Author: Miguel A. Ballicora
Date: 08:24:39 10/09/01
Go up one level in this thread
On October 09, 2001 at 07:49:17, Slater Wold wrote: Thanks Slate. >1q2kb1r/1r1n1ppp/2Qpp3/8/N7/8/1PP2PPP/R1BR2K1 w - - acd 17; acn 1326517364; acs >3606; bm Nc5; ce 344; id wcsac0055; pv Nc5 Rc7 Nxd7 Rxc6 Nxb8 Rxc2 b4 Rc8 Na6 >Be7 b5 Bf6 Rb1 Ke7 b6 Rhd8; I remembered this position was a Smyslov game and I was right. Last night I checked the "My best games" book and it was there. Smyslov-Kottnauer. Black did not play Rc7 (played dxc5) and lost quickly. The pv is very convincing and it does not look there is an alternative; so, this position stands. >rrq1k3/2p2p2/3pb1p1/3N2Pp/p1PpPR2/P2P3P/1P6/K2Q1R2 w q - acd 18; acn 417073342; >acs 3603; bm Rxf7; ce 398; id wcsac0145; pv Rxf7 Bxf7 Qf3 Be6 Qf8+ Kd7 Qg7+ Kc6 >Ne7+ Kb7 Nxc8 Rxc8 h4 Rg8 Qxd4 Kc6 Rf6 Kd7 c5 Ra5 c6+ Ke7 Kb1; >8/3qNrpk/1bpnRpNp/pp1p1P2/3P2P1/P1P4P/1P2Q1K1/8 w - - acd 16; acn 548239247; acs >3603; bm Nxd5; ce 325; id wcsac0216; pv Nxd5 cxd5 Rxd6 Qxd6 Qe8 Qd8 Qxf7 Qg8 >Qxg8+ Kxg8 Ne7+ Kf7 Nxd5 Bd8 Kf3 Ke8 Ke4 Kd7 h4 Kd6; Nxd5 is obviously pretty good. However, I have the feeling that there are alternatives (cooks) in this position. My program Gaviota did not solve it in a long time, so I let it run overnight in this position and sticks to 1.g5 with a high score >2.0 pv was something like 1.g5 fxg5 2.Qd3 Ne4 3.Ne5 and White ended up the exchange and a good position. Can you confirm this? the question is white plays 1.g5, can black find something to save the game? >1r3r2/1b4kp/1q1p2p1/p1n1P1p1/1p2P3/7P/PPPQNNBK/R4R2 b - - acd 18; acn 623805900; >acs 3606; bm Rxf2; ce -223; id wcsac0257; pv Rxf2 Rxf2 Nxe4 Qd4 Qxd4 Nxd4 Nxf2 >Ne6+ Kh6 exd6 Bc8 Nc5 Bf5 Rf1 Rc8 d7 Bxd7 Nxd7 Rxc2 Kg3; >r1q2rk1/pbpnbppp/1p6/4p3/2P5/1PQ2NP1/PB2PPBP/R2R2K1 w - - acd 20; acn 656525154; >acs 3606; bm Rxd7; ce 442; id wcsac0325; pv Rxd7 Qxd7 Nxe5 Bf6 Bxb7 Rad8 Nxd7 >Bxc3 Bxc3 Rxd7 Kg2 Rfd8 Kf3 h5 Kf4 Kh7 e3 g6 e4; >2qrr1n1/3b1kp1/2pBpn1p/1p2PP2/p2P4/1BP5/P3Q1PP/4RRK1 w - - acd 15; acn >976000576; acs 3603; bm Qh5+; ce 32748; id wcsac0398; pv Qh5+ Nxh5 fxe6+ Kg6 >Bc2+ Kg5 Rf5+ Kg6 Rf6+ Kg5 Rg6+ Kh4 Re4+ Nf4 Rxf4+ Kh5 g3 Bxe6 Rh4#; This position is certainly a solid test, ends up in mate and no other alternative can come closer (I think). >r4rk1/1bp3pp/p3Pp2/1pb2P1Q/5B1N/1Pq1p1PK/P6P/R2R4 w - - acd 15; acn 226843577; >acs 3603; bm Rd7; ce 388; id wcsac0408; pv Rac1 Qb4 Rd7 Bc8 Rxc7 Be7 Bxe3 Qa3 >R1c5 Qxa2 Rxe7 Qxb3 Bf4 Qd3; I can see that this position is problematic. Rd7 is certainly the best move when I checked it, IIRC 1... Qxa1 2. Rxg7+ Kxg7 3. Bh6+ Kh8 4. Ng6+ hxg6 5. fxg6 and White threats Bg7 and black cannot stop it but delaying it throwing material). However, Rac1 is a winner too. The only problem is that I like this position because I believe that we can see if a program has more or less problems with horizon effects. I will check if I can modify slightly the position to avoid the cook (I will try a white pawn in c2 rather than in a2) >r4k1r/ppp1n2p/2n2N2/4p1N1/1q1b2Q1/8/P4PPP/R1B1R1K1 w - - acd 12; acn 615027542; >acs 3603; bm Ba3; ce 347; id wcsac0429; pv Ba3 Qc4 Nfxh7+ Rxh7 Nxh7+ Ke8 Qg7 Kd7 >Rad1 Qxa2 Rxd4+ exd4 Nf6+ Kc8 Qf8+ Ncd8 Bxe7; >4r2k/p2qr1pp/1pp2p2/2p1nP1N/4R3/1P1P2RP/1PP2QP1/7K w - - acd 19; acn 941499465; >acs 3602; bm Rxg7; ce 302; id wcsac0442; pv Rxg7 Rxg7 Nxf6 Qe7 Nxe8 Rg8 Qf4 Rxe8 >f6 Qd6 f7 Rf8 Rxe5 a5 g4 a4 Rf5 Qxf4 Rxf4 axb3; >r2r2k1/1b2bp1p/p4npQ/1p1Pp1N1/Pnq1P3/R5N1/5PPP/1BBR2K1 w - - acd 13; acn >1530814052; acs 1057; bm Nh5; ce 32754; id wcsac0450; pv Nh5 gxh5 Rg3 Qc3 Bd3 >Qxd3 Rdxd3 Bf8 Nxh7+ Kh8 Nxf8+ Nh7 Qxh7#; The rest of the positions looks good. I will look at them again tonight. Once we are happy with this 10 (or less if we throw out one) we can move to another 10. Hopefully, we can get at least 30 good clean tactical test positions that are not extremely easy or difficult out of the 44 I selected from WCSAC. The test "as is" with the 1000 positions is unusable. 800 positions are too easy and there are _many_ with cooks and transpositions. Regards, Miguel > > > >Slate
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.