Author: Uri Blass
Date: 21:54:58 02/09/02
Go up one level in this thread
On February 09, 2002 at 23:13:30, K. Burcham wrote: >This turned out to be a very difficult test position. >Deep Shredder6 in over 22 hours on 3100 mhz will not play Nxf2. >Chess Tiger 14.0 1000mhz in over 21 hours will not play Nxf2. >Fritz7 1507 mhz in over 20 hours will not play Nxf2. > >if i make the first move Nxf2, then Deep Shredder6 will make the next move in >this line Nxf2 Kxf2 Bh4 g3 f4. and will play each move in this line. so Deep >Shredder6 agrees with this line. > >So with this line, it seems the depth is about 10 until Shredder6 sees the line >is winning for black. Nxf2 Kxf2 Bh4 g3 f4 4.Kg2 fxg3 5.hxg3 Bxg3 6.Rh1 >Bh4 7.Nf1 Qf6 8.Bxh7+ Kh8 9.Bd3 Rae8 10.Kg1 Be2 11.Bf5 (224.630.445) 575.2 >13.01 6:30 -0.85 (did i count right here) but with 20 hours at depth 19 >Deep Shredder6 will not play Nxf2. From your analysis deep shredder needs depth 13 to see negative score after the first 5 moves but shredder saw even bigger positve score in the root position and only at depth 15 the score was significantly better for black than the score of the root position. > > >[D] r2q1rk1/ppp1b1pp/2n5/3p1p2/3Pn1b1/2PB4/PPQN1PPP/RNB1R1K1 b - - 0 1 > >black won by Nxf2 Kxf2 Bh4+ g3 f4 >I will try to give a more detailed analysis of this position but I remember that >black can win it against every defence of white >It is difficult even after f4 for programs to see that white is losing and >Junior5 after some hours in the position after f4 evaluates the position as 0.19 >pawns advantage for white. > >Uri Note that this post was from december 1998 and I used p200 for Junior5. <snipped> >Fritz7: 1507 mhz 256 hash >1...Bd6 2.Nf1 Qh4 3.Re3 Nf6 4.f3 f4 5.Re2 Bh5 6.b3 Rae8 7.Ba3 > (-.38) Depth: 18/48 05:47:07 14824155kN >1...Bd6 2.Nf1 > (-.41) Depth: 19/52 14:00:51 36310193kN > (over 20 hours did not update analysis) Fritz7 needs depth 15 after the first 5 plies so maybe it needs depth 20 to find the move. 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.