Author: Uri Blass
Date: 18:16:05 08/18/02
Go up one level in this thread
On August 18, 2002 at 19:53:18, Albert Silver wrote: >On August 18, 2002 at 06:27:08, Uri Blass wrote: > >>On August 17, 2002 at 23:14:12, Vincent Diepeveen wrote: >> >>>On August 17, 2002 at 17:43:15, Mike S. wrote: >>> >>>i hope you realized i put serious time in some positions >>>of their testset, in order >>>to find out that i have put more time in the positions than >>>they have when i started emailing them about it. >>> >>>If they use the word time. they mean 'computer time'. >> >>I think that only computer time is needed to generate a good test suite. >>I try to generate a good test suite based on games of movei in the 3th division >>of the winboard programs but it is going to take time. >> >>I am going to put only mistakes when there is a difference of more than one pawn >>between the score of yace after the move and the score of yace before the move >>after long analysis. > >I don't think you need long analysis for this. Basically, if the eval drops >significantly even after short analysis (10-15 seconds per move) you can be sure >something happened either the last move or a few moves ago (maybe it only saw >the problem a few moves after the blunder). When you see this sudden change, you >can use your own judgement, with the help of an engine, to see what happened. >You could then determine what the best move should have been, and whether it >constitutes a good test move. > > Albert There are cases when I do not need long analysis but there are cases when I need long analysis. You are right that if I want a test suite I can ignore the cases when I need long analysis but if I want the test suite to include all the tactical mistakes that computers can find in the games of movei I need often long analysis. Here is one example(I went backward in the game so the analysis may be based on some learning) [D]8/8/B7/8/KN1q1n2/P2p4/1r2k3/7Q w - - 0 81 Black is clearly better but I cannot be sure that white is losing without deep analysis because white may get chances for perpetual threats. I believe that the losing mistake was earlier but I am not sure and I consider to include the position as a test position. Yace says -2.14/11,-2.19/12,-2.19/13,-2.39/14 Movei 0.0.72h - Bestia 0.88-b2 8/8/B7/8/KN1q1n2/P2p4/1r2k3/7Q w - - 0 1 81.Qh4 Qd7+ 82.Ka5 Qf5+ 83.Ka4 Kf3 84.Bb7+ Ke3 85.Qe7+ Qe6 86.Qg7 Qb3+ 87.Kb5 Rf2 88.Qe7+ Kd2 89.Be4 Re2 90.Qe5 Ne6 91.Qd5 -+ (-2.19) Depth: 13 00:55:14 1248860kN 81.Qh4 Ke3 82.Qe1+ Kf3 83.Bb7+ Kg4 84.Bc8+ Kg5 85.Qe7+ Qf6 86.Qc5+ Kh4 87.Be6 Nxe6 88.Qc4+ Kg5 89.Kb5 -+ (-2.39) Depth: 14 04:22:55 1202799kN (blass, tel-aviv 19.08.2002) Maybe white is losing but it may be a long struggle(the main line of yace at depth 13 has no conversion) and I suspect that the position may be used as a test position when the target is to avoid a tactical mistake from computer point of view. In the position after 81.Bb7 Kd2 Yace showed the following line 82.Qh6 Ra2 83.Bd5 Rc2 84.Kb5 Rc3 85.Qh2+ Ne2 86.Qh6+ Qe3 87.Qg6 Rxa3 88.Qg2 -+ (-3.41) Depth: 12 00:18:54 398049kN In the position after 82.Kb5 The following line is based on some learning(it learned that 83.Qg2 is the best when in the game 83.Bd5 was the immediately losing move) 82...Ne2 83.Qg2 Rb3 84.Bc6 Kc1 85.Bf3 Kb2 86.Bxe2 Qe5+ 87.Ka6 dxe2 88.Qh1 e1Q 89.Qh6 -+ (-5.04) Depth: 12 00:12:56 305376kN 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.