Ah you are "chalk and cheesing" again. ZL to KL7 is
a sea path lower attenuation probably higher power ( I think emergency sets ran
250W to 500W except in lifeboats). The ZL to RU (Vladivostok) (also a
north-south path all in darkness at the same time) was one of the first long
ones to be cracked on 136kHz but required a lot more power than you are running
over a path with light and dark plus summer in VK. For some reason the VE7/KL7
to ZL was never conquored on 136kHz as far as I remember.
I dont see how 8sec difference to GM validates it
unless you know the timing accuracy at both stations. Other weak ids have have
matched element start times. There are no seconds quoted on that spot, the other
three UK spots are 2mins before the GM deep search not 4 mins as you were
quoting earlier.
Was anyone else transmitting? if not the shortens
the odds to chosing your call. I am surprised the GM needed deep search at
180miles when the Essex station copied 27dB stronger without deep search. Ok
there could be some fading in that but no earlier or later straight copy?? The
system has fired off too many impossible spots to be taken too seriously on a
couple of optimistic ones.....it statistics!
You are doing a statistical evaluation. Most
statisticians would want to know the probability of a genuine match together
with, and just as important, the probability of a false match. Even then they
would require more than "one data point" for claiming a result.....remember the
LHC's much quoted 5sigma ??
3sigma says if you repeat the experiment 100 times
you will get the same result in 98. At one-sigma you will only get slightly more
similar results to completly different ones (its around 60/40) and is not too
convincing.
The only reliable way to prove a contact has always
been to exchange unknown information. I doubt you will convince many that this
is more than (generously) "a possible" (as the Whizzo-prang boys of the RAF used
to say, and they were notoriously optimistic :-)) )
Alan
----- Original Message -----
Sent: Sunday, January 25, 2015 9:24
PM
Subject: Re: LF: Just like the 10 Bus ,
nothing for years , then 477 UK>VK :)
The unknown ..
The stand alone Opera APP is
responsible for the off air detection
, the web linking distributes the
spots to other users and the psk-map
With the introduction of the dynamic option
, web data is used to populate the
search lists from Tx beacons and x 2
same call decodes the detection/decode time
stamps , originated by the Tx/Rx
stations , which all ways existed , are now
being utilised to test for time
windows
Worst case are loan wolf Tx and
Rx , then its down to time alone to
validate the deep-search ,
opera normal is a data Tx so
extracts the call from the Tx ,
false decodes have real call
formats with random values , false
detections have real calls but random
time stamps ,
eg vo1na > ku4xr last night , from the
data base 1 spot can be seen to be false
, others are validated by UK
decodes , the later G8HUH spots
would need access to the TX end times
to evaluate , qsb bringing opera into
range at times , handy as a time
stamp for no web linked Tx stations
73-G,
Sent: Sunday, January 25, 2015 8:54 PM
Subject: Re: LF: Just like the 10 Bus , nothing for years , then
477 UK>VK :)
Hi Graham,
Thanks for
clarification.
Am 25.01.2015 21:41, schrieb Graham:
Well Wolf , there is a good paper trail
in the system for this one ..
The system history shows , GM4 and
VK spots are time stamped less than 8
seconds difference, so validated the spot
also VK shows no decodes or
detections in the previous 24 Hours or since ,
opera has a very low false detection level
Ok, I would be convinced if Opera didn't exchange so
much information about who actually transmits, and where, and when, over the
internet.. I wouldn't say it's impossible to that this """decode""" is real,
but I wouldn't call these """decodes""" from this "closed source black
box named Opera" too reliable.
"Just because the computer says
something doesn't mean we should take it for
granted".
Cheers,
Wolf
.
