Author: Lanny DiBartolomeo
Date: 20:20:19 12/13/98
Go up one level in this thread
On December 13, 1998 at 22:46:49, Laurence Chen wrote: >> >>Read comment made by Lanny DiBartolomeo: >>And your basing this on a cm2000 program (which is a totally different engine) >>and 1chess position from GM kasparov that cm6000 couldnt find? I can come up >>with plenty of examples in which it can find, as a matter of fact there was a >>post of a position in which fritz5 didnt find and cm6000 and rebel10 found. >>No one said that cm6000 finds all the moves no one, not even GMkasparov could. >>But I do Know about static advantages and dynamic play and i know most people >>here do! And cm6000 plays very well it plays the position and what the position >>calls for it will find a majority of the time be it static or dynamic it has >>given exchange sacs on me a number of times and continued with play why? not >>because it saw it would win back material but it created enough pressure on the >>board for the sac! SO, sorry but you are not talking about the same program i >>play against. >Yes we are talking about the same program, I use CM 6000, and I see it make >exchange sacs lots of time, but however, whenever it did, it did as a pseudo >sacrifice, not as a real sacrifice, it was able to win material back in a few >moves later. So, why is it that CM 6000 won't make a positional sacrifice in >this position?: >1rb1r1k1/3n1ppp/p1p1p3/q3P3/7P/2N1Q1R1/PPP3P1/2KR1B2 w - - 1 0 >It chooses 18. a3 as the best move after I left the CM 6000 to analyze the >position for 6 hours in the infinite level. Fritz finds the move 18. Rxd7, but >it gives an evaluation of 0.00, it prefers the other move 18. h5. This position >came from the game between GM Keres and GM Szabo, Budapest 1955. GM Keres won >the game in this fashion: 18. Rxd7! Bxd7 19. Bd3 h6 20. Qf4 (Threatening both >21. Qf6 and 21. Rxg7+ etc.) 20. ... Kf8 21. Rxg7! Kxg7 22. Qf6+ Kf8 23. Bg6! 1-0 >(23. ... Re7 24. Qh8#). Even Junior 5 does not like the move 18. Rxd7, so what >is wrong then with this move? I believe that GM Szabo defended poorly, usually, >there is a psychological shock which a player gets when one receives a sac on >board. I wonder how would Keres continued if Szabo played 19. ... g6 instead. >You see, CM, Fritz 5.16, Junior 5 reject the sac because it is unsound, it >doesn't see any gain of material, nor any prospect for a mating attack. GM Keres >sacrificed the exchange in order to remove the most active defender on board, >and he did it for a gain in initiative, and also because of the psychological >effect the sac would have on his opponent. CM 6000 is not the only engine which >performs sac on board, all the other engines in the top ten in the SSDF list are >capable of performing exchange sacrifices. But I have yet to see one engine >which performs a real sacrifice, not a pseudo-sacrifice. If CM 6000 did not >perform any positional sacrifices, then no one would buy a copy of CM !!! I used >the default CM personality to analyze this position. Next I am going to try the >logging player you suggested and let you know if the results are any different. The thing with cm6000 is it does perform real sacrifices but rarely was it wrong in this, most times it is not to gain back the material (right away) but for extra pressure and initiative , Also a real sacrifice is not to say oh ill do this and if Im lucky... but it is because dynamically it will be made up for and because would create better prospects than what you had,My point is that a real sacrifice is not without a real soundness behind the move.But Im off track here my responding to these post is to let you know that cm6000 is not a static searcher the king is an activity seeker and it performs sacrafices not just for tactical reasons but for positional as well which a sacrifice for position is a true sacrifice it doesnt gain anything back for material but has a better grip to the game, it has done this to me quite often and it didnt get anything back until way later in the game I had to just defend move after move after move there was no way for it to see that on say move 13 im giving up a pawn and by move 30 it was choking my King.
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.