Hello, recently I got Emanuel Lasker's old book on board games from 1931. It has a whole Chapter on variants of Backgammon (Puff, TricTrac, Gammon). Lasker discusses what might be good decisions in typical situations. Then he writes one philosophical paragraph.
A scan of the original page 239 is online at http://www.althofer.de/lasker-monte-carlo.jpg *********** original text in German ***** Einzelne der obigen Zahlenwerte habe ich geschätzt und ich rate meinen Lesern, sie zu kontrollieren, indem sie Versuche machen. Es ist das überhaupt eine gute Methode. ... ... Mit einem grosssen mathematischen Apparat liessen sich ja diese Fragen bis aufs Tüpfelchen über dem I erledigen, aber für den Spieler ist es ratsamer, sich der Methode zu bedienen, die im Leben von Nutzen ist, und da ist es weniger die mathematische als die statistische Methode der Beobachtung, die er sich anzuzüchten hat. *************************************** Translation to English by Ingo Althoefer. From Emanuel Lasker: Board Games of the Nations, 1931. Chapter "Puff und Tric-Trac (Gammon variants)", middle of p.239; Some of the numerical values above are estimates by me. And my proposal for the readers is to control them by making "trials" [I avoid the term "simulations" instead of "trials", I.A.]. It [="making trials", I.A.] is in general a good method. ... ... By a large mathematical machinery these questions might be answered to the very detail. But for the player it is more advisable to use "the" method that helps in real life. And this method is not the mathematical one but the one with statistical observations which he should acquire. ********************************** Reading this, one has to keep in mind that of course there were no computers around in 1931, and that the name "Monte Carlo method" had not been coined. Does someine here know of even older proposals of Monte-Carlo for games? Ingo. _______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
