Hi Charlie,
Sorry, I seem to have overlooked this message yesterday.
> I have spent a few minutes this afternoon familiarising with averaging in
> r6563.
>
> TX source was second instance of WSJT-X (r6169) on same PC with hardware
> audio loopthrough.
>
> Selected some spreading (30) and signal level (-24) to (at least
> initially) not have any single decodes. These settings gave only average
> decodes after 2-3 periods nicely. Message was call call report - wanting
> to try out some unknown info.
>
> I then switched the TX program to calls calls R-report, and noticed the
> average decode stayed with the previously decoded calls calls report, ie
> was not seeing the R. Maybe the changed report then needs a lot of
> repetitions to overcome the existing average?
As things now stand, you need to click "Clear Avg" to start a new
average. Possibly we will want to use the alternate (reversed) sync
pattern, as we did in JT4, so that the receiving software can recognize
the message has changed, even before it can be decoded?
> I then selecting enable deep search resulted in message appearing
> call3.txt v short or missing. It is very short in my case (just the one
> call at this point). Decode appeared to remain lit.
I answered this already, to Charlie. For the benefit of others, here's
the scoop:
Somewhat arbitrarily, the r6563 code insists on having at least 10 calls
in the CALL3.TXT file. I inserted this test so that I would avoid
wasting simulation time by mistakenly running the program with nothing
but the default (empty) file present. My simulation tests used a file
with 5683 calls.
-- Joe
------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785471&iu=/4140
_______________________________________________
wsjt-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/wsjt-devel
