Author: Uri Blass
Date: 22:45:07 11/26/02
Go up one level in this thread
On November 26, 2002 at 17:01:04, Aaron Gordon wrote: >CM9000 looks nice indeed. Takes Deep Fritz 7 quite bit longer on hardware over >twice as fast.. here's DF7's analysis (w/ 350mb hash) after an hour on an >AthlonXP 1900+ (1.6) @ 1.8GHz... > >Analysis by Deep Fritz 7: > >25...Nc2! > = (0.00) Depth: 6/24 00:00:00 223kN >25...Nc2 26.Qd3 Nb4 > = (0.00) Depth: 7/23 00:00:00 288kN >25...Nc2 26.Qd3 Nb4 27.Qd4 > = (0.00) Depth: 8/23 00:00:00 368kN >25...Nc2 26.Qd3 Nb4 27.Qd4 > = (0.00) Depth: 9/29 00:00:00 692kN >25...Nc2 26.Qd2 Be7 27.Bxb3 Na1 28.Ba2 > = (0.03) Depth: 10/28 00:00:01 1688kN >25...Nc2 26.Qd2 Be7 27.Bxb3 Na1 28.Ba2 Rb8 29.Kb1 Rxb6 30.Kxa1 Qa4 31.Qd4 Qb5 > = (0.03) Depth: 11/28 00:00:03 3423kN >25...Nc2-- > ² (0.31) Depth: 12/33 00:00:07 7009kN >25...Nc2-- 26.Qd8+ Rxd8 > ² (0.31) Depth: 12/33 00:00:07 7465kN >25...Nc2-- > ² (0.59) Depth: 13/28 00:00:14 13475kN >25...Nc2-- 26.Qd8+ Rxd8 > ² (0.59) Depth: 13/36 00:00:16 16208kN >25...Nc2-- > ± (0.87) Depth: 14/31 00:00:32 31189kN >25...Nc2-- 26.Qd8+ Rxd8 27.Rxd8+ Ke7 28.Rgd1 f5 29.e5 Nd4 30.R1xd4 Qh1+ 31.Rd1 >Qxd1+ 32.Kxd1 h5 33.Ra8 Kd7 > +- (6.19) Depth: 14/42 00:00:35 34752kN >25...a5! > +- (6.16) Depth: 14/42 00:01:04 60665kN >25...a5! > +- (6.00) Depth: 14/42 00:01:06 62086kN >25...a5! > +- (5.69) Depth: 14/42 00:01:09 65689kN >25...a5! > +- (5.06) Depth: 14/42 00:01:16 72231kN >25...a5! 26.Kb1 Rb8 27.cxb4 Bxb4 28.Ba7 Rc8 29.Rc1 Ke7 30.Bxb3 Qd6 > +- (3.81) Depth: 14/59 00:04:51 271917kN >25...Rb8! > +- (3.78) Depth: 14/59 00:05:10 290199kN >25...Rb8! > +- (3.62) Depth: 14/59 00:05:15 294665kN >25...Rb8! 26.cxb4 Be7 27.Bc5 Rd8 28.Qe3 Rb8 29.Kb1 a5 30.Bd3 axb4 31.Bxe7 Kxe7 >32.Bc4 Rhc8 > +- (3.56) Depth: 14/59 00:05:36 314368kN >25...h5! > +- (3.53) Depth: 14/59 00:06:44 378347kN >25...Be7! > +- (3.50) Depth: 14/59 00:07:07 399866kN >25...Be7! > +- (3.34) Depth: 14/59 00:07:08 400923kN >25...Be7! > +- (3.03) Depth: 14/59 00:07:11 404606kN >25...Be7! > +- (2.41) Depth: 14/59 00:07:16 409733kN >25...Be7! 26.Bxb3 Rb8 27.Ba7 Nd3+ 28.Rxd3 Rxb3 29.Rgd1 Rb7 30.Bc5 0-0 31.Bxe7 >Rxe7 32.Qd6 Qb7 33.b4 f5 > ± (1.06) Depth: 14/59 00:07:43 435555kN >25...Be7 26.Bxb3 Rb8 27.Ba7 Nd3+ 28.Rxd3 Rxb3 29.Rgd1 Rb7 30.Bc5 0-0 31.Bxe7 >Rxe7 32.Qd6 Qb7 33.b4 > ± (1.06) Depth: 15/42 00:08:55 505123kN >25...Be7-- > ± (1.34) Depth: 16/42 00:11:20 649265kN >25...Be7-- 26.Bxb3 Rb8 27.Ba7 Nd3+ 28.Rxd3 Rxb3 29.Bc5 Rb7 30.Bxe7 Kxe7 31.e5 >Rhb8 > +- (1.66) Depth: 16/65 00:15:09 865619kN >25...Be7 26.Bxb3 Rb8 27.Ba7 Nd3+ 28.Rxd3 Rxb3 29.Bc5 Rb7 30.Bxe7 Kxe7 31.e5 Rhb8 >32.exf6+ gxf6 33.Rd2 Rb4 > +- (1.47) Depth: 17/47 00:26:12 1517784kN >25...Be7 26.Bxb3 Rb8 27.Ba7 Nd3+ 28.Rxd3 Rxb3 29.Bc5 Rb7 30.Bxe7 Kxe7 31.e5 Rhb8 >32.exf6+ gxf6 33.Rd2 Rb4 > +- (1.47) Depth: 18/53 00:49:33 2891072kN > > >Any idea as to what is going on here? Some sort of knowledge that needs to be >implemented perhaps? Deep tactics/positional stuff? I don't play chess myself so >I can't tell how this position differs from any other position. Thanks (and to >the others who did also) for posting the eval of your engine. It is tactics The main problem is to see Qd8+ after Nc2 Fritz can see it in it's main line at depth 12 but it is lazy to solve the fail low at depth 12 and at depth 13 and solves it only at depth 14. Uri
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.