Author: Angrim
Date: 23:44:31 12/16/01
Go up one level in this thread
On December 15, 2001 at 12:16:00, Will Singleton wrote: >On December 15, 2001 at 03:50:10, Gian-Carlo Pascutto wrote: > >>On December 15, 2001 at 01:04:22, Will Singleton wrote: >> >>>played a few games on ICC, and it's pretty damn strong. >>> >>>Here's an example. In a 5 5 game tonight, Yace went from a negative score to >>>like +2 in 10-15 seconds, like right out of book vs Rebel. How long does it >>>take your prog to go positive in this position? >>> >>>[D]1q2kb1r/3n1ppp/2b1p3/1Np3N1/1p1Pn3/1B6/1P2QPPP/2B2RK1 w k - 0 1 >> >>3.0 seconds on an Athlon 1000 >> >>Sjeng: go >>Opening phase. >>Time for move : 1200000 >> 2 -96 12 297 Nxe4 Qxb5 >> 2 -96 12 1113 Nxe4 Qxb5 >> 3 -96 13 2023 Nxe4 Qxb5 Bc4 >> 3 -96 15 6426 Nxe4 Qxb5 Bc4 >> 4 -96 16 9695 Nxe4 Bxb5 Bc4 Bxc4 Qxc4 >> 4 -96 24 26691 Nxe4 Bxb5 Bc4 Bxc4 Qxc4 >> 5 -90 26 30692 Nxe4 Bxb5 Bc4 Bxc4 Qxc4 Nb6 >> 5 -90 43 69967 Nxe4 Bxb5 Bc4 Bxc4 Qxc4 Nb6 >> 6 -84 48 84218 Nxe4 Bxb5 Bc4 Bxc4 Qxc4 Qc7 dxc5 Bxc5 >> 6 -83 66 121875 Nxf7 ! >> 6 -57 79 152979 Nxf7 !! >> 6 -15 152 310066 Nxf7 Kxf7 Qg4 Qe8 Nc7 Ndf6 Qf4 Qd8 Bxe6+ >> 6 -15 175 368981 Nxf7 Kxf7 Qg4 Qe8 Nc7 Ndf6 Qf4 Qd8 Bxe6+ >> 7 6 300 631127 Nxf7 Kxf7 Qg4 Qe8 Qf5+ Nef6 Nc7 Be4 Qg5 Qd8 Bxe6+ >> 7 6 374 835975 Nxf7 Kxf7 Qg4 Qe8 Qf5+ Nef6 Nc7 Be4 Qg5 Qd8 Bxe6+ >> 8 39 526 1242897 Nxf7 !! >> 8 132 993 2251372 Nxf7 Kxf7 Qg4 c4 Bxc4 Qe8 Qf5+ Ndf6 Nc7 Nd6 Bxe6+ Ke7 >>Qe5 >> 8 132 1149 2601088 Nxf7 Kxf7 Qg4 c4 Bxc4 Qe8 Qf5+ Ndf6 Nc7 Nd6 Bxe6+ Ke7 >>Qe5 >> 9 165 1529 3567486 Nxf7 !! >> 9 174 2784 6158155 Nxf7 Kxf7 Qg4 Bd5 Bxd5 exd5 Qxd7+ Kg6 Qc6+ Nf6 dxc5 >>Qd8 Bd2 b3 >> 9 174 3142 6990592 Nxf7 Kxf7 Qg4 Bd5 Bxd5 exd5 Qxd7+ Kg6 Qc6+ Nf6 dxc5 >>Qd8 Bd2 b3 >> >>-- >>GCP > >Hey! Sjeng isn't supposed to beat up on Yace, especially on my new fast hw :( >You're spoiling my fun. > >Seriously, Sjeng seems to have improved dramatically from the days when Amateur >seemed to handle it pretty well. See below. (I wonder why the CCC search It looks like Sjeng has improved dramatically since version 11, which isn't all that old. Not only better eval or extensions, but much faster. Here are some results for Sjeng 11.0 on Athlon 1.2ghz.. <snip> 8 -45 1564 926399 Nxf7 ?? 8 -139 2031 1217255 Nxf7 Kxf7 Qg4 Qe8 Nc7 Ndf6 Qxe6+ Qxe6 Bxe6+ Ke7 d5 8 -138 2824 1693224 Nxe4 ! 8 -99 3203 1959513 Nxe4 Qxb5 Bc4 Qb7 Nxc5 Nxc5 dxc5 Bxc5 Qe5 9 -100 3561 2214145 Nxe4 Bxb5 Bc4 Bxc4 Qxc4 cxd4 Qxd4 e5 Qd1 <Be7> 9 -99 4448 2752556 Nxf7 ! 9 -64 4725 2970835 Nxf7 !! 9 -18 7925 4998663 Nxf7 Kxf7 Qg4 c4 Bxc4 Qe8 Qf5+ <Ndf6> <Nc7> <g6> <Qe5> 9 -18 8180 5284339 Nxf7 Kxf7 Qg4 c4 Bxc4 Qe8 Qf5+ <Ndf6> <Nc7> <g6> <Qe5> 10 17 10707 7973900 Nxf7 !! So 11.0 first goes positive at 107 seconds and ply 10 vs 3 seconds and ply 7 for the new version. My own chess engine was still at around -140 at ply 10, after 1/4 hour :) Angrim(good thing I don't take my chess engine too seriously)
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.