[Skimmertalk] CW Skimmer Version 1.4 - improved accuracy for contesters, new free trial period

[Jeff Maass]

Is there a revision history file associated with CW Skimmer?

Other than the improved performance, what changed in v1.4 over v1.3?


73,  Jeff  K8ND


> -----Original Message-----
> From: skimmertalk-bounces at contesting.com 
> [mailto:skimmertalk-bounces at contesting.com] On Behalf Of Pete Smith
> Sent: Sunday, January 18, 2009 11:58 AM
> To: skimmertalk at contesting.com
> Subject: [Skimmertalk] CW Skimmer Version 1.4 - improved 
> accuracy for contesters, new free trial period
> 
> I have some good news to report for contester users of CW 
> Skimmer.  I have just finished a series of tests with Skimmer 
> 1.4, to measure the improvement in two areas important to 
> contesters, in particular
> 
> . An excessive number of busted spots
> . An excessive number of cases where a caller was 
> mis-identified as a runner and spotted when that should not 
> have happened (called CQ/DE errors below).
> 
> A little background first.  Skimmer was originally intended 
> primarily as a DXing tool, to facilitate cracking pileups.  
> In this application, speed of identifying the desired station 
> is a lot more important that whether or not the program also 
> generates some busted spots.  Last summer, Alex released a 
> version of Skimmer that was intended to let users tailor 
> Skimmer's callsign validation to fit their preferences.  Four 
> levels of validation were
> offered:  Minimal and Normal, intended primarily for DXers, 
> and Aggressive and Paranoid, intended largely for contesters.
> 
> Now here's the good news.  I tested version 1.3 and version 
> 1.4 against a 10-minute excerpt from a recording of 40M, 
> during the CQWPXCW last May, using a May 2008 master.dta file 
> and Aggressive call validation.  I listened to the recording 
> for each spot that Skimmer made, and classified them  as OK, 
> busts, and CQ/DE errors.
> 
> Here are the data:
> 
> Version             1.3                 1.4             Improvement
> Stations missed*    --                 11 (of 92)       --
> Busts               21                  0 (!)          >99%
> CQ/DE errors        15                  3               80%
> 
> *legitimate CQ's missed by the new version; in most cases, 
> these stations were very weak or were only heard to CQ once 
> during the 10-minute sample, for whatever reason.
> 
> For anyone who wants to try this for himself, Alex has made 
> Version 1.4 available for download from 
> <http://www.dxatlas.com/>www.dxatlas.com for a 30-day free 
> trial.  People who earlier downloaded a trial can do so again 
> with this version.
> 
> 
> 
> 73, Pete N4ZR
> the World Contest Station Database is back, and updated daily 
> www.conteststations.com
> 
> 
> _______________________________________________
> Skimmertalk mailing list
> Skimmertalk at contesting.com
> http://dayton.contesting.com/mailman/listinfo/skimmertalk
>