[ARG] The Pizza Code Mystery

[Ammat_the_Ankh]

@Astij

I’m trying out several anagrams for things on KXBM after JohnnotJohn solved a few (nice work by the way), things that stick out are the name Major Frank Abrams and localdesertsingles, the former in particular because it contains the letters BMRF. Anyone got any ideas?

Nothing real, I’m afraid. Even if you isolate “BMRF”, you can’t get much out of “Major Frank Abrams”. The only interesting anagram I could find was “Fan Jar Marks BMRF OAR” which I don’t think means much.

“localdesertsingles” on the other hand produces too many results to effectively comb through. Picking out a few key words I got “Code” “laser” “sent”. “Sign Tells Laser Code”?

[Steven_Wells]

I hit the inspect element button on Chrome while hovering over the radio ad on kxbm news article and found this line of stuff.

https://www.kxbm.net/article/?view=all-clear-given-for-black-mesa-evacuation-zone-10360

I do not understand it or its relevance but for some reason it made me really want to turn on the lights and never sleep again.

Also, what does HALOS stand for? Or does it stand for anything? H could very well stand for Horn. “Horn’s Automated Logarithmic Operating System” or something.

[DeathTBO]

@Steven Wells

I hit the inspect element button on Chrome while hovering over the radio ad on kxbm news article and found this line of stuff.

https://www.kxbm.net/article/?view=all-clear-given-for-black-mesa-evacuation-zone-10360

I do not understand it or its relevance but for some reason it made me really want to turn on the lights and never sleep again.

Also, what does HALOS stand for? Or does it stand for anything? H could very well stand for Horn. “Horn’s Automated Logarithmic Operating System” or something.

Already been found.

On the subject of HALOS being an acronym, that is an interesting question. Perhaps it could give us a little insight on where to look.

Edit: Check the wiki if you think you found something: https://thepizzaisalie.wikia.com/wiki/

[Ammat_the_Ankh]

I assume HALOS is a cross between HAL 9000 and GLaDOS

[Mack0438]

@thewizard99

@DeathTBO

I think I’ve finally went through every monitor, but either we have already seen them or they seem useless.

This one caught my eye:

Tons of numbers, but I don’t think it’s useful. There was another one very similar, but I couldn’t read it.

The Data on the right does not correlate to the data on the graph ( as one would assume) maybe they are cord’s in longitude and latitude?

It does correlate. For every blue dot on the graph there is a coordinate on the right. The data on the right has to be read from top to bottom and the graph from left to right.

[Mack0438]

Okay I think I have something for the 752 Hex code.

What if “BENALOHPAILLIER” isn’t exactly the password for the code but a hint towards how we can decrypt it (nothing new).
And the “21 goes into 1” could mean that we have to decrypt the first 21 numbers with Benaloh decryption (first part of the password) and the rest with Paillier.

To decrypt with Benaloh we need four numbers (p,q,r,y) and we have four: “21.32 , 16 , 129.52 and 1997.37” or Code A - D.
But I don’t get how to decrypt with Paillier and how to decrypt a hex code with Benaloh yet.

[JohnNotJohn]

I’ve been looking over some of the earlier pages of the thread, and I noticed that Storm said we were close two years ago, when the HALOS file was found. All of the work that has been done since then has focused on ultimately solving that one puzzle. Meaning everything Storm has done since then must have been subtly hinting at the solution to that puzzle. However, he might just be messing with us. Also, I personally recommend the 80s and 90s, corresponding to about a year ago. Lots of interesting stuff, including acronyms, anagrams, analyzing static, large amounts of numbers, and more. Old thoughts sometimes inspire new thoughts. See if anything occurs to you while re-reading.

Just keep your eyes peeled, and you might find something.

[Mack0438]

Okay I’ve written a little bit of code in Java to decrypt with Benaloh. It is based on some other code I found:

https://pastebin.com/TMiwq0mR

[code]import java.math.BigInteger;

public class Benaloh {
//Code to decrypt
private static BigInteger halos;

public static BigInteger decrypt(BigInteger ch, BigInteger ph, BigInteger qh, BigInteger rh, BigInteger yh){
    BigInteger nh=ph.multiply(qh); 
    //Objects created for encryption don't necessarily contain the private key
    if(ph==null || qh==null)
        return null;
    BigInteger exp=ph.subtract(BigInteger.ONE).multiply(qh.subtract(BigInteger.ONE)).divide(rh);
    BigInteger dexp=null;
    BigInteger ymp1q1=ch.modPow(exp, nh);
    //Find if m is zero. By construction we have y^exp is not congruent to 1 mod n, since u^phi(n) mod n is 1, then y^exp
    //can only be 1 iff m=0. (y^0)*u^phi(n) mod n
    if(ymp1q1.equals(BigInteger.ONE))
        return BigInteger.ZERO;
    //The decryption will be done using exhaustive search
    for(BigInteger m=BigInteger.ONE;m.compareTo(rh)<0;m=m.add(BigInteger.ONE)){
        dexp=m.multiply(exp);
        if(ymp1q1.equals(yh.modPow(dexp, nh)))
            return m;
    }
    return null;
}

public static void main(String[] args) throws Exception {    
    //HalOS 752 Hex Code Mystery
    halos = new BigInteger("b32b003a35badd66577c2", 16); // only the first 21 numbers of the halos file
    
    BigInteger ph = new BigInteger("1001"); //21.32 , 16 , 129.52 and 1997.37  ||  Code A - 1001 - Code B - 0851 - Code C - 3914 - Code D - 0914
    BigInteger qh = new BigInteger("0851");
    BigInteger rh = new BigInteger("3914");
    BigInteger yh = new BigInteger("0914");
    
    System.out.println("The decrypted halos file: " + decrypt(halos, ph, qh, rh, yh));
}

}[/code]

Feel free to use it as you wish.

Furthermore I checked the 21 first numbers with Code A - D in every combination but it doesn’t decrypt …

[Cranken]

@Mack0438

Okay I think I have something for the 752 Hex code.

What if “BENALOHPAILLIER” isn’t exactly the password for the code but a hint towards how we can decrypt it (nothing new).
And the “21 goes into 1” could mean that we have to decrypt the first 21 numbers with Benaloh decryption (first part of the password) and the rest with Paillier.

To decrypt with Benaloh we need four numbers (p,q,r,y) and we have four: “21.32 , 16 , 129.52 and 1997.37” or Code A - D.
But I don’t get how to decrypt with Paillier and how to decrypt a hex code with Benaloh yet.

Well if you look at the HALOS File you can see that there are letters in the blocks and afaik you can’t decrypt them with the benaloh/paillier cryptosystem.
IMO it has to be another encryption system.

[Mack0438]

@Cranken

Well if you look at the HALOS File you can see that there are letters in the blocks and afaik you can’t decrypt them with the benaloh/paillier cryptosystem.
IMO it has to be another encryption system.

But you can convert it to decimal or systems you can decrypt with Benaloh.

[Gunsrequiem]

We should all focus on what JohnNotJohn, says, especially if he’s a Welshman ; )

[pointless]

So, I’m late to the party (which ended around page 130). Nevertheless, I went through the source code myself and here’s a completely deobfuscated version of the javascript code from the KXMB article (which hasn’t been posted so far):

$.hit = function(event, rect) { var elem = $(event.target), X = event.offsetX, Y = event.offsetY; if (X === undefined) { X = event.pageX - elem.offset().left; Y = event.pageY - elem.offset().top; } return X >= rect.left && X <= rect.right && Y >= rect.top && Y <= rect.bottom; }; (function() { 'use strict'; function switchRadio() { on = !on; audio.volume = on ? 0.05 : 0; image.attr('src', on ? './assets/SSBuZXZlciBzdXNwZWN0ZWQgc3VjaCB0aGluZ3MgY291bGQgYmUuLi4=' : './assets/SWYgeW91IGludGVuZCB0byBnbyBvbiwgdGhlbiwgSSBiZWcgb2YgeW91LCBwcm9jZWVkIHdpdGggZXh0cmVtZSBjYXV0aW9uLg=='); } var Context = AudioContext || webkitAudioContext; if (Context !== undefined) { var on = false, image = $('img[src="./assets/SWYgeW91IGludGVuZCB0byBnbyBvbiwgdGhlbiwgSSBiZWcgb2YgeW91LCBwcm9jZWVkIHdpdGggZXh0cmVtZSBjYXV0aW9uLg=="]'), context = new Context, gain = context.createGain(), audio = new Audio('./assets/RGllLi4ueW91IGFsbCBkaWUuLi55b3UgYWxsIGRpZS4uLg=='), quantum = 52, chars = [[' ', '.'], ['/', '-']], quanta = {'.': quantum, '-': 3 * quantum, ' ': 3 * quantum, '/': 7 * quantum}, code = '64ab95eca3a4c9d56841ec8f55c641f4a2a16924c08ca862584605736908c0aca8e296ac07f1e70ec28aecc05c882d537ecc085e624b096bc05af17d260ca42c51e96282ee60c0ad0c762767c2868a1acf82490585c0ce085e62c62505ab719e6624e29..., cachedImage = new Image; cachedImage.src = './assets/SSBuZXZlciBzdXNwZWN0ZWQgc3VjaCB0aGluZ3MgY291bGQgYmUuLi4='; gain.connect(context.destination); gain.gain.value = 0.25; audio.volume = 0; audio.loop = true; audio.play(); (function audioLoop(n) { if (n === code.length) { requestAnimationFrame(function() { audioLoop(0); }); return; } var oscillator = context.createOscillator(), d = parseInt(code.charAt(n), 16), l = d % 2, t = d < 4 ? 0 : 1, char = chars[l][t], stop = false, dur = quanta[char]; oscillator.frequency.value = 700; oscillator.connect(gain); setTimeout(function() { if (on && ' ' !== char && '/' !== char) { oscillator.start(0); stop = true; } setTimeout(function() { if (stop && ' ' !== char && '/' !== char) oscillator.stop(0); audioLoop(n + 1); }, dur); }, quantum); })(0); image.on('click', function(event) { if ($.hit(event, {left: 158, right: 168, top: 74, bottom: 83})) switchRadio(); }) } })(); (function() { 'use strict'; $('#footer img').on('click', function(a) { if ($.hit(a, {top: 126, bottom: 141, right: 124, left: 58})) window.open('https://steampowered.com') }) })(); console.log('[HEADCRAB ART]');
The morse code is completely extracted from the hex string hardcoded in the source code and has nothing to do with the noise sound file (see the lines 54-57). What’s interesting is that the format of this string is the same as the HALOS file we are stuck at. The string is treated as a sequence of hex digits, a sound is played when the digits is at least 4 and the effect is long and when the digit is odd. The digits are chosen so that it looks random at the first sight.

This suggests we are not dealing with cryptography, but with steganography. This goes well with Stormseeker’s comment that CIA couldn’t brute force it. I tried looking at the binary representation of each hex digit of the HALOS file, but I couldn’t find any pattern which would prove it’s not random (this is not a problem in the code from the KXBM article, disregarding the obviously nonrandom ending). It’s been randomized really well and there’s just too little data to do any statistics on. I’m out of ideas for now.

[Mack0438]

@pointless

So, I’m late to the party (which ended around page 130). Nevertheless, I went through the source code myself and here’s a completely deobfuscated version of the javascript code from the KXMB article (which hasn’t been posted so far):

The morse code is completely extracted from the hex string hardcoded in the source code and has nothing to do with the noise sound file (see the lines 54-57). What’s interesting is that the format of this string is the same as the HALOS file we are stuck at. The string is treated as a sequence of hex digits, a sound is played when the digits is at least 4 and the effect is long and when the digit is odd. The digits are chosen so that it looks random at the first sight.

This suggests we are not dealing with cryptography, but with steganography. This goes well with Stormseeker’s comment that CIA couldn’t brute force it. I tried looking at the binary representation of each hex digit of the HALOS file, but I couldn’t find any pattern which would prove it’s not random (this is not a problem in the code from the KXBM article, disregarding the obviously nonrandom ending). It’s been randomized really well and there’s just too little data to do any statistics on. I’m out of ideas for now.

Okay I used the exact same method in that code for the 752 Hex Code and here is what I got:

-…-…-.-----------.–.----.-.–.—…--------.—…-…-.----.—…–…--.—.-.–.-.-.--------------…-.-…-----.-----…-----…----.-.—.-.-…—.-------.--------.----.-.------…–.—…---.--------------.---------.–.-------…-.----…–…-.-.—.-----------.----…–.—.-.—.---…–.—…-.–.--------------.–.----.—…---…—.-----.-.—.-.–.—.–.-.-…----.--------…-----.-.-----.-…-----.-----.-.----.–.—.------.-.-.-.------.-----.----…----.--------…—…-…–.-.-.------.-…—.----.–.-----.–.--.-----…-.------.------…–.-----.-----.–.--.–.-----------.------…-----…–…--------.-.–.-…–.-.-.-.-------.-…-----------------.--------------…-------…-.----.—.---.-.—.-----.—…------------.–.—…---.—

But I don’t know how to seperate them.

EDIT: nvm I did a mistake. The morse code is not correct.

[Pro-account-2000]

Good stuff pointless!

@Mack0438

Okay I used the exact same method in that code for the 752 Hex Code and here is what I got:

here’s what I got

-  -   . --.--..---. .. .-.- - .. ..-   .-.-.--. --.  .   . ---- .--  -.  -. .-. . .- . - .---.-.-...-.-  . -  -.... ..-.-  .-.--     -... . .-. . -  -.- ..--.-. -.------ --.. - -.--..    -. -..  -.- .-.----..----. .--.-..-. .. ---.--.  - ----  -.  - . ... ....-----.- .-.-  -. -.. . -.- ...   .. ---   . -. -.-..-.-.....- .- .-.. .-.  ...  .-- .-... . ..- . .. --- -. . .  -.-- --.--...  ----- - -.-.. -  .---- ..... - ..-. -. -.- ..-.-- - . . .-..-- -.--- -..-  ..-. -..----.  ..-    .   .- - - -....- .  -.. .-.- -. -.... -- -. --.--  . -...-- .-.---  .. .-..- ..-.. -- .. .. ..-...--.-. ---.--  -...-  -.  ....-.-. - -- -   -- . . . ..-...- .  .--..--..-..-.--- .-..-.-.-..---   --...-.  . -.-. -.- -.- . ..- .--.. -.-  ..---.---.-. -- -..  ..- ---

Unfortunately it translates to gibberish (I tried swapping dots and lines with each other but that doesn’t produce anything interesting either.

[pointless]

Mack0438, Pro-account-2000: The exact same method doesn’t work and it’s been already tried. I fact, it would be rather disappointing if it worked. I think this is just a hint and something different must be tried.

[Mack0438]

Okay I rewrote my code and now I get this for the 752Hex code:

Output ==>

.-./ /.- / /./ -./.-.- -.-.- - -.-.-.- - / - -./ -.- -.-./.- / - -./ - -.-./././ -.- -.- -.-.- - /.-.- - /./ - / / / -./.-.-.-.- / -.-.- /./.- - /./.- - / -.- -./ - / -.-./ -./.- / -.-.-.- -.- -.- - - -.- -.- / / -./.-./ /.- - - - -./ - -.- -.-././ -.- -.-.-././././ /.- - - - / - / -.- - / -./.-././.- -.- / - -.-.- -.- - /.- -.-.-.-.-.-.-./.-.- - -./.- /.- -.-.- - -./ / /./.- - /.- - - / /.- -.-./ -.- -.-.-.-.- - -.-.-.-.- - / -.-.- -.- - -.- -./ - - /.-.-.- -.-.- -./ /.- /.-.-.-.-././.- - /./.-./ -./ - - -./ - - - -.-.-.-.-.- -.- / -.- -.-././.- -./.- - - / -./.- -.- / - - -././ / - - /.-.-.- / / / -./.- - /.- -.- - -.- -.- - - - - -.- / -.-./ -.- - - / -.- -././ - - -././ -.-.- / -.- - - - / - / - -.- / -./ - - /.-.-.- /.- -./ -./ - / /.- -.-.-./.-.- -.-.- - - - / /.-.-.-.-.- /.-./.- -.- - -./.-././ -.-.-.-.-./ - - - - -./.-./ - -.- -./.- -./.- -.- / - -.- -.-.-./.-./ - / - / -.- - -.-.-./.- -.-.-.-./.- - -.-././ - -.- -./.- - -.-.-.-.- - / / - -.- / / /./ -././ / -.-./.-./.- /.- - - - -.-./ -././.- - -./ -.- -.- /.- -./.- - - - -./.-.- /.- - /.-.- -.-.-././ - /.- - - -.-.- / -.- -.-.-.-./ / - - / -.- - -.- / - -.- - - /.-.- / - - / - - / - -.- - - -.-.- -.- -./.-.-.- -.-.-././.- - - -.- / /.- -././ - - - -.- -.- - /.-./.-.- /.-./././.-.- / - / -./ -./ - -.- - - -.- / - /./ -.-.- - -.-.- - -.- - -.- -.-.-.- / -.- - -.- -.- -.- - -.-.-.-./ / /.-.- - - -.- -././ - /.- -.- - /.- -.- /.- -.- / -./ - -.-./ -.-.- - -./.- -.- /./ - -.-.-.- -.-.-.- -.- - /.-.-./.- - - /./ - -.- /.-.-.- 

EDIT:
Okay sorry didn’t see it

[Skuros]

Has anyone considered that “tempus omnia revelant”, instead of meaning “given time, all will be revealed”, might mean something more along the lines of “the time associated with (something) will reveal all”? Especially given that the interval on which “…aren’t you alone…” repeats is defined so conspicuously, it could be a viable reading of the phrase.

Granted, I don’t know what specific times there are associated with the ARG. The only ones that come to mind are the half-lives of the Niobium isotopes given on the wiki page, the date at the end of the page (12-12-12), and maybe the date that HALOS.txt showed up (November 7, 2012?). Maybe one of these could be used to decrypt the 752 hex file?

[TornadoAP]

@faed

It probably doesn’t mean anything, but the 2000 millisecond interval on the “…aren’t you alone…” is actually defined in code as:

21.32*16/129.52+1997.37;

Which actually is 2000.0037245213093 milliseconds.

This definitely seems important.

One thing I notice instantly is the 1997, which could very easily be a date.

This is the number sequence in full by the way.

21321612952199737

Now what becomes immediately apparent in the 21 in several parts of that code.

Now for straight just changing the numbers into letter it comes out as this.

2 1 3 2 1 6 1 2 9 5 2 1 9 9 7 3 7 B A C B A F A B I E B A I I G C G

I’m going to try to alter it to see if it come out with anything actually useable.

[ramonster]

I’ve tried converting the 752 code to morse using the same codec that kxbm uses for it’s morse signal(I know it’s already been done but I was bored).
It was complete gibberish.
At the time Stormseeker has said “someone is already close” Gunsrequiem suggested that each hex character must be shifted/rotated.
So I’ve tried shifting every character with 1…15, before converting it to morse. Still nonsense. BTW if you want to poke around with the morse codec, here’s the code.

I’ve been thinking about what’s relevant for the hex code and what’s not.
We can assume that all things we need to solve the code was already present at the time we found the hex code, or not if you interpret the ‘Time reveals all things’ in a way that we just have to wait until some clue pops up.
Stormseeker’s mention that someone was already close sounds to me that all clues we need were already present.
kxbm.net wasn’t in the air during the time we found the code, which means that we might not even need the website for the hex code.
Moreover, how do we know that the website is even related to the ARG? Sure it has some cool hidden messages, but none of them are really pointing at something ARG-related.
‘Thomas Lee’ is indeed a anagram of ‘meet halos’, but so is of ‘has omelet’, ‘hole steam’ and ‘metal hose’.

I think in order to solve the hex code, we need to stop looking at the site and look at the things we already had 2 years ago.
Assuming that someone actually was really close just as Storm has mentioned then we have to look at things that we were doing 2 years ago.

[TornadoAP]

@ramonster

I’ve tried converting the 752 code to morse using the same codec that kxbm uses for it’s morse signal(I know it’s already been done but I was bored).
It was complete gibberish.
At the time Stormseeker has said “someone is already close” Gunsrequiem suggested that each hex character must be shifted/rotated.
So I’ve tried shifting every character with 1…15, before converting it to morse. Still nonsense. BTW if you want to poke around with the morse codec, here’s the code.

I’ve been thinking about what’s relevant for the hex code and what’s not.
We can assume that all things we need to solve the code was already present at the time we found the hex code, or not if you interpret the ‘Time reveals all things’ in a way that we just have to wait until some clue pops up.
Stormseeker’s mention that someone was already close sounds to me that all clues we need were already present.
kxbm.net wasn’t in the air during the time we found the code, which means that we might not even need the website for the hex code.
Moreover, how do we know that the website is even related to the ARG? Sure it has some cool hidden messages, but none of them are really pointing at something ARG-related.
‘Thomas Lee’ is indeed a anagram of ‘meet halos’, but so is of ‘has omelet’, ‘hole steam’ and ‘metal hose’.

I think in order to solve the hex code, we need to stop looking at the site and look at the things we already had 2 years ago.
Assuming that someone actually was really close just as Storm has mentioned then we have to look at things that we were doing 2 years ago.

NO.

KXBM is completely related to this ARG. bmrf.us, a confirmed site connected to the ARG, redirects to kxbm.net. KXBM also has references to previous clue’s we’ve found, such as AI5 among others.

And so what if KXBM.net was not on air during the original search, that doesn’t mean it can’t contain clues to the solving the ARG.

I agree with what you say about our priority being solving the Hex code, but that isn’t going to stop me or anyone else from searching KXBM.