[Dx4win] V8.01 Observations (long)

[Ing. Miloš Prostecký]

Yes, Randy,

I find same problem with DXCC CW statistic, that majority of checked
countries is only as confirmed! So I hope it will be solved.

73, Milos, OK1MP


Message: 9
Date: Tue, 04 Aug 2009 20:27:57 -0500
From: Randy Farmer <w8fn at tx.rr.com>
Subject: [Dx4win] V8.01 Observations (long)
To: dx4win at mailman.qth.net
Message-ID: <20090805012758316.FDDC18482 at cdptpa-omta01.mail.rr.com>
Content-Type: text/plain; charset="us-ascii"; format=flowed

I downloaded and had a look at the new version today. I don't yet 
have my license key, but I was able to import my version 7 log and 
check out some of the new features. Importing the Version 6/7 format 
log file into Version 8 was absolutely trouble free. I did have to 
re-do some of my custom Logbook Report formats to account for the new 
fields, but this was no problem at all. All of my Selections and 
Group definitions imported perfectly.

I'm especially glad to see the improvements to the LOTW record 
keeping and the elimination of the need to explicitly filter the log 
by date selection to get accurate DXCC CW statistics. The testing 
I've done so far has concentrated on checking out the DXCC CW 
reporting. Unfortunately, I have found some problems.

I have around 56K QSOs in my log, the vast majority being CW. The log 
files go back to 1965, so there are a good many CW QSOs in the log 
that were made prior to being eligible for CW DXCC credit. Some of 
these have been submitted for Mixed and Band DXCC credit. I also keep 
track of my DXCC records using a fairly elaborate Excel spreadsheet 
and make sure it's updated every time the DX4WIN log changes and/or I 
receive new DXCC credits from ARRL. My standard procedure is to do a 
quick compare of the DX4WIN DXCC Summary reports against this 
spreadsheet each time after importing LOTW data or making any other 
changes to the log.

I checked out the Version 8 DXCC Summary reports and found right away 
that regardless of the selection made in the dropdown boxes the 
Summary report ALWAYS gives the stats for All Bands / All Modes. The 
good news is that these stats are completely accurate, including the 
CW numbers. I tried all combinations of the "Modes" and "Bands" 
dropdown choices and always got the All modes / All bands report.

I then took a look at the Listing reports for various band and mode 
combinations.To use the Listings data, I use the Write to File option 
to dump the report data to csv format to import into Excel. It's 
disappointing that while the Listing report format has changed 
slightly, it STILL uses the asterisk (*) character to designate 
"Checked" status. The asterisk is a special wild card character in 
Excel, and directly importing the csv file makes it impossible to 
edit and manipulate the data -- weird things happen. This means I 
still need to first run the csv file through Wordpad or some similar 
text editor to do a search and replace of the asterisk with some 
other character so I can use the file in Excel. It would be real nice 
if the report output used some other character to show "Checked" status.

  All of the DXCC reports except for CW were completely correct. 
There were multiple discrepancies for CW DXCC, which upon close 
examination all turned out to be the result of counting pre-1975 QSOs 
for CW credit. If these old records were excluded from the 
statistics, the CW report would be totally accurate. I have taken 
great care in setting the Mixed, Band and Mode credit flags for these 
older QSOs in the log file, and none has the Mode credit flag set, 
yet the report claimed that they were credited for CW. Apparently the 
internal logic of the program looks at neither the QSO date nor the 
status of the DXCC Mode flag when it generates the DXCC report.

So we're almost there... The Summary gets the CW statistics correct, 
so there's clearly some logic that runs on QSO date, but it's 
obviously incompletely implemented.

Also regarding the Listing report, when including Deleted entities 
the new version sorts its output in alphabetical Prefix order without 
regard for whether the prefix is for a Deleted entity or not. The 
older versions put all the Deleteds in alphabetical order at the 
bottom of the prefix list, a MUCH better way to present the data. I 
had to immediately re-sort the imported csv files to get them in the 
same order as the data from Version 7. Please restore the Version 8 
Listing report to the same format as the Version 7 and earlier listings!

There is also some sort of internal coding error that caused me a bit 
of hair pulling. For some reason, while the prefix for Minami 
Torishima is JD1/M, the Listing report renders it as JD1/W and thus 
interchanges the lines for Minami Torishima and Ogasawara, JD1/O. It 
was no fun dealing with this until I took a close look at the report 
and saw the incorrect prefix.

Apart from these small problems, I like the looks of the enhancements 
to this new version of DX4WIN. I'm looking forward to getting my 
license key and doing more testing with live hardware. Thanks to Paul 
for a significant improvement in this new generation of the program. 
With a little tuning it will be a great leap forward.

73...
Randy, W8FN




No virus found in this outgoing message.
Checked by AVG - www.avg.com 
Version: 8.5.406 / Virus Database: 270.13.43/2281 - Release Date: 08/04/09
18:01:00