Author: Peter Berger
Date: 15:13:59 12/17/00
Go up one level in this thread
On December 17, 2000 at 15:41:32, Uri Blass wrote: >On December 17, 2000 at 14:08:33, Peter Berger wrote: > >>[D]r5k1/1pqn3p/r4b2/p2p4/8/RP6/RP1BBPP1/3Q1K2 b - - 0 1 >> >>In this position Black seems to be in serious trouble that calls for drastic >>measures . The difficulty is that the key move ..Qh2 ( with idea of ..Rxe3 after >>..Re8 Be3 and the chess on h1 ) doesn't reach perpetual . Still it offers by far >>the best fighting chances as White's rooks are (temporary?) out of game and >>black gets ample compensation for the exchange . The usual question : how long >>does your(?) program need to find ..Qh2 and which program can find this at all ? > >A lot of program can find Qh2. >It is a positional problem. Maybe I tried to write in a way that is over my abilities when it is about language :-) a.) Qh2 might be discovered by many programs ! -- agreed ! b.) It is a positional problem ! -- agreed ! c.) The perfect program will find Qh2 with negative score as the alternatives are bad -- this is much more about the bad alternatives than anything else. As far as I can see ; the comet and gambittiger's solutions look pretty good to me , comet being a little slow . It is very well possible that Qh2 loses in the end ; but the kind of compensation black has is pretty interesting . Bringer for example judged over +2.0 all the time which let it die easily as the compensation got better ; Fritz has much better evaluation here ; I suspect it is very much about the immobile rooks . For the compensation it might be interesting to play through the game ; White didn't play too well but one gets an idea about how difficult it is to play and +2.0 will make it very easy for you to even lose it where +1.0 might help you in case blacks threats get more serious to find an escape . > >I suspect that Fritz6a can play a different move if you give it more time based >on the evaluation in later moves. Yes , I suspect just the same . Playing it with 0.0 is risky ( Fritz probably was a little lucky here ) , others already posted better results . It is about multiple things : a.) Position is basically bad b.) there is a chance for eternal check c.) eternal check won't work d.) Qh2 is the best anyway > >It simply does not search deep enough to see that Qh2 lead to white's adavnatge >and not to a draw. > >After Bxa6 it suddenly understand that white is better. > >Uri Itis also a tough position as stated ; it might be wrong in the end ; anyway it might be interesting to discuss ; the results posted seem to suggest that I wasn't all wrong with my evalutation .
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.