UA - Yet another method

Pendra

P
So, this was supposed to be a post in one of the UA threads, but it turned out to be more interesting, so why not create a new thread.

So after I checked the recordings, I (as others before) concluded that there is no sequence with the same sound three times after each other.
No --- or |||.

Let's see the elements by length.

1 digit: 2 elements
-
|
Binary

2 digits: 4 elements
--
-|
|-
||
Still pretty binary

3 digits: 6 elements
--|
-|-
|--
-||
|-|
||-
We miss 2 elements already

4 digits: 10 elements. Assigned arbitrary single length markings to each variation
--|- : 0
--|| : 1
-|-- : 2
-|-| : 3
|--| : 4
-||- : 5
|-|- : 6
|-|| : 7
||-- : 8
||-| : 9
Decimal? Huhh

5 digits: 16 elements. Again, assigned arbitrary single length markings to each variation
--|-- : 0
--|-| : 1
--||- : 2
-|--| : 3
-|-|- : 4
-|-|| : 5
|--|- : 6
|--|| : 7
-||-- : 8
-||-| : 9
|-|-- : A
|-|-| : B
|-||- : C
||--| : D
||-|- : E
||-|| : F
Hexa?

6 digits: 26 elements. Again, assigned arbitrary single length markings to each variation
--|--| : 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
English alphabet.


You know what? Even if this has nothing to do with the solution, I freakin like this encoding method. You can go from Binary to Decimal then Hexa then to Alphabetica just by increasing the segment length.

7 digits : 42 elements (THX avgroenink). It rings one bell, but I doubt it is the right one.
--|--|- : 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
|-|--|| : 0
|-|-|-- : 1
|-|-|-| : 2
|-||--| : 3
|-||-|- : 4
|-||-|| : 5
|-|-||- : 6
||--|-- : 7
||--|-| : 8
||-|--| : 9
||--||- : .
||-|-|- : ,
||-|-|| : ?
||-||-- : !
||-||-| : :
-|--|-- : #
Oh well, extended alphanumerical?


I checked the recording from here https://forums.frontier.co.uk/showthread.php?t=143220 and I believe the segments are 5 digits (now I belive 6 or 7). The whale noise + the silent part is the divisior (or not, see below). There are few segments with 6 digits, but they look wrong to me. Either the sound is too "weak". If I wrote an Analog -> Digital conversion, it would not register. Or it is completey inside the howling.

I put down 6 parts as complete as I could plus some segments from other parts. Oh BTW, most parts contained 11 segmentes in the recording I had. Maybe they are 5 mins long or something happened in every 5 mins tha broke the recording.
Only clear sections were noted. Nothing from whale noise or "weak" part. There is a definite frequency of some patterns.

--|-- : 0 > 5
--|-| : 1 > 3
--||- : 2 > 7
-|--| : 3 > 10
-|-|- : 4 > 2
-|-|| : 5 > 3
|--|- : 6 > 1
|--|| : 7 > 7
-||-- : 8 > 3
-||-| : 9 > 11
|-|-- : A > 2
|-|-| : B > 6
|-||- : C > 5
||--| : D > 6
||-|- : E > 2
||-|| : F > 2

If we used the 6 digits version and English letter frequencies, I would say the 9 is E, the 3 is T, 2 and 7 are A, I or O etc.
I need to sleep, you continue :D.

Edit:
The howling (or whale noise) is a parity. There are two types of tones (_-_ and -_~)and they follow each other. 1...2...1...2...1...2...1...2....
It can also be "sent" and "received".
Or a metronom.
Or it can be an itiner how to put the sections together. You can't put a setction that starts with a 1 next to a section that ends with a 1.

They seem to be superimposed over the high-low notes. You hear 1 or 2 in every 28 secs, exactly. Regardless of the high-low tones.

The weird thing is that high-low tones are slightly longer in frequency (~30 secs), which creates a shifting. That is why the 6th tone is sometimes outside of the howling. The frequency of the high-low tones are changing slightly between 3,7-4 secs. But some segments are ~27 secs long, others are ~30. As some were missing an element. And I'm not talking about the ending or the starting segment of a section but segments in between.
This change of length also ensures that the notes and the howls don't go massively out of sync.
 
Last edited:

Tephlon12

Tephlon12
Well, I don't know much about breaking codes or codes in general. But something I thought might be useful was to examine a visual represtentation of the audio. So I took some audio of the artifact and loaded the file into Audacity and I came away with the following information. There is a long "whale-like" noise and that always seems to be followed by five sets of the "growling/clicking noise. I'll post the screencapped image from audacity here. Its something I though may be helpful, if its not, or if its old news then feel free to just ignore me. lol

EDUknownArtifact.jpg
 

Pendra

P
Also, if you check the pattern, you see "fat" dots and "thin" dots. The fat dot is the down note, the thin dot is the high note. Using this, you can write an Analog to Digital conversion with relative ease.

I can't fight the feeling that the whale noise is more than just a divider, but couldn't find a pattern to it yet.
 
Last edited:

Petersen

P
Pendra said:
So, this was supposed to be a post in one of the UA threads, but it turned out to be more interesting, so why not create a new thread.

So after I checked the recordings, I (as others before) concluded that there is no sequence with the same sound three times after each other.
No --- or |||.
Click to expand...

https://www.youtube.com/watch?v=jxxZB9ytxQw

The first 5 purrs in this sure sound like 10001, with the last 0 at the beginning of the howl, and the last 1 being just after the howl ends.

I'm not confident enough in my hearing to say it for certain, but it certainly sounds like it to me.

I had the same revelation that binary combinations of length six which forbid three or more adjacent same-symbols amounts to 26 combinations, which I think is remarkable. Unfortunately, most of the sets are of seven length. =-\


EDIT:
Valethar said:
Have you filtered it for salinity and temperature?
Click to expand...


I thought it might indicate the 'polarity' of that set of bits. But I don't know how to resolve that. What about the 'bits' in the howl? Which howl should denote the inversion? I tinkered around with it each way, but I got very frustrated not finding anything and stopped. I also didn't have a whole lot of data to work with (because I find transcription unbearably tedious, and I have low ear-confidence for resolving obscured bits).
 
Last edited:

Pendra

P
The howling is a parity. There are two types of tones (_-_ and -_~)and they follow each other. 1...2...1...2...1...2...1...2....
It can also be "sent" and "received".
Or a metronom.
Or it can be an itiner how to put the sections together. You can't put a setction that starts with a 1 next to a section that ends with a 1.

They seem to be superimposed over the high-low notes. You hear 1 or 2 in every 28 secs, exactly. Regardless of the high-low tones.

The weird thing is that high-low tones are slightly longer in frequency (~30 secs), which creates a shifting. That is why the 6th tone is sometimes outside of the howling. The frequency of the high-low tones are changing slightly between 3,7-4 secs. But some segments are ~27 secs long, others are ~30. As some were missing an element. And I'm not talking about the ending or the starting segment of a section but segments in between.
This change of length also ensures that the notes and the howls don't go massively out of sync.
 
Last edited:

Phyl

P
Last edited:

Pendra

P
I analyzed the 1st segment very carefully.
H = High note
L = Low note
h = faint high note
l = faint low note
P=Parity. There is a waving pattern and a decaying pattern. They alternate.
The TLen is the total time length of a note segment.
The ULen is the time between two notes in the segment. As you can see, it speeds up over time.
0 is a filler to make the chart nice.
There is always a "space" between two segments. The space is the unit length of the segment.
There are two exceptions. Both had very faint notes next to each other. As if the fade effect malfunctioned. There, I can't decide which one is a space and where the new segment starts. It is possible that both faint notes are irrelevant. It is also possible that only one of them is irrelevant. That would maintain the 6 or7 notes then 1 blank theory.


1 2 3 4 5 6 7 P TLen ULen Remark
H H L L H H 0 ~
H L H H L L 0 \ 27,0 3,90
H L L H L H H ~ 31,0 4,00
L H H L L H 0 \ 27,0 3,90
L L H L L H l ~ 27,5 3,90 faint
l H H L L H H \ 30,0 3,70 faint
L H L H H L L ~ 30,0 3,80
L H H L H H l \ 26,2 3,70 faint
h L L H L H L ~ 29,6 3,70 faint
H H L L H L L \ 29,1 3,60
L H L L H H L ~ ENDS 3,70

 
Last edited:

woodhouse

woodhouse
I replied in the other thread as well.
There are 42 7-bit combinations, not 41. You appear to have missed one between H and I.
The number of possible sequences is always even, as every possible solution can be inverted.
 
You must log in or register to reply here.
Back
Top Bottom