Author: Robert Hyatt
Date: 18:13:23 02/25/04
Go up one level in this thread
On February 25, 2004 at 20:02:12, Vincent Diepeveen wrote: >On February 24, 2004 at 11:08:40, Robert Hyatt wrote: > >>On February 24, 2004 at 10:37:48, Vincent Diepeveen wrote: >> >>>On February 23, 2004 at 14:19:35, Jaime Benito de Valle Ruiz wrote: >>> >>>>Ok, I know it was my choice, but several people have already pointed out that >>>>it'd be a more sensible idea to give the scores for white (positive = white >>>>wins, negative = black wins). >>>>I thought that it'd be better to suggest this change now that we have just >>>>started. >>>> >>>>Anyone has any strong argument against it? (hope not) >>> >>>Yes. >>> >>>It's better to make all positions white to move, that solves the problem easier >>>than this 'rule'. >>> >>>If people have asymmetric eval then which takes care they misevaluate the >>>position, bad luck for them :) >> >> >>That makes no sense. A position with black king at e7, white king at e5, white >>pawn at e4 makes a great deal of sense to have either white to move or black to >>move, and changing it makes no sense. Black to move and white wins. White to >>move and it is a draw. The eval ought to handle that correctly... > >Nowadays i've got KPK egtb, not sure about you. >Even without them, 40 plies of search there will do too :) >But well when you still live in the 80s i'm sure you want to 'handle' it >correctly :) > I want to handle it correctly, because I want to handle it correctly. Nothing more, nothing less. I won't always have tables handy. As far as living in the 80's, perhaps that is far better than living in never-never-land where you reside? As far as the 40 plies of search, I don't quite see how you do that when we are evaluating _tip_ positions. But of course, that point is too subtle for you and simply flies over your head...
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.