Author: John Roberts
Date: 23:36:50 01/15/04
Go up one level in this thread
On January 15, 2004 at 20:24:57, Brian Kostick wrote: >On January 15, 2004 at 07:59:20, John Roberts wrote: > >>On January 15, 2004 at 07:15:05, José Antônio Fabiano Mendes wrote: >> >>> Please see ==> http://www.chessgames.com/perl/chessgame?gid=1272141 >>> >>>AdrianP: 17. Qc1N! is a fantastic Adams novelty... impossible to spot, >>>especially for a computer, but once you've seen it, it makes perfect sense. The >>>Q dropping back kills off most of B's chances of counterplay but is ready to be >>>re-deployed without much difficulty along the c1-h6 diagonal or along the 1st >>>rank. But those are just incidental bonuses - the main point is clear when you >>>look at the juicy N outposts on e4 and f6... to clear the d2 square for the b3 N >>>(also slowing down B's potential counter-attack with the advance of the a-pawn). >>>Beautiful move! It is very fitting that Adams converts his advantage in such a >>>precise manner. >> >>Why don't we run some evals on it? Maybe it'll turn up in the 22nd ply or at >>least come in 2nd-5th, etc. >> >>Anyone up for it? I think it's far from impossible! > > >This is multi line infinite analysis after 15 ply. Qc1 as seen here is ranked >6th by DF7 while a couple of other engines I tried didn't even give it that. We >do know that it won against Zhong this day and I look forward to doing full game >analysis. Regards, BK > >From: M. Adams - Z. Zhong, Corus 2004 >[D]r1r2bk1/1pqn1p1p/p2p2p1/3Pp1Pn/8/1N2BP1B/PPPQ3P/1K1R3R w - - 0 1 > >Analysis by Deep Fritz 7: > >1. ± (0.88): 17.Nc1 Bg7 18.Bg4 a5 19.Ne2 f5 20.gxf6 Nhxf6 21.Be6+ Kh8 22.Nc3 a4 >23.Nb5 >2. ± (0.88): 17.Na5 f5 18.gxf6 Nhxf6 19.Be6+ Kh8 20.Rhf1 Bg7 21.Rde1 Rf8 22.f4 >Ne4 >3. ± (0.78): 17.Qd3 Bg7 18.Nd2 Nc5 19.Qc3 Nf4 20.Bxc8 Nxd5 21.Qa3 Nxe3 >4. ± (0.78): 17.Bg4 a5 18.Bxh5 gxh5 19.Qd3 a4 20.Nd2 a3 21.b3 Qc3 22.Qxc3 >5. ± (0.78): 17.Rhe1 a5 18.Nc1 a4 19.a3 Bg7 20.Bg4 b5 21.Nd3 >6. ± (0.75): 17.Qc1 Bg7 18.Nd2 b5 19.Ne4 b4 20.Qd2 Rab8 21.Bg4 f5 22.gxf6 Hi Brian, as you can see from my post at: http://www.talkchess.com/forums/1/message.html?342494 I like analysing for weeks on one position, and this is from my 3rd fastest home computer, my 2nd fastest one is now on 1257 billion kNs. I do this for the fun of it, and I realize people will say, "don't you know the limitations of this...what a waste of time" but I'm having fun. I would not be too surprised if: a)Qc1 becomes the best line by 22 or 23 plies; b)that Adams probably found it by running some program for weeks. Certainly no one would have a problem believing this of Kasparov. Try running just one line for weeks, then with the hashtables intact, you can try two.
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.