Author: Jaime Benito de Valle Ruiz
Date: 10:59:33 06/16/04
Go up one level in this thread
On June 16, 2004 at 13:46:22, Ingo Bauer wrote: >On June 16, 2004 at 13:17:42, Jaime Benito de Valle Ruiz wrote: > >>On June 16, 2004 at 13:12:55, Ingo Bauer wrote: >> >>>Hi >>> >>>[D]r4rk1/1b3pp1/p5qp/1p1nP1N1/4N2P/3Q4/PP3PP1/R3R1K1 b - - 0 21 >>> >>>Hi >>>S8; Athlon@2500MHz, 256 MB Hash, 53s - Chess is more than BLITZ! >>>(It might be the anti trojan, Shredder just needs a certain depth to eliminate >>>this!) >>> >>>Engine: Shredder 8 (256 MB) >>>by Stefan Meyer-Kahlen >>> >>>11/21 0:01 -0.38 1...Nf4 2.Qg3 Bxe4 3.Qxf4 Bd5 4.Ne4 Rad8 >>> 5.a4 bxa4 6.Rxa4 Rfe8 7.Rc1 Bxe4 >>> 8.Rxe4 (785.430) 452 >>> >>>12/26 0:03 -0.38 1...Nf4 2.Qg3 Bxe4 3.Qxf4 Bd5 4.Nf3 Rad8 >>> 5.Rad1 Rfe8 6.b3 Bxf3 7.Qxf3 Rxd1 >>> 8.Qxd1 Qf5 (1.433.551) 458 >>> >>>13/26 0:05 -0.41 1...Nf4 2.Qg3 Bxe4 3.Qxf4 Bd5 4.Ne4 Rad8 >>> 5.a4 Bb7 6.axb5 Rd4 7.f3 Bxe4 8.Rxe4 axb5 >>> 9.Rxd4 (2.695.846) 468 >>> >>>14/29 0:10 -0.40 1...Nf4 2.Qg3 Bxe4 3.Qxf4 Bd5 4.b3 Rac8 >>> 5.Rad1 Rfd8 6.Ne4 a5 7.Rd2 Kh7 8.Ng3 Kg8 (4.882.285) 472 >>> >>>15/38 0:20 -0.65-- 1...Nf4 2.Qg3 Bxe4 3.Nxe4 Qxg3 >>> 4.Nxg3 Nd3 5.Re2 Rfe8 6.Rd1 Nxe5 >>> 7.Rde1 Rac8 8.Rxe5 f6 9.Rxe8+ Rxe8 (9.635.907) 474 >>> >>>15/38 0:23 -0.66 1...Nf4 2.Qg3 Bxe4 3.Nxe4 Qxg3 >>> 4.Nxg3 Nd3 5.Re2 Rfc8 6.a4 bxa4 >>> 7.Rxa4 Re8 8.Ra5 Reb8 9.Nf5 Rxb2 >>> 10.Rxb2 (11.503.583) 485 >>> >>>16/43 0:47 -0.72 1...Nf4 2.Qg3 Bxe4 3.Nxe4 Qxg3 >>> 4.Nxg3 Rfe8 5.Rad1 Rad8 6.Rxd8 Rxd8 >>> 7.Re4 Rd1+ 8.Nf1 Nd3 9.e6 Nxf2 >>> 10.Kxf2 (22.932.705) 485 >>> >>>16/43 0:53 -0.71++ 1...hxg5 2.hxg5 Nb4 3.Qb1 Bxe4 >>> 4.Rxe4 a5 5.a3 Nc6 6.Qa2 (26.045.478) 488 >>> >>>16/43 1:06 -0.62 1...hxg5 2.hxg5 Nb4 3.Qd2 Bxe4 >>> 4.Qxb4 Bd5 5.Qh4 Rfe8 6.Re3 Be6 7.a4 bxa4 >>> 8.Qxa4 Qxg5 9.Qxa6 Rxa6 (32.689.455) 492 >>> >>>17/38 1:31 -0.63 1...hxg5 2.hxg5 Nb4 3.Qd2 Bxe4 >>> 4.Qxb4 Bb7 5.Qf4 Qc2 6.b3 Rad8 7.Qe4 Bxe4 >>> 8.f4 Bxg2 (44.537.016) 487 >>> >>>18/38 2:31 -0.45 1...hxg5 2.hxg5 Nb4 3.Qd2 Bxe4 >>> 4.Qxb4 Bd5 5.Qh4 Rac8 6.b3 Rc2 >>> 7.Rad1 Rd8 8.a3 Qh7 9.Qf4 Qh5 10.e6 fxe6 >>> 11.b4 (73.230.913) 483 >>> >>>19/46 4:18 -0.45 1...hxg5 2.hxg5 Nb4 3.Qd2 Bxe4 >>> 4.Qxb4 Bb7 5.Qg4 Rad8 6.Rad1 Rxd1 >>> 7.Rxd1 Qc2 8.e6 fxe6 9.Qxe6+ Kh8 >>> 10.Qh3+ Kg8 (123.315.370) 477 >>> >>>best move: h6xg5 time: 5:11.781 min n/s: 476.379 CPU 99.9% n/s(1CPU): >>>476.855 nodes: 148.526.185 >>> >>>regards >>>Ingo >>> >>>PS: I do a longer analysis to see if eval rises! >> >>Yes, it eventually finds the right move, but that's not the issue; after four >>minutes it still comes up with only -0.45. Why does it gives a more realistic >>score (less than 2 pawns) if you force one move and wait for a small fraction of >>a second (a few plies)? >>I've only tried this position with Ruffian 1.0.5, and it gives less than -2.00 >>from ply 1. >>Something's not right here, if you ask me. > >If you argue about the eval the other engines are nearly as wrong as Shredder >because it seems to be a won position and the eval should be Matt in X. If you >argue about the move, you have to be a bit more patient. > >As Günter Simon (and I) allready wrote: It ->might<- be the "Anti-Trojan" code. >Therefore Shredder is pretty clever in some anti-computer positions! It is a >trade off, thats all. > >Bye >Ingo True; It could be a trade off. You cannot get everything perfect all the time. I wish my engine had as many "mistakes" as Shredder :) But, without comparing Shredder with other engines, it's not normal to see Shredder changing its eval so radically after making a move... so quickly. I bet Stefan might solve this glitch in no time if he had a look at it. It's still my favourite engine anyway. Regards, Jaime
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.