Author: José Carlos
Date: 07:41:11 07/18/03
Go up one level in this thread
On July 18, 2003 at 10:03:48, Sune Fischer wrote: >On July 18, 2003 at 05:42:38, José Carlos wrote: > >>On July 17, 2003 at 18:08:33, Sune Fischer wrote: >> >>>Can your engine do what Crafty, Yace and Ruffian can't? >>> >>>[D]8/3K4/6Q1/8/8/8/5k1p/8 w - - >>> >>>Tsk tsk come on now, it doesn't get much simpler than that. >>> >>>-S. >> >> Yes, that's trivial knowledge, if you think about it: > >Right, it is trivial to implement it as a special case. >But to make the engine understand the same concept in more complicated positions >is a little more tricky. > >For instance like this one: >[D]8/3K4/6Q1/1p6/bP6/8/5k1p/8 w - - 0 1 > >It's really annoying that they cannot make the obvious connection. Mmm, are you sure that's a draw? I mean, black doesn't have the stalemate trick anymore... But of course you're right in the idea, we (programmers) can't generalize the pattern, but only cover up to certain clear generalizations. >>Anubis v0.00.0012 >>(c) Jose Carlos Martinez Galan >> >>1 16.00 0.02 189 g6h7 >>2 0.00 0.06 1255 g6h7 f2g2 >>3 0.00 0.11 4889 g6h7 f2g2 h7g8 g2h1 >>4 0.00 0.35 14986 g6h7 f2g2 h7g8 g2h1 g8f7 >>5 0.00 0.80 38704 g6h7 f2g2 h7g8 g2h1 g8f7 h1g1 >> >>etc... >> >> José C. > >That's certainly better than most! :) Thanks ;) >-S. José C.
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.