Assume reception of CW SKIMMER spots on a popular band like 20m HF.
I prune the DB at 1000 records, so definitely I do not have 1000 unique callsign (records). In case I want to go towards the objective of having more unique callsigns, then I have to prune the DB at 2000 or perhaps even more. This make SC program way too slow for my deployment ( I have already applied several techniques and setups for making it faster); the DB size is the major factor for program slowness.
Now, the more unique callsigns I have in DB the more accurate (statistically - informatively etc) are the reports generated out of SC. So, consider a use case where the DB has 1000 records all of which are unique callsigns (e.g. by keeping the latest & greatest record of each callsign). This would be the ideal input for me and perhaps other hams.
+ It has never been an objective of DXLab to assist an operator seeking to maximize the number of unique callsigns worked or confirmed, though DXKeeper does provide a "Uniques" function that lists unique callsigns logged.
+ SpotCollector is designed to be used in three mutually supportive ways:
1. to identify currently-active stations with whom a confirmed QSO would advance your progress towards the DXCC, IOTA, Marathon, VUCC, WAS, WAZ, and/or WPX awards on the bands and modes that you're pursuing them
2. to reveal the activity pattern of one or more specific stations with whom a confirmed QSO would advance your progress towards any award you're pursuing
3. to reveal propagation openings between your station and one or more specific stations with whom a confirmed QSO would advance your progress towards any award you're pursuing
+ #1 doesn't require maintaining more than an hour's worth of Spot Database Entries
+ #2 can be accomplished with a similarly modest Spot Database by inserting the pursued stations into SpotCollector's "Special Callsign List" with a NoPrune tag
+ #3 requires 2-3 days of Spot Database Entries to be effective
+ The relationship between a Spot Database's time span and size depends on the number and character of your spot sources. CW Skimmer during an ARRL HF CW contest will create many more Spot Database Entries per hour than than a single DX Cluster node during a weekday. As described here
+ the size control panel provides insight the size of your Spot Database and the rate of incoming spots.
+ For reference, the 12-year-old desktop in my radio room has an Intel i7 CPU and 16 GB of RAM. It supports SpotCollector with 6 spot sources (4 DX Clusters, DX Summit, and a local instance of WSJT-X) and a Spot Database with 20K entries while generating propagation forecasts for each Spot Database entry, with the reset of the DXLab Suite, Microsoft Outlook, Google Chrome, and the DXLab software development environment all running. If you're experiencing poor performance with a modest-size Spot Database, then I suggest using a performance analysis application to determine where your CPU is spending its time; there could be background processes silently consuming significant amounts of CPU time or RAM.
? ? ? ? 73,
? ? ? ? ? ? Dave, AA6YQ
