Hi, Andreas Lochmann wrote:
you should also use the harmonic average instead of the arithmetic one:
You are right - by taking the harmonic average of all scores of "professionals" and cutting of those that are 50% above that average we should get the ambitious-played professional scores. Taking again the harmonic average of these remaining scores should be a usable PAR.
We could ask the players from mag-heut.net to send us their complete 0.92-enigmarc2's. However, this won't work for most of the new levels.
As Ralf seems no longer reachable we should first write the evaluation code and submit the request for score files near to the next release. Don't worry about the new levels. I try to make the ratings.xml online updatable - so rating updates will no longer be bound to application releases. Ronald _______________________________________________ Enigma-devel mailing list Enigma-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/enigma-devel
