Author: Robert Hyatt
Date: 20:40:05 12/17/03
Go up one level in this thread
On December 17, 2003 at 19:12:09, Sune Fischer wrote: >On December 17, 2003 at 18:28:49, Martin Andersen wrote: > >>On December 17, 2003 at 15:11:04, Peter Berger wrote: >> >> >>>Crafty 19.4 - Junior 8.0.0.2 >>>[D]2q1r1k1/5p1p/2P5/p1rn2p1/Qb1Nn3/6PP/1P4BK/R1B2R2 b - - 0 1 >>> >>>Not a bad testposition at all - 26. ..Nxg3 is not so easy to find for some >>>engines. White is toast. >>> >>>Crafty itself can't find it in an acceptable time on my Athlon 2600+ at all. >>>Ruffian and Yace need some time too. >>> >> >>Crafty 19.7 Linux, Athlon 2400+ takes 23 min to prefer 26...Nxg3: >> >>Depth: 13 Nodes: 1272643K (911 kn/s) >>Score: -0.40 Time: 1396.55 seconds >>1. ... Nxg3!! >> >> >>Martin. > >Hmm, I don't understand why Crafty can't find that, even latest frenzee without >dubious speculative patzer eval seems to find it in about a minute (@2GHz AMD): I think this is the obvious bad trade problem. 99% of the time this is bad. Here it seems to work. IE if you evaluate a piece for multiple pawns as not a good idea, this is harder to find. >9 29 313 2342291 1...Qc7 3.Qb3 Qe5 4.Qf3 Nxg3 5.Qxg3 Qxd4 6.Qxg5+ Kf8 7.Qh6+ >Kg8 >10 14 1472 11042144 1...Qc7 3.Qb3 Ndf6 4.Rb1 Qd6 5.Nf5 Qxc6 6.Nh6+ Kg7 >7.Qxf7+ Kxh6 8.Rxf6+ Qxf6 >11 0 3500 27628082 1...Qc7 3.Qb3 Ndf6 4.Qd3 >11 0 5878 47430265 1...Nxg3 >11 -50 6519 52479004 1...Nxg3 3.Kxg3 Qc7+ 4.Kf2 Qe5 5.Rd1 Rc4 6.Qb3 Rxd4 >7.Bxg5 Nf4 8.Rxd4 >12 -74 10131 80986204 1...Nxg3 3.Kxg3 Qc7+ 4.Kf2 Qe5 5.Rd1 Rc4 6.Bxd5 Be1+ >7.Kg1 Rxa4 8.Rxa4 Qxd5 > >-S.
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.