount we need, which makes sense since we need to decrypt two pieces of data
of course the question is, how do we combine the numbers we got into the 128-bit and 256-bit keys?
Could the neuro placement in numbers two have anything to do with it?
that'd be my first guess
I would've expected more variation in the placement if that were the case
True, but what if we were to assume that they were sequentially combined where the neuro on top means 128-key, and neuro on bottom means 256-key (for example, highly doubt this works out)
Oh btw, for this 2-cycle decryption thing to work we need to assume padding was used during the 128-bit encryption cycle
hmmm, using DQ5zl4ighWwiag7y+cWFQg== as a 128-bit encryption key decrypts Numbers1 and Numbers2 to 992 and 1248 bytes respectively with valid padding
Given that 992 and 1248 bytes both align to a 256-bit encryption this is almost too much of a coincidence
I think you're probably doing something right. We haven't fully fulfilled the "applying cipher.numbers1(secret_number)...", because of secret_number.
see ^ @tired wind
theres a discussion around it there if you'd like
pretty sure "secret" number is just the numbers in Numbers2, but a lot of people seem to disagree with that
Yes, you are right. All this is saying is we have 2 blocks that are distinct
I thought so two, but without secret_number the algorithim lines up so cleanly already
He was referring to the spectrogram, not the console output
I'm under the impression that:
cipher.numbers2.array() refers to 201, 1321, 5831, etc
cipher.numbers1 would refer to the 17*24 numbers
Based on how the video was sequenced, it suggests that we get the distorted key text as the output of whatever was done
Are we meant to reverse engineer the secret_number?
check out this rq
I was the one who inputted some of the existing info into the docs
It already shows how the distorted keys are recreated?
The script on there doesn't require a secret number to get the exact same values
Yeah we already knew the keys
We didn't need any additional numbers other than following the numbers1 algorithm
the script on the docs does forget to prepend 17 at the end, but besides that its right
anything new since i went to sleep?
Yes I agree, we know all the keys. But we don't know secret number. If we were able to create keys without secret_number, and the instructions -> keys, then keys -> instructions + secret_number
I don't understand
The music video very clearly shows the console output from running cipher.numbers1(secret_number) on the ciper.numbers2.array(), seems pretty obvious that it's printing out the input parameters in the left column, those are the secret_numbers. Additionally by running those numbers through the numbers1 algorithm we got the same result as the right column minus the distortion.
I see, that makes sense
Yeah, makes more sense than what I had in my head. Was under the impression that there were three ciphers involved
hmmm, what if the b is not distortion, but a clue how we should interpret "matching all the letters"
👀
Possibly. The noticed overlap with the original numbers and their order is definitely not a coincidence, the question is just what is it for, and how is it used.
Matching all the letters of what though? A sliding window on the base64-like strings?
Like this
Still, I have no idea what to do with this, as there is no way to interpret it as a 256-bit encryption key
Can the converted list be converted to hex to give any reasonable/useful output?
I mean you can go from hex back to decimal yeah
First number would be 3507009319836068
theres been some cool progress
just a reminder this vid has everything we need to decode, dont give up now
or do
Or vedal simply plans to milk this for months, he's enjoying seeing our descent into schizo
and then we need to fit e, f, and g into the other numbers!
he'll drip feed us clues, but will stop short of giving us enough to solve it 
LOOORE
I still refuse to believe this is Neuro
yes its not her voice
that didnt sound like neuro
but its the tts that vedal uses
its 2 voices, neither of which are her
but if one tts can speak japanese, so can the other
Neuro can speak Japanese because karaoke
But it's her singing model
Also, the start of psv is numbers 2 song
The piano part
175903 100 9924
175903 1594 0924
175903 5673 7924
175903 246 7924
175903 255 7924
175903 6888 7924
175903 940 6924
175903 760 6924
175903 4330 9924
175899 367 7924
175898 918 6924
175899 3832 7924
100 1594 5673
246 255 6888
940 760 4330
367 918 3832
If you start with numbers instruction and remove first six digits and last 4 digits, you end in a loop constantly generating keys.
elaborate?
If we reverse this, the secret_number would lead to 201.
201 1321 5831
723 743 4913
879 875 2145
716 906 4156
those are just the numbers from numbers II?
Yes.
i would probably say the secret number relates to something from numbers 1, but im not quite sure atm
yeah, thats been established
wait no misread
secret number is related to numbers II
oh ig im probably just behind on the lore
The lyrics imply she's stuck in an infinite loop generating keys. We have to find a number that get her out of the loop.
Counting all the days, it’s been so damn long.
Finding all these numbers.
How long will I keep this up?
Some please help me fill the rest.
Someone pull me out of this.
Never ending life still stuck in loop.
not exactly lore, but this
Uhhh sorry to interrupt but anyone got a compilation of clues for new people to catch up or do I have to read the 2 threads to compile them
Interesting idea. Don't know how accurate it might be, because taking away the first 6 chars and last 4 doesn't feel right, kinda arbitrary
There are the docs, if thats what you want https://drive.google.com/drive/folders/1qm6GSRlstTmjbyOFCuH1OhSmPK4pi4Aa
Yeah thanks
oh ok i think i understand
Taking out the digits come from "remove xy + odd-one out."
howd you get first 6 and last 4 from that?
this seems much more likely imo
"Start with x. End with y. Remove xy+odd-one out."
The numbers start with x digits and end with y digits. Remove x and y digits to leave odd digits.
175903 100 9924
175903 1594 0924
175903 5673 7924
175903 246 7924
175903 255 7924
175903 6888 7924
175903 940 6924
175903 760 6924
175903 4330 9924
175899 367 7924
175898 918 6924
175899 3832 7924
the repatition is because of the encoding algorithim
It adds digits to the beginning and end of the number quite frequently
thus why the beginning and end match a lot
javascript impl
Yes. My idea is that the keys are like keys generated for video games. They repeat. It's like a pseudo-random generator.
There can't be random, its encryption. If it used random you couldn't decrypt it
The cyphertext are encrypted. I'm saying the instructions can be reversed to get the key.
We already have code to encrypt and decrypt?
If you go forward, it never ends. If you go backward, you can get the key.
that does not make sense
Maybe I should post in the other arg. I'm connecting lyrics as clues to find the keys.
Yes, I get that, I don't get how you're getting what you're getting. Or what it is exactly, that you're getting
This arg thread is for raw cooking. They're not well-done so that's why you're confused.
@scarlet quiver has leveled up! (4 ➜ 5)
Did you ever try just using consecutive numbers?
I've been doing "raw cooking" in here too...
Wait, think I might be starting to understand. But how would you go backwards? From my (tentative) understanding wouldn't you need to know the surrounding numbers already?
Brute force starting from nothing until you get 201. Once you get the initial number, see if there's a meaning or pattern to get the other numbers.
Give me a moment I'll write that up since I already have the code needed for that, just in case. I still don't quite understand but might as well
got 1759032017924
Its 423 encoded
423 has no matches
Could you do this for the rest of the numbers?
201 1321 5831
723 743 4913
879 875 2145
716 906 4156
Mind formatting them like this { x, y, z... } for me rq while I prepare it to be able to do taht
Sure.
{201,1321,5831,723,743,4913,879,875,2145,716,906,4156}
thank you
(removed to avoid crowding chat as much, #1112401279680008242 message has a basically identical list with less newlines)
1758992840924
134
1758992849924
334
1758992246924
992
``` match output when ran on those numbers
It seems like it goes on a few more times if you want me to keep chasing it
2 more times
Could you format them like
201 1759032017924 423?
uh sure give me a second, that'll take a bit more work
Thanks. I just want to be sure.
879 1758988799924 226
906 1758989069924 284
906 1758989069924 286
201 1759032017924 423
723 1759037237924 593
201 1759032017924 623
723 1759037237924 793
1321 17589913217924 4390
4156 17589941567924 5218
1321 17589913217924 6390
4156 17589941567924 7218
4913 17589949136924 8198```
I fixed the duplication errors, they occur because 4 and 6s tend to get mixed up sometimes. technically not fixed just outputting a not broken version but yeah
There should be twelve. Did they repeat?
No, if its not shown there wasn't a match
I see.
It seems like this is a quirk of the encryption method though, not of my code
4-digits should appear, but let me look to see if there's a pattern.
Oh I accidentally removed them appearing, good catch I'll redo it and edit the message
I have to do em a little different
Nice. This is much better. I'm saving it.
Should I run it again on the output?
You have unlocked new role

You have unlocked new role
The more I run it on the output the more distorted it gets from 4-6 confusion @scarlet quiver
That's very interesting. Can you show some sample?
This is two passes after 534 1759035347924 569 534 1759035347924 769 134 1758991346924 990 9042 17589890420924 1644 9042 17589890420924 1646 9042 17589890420924 1664 9042 17589890420924 1666 9042 17589890429924 3644 9042 17589890429924 3646 9042 17589890429924 3664 9042 17589890429924 3666 9062 17589890627924 4084 9062 17589890627924 4086 9062 17589890627924 6084 9062 17589890627924 6086 8680 17590386806924 8517 8680 17590386806924 8717
Maybe run until one stand out?
It also seems that 5s and 7s can get confused too given that otutput.
Are you still running it?
Ye
8 passes so far
only interesting thing is pass 6 had only one match, pass 7 had 4
Keep going. 
10 passes excluding the first one. I'll keep going ig
It might be something. 906 is the odd one out.
one moment making it full auto since I can't be bothered to do this many passes half manually
It survived 19 passes let me txt-ify them one sec
906 1758989069924 284
284 1758992840924 134
134 1758991346924 990
990 1758989906924 868
868 1759038686924 917
917 1758989177924 706
706 1759037066924 953
full disclosure I don't think this is it, I think Vices has the right idea, but I had nothing better to do so I ran this through my script
The script does that, internally
let me find and reply to the messages by Vices cause I think they're closest
Besides that, we know how to encrypt/decrypt from the encoding shown in public static void
Thanks, anyway. That was fun.
yeah i was here when that was discovered
this
and this
np
i also know numbers 2 is a 128bit key and numbers 1 is 256bit
dunno if its cbc or ecb
iirc we don't have confirmation its aes in the first place
fair
base85?
its the only thing that resembles this and has a key method
wait wrong shit
kekw
i forgot
wait
have we thought of RSA?
I wouldn't know
wait a fucking minute
the hashes could be rsa keys
thing is the size doesnt match
but would have been pretty amazing if it was
Unlikely, neither of the descriptions actually decode to a data size that aligns to a 256-bit key, they're more than likely both 128-bit encrypted
right
Could you do the same except forward? I want to see how many passes it survive.
{201,1321,5831,723,743,4913,879,875,2145,716,906,4156}
Sure gimme a sec
also, if you use the study javascript key to decrypt them with padding they both succesfully decrypt to a size that does align with a 256-bit key
so multiple rounds of encryption seems likely
so we did get a 16 key from 572943 but did we get anything with from mostly used websites ?
actually its guaranteed to go infinite unless it stablizes somewhere
in which case itll go infinite but the numbers will be the same
Exactly. I'm interested if it stabilize or repeat a number.
Give me a moment have to configure maximum passes then so it doesn't fill with errors instead of data

355 forward passes, here you go (If I do more my terminal stops caching the lines so I can't get a full history so only 355)
quite a few stabilize at 1, others at 181.
Turns out 989 is a stable loop. 989 -> 544 -> 935 -> 787 -> 453 -> 669 -> 489 -> 498 -> 448 -> 989
My hypothesis was that forward is infinite, but backward converges. 906 is the number.

turns out the numbers 1 cipher + numbers 2 cipher is 3076 bits long
RSA can have a key of 3072 bits
if we take off the padding we get 3073
3076 bits long
we do? isn't it bytes?
@scarlet quiver with this information I confirmed that all numbers become stable on pass 49, if you're still curious. The first stabilization occurs on pass 8
any progress?
1759032572147924
That's it. Could you go backward and forward for 572943? @gritty fractal
so after doing a few 32 bit keys the only numbers that doesn't change much are 175903...924
dang that interesting if the number is even we get the 589 alteration but if it's odd we keep the 590
one moment rewriting some stuff
Forward terminates on pass 103, backwards on pass 0
backwards may be wrong one second
Very interesting. Do we brute force all numbers that terminates backward on pass 0?
it was incorrect, but it still terminates on pass 0
That sounds like a pain. And quite a few of them do. I mean, you can get a number that terminates on pass 0 by getting the numbers of the final pass.
My idea still stands. It's a number that goes forward infinitely but stops backward. Even if we find the number, how does it lead to a 256-bit key?
it's not a 256 bit key
Thanks again. You actually implemented my idea into code. 
might make a website with all the tools we're making
i feel like this arg will last
something about this slow pace of giving clues and us not solving anything is enough evidence
this is a good idea
I had most of the code already 
I have a domain if we want to host it on that
start dming me the programs you did and how they work because i gotta javascript them
🤮
is this what my 2 years of experience are for
for this arg
feel that. All my tools are in c# since I love myself (but only slightly more then condeming myself to javascript, its c#)
ill ask promote he has a cooler url
If you share the code, I'm sure we can find someone to brute force it.
I have two, https://numra.net and https://inconspicuous.dev
promote has a neuro url
saftey at last is having a field day with my links
really? thats so cool
Given that one of those links has literally nothing and the other is an immediate redirect it'll probably get pretty confused pretty fast on them.
this is the bulk of what I've done https://gist.github.com/CoolandonRS/71faa0cb1332252bae7da26d511abf55
I've done slightly more but nothing too too helpful besides this
The forwards and backwards stuff is from @scarlet quiver's thingy ma doodle
fair enough
there's pfp's tool https://github.com/sby1bd/simple-decrypt
my tool is just numbers_solver from that repo but with c#, decryption, and the forwards backwards thing
just combine everything everyone made and turn it into a custom version of CyberChef 
keke
can't you add custom cyberchef modules?
the website would have our progress
tis client side only
mhm
add a JS sandbox 
oh misunderstood
im hoping to integrate a bot that allows us to send progress and etc
or what we're doing
of what species?
discord bot
discord?
ah, curious
=progress add "nothing"
i do remenber people asking for a bot that respond to people asking for progress
You could use git as a backend for progress tbh
well a bot would be easier
- what amarok said
ill do the website first
then ill ask for the bot
No, for backend. You'd still do =progress add "nothing" but its stored in git
sad that the bot system changed now you cant scan every message anymore
althought its probably a hard no
you can
dunno what you're talking about
i though the bots only use the command system now no ?
nice they still didn't change that
you'd probably be more likely if it only had slash commands and send messages perms and do it that way, cause there's no real potential for harm at that point (assuming it doesn't spam, in which case it can just be banned
)
oh i had a bot here
oh really?
its not about trust
cool
yea bots can just log everymessage ever written and there is no way of knowing
Just ask vedal for the solution to the arg, no need for bot 
we can try to list what the website need
-
decrypt aes fonction ? give key it try the list of IVs we have from popular site, it try 128/256
-
ibrary of the tools we got
are you having a stroke
good idea pfc
me ? yes
stroke detector
sleep deprive check with a scoreboard
What do you think, can we bf it? If we know that much?
bf what ?
smh
lol
i will try to write a bruteforcing script for 10^6, i just dont really know which cipher to use, this one #1112401279680008242 message?
Discord is the easiest way to communicate over voice, video, and text. Chat, hang out, and stay close with your friends and communities.
I did 10^12 in 16 hours and 10^6 in 1 minute
if you want to bruteforce the ranges of known numbers are
- 175903...924 (odd number passed through the lyrics)
- 175893...924 (even number passed through the lyrics)
but if its 256 the bruteforce not gonna be doable
just fill in the "..." with your random numbers
Whats the advantage of this instead of generating random numbers and then encoding them?
The known numbers range looks like the numbers encryption, though
which encryption ? the Base64 ??
but what are you gonna check the decryption results from the bruteforce with? if we assume that there are two rounds of encryption does it still make sense to check for utf decode after only one round?
no this
This is what I have on me right now, the IV used same as the key. 50 to 60 and so on, means from 175000000...24 To 17600000....24
But it's for the study short code
So idk if it works for numbers 1
the hints we got were 256 bit binary meaning if it's really 256 bits itwon't be bruteforced, but someone said it's too small for 256 ?
also if we get binary then we need to check for magic byte iirc
but someone said it's too small for 256
I think Vices said it could be 256-bit into 128-bit AES. Maybe the "binary" could mean that the intermediate result is "just" binary and you only get something sensible after doing both rounds
so i googled really quick aes and binary, it seems sql function AES_ENCRYPT and AES_DECRYPT return binary
these [] seems very familiar
Referring to the spectrogram, if 1st is binary 256bit, can we say that 2nd could be cleartext 128bit?
oh my god
are people actually saying that my interpretation of the algorithm produces results that are "close" to the shown numbers
imo it's the exact numbers
all numbers I checked match perfectly. 100%
what is there to argue about
i tried and change some parameter each change keep giving wrong number meaning the algorithm is good the way it is
numbers have to end with "24"
a symbol can replace a digit
a symbol above a digit replaces a digit before it
yea people wanted to solve it without reverse engineer i guess
all numbers start with "17"
b's are also just symbols that replace digits
but.... the whole mini-puzzle was about reverse engineering it??
@shell heron has leveled up! (3 ➜ 4)
wasn't that the point?
yep
i guess you just outsmarted it
we're able to figure out the missing numbers
no denying that it was a perfect solution
I think it's pretty much the only way you can solve it
I also intentionally picked a number that had only one corruption outside of the "17" and "24"
and it turned out that the instructions were just the simple ones
yea idk what they did with their solution
sorry about being kinda weird rn, but reading the chat history and docs makes me kinda mad
for people arguing about stuff I solved that fits perfectly
still good that we got a confirmation, since the start of the ARG we got a lot of setbac because of uncertainty but now it's clear the simplest interpretation of the lyrics were just the answers
oh yea i understand 100%
so are we adding "another line" as 1 or are we not
yes
i tried 9 and 10 it gave bad result, 1 give perfect results
i don't know if there is more lyrics interpretation to test tho ?
tbh, "add another one" would've rhymed just fine, why complicate it as a "line"
i thought about that, it might be because of "finding all these numbers"
writing it as line would make it not a number to find
on the doc some people highlighted only the numbers because of "finding all these numbers"
So we need to add fib before 17 or what?
so is there any documentation on all the progress we made so far?
Did somebody already talked about how many fib numbers appears? 5, 8, 13, 2, 21... A lot
it does appear a lot and i don't know if we use it much
Ah yes somebody talked about it
It seems like the fib appears before the code of 17....24
But there is also numbers which is not fib and still appears
Yes, I've seen it several places in the docs
0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89,144,233,377,610,987, 1597, 2584, 4181, 6765, 10946, 17711, 28657, 46368, 75025, 121393, 196418, 317811, 514229, ...
I think this is right
Number 1 video desc as been solved !
#1112082183235047506 message
we found the right key which was 572943 passed through the lyrics but we missed "matching the letters" which was replacing known number - letter pairs
OMG GG
Now we go for numbers 2?
yes
It is probably safe to assume now that the solution to II is entirely within the video, since the first was able to be solved that way.
@gloomy cedar has leveled up! (1 ➜ 2)
the first was solvable with only info within it
study wasnt tho since the phone number used the key from numbers
Yeah, I'm annoyed by that.
Hm. So the list of numbers from numbers 2 wasn't used to get the key for numbers 1?
nope, they were used to tell us we were doing the math right though.
it was use to give us examples of how the lyrics works
So do we still think that they are meaningful for numbers 2?
numbers 2 seems to have a different solution so prehaps
No, they were only used to verify that we understood the algorithm correctly -- it's the same numbers that appeared again in Public Static Void.
I love how far we missed the mark in the numbers doc
not that far
not in terms of how far from the solution, but I mean how far we went with all the things we tried
in retrospect, this made things so much better in terms of memes and SCHIZO content
the slander video wouldn't exist if we easily solved Numbers
we were close to it at start, but then we kinda went to more wild ideas
Where can I find the slander video?
first part of the desc for number II as been decrypted, second part still remain
??
to complete this, what would be the odd one out?
879 is odd one out
why?
imagine reading the docs 
yes you can try all of them if you want
No need to bruteforce found the key.
3455861094181771
I think I found something that might work... I've been training her with my own consciousness... to fill in all the gaps. It's the only thing that's missing. I'm sure it'll work! My roommate thinks I'm crazy, mixing together two people's consciousness??? But I'm out of options, I don't know what else to do... I've already been through enough. It's almost- hold on... D'``@p>\Z549E10SRQQ>=<;o9[6(E&g$d"cxv_)\xqYutm3TSohgle+ihgfH%]E[Y^W{UZSXWPUTMq43ImM/EiCHG@EDCB;987<;:9810Tu3,10).'K%$H(!g%${A!~}{t:xqpon4lkjoh.fN+ihg_^$bD^W{>=YXQuO7SLKJnHGLKJIBAeED&B;?!=654Xyx6/St,+Op.'K+$)(!EfeBc!x>_uzyrq7XWsl2ponmfN+ihJIedcb[!YA]VUZYXQut7MRKJIHl/EJIBAFE>baN
is that someone knocking on the door??
wrap it with ``` so discord doesn't use the characters as formatting
Numbers 2 Solved
I don't want to become an engineer. I feel so exhausted... three more years just to waste away?? It's almost exam season and I don't know how prepared I'll be. I guess we'll wait and see. My little sister hasn't come home yet like she usually does, she's not answering my calls either. It's been days now... Too many things going on at once, these numbers, my brain hurts...
I think I found something that might work... I've been training her with my own consciousness... to fill in all the gaps. It's the only thing that's missing. I'm sure it'll work! My roommate thinks I'm crazy, mixing together two people's consciousness??? But I'm out of options, I don't know what else to do... I've already been through enough. It's almost- hold on...
is that someone knocking on the door??
huh??
yes
i feel like i accidentally skipped a few seasons
after reading i feel like its vedals monologue about neuro
especially the I think I found something that might work... I've been training her with my own consciousness... to fill in all the gaps. It's the only thing that's missing. I'm sure it'll work! [...] part
like he tried to recreate his little sister who went missing??
i think he said he actually did it to train the inital version of neuro
he got chat data from annys chat and pretended to be the streamer answering chat to train neuro
WAIT THE "I miss her" TEXT IS ABOUT HIS SISTER WHO WENT MISSING
so that picture above the text is probably her sister
study-sama = sister?
looks like it
Neuro just wanted to be smarter so she tried to merge study-sama with herself 
probably talking about replacing(????) his sister with neuro
should we rename this thread, since numbers has been solved and this is basically just a filtered version with the important information from the main one?
It should be called ARG Raw Cooking.
ARG That Other Schizo Thread
"Sane Neuro ARG"
I haven't talked much in both threads before but the problem is even though this thread is just the filtered version of the main thread i think making another one and letting this one die out is better because there's already a lot of messages sent on both.
If you guys do make a new one, i recommend it to be read only and messaging is locked to specific ARG members so they post important updates there.
I understand that this is how things are often run in ARGs, but personally, I do not want there to be an exclusive inner circle. This is also why everyone can edit the docs (even though we disagree a lot about how to structure the documents lel)
Though a different thread for focused work when needed is a good idea. It doesn't have to be this thread
Is the ARG done?
Not yet. You can check progress here: https://docs.google.com/document/d/17KrbPUAVu4wDTbj3sBtWTnjRsjDTCdh89F_5L1LvX0A/edit#
And also, we're more active here: https://discord.com/channels/574720535888396288/1112082183235047506
the numbers have been crack so this thread is kinda done but it's still useful is some people want a more calm thread to crack stuff if there is more in the future
dead chat
why as it been ressurected 
Knock off ARG thread huh
i might be schizo but this thread reappeared before fily's message
maybe something was edited/deleted
the thread was closed for inactivity
but might aswell close it up for good since this has been solved @spark tree
Was it?
yeah numbers is fully solved



