Author: Robert Hyatt
Date: 17:47:21 03/06/98
Go up one level in this thread
On March 06, 1998 at 13:08:48, Ren Wu wrote: > >On March 06, 1998 at 07:27:46, Kai Lübke wrote: > >>The following position (probably not unknown to many who followed >>the WMCCC in Paris) is from the Dark Thought vs. MChess 7 game in Paris. >>In this situation, MChess 7 played Rc4!! >> >>2r2kn1/1p3pr1/pq2p3/3pP2N/b4P1P/PnPQ4/1B4P1/1R2KB1R b k - ; bm Rc4; id >>"shep06" >> >>I have tested all current top programs on this, and they all find >>the solution between 1:52 (Hiarcs 6) and 13:58 (Rebel 9, all on P6-200). >>Their evaluations are between -0,62 (Hiarcs 6) and -0,25 (Fritz 5, Rebel >>9). >>MChess 7 itself takes 4:04 and scores it as -0,53. >> >>Now Junior 4.6 (engine) finds this in 4:08, and after 4:50 it is the >>only program so far to return a positive score for black: +0,23. >> >>Maybe we'll soon have a program that actually scores this as a clear >>win. At least we're one step closer now. :) >> >>Another "Hooray!" for Amir! >> > >This is not a hard problem for my program Initiative. On a K6-200, with >24MB hash table, mine find the move Rc4 (fail high) after 52 seconds, at >ply 10, and get the score .21 at 60 seconds. It finished ply 11 after >139 seconds, with score .99. > >the log file looks like > >10. -52 37.4 4070305 b3c5 d3d2 g7g6 h1h3 g8h6 d2c1 h6f5 h5f6 c5e4 >10. -51 52.4 5599675 c8c4 h5g7 c4e4 f1e2 a4b5 c3c4 b5c4 d3g3 e4e2 >e1f1 b6f2 g3f2 e2b2 f1g1 b2f2 >10. 21 60.6 6415213 c8c4 h5g7 c4e4 f1e2 a4b5 c3c4 b5c4 d3g3 c4e2 >g7e6 f7e6 >10. 31 65.9 7012326 c8c4 h5g7 c4e4 f1e2 a4b5 c3c4 b5c4 d3e4 d5e4 >e2c4 b6e3 c4e2 e3g3 e1d1 g3g7 >10. 31 71.6 7631678 c8c4 h5g7 c4e4 f1e2 a4b5 c3c4 b5c4 d3e4 d5e4 >e2c4 b6e3 c4e2 e3g3 e1d1 g3g7 >11. 99 110.6 11351981 c8c4 h5g7 c4e4 f1e2 a4b5 g7e6 b6e6 d3h3 e4e2 >e1d1 e2d2 d1e1 e6g6 f4f5 d2e2 e1d1 g6g2 >11. 99 138.8 14307908 c8c4 h5g7 c4e4 f1e2 a4b5 g7e6 b6e6 d3h3 e4e2 >e1d1 e2d2 d1e1 e6g6 f4f5 d2e2 e1d1 g6g2 > >Be a ameture program, I am glad it can do something better than >professionals! Alrought I know there are many more program can do this >too. > >Ren. (renw@iname.com) I ran this on Crafty. I believe that it found Rg3 to be just as good, and it liked this move from ply=1 to the end, with the eval getting higher search by search... IE the same move that Dark Thought plays now..
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.