Hi, i cropped your Sound a bit and get this: https://www.dropbox.com/s/md1ifplr1340tgm/SHORTER_Echotwotwo_one_fivefive-RedWizzard-Lugh.wav?dl=0
UNKNOWN ARTEFACT: Sound Engineers, apply here!
- Thread starter sohara
- Start date
Hi, i cropped your Sound a bit and get this: https://www.dropbox.com/s/md1ifplr1340tgm/SHORTER_Echotwotwo_one_fivefive-RedWizzard-Lugh.wav?dl=0
Last edited:
Very interesting and pretty logical theory. But, there is just one flaw - there are "000" and "111" sequences present in the signal (000 in Murp's recording, 111 in saman's recording). But, it may be also interesting to briefly test other codes you mentioned. They are human-invented, but as there is no known purpose of unknown artifact, it doesn't eliminate a possibility that thargoids are mimicking it...
I've taken a look at one of your transcript that has a 111 (Saman's), and I disagree with your transcription - i think it is 101, making the first sequence 0010101 and therefore conforming with the adjacency rule. I have not been able to listen to it as I'm in a library and have no headphones, but I have processed the file to remove the howl and it can be seen the obscured purr has the signature of a 0 not a 1:
There is a processing artifact that shows a 1 signature shortly after the 0, but you can see the timing does not fit for this to be the purr in the original source, and ther eis no sign of this in the original signature. This is due to the way I have deleted the howl - by taking a matching howl from the same recording, inverting the waveform, aligning it with the one to be removed and adding the tracks togther. The artifact was present on the howl I used to do this. This is an effective way to remove the howls when they are obscuring a purr, but it is time consuming to get the waves completely lined up so I'm not planning to do this except for in exceptional circumstances.
I don't mean this as any kind of critisism - what you are doing is brilliant and REALLY useful and I don't underestimate the time you are putting in. However as this claim of a 111 sequence was conflicting with the established knowledge I wanted to validate it myself. I may get to look at the 000 sequence later - will have to see how time pans out!
Keep up the good work commander, we will get there
Last edited:
We know for sure that there are some purposefully planted clues, AND that they have great sound engineers at FD.
Even if in the end the message was that there is no message, because we cannot understand how these artifacts would work, it would still be something.
Edit: And we know that they are tech enthusiasts.
Last edited:
HELL yeah ! I was Thinking the same. Too many theories for this one (I dont say this isn't good to do, but imo it's too much Thinking) I think it's just their thargoids nav beac or radio or something like that, and it's in the wrong place. When it will be in the right galaxy maybe something will happen. We know these alliens love living in weird kind of space, like blackholes and . Maybe we have to try there first...
Good job! It's always useful to revalidate it and find possible mistakes. In this case, I am not very sure if it is 0, too. I also briefly filtered it and I believe there is 1. But, we may set is as an unknown bit, it doesn't harm anything. Let's see what you will discover in Murp's recording...
But, even with all those samples, we are not still able to get at least partial results and we are just still eliminating dead ends, that's embarrassing...
But, even with all those samples, we are not still able to get at least partial results and we are just still eliminating dead ends, that's embarrassing...
Every dead end explored reduces the search space
I think there are still some very interesting routes for investigation. Sadly this has dropped right in the middle of exam time for me, so I've not be able to put much time to it. However they'll be over on Monday and I'll crack on then
At the moment I think there are hints of a grammar in the messages, but I need to write a program to analyse that further. Key to this will be the number of transcriptions and their accuracy, so I will also add my interpretations to the mix - the more transcriptions we have the more likely we are going to be able to reduce errors.I don't think we have the answers to solve this at the moment, but I do think we can understand a lot more about the nature of the puzzle which will give us a head start when the next clues drop
linking my find here as well, maybe it get's you guys started:
https://forums.frontier.co.uk/showthread.php?t=141038&p=2211116&viewfull=1#post2211116
https://forums.frontier.co.uk/showthread.php?t=141038&p=2211116&viewfull=1#post2211116
Reading through the thread nobody has mentioned this yet... Another idea would be something like GCR encoding -- since it's always 5 "bits" that are clearly audible. (Any commodore veterans here? )
)Good job! It's always useful to revalidate it and find possible mistakes. In this case, I am not very sure if it is 0, too. I also briefly filtered it and I believe there is 1. But, we may set is as an unknown bit, it doesn't harm anything. Let's see what you will discover in Murp's recording...
But, even with all those samples, we are not still able to get at least partial results and we are just still eliminating dead ends, that's embarrassing...
I agree with Phyl here (though I'm not super-confident in my ears), but I think Murp's is pretty clearly 000. X_X
Alright, I set it as an unknown bit in this sample and also fixed few transcription errors in other samples. Revised transcription is there again: http://inara.cz/uatranscrib.txt
Also, all data are available in JSON format (data structure is noted in .txt file): http://inara.cz/uatranscrib.json
Ericlas: Hard to tell what it means, it's so vague that it hardly helps to decode signal directly...
Also, all data are available in JSON format (data structure is noted in .txt file): http://inara.cz/uatranscrib.json
Ericlas: Hard to tell what it means, it's so vague that it hardly helps to decode signal directly...
Last edited:
Gentlemen,
as I am stopping any decoding effort on this and I will probably try to look again on it with fresh mind later, I summarize tests done by me:
- translation of values to some kind of coordinates - no results, also unknown metrics
- using binary as Arecibo-like pictogram, for single and pair values - no results (also there is unlikely it's something like this, none 00000/1111 sequences for empty space)
- using value pairs as 2D coordinates - no results (it's pretty stupid idea anyway, but I tried)
- using XOR between value pairs and other bits - no results (there is no reason why to do it, too)
- substitution cipher (based on values frequency) - no solid results
- using binary tree/Huffman with some testing dictionary - no results
- comparing binary with database of pulsar periods - no results (but may be tried again if anybody has a database with more than 150 pulsars)
- trying to "stitch" all known samples - no results (no adjacent parts found, all known samples seems to be unique)
- various other minor a pretty naive tests (just for a case) - no results
Only result is, that these are not classic random numbers. Frequency analysis was published there few times, but visual graph make it more obvious: some values are very frequent and also values distribution tends to be in clusters. (left axis: number of occurences, bottom axis: values)
I dare to say that if we will find what algorithm or principle is causing this kind of distribution, we will be much closer to solution. If anybody will have a theory, please test it. Throwing ideas is always nice, but someone must do the work too. I am taking a "vacation" for a while...
as I am stopping any decoding effort on this and I will probably try to look again on it with fresh mind later, I summarize tests done by me:
- translation of values to some kind of coordinates - no results, also unknown metrics
- using binary as Arecibo-like pictogram, for single and pair values - no results (also there is unlikely it's something like this, none 00000/1111 sequences for empty space)
- using value pairs as 2D coordinates - no results (it's pretty stupid idea anyway, but I tried)
- using XOR between value pairs and other bits - no results (there is no reason why to do it, too)
- substitution cipher (based on values frequency) - no solid results
- using binary tree/Huffman with some testing dictionary - no results
- comparing binary with database of pulsar periods - no results (but may be tried again if anybody has a database with more than 150 pulsars)
- trying to "stitch" all known samples - no results (no adjacent parts found, all known samples seems to be unique)
- various other minor a pretty naive tests (just for a case
) - no resultsOnly result is, that these are not classic random numbers. Frequency analysis was published there few times, but visual graph make it more obvious: some values are very frequent and also values distribution tends to be in clusters. (left axis: number of occurences, bottom axis: values)
I dare to say that if we will find what algorithm or principle is causing this kind of distribution, we will be much closer to solution. If anybody will have a theory, please test it. Throwing ideas is always nice, but someone must do the work too. I am taking a "vacation" for a while...
Last edited:
I thought people were just RP'ing the Artifacts but this is an actual thing. Cool. I was wondering, does it matter where you jettison it, does the message change? Does anything happen if you put a bunch of them together in the same area or specific area? Does anything strange happen when you combine it in the hold with other rares? The Jaradharre Puzzle Box apparently emits radiation.
Already tried all of that, no luck :/
Could be some kind of partial key for a substitution cypher EB -> P. Seems unlikely to me though, as it would be unusual to map plaintext of 2 characters to cyphertext of 1.