Author: Uri Blass
Date: 10:15:06 12/30/05
Go up one level in this thread
On December 30, 2005 at 12:03:57, Zappa wrote: >Well, its been a while since I had to write a tournament report congratulating >someone else. I hope this doesn't become a habit :) Anyway, congrats to Vasik! > Rybka simply played the best chess of the tournament and deserved to win. As I >have said before, I am a big fan of Rybka, because whatever he is doing, its >fundamentally different than the history-pruning-mobility-based-fruitish engines >that everyone else seems to write (even Zappa is fundamentally a mobility based >engine). > >To be honest, I was not real optimistic going into the tournament. I have >changed a lot of things in the latest Zappa versions, and I feel that I haven't >managed to bugfix and tune all of them properly (example: the game against >Ikarus). I feel most of the knowledge is correct, but its just a matter of >getting the values right. People have wondered for months now why Fruit is so >strong. My opinion is that he simply has so few parameters to tune that he can >get them exactly right :) My main hope for the tournament was that the >supercomputer would save a few points for me, but that turned out to be a comedy >of errors. I remember that you said that you are interested in parallel search and not in better evaluation function so I do not understand why you added a lot of things in the latest zappa versions. Adding parallel search correctly was probably enough to win the competition. Uri
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.