Author: Mike Byrne
Date: 18:19:46 01/29/06
Go up one level in this thread
On January 29, 2006 at 19:18:07, Vasik Rajlich wrote: >On January 29, 2006 at 10:42:03, Robert Hyatt wrote: > >>This reminds me of a "hobby" of mine, playing blackjack. >> >>Card counters often discuss the many "basic strategy departure index plays" and >>ask "how important is it to learn the indexes for the uncommon plays like when >>to double 8 vs 6 and the like?" >> >>The answer is in two parts: >> >>(1) the hands are not very common, which means playing them correctly or >>incorrectly will not have a great influence on your long-term winning edge; but >> >>(2) when the situation comes up, and you have a big bet on the table because of >>the positive count, suddenly that "not very important play" can be the >>difference between a couple of hundred bucks and zero. >> >>So while they are not used often, when they are used it is sometimes critical. >>I have seen Crafty win many KRP vs KR endings where its internal evaluation >>thinks it is a draw because the enemy king is too close to the promotion square, >>but due to a subtle rook move it is exactly one square too far away. This is >>nice to know if you enter some long combination where the final position is the >>resulting KRP vs KR ending, and you just traded everything away in a winning >>position to reach what you hope is a really winning position. > >I agree that tablebases are nice to have. Sometimes they help, and they never >hurt. If only every engine change was like this :) > >However, it looks like they help too rarely to show up on the Elo charts. > >Vas oh they show up all right --- my guess is that they are worth about 5 elo points -- now if we went to 32 man EGTB's, things might be a little different ;>) also great work with Rybka -- an amazing engine imo .... best, Michael
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.