Author: Richard A. Fowell (fowell@netcom.com)
Date: 21:13:36 03/29/98
Go up one level in this thread
On March 29, 1998 at 02:06:58, Howard Exner wrote: >On March 29, 1998 at 00:35:14, Havergal Brian wrote: > >>Does anyone have an explaination for Hiarcs "wild" knight sac in this >>game? If you do please enlighten me...thanks. >> <snip> >I posted this position in epd format a while back but got no replys >on how other programs would play here. I'm wondering if it was played at >first because Hiarcs saw that it could draw then realized later it was >winning. >A Hiarcs owner would have to check this one out. It's a nice move. <snip> The EPD for this position is: 3rnrk1/1b2bp1p/pqn1p1p1/1p2B1N1/3P1Q2/P1N5/1P2BPPP/R2R2K1 w - - Here, HIARCS played Nxh7! and later won. I think Harold's suspicion is right - HIARCS seems to have played this move because it was the best move it could find, not because it saw it's way clear to the win. Here's HIARCS analysis (HIARCS 6.0 Macintosh, 256K cache): Ply Seconds PV ------------------ 4»18 0 d4-d5 c6xe5 f4xe5 e7-d6 e5-d4 b6xd4 d1xd4 e6xd5 a3-a4 = -146 5»19 8 e2-g4 e7xg5 = -52 6»23 21 e2-g4 e7xg5 f4xg5 f7-f6 g5-h6 f8-f7 e5-f4 c6xd4 = -66 7»25 52 e2-g4 e7xg5 f4xg5 f7-f6 g5-h6 f8-f7 e5-f4 c6xd4 h6-h3 = -60 7»25 87 g5xh7 g8xh7 = -34 8»27 168 g5xh7 * -2 8»27 210 g5xh7 g8xh7 d1-d3 c6xe5 d3-h3 = 0 9»29 548 g5xh7 * -32 9»29 855 g5xh7 g8xh7 d1-d3 f7-f6 d3-h3+ h7-g7 f4-h6+ g7-f7 h6-h7+ = -36 10»30 5120 g5xh7 * -4 10»30 5968 g5xh7 g8xh7 d1-d3 f7-f6 d3-h3+ h7-g8 f4-h6 g8-f7 h6-h7 = -2 HIARCS switches to Nxh7 at ply 7, and less than two minutes. Hoever, even after 10 ply (and over an hour) it evaluates it's position as slightly inferior. Note that the continuation HIARCS shows is the game continuation: 19. Nxh7 Kxh7 20. Rd3 f6 21. Rh3+ Kg8 22. Qh6 Kf7 23. Qh7+ fowell@netcom.com
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.