Spoilers Beyond 3.0 Open Beta: Aliens Spoilers

Status
Thread Closed: Not open for further replies.

Maligno

Maligno
I had a look at 183 data packets in the 3.0 Beta. The statistical characteristics of the data packets are pretty much the same as in the live game. The curious spike at Length=9 is still there (previously seen in this study: https://canonn.science/codex/analysis-of-guardian-obelisk-audio-data-packets/). I should note that none of the data packets were seen to repeat in the Beta, which is the same result we got in 2.4.

This is one of the first things I'll look at when 3.0 is released. There's still a chance they'll inject easy-to-decode meaning into the obelisk audio in the 3.0 release.


TnDpWdd.png
 
Last edited:

CMDR Dreamstate

CMDR Dreamstate
Maligno said:
I had a look at 183 data packets in the 3.0 Beta. The statistical characteristics of the data packets are pretty much the same as in the live game. The curious spike at Length=9 is still there (previously seen in this study: https://canonn.science/codex/analysis-of-guardian-obelisk-audio-data-packets/). I should note that none of the data packets were seen to repeat in the Beta, which is the same result we got in 2.4.

This is one of the first things I'll look at when 3.0 is released. There's still a chance they'll inject easy-to-decode meaning into the obelisk audio in the 3.0 release.


https://i.imgur.com/TnDpWdd.png
Click to expand...

Good going mate, that looks good.

I had an idea while staring at the lines. I took your image and royally butchered it (a thousand appolagies in advance ;) ) and thought the spectrograph looked... odd, once I had done it.

Here's how the unedited chop looks like, below.

fUADosL.jpg

As you can see from the above image, only two of the bars are seemingly disjointed from the rest on the right-hand side...

Why this is, I don't know.

Here are the patterns I noticed:

70uO6kG.jpg

And when looking closer at the thin yellow bars, could it be text or digits?

ekqZ36n.jpg
 
Last edited:

Maligno

Maligno
CMDR Dreamstate said:
Good going mate, that looks good.

I had an idea while staring at the lines. I took your image and royally butchered it (a thousand appolagies in advance ;) ) and thought the spectrograph looked... odd, once I had done it.

Here's how the unedited chop looks like, below.

https://i.imgur.com/fUADosL.jpg

As you can see from the above image, only two of the bars are seemingly disjointed from the rest on the right-hand side...

Why this is, I don't know.

Here are the patterns I noticed:

https://i.imgur.com/70uO6kG.jpg

And when looking closer at the thin yellow bars, could it be text or digits?

https://i.imgur.com/ekqZ36n.jpg
Click to expand...

Thanks for having a closer look at that spectrogram.

The areas with horizontal lines are ones that I call "Sparce Spaces" (see https://canonn.science/codex/analysis-of-guardian-obelisk-audio-data-packets/ for more details). They are regions that are not completely empty and are comprised of around 10 horizontal lines. There are about 11 unique Sparse Spaces, 5 of which happen exclusively in the end-of-packet sequence, 2 that happen before or after the regular bars (L and N to be exact), and 3 others that can be found in the main packet sequence, usually between N-L-S-T bars (plus a tiny one that always accompanies one of those 3). We've found that those 3 Sparse Spaces follow similar rules to N-L-S-T bars, i.e. they don't like to occur next to each other and there must be at least 1 of each in a data packet. We believe that those 3 Sparse Spaces could very well convey information. We still don't have a clue what sort of numerical system (or meaning) these patterns map to.

The pattern that you highlighted in green is simply an effect that has always been in the obelisk audio. Now that the data packets have been shifted down in frequency they are subjected to that "inverse noise" patch.

Regarding a possible modulation in the horizontal lines found in the Sparse Spaces, well, I'll need to have a closer look at that. I doubt there's anything there because they are rather noisy.
 

MyklAtrum

MyklAtrum
Just a quick note that I will be locking this thread and unpinning it when Beyond releases this week. But not deleting so that the information is still available to those who need it. :)
 

Maligno

Maligno
Just wanted to jot down a few thoughts before the thread is locked.

I enhanced the Automatic Data Packet Recognition algorithm to extract all information-carrying signal components, including the areas with horizontal lines, which I will call Codons (photon, proton...get it? :) ) from now on (I used to call them Sparse Spaces, but that's quite a mouthful).

The image below shows a typical data packet with all Bars and Codons extracted (collectively, they are known as Tokens, a term coined by CMDR Clanga farther upthread).

436BKSp.png

It appears that N bars in the main part of the signal are longer than they used to be. In 2.4 they used to be just as long as the other N bar that always appears at the end of the signal. Others have suspected that N bars act as signal separators, and given the way they look now I think I'm convinced that's the case as well.

For each data packet I extracted all of the sub-packets within the N bar separators. The figure below shows a histogram of the number of sub-packets across all 183 data packets I recorded in the 3.0 Beta. The histogram resembles a Poisson distribution...so something we would expect in certain natural processes, such as the number of decay events per second from a radioactive source

g30Yq9D.png

The figure below shows a histogram of sub-packet length (i.e. number of Tokens). I think this particular histogram appears to follow a Power-Law distribution...again, something that happens in certain physical or biological phenomena.

2TssIxi.png

Now, the figure below shows a histogram of data packet length. This is similar to the one I showed a few posts ago, except it now includes the number of Codons in a data packet. The statistics are not high enough to draw any kind of conclusion, but just like in other cases I looked at in the past the distribution has suspicious spikes.

I8cbAlM.png

Currently, I don't think these data packets are encoding any meaningful information. At least these figures will provide a baseline that we can compare to in the future. It's possible the fdevs will decide to "enable" meaningful information in the data packets in the months to come.
 
Last edited:
Status
Thread Closed: Not open for further replies.
Back
Top Bottom