Author: Slobodan R. Stojanovic
Date: 12:14:27 05/21/03
Go up one level in this thread
Kurt, I don´t understand all this theatre about testing/not-testing The King in the future. And if NOT-TESTING why all these summary of results and experiences of previous testings? Aren´t they already at your web site? If you were testing the King ONLY for fun, or ONLY in order to transmit your experience to other The King´s funs, why are you stoping with further testings? What does the King´s author opinion matter? If you think the King´s author doesn´t agree with your results, and you are offended, it would mean that you were expecting some kind of recognition. But if you think Johan doesn´t appresiate your effort, why are you insisting with all this roll of your experiences and results? Sincerily, there are many contradictions in your posts. And there are so many post of yours treating the same subject that one could think that your conscience is not completely clean. Sorry Kurt. But these were my impressions. Regards. SL. On May 21, 2003 at 14:33:40, Kurt Utzinger wrote: >******* >English >******* > >After further reflection and one night sleep I have [after consulting >chessfriend Rolf Bühler] come to the following decision: > >1.) >The section "Chessmaster" on Homepage "Kurt & Rolf Chess": >http://www.beepworld.de/members39/utzinger >will again be opened as we do not want to "punish" the computerchess community >and the Chessmaster fans. This site will however no longer be updated. > >2.) >Chessmaster 9000 or The King 3.23 [in particular CM9_SKR] may still be >integrated in our tournaments, but the program is being dealt with like all >other engines [no special comments]. > >3.) >We abstain from creating/testing new CM9-settings. > >4.) >No comments whatsoever will be published about Chessmaster [The King] in the >future. > >5.) >Final remarks: > >From our long experience with the great [and often underestimated] TheKing >engine we can say the following: > >- default parameters may be best as overall setting >- there are some better settings than default at longer time controls [40'/40 >upwards] >- strong settings at blitz level are seldom worth testing at longer time >controls >- strongest settings at higher time controls are in most cases less reliable at >blitz level >- different methods of implementing The King/settings under Fritz-GUI lead to >other results >- play at least 300 games [50 games vs 6 engines] to find out if a setting is >good >- do not play CM9_X vs CM9_Y to determine strongest CM-setting >- even slightest amendments of CM9-params may have greatly influence playing >strength >- 32 MB hash is enough and perhaps best for The King [anyway not more than 64 >MB] >- reproduction of moves played during games is often not possible >- best CM settings for analysis [test positions] may not be best for tournament >play >- and just our opinion/feeling: The King plays stronger outside the CM9-GUI > > >****** >German >****** > >Nach reiflicher Überlegung und einer Nacht Schlaf, habe ich nach Rücksprache mit >Schachfreund Rolf Bühler folgenden Entscheid gefällt: > >1.) >Das Menu "Chessmaster" der Homepage <a >href="http://www.beepworld.de/members39/utzinger" target=_top>Kurt & Rolf >Chess</a> wird wieder eröffnet. Wir wollen die Schachgemeinschaft und die vielen >Fans von Chessmaster nicht bestrafen, weil wir die Lust verloren haben. > >2.) >Chessmaster 9000 bzw. The King 3.23 [insbesondere CM9_SKR] wird bei Bedarf >weiterhin in unsere Turniere integriert, jedoch bezüglich Kommentierung nicht >mehr speziell erwähnt oder besonders hervorgehoben. > >3.) >Ab sofort verzichten wir darauf, neue CM-settings zu kreieren und zu testen. > >4.) >Über Chessmaster werden wir keine Kommentare [welcher Art auch immer] mehr >abgeben. > >5.) >Einige abschliessende Bemerkungen: > >Aufgrund unserer langen Erfahrungen mit der grossartigen [und oftmals >unterschätzen] The King engine, können wir das folgende anmerken: > >- Einstellung "default" ist u.U. das beste Setting, brauchbar für jede >Spielstufe >- es gibt bessere Settings als default für längere Zeitkontrollen [40'/40 und >aufwärts] >- starke Settings für Blitz bringen selten dieselben guten Resultate mit höheren >Bedenkzeiten >- starke Settings auf höheren Bedenkzeiten sind meistens für Blitz weniger >geeignet >- unterschiedliche Methoden der Einbindung von TheKing/Settings bringen nicht >dieselben Resultate >- mindestens 300 Partien [50 games vs 6 engines] spielen, um herauszufinden, ob >Setting gut ist >- Verzicht auf Matches CM9_X vs CM9_Y, um das beste Setting zu testen >- selbst geringfügiste Änderungen der Parameter können die Spielweise von The >King beeinflussen >- 32 MB hash sind genug und wahrscheinlich am besten für The King [jedoch nie >mehr als 64 MB] >- oftmals ist es schwierig, in Partien gespielte Züge zu reproduzieren >- starke CM9 Settings für die Analyse [Testpositionen] sind u.U. weniger gut für >das Turnierspiel >- unsere Meinung/unser Gefühl: The King spielt etwas stärker ausserhalb der >CM9-GUI
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.