Yep it's something I've been trying to note down recently. If you speed the sound up 2x or so you can hear distinct tones, but after a while you get used to picking them out at 1x speed. One is slightly shorter than the other, and very slightly higher pitched.
For CMDR Branch's Sirius recording (
https://www.youtube.com/watch?v=M9SKX5vz7Qo) I get (counting only clear purrs before the howl):
?|-'--'|'--''|-''--|'--'-'|'-''-|'-'-'|'-''-|-'--'|--''-?<scooped>
The bars represent the howls, and the question mark bit represents blocks I couldn't hear between the start of the UA drop and the point at which the audio was clear enough to hear. But there are probably no more than 1 or 2 blocks missing. Mostly groups of 5 with on group of 6 in there.
It might be interesting to have someone drop the UA and the other listen using the camera drone to ensure the no data is missed when it first appears. Then run several scoop/drops to get end-to-end sequences.
Good thinking on the MC numbers as they are 5 tokens long, but unfortunately they don't seem to line up with what I've found so far.
1 '----
2 ''---
3 '''--
4 ''''-
5 '''''
6 -''''
7 --'''
8 ---''
9 ----'
0 -----
There's another thread around on investigating the sounds here:
https://forums.frontier.co.uk/showthread.php?t=142912
Given the recent blocks of 3/2 purrs in Ratking15's tests I think some of the summary on the first post might need updating, but it's a good start on what people have looked at so far. I'm going to dig through that today to make sure I'm not just running over ground already covered by someone else.
That said, I think during this thread activity there was a post from a member of FD suggesting we might be over thinking the problem, so might not be anything in the purrs after all. Still it's kinda fun to investigate it given it seems to have structure to it
As for possible encodings, recently it was pointed out to me (by a CMDR Snyper) that since they're in groups of 5 tokens they might be a form of
Baudot Code, which was quite interesting as it's in line with old telecoms encodings, but unfortunately I've not found enough data to support it yet (didn't seem to decode to anything meaningful). Also that doesn't fit with the groups of 3/2 purrs found in ratking15's recordings.
One interesting property of the (pre-howl) purrs so far is that the ' and - never seem to be longer than runs of 2 in any given block. So if you have two -- then the next token always seems to be ' (so far), and for '' then next always seems to be - (so far). At least as far as the recordings I've heard anyway, so might be nothing.
