FT Roundup - 2022
Call: AB1J
Operator(s): AB1J
Station: AB1J
Class: SO Unlimited LP
QTH: Waltham MA
Operating Time (hrs): 21.3
Summary:
Band QSOs
------------
80: 68
40: 149
20: 169
15: 208
10: 76
------------
Total: 670 State/Prov = 49 Countries = 33 Total Score = 54,940
Club:
Comments:
This is a CHECKLOG.
I experimented running WSJT-Z in auto mode. Automated (robotic) operation isn't
allowed under the contest rules, hence the CHECKLOG.
All attended operation. I did the same last year, but less intensively.
For the first 12 hours I pretty much let WSJT-Z run entirely on its own using
Auto CQ and Auto Call and intervened only to save some QSOs from disaster. I
didn't want anyone else to suffer from my experiment. I apologize for any
problems I caused to others.
In that 12 hour period I had 335 QSOs and 30 interventions or 9%. Guess you
could say WSJT-Z is 91% contest ready. Or whatever.
It was extremely frustrating to sit there and watch WSJT-Z fiddle around and
waste time. Running out the watchdog timer, for example. There needs to be a
human op to make things go smoothly and make rational decisions, like when to
hold 'em, when to fold 'em, when to move on. When to go split, when not to.
Etc.
For the rest of the contest I used an intensive combination of manual and
automatic operation. This went very smoothly and I was quite happy with it. My
rate went up a bit with manual+auto operation even though I slowly ran out of
stations to work.
In manual operation I mostly used Auto Call for S&P, which was the
equivalent of "Best S+P" in FT4. I didn't use the FT4 "Best
S+P". In most cases I liked WSJT-Z making the calls for me. It never gets
tired or loses concentration.
I kept a separate contest ALL.TXT file so I can take a look at any anomalous
events.
With my station I max out at a little over 700 QSOs in a 24 hour FT contest.
That's pretty consistent, contest after contest. I kept in that ballpark this
time if you adjust for the 21.3 hours. There wasn't as much activity as I
expected and I spent a lot of twiddling my thumbs. Score-wise, I was happy with
the result since my station isn't really competitive.
One thing for sure is that effective FT contest automation is a ways off. It
would be fun to enhance WSJT-Z to make it more efficient in contests, but since
it's against all contest rules there's little incentive to do so. (One exception
I know of is the YBFTDX, and that's probably an oversight). I personally don't
find automated contest operation scary or objectionable. The best chess and Go
players are computers, but that doesn't stop people from playing those games.
I do think that if I worked aggressively in partnership with WSJT-Z I could do
much better than with WSJT-X. Maybe next year I'll hit it harder with WSJT-Z. I
took a lot of notes about how to do things better and need to put them into
practice.
Clearly, this is a personal interest and I'm competing with myself. Am I sorry I
spent all this time in the Roundup for nothing? (CHECKLOG) Not really, as it is
an interesting ongoing experiment. I became disenchanted with this contest
several years ago due to its bait and switch alcohol policy so I decided just
use it as a test bed for automation experiments.
73,
Ken, AB1J
Flex 6500
WSJT-Z in Auto modes
N1MM Logger+
Attic dipoles on 20m, 15m & 10m
Outdoor 66' stealth wire on 80m & 40m with an indoor counterpoise
LoTW eQSL ClubLog
Posted using 3830 Score Submittal Forms at: http://www.3830scores.com/
______________________________________________
3830 mailing list
3830@contesting.com
http://lists.contesting.com/mailman/listinfo/3830
|