Hi Philip

One possibility is multiple decodes from extra strong signal and likely an 
overdriven transmitter.  That +13 next to a -15 is a good indicator that the 
-15 is a splatter kind of signal.  I have seen some with 3 or 4 decodes, all 
within a few hundred hertz.

I also get it when I have multiple laptops on the same radio line out cable or 
multiple WSJT-X instances running on the same computer. The monitor-only 
instances decoded during the transmit period and are reporting out.  In my case 
it was +18 though usually without sidebands here. After seeing this I now turn 
off PSK reporter in the monitor instances, as well as in JTAlert for those 
instances.   If you search for K7MDL spots in the last week I have been 
experimenting using my sub receiver with a 2nd instance of WSJT-X and JTAlert. 
You will likely find some odd 6M spots matching my transmissions on the low 
bands.

Additionally, since the 2nd and 3rd+ instances are not connected to the radio, 
they do not change band info, so more than once I had the wrong band reported 
for the spots it did make.  Now I try to make it a routing to disable spotting 
for non-rig connected instances.


  *   Mike K7MDL

From: Philip Gladstone <[email protected]>
Sent: Sunday, January 20, 2019 22:34
To: [email protected]
Subject: [wsjt-devel] Puzzling behaviour in packets sent to pskreporter

I notice that I am getting a bunch of duplicates in the packets (as seen in the 
raw packet captures) being sent to pskreporter. For example:

WE6Z reported seeing KG7VOR twice at Jan 21, 2019 03:03:29. The only 
differences were:

One entry had Frequency 1841279 with snr -15
Another had Frequency 1841400 with snr of 13

In fact there were 9 spots of KG7VOR in the same packet (at five difference 
timestamps).

This is not limited to a few cases, but it appears a fair amount.

At the same time, I'm getting comments from people that not all of their spots 
are being reported -- so I'm wondering if this duplication is actually a 
symptom of something more serious....

Puzzled

Philip
_______________________________________________
wsjt-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Reply via email to