Author: Robin Smith
Date: 21:38:59 12/21/04
Go up one level in this thread
On December 21, 2004 at 18:25:06, Uri Zlatnik wrote: >The recognition of winning pawn structure is one of the problems of chess >softwares (including top ones like Junior 9 and Shredder 8). > >Many times there is a sound sacrifice (usually winning) which the program fails >to "see" immediately. This will take Kasparov not more than 1 minute to figure >out. > >For example take the game of Kuzmin-Grospeter, Kusadasi 1990: > >1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be2 e5 7. Nb3 Be7 8. 0-0 >0-0 9. Kh1 b5 10. Nd5! Nxd5 11. Qxd5 Ra7 12. Be3 Be6 13. Qd1 Rd7 14. a4 b4 15. >f4! Qc7? 16. f5! Bc4 17. Bxc4 Qxc4 I don't think that in this case most chess software has trouble because of "recognition of winning pawn structure". In this position both Fritz and Junior want to play for the identical exchange sacrifice, and identical pawn structure, via 18.Nc2 Qc8 19.f6 Bxf6 20.Rxf6. So the reason programs don't find 18.f6! cannot be because of pawn structure problems; it is search depth and search extension problems. -Robin >At first glance we notice that Black is attacking an e-pawn which is difficult >to defend comfortably (18. Qf3 Qxc2). Then we might appreciate that white has a >clear positional advantage thanks to the pawn structure: The d5-hole gives White >an excellent outpost and the f5-pawn gives him the basis for kingside attack. > >But protecting e4 through normal means will vaporize those advantages (18. Qd3? >Qxd3 19. cxd3 Rc8 or 18. Nd2 Qc6 and 19... d5). Moreover, White understands that >whatever advantage the structure offers him will be wiped out once Black >activates his poorly placed pieces (... Rc7 and ...Nd7-f6). Putting it all >together, White concludes his advantage is fleeting and he must use it >immediately regardless of material consequences. And so he finds: > >18 f6! Bxf6 19. Rxf6! gxf6 20. Qg4+ > >The attractive 20. Bh6, threatening 21. Qg4+ and mates, would allow 20 ... >Qxe4!. > >20... Kh8 21. Qh4 Qe6 > >Black cannot allow 22. Qxf6+ and 23. Bh6, with mate to follow on g7. But now... > >22. Nd4!! Qe7 > >After 22 ... exd4 23 Bxd4 Black cannot defend against Bxf6+ e.g., 23 ... Kg7 24 >Qg5+) > >23. Nf5 Qe6 24. Rf1 d5 > >At last, desperate attempt to get his pieces into defensive (...Rd6, ... Nd7) >action. > >25 Ng7! Qe7 Or 25 ... Kxg7 26 Bh6+ Kh8 27. Bxf8, threatening 28. Qh6 > >26. Bc5! Qxc5 27. Nh5 Resigns
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.