Author: John Dahlem
Date: 16:07:29 04/10/01
Go up one level in this thread
On April 10, 2001 at 16:43:06, John Merlino wrote: >On April 10, 2001 at 16:03:29, Rajen Gupta wrote: > >>hi john:jonas cohans(i think it was he)has also posted 2-3 new settings for >>cm8000. maybe these could be now be tested with a constant opening book.(perhaps >>the cohanas opening book as this has so far proved to be better than the default >>books)the next step would be to test these against other progs and hopefully one >>could arrive at a stronger setting for cm8000 >> >>rajen >> >>On April 09, 2001 at 14:05:27, John Merlino wrote: >> >>>I have finished some weekend testing of the books provided to me by Jonas >>>Cohonas and Brian Kostick. The books were tested with (and against) the default >>>personality. Here are the results: >>> >>>In 32 G/10 games: >>>Kostick 18.5, CM8000 13.5 (+12 =13 -7) >>>Cohonas 18.5, CM8000 13.5 (+14 =9 -9) >>> >>>In 32 G/60 games: >>>CM8000 18.0, Kostick 14.0 (+9 =18 -4) >>>CM8000 20.5, Cohonas 11.5 (+13 =15 -4) >>> >>>I am sending the games to Jonas and Brian, but if anybody else wants them, let >>>me know.... >>> >>>jm > >You can see from my testing results that the Cohonas book has, in no way, >"proved to be better than the default book". > >Also, testing with both a changed book AND a changed personality is not really >proper testing procedure. If the modified personality is better than the >default, is it because the book is better, or the personality, or a combination >of both? I don't think it would matter, as long as it was better (and since it has been proven the book is better, it's worth a try)! john > >jm
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.