Hi Folks:

Lee, VE7CC, could comment more authoritatively on this than I can but let me try.

The three of us whom Robert cites use CWSL_Digi 0.88 to decode FT8 and FT4.  The author, Alex Ranaldi, gives us a variable with three levels to control the "depth" of the decoding by the WSJT engine.  The default is set for "3", which I understand calls for the WSJT "Deep" sensitivity.

Ever since beginning to skim FT8 I have used "Deep" decoding and some percentage of the resulting data is in error.  But in response to Robert's suggestion, I've lowered this variable to "2" and let's see if it makes a significant difference in the error rate.  I'm not certain whether it is simple for VE7CC to give us any data on busted WSJT calls, or not.  

I've been averaging about 10,000 PSKReporter decodes each day.  It will be interesting to see how that changes.

But it seems to me that there is a more fundamental issue here.  There appear to be two distinct user communities for FT8 skimming with different expectations: the PSK Reporter community and those DXers who have used telnet DX Spot for many years.

I often receive appreciative emails from the PSK Reporter community who have found the FT8 skimming useful to their hobby.  Some of the emails from DXers who have used the CW data from the RBN for many years expect error rates for FT8 which I don't know how to achieve.  At times after particularly exercised emails from fellow RBN users, I've simply turned off FT8 posting to the RBN system.  And then a digital contest comes around and I turn it on again.

I am at a loss to know what the best course for the hobby truly is but I'm willing to experiment.

Wes WZ7I

On Mon, Dec 16, 2024 at 9:25 PM Pete Smith N4ZR <pete.n4zr@gmail.com> wrote:

Good point, Robert, though I don't know what the people you cite are using (DEEP vs NORMAL).

73, Pete N4ZR
On 12/16/2024 7:04 PM, Robert W5AJ via groups.io wrote:
 
This is not issue with CC - but skimmers that are feeding into the DX spots.
Today's examples include:   WZ7I-# spotting 9E8VB, KM3T-# spotting P7VUM, WC2L-# spotting P63NHS 
None good.   For me and I suspect others, these show up in the ALARM page!
These do tend to come more from certain clusters/skimmers.
I've locked out (best possible) some skimmers from displaying in my CC Cluster with the use of "Keywords" setting and inserting their skimmer callsigns.
This stops both good and false decodes from these folks.
 
FINDINGS:   WSJT, earlier versions do, although rare, false decode when set in "DEEP" mode.   Seems skimmers should use the "NORMAL" setting there to stop false decodes.
Hopefully some of the skimmers monitor this group and will pay attention....
 
73 Robert W5AJ
_._,_._,_
Groups.io Links:

You receive all messages sent to this group.

View/Reply Online (#10606) | Reply to Group | Reply to Sender | Mute This Topic | New Topic
Your Subscription | Contact Group Owner | Unsubscribe [pete.n4zr@gmail.com]

_._,_._,_
_______________________________________________
Skimmertalk mailing list
Skimmertalk@contesting.com
http://lists.contesting.com/mailman/listinfo/skimmertalk