Author: Robert Hyatt
Date: 20:06:00 03/15/03
Go up one level in this thread
On March 15, 2003 at 00:28:20, Ricardo Gibert wrote: >On March 15, 2003 at 00:01:37, Robert Hyatt wrote: > >>On March 14, 2003 at 16:58:15, Uri Blass wrote: >> >>>On March 14, 2003 at 14:19:28, Ed Panek wrote: >>> >>>>On March 14, 2003 at 13:35:55, Will Strickland wrote: >>>> >>>>>Hi all. Just curious as to what some of you feel are the best current programs >>>>>for evaluating closed positions(locked pawn chains and semi-locked pawn chains, >>>>>etc...). And which of the top programs demonstrate a major weakness in this >>>>>area? I know that programs, in general, suffer in closed positions compared to >>>>>open ones. But how would you rate the performance of Fritz, Shredder, Tiger, >>>>>Junior, Hiarcs, Chessmaster, Crafty, Ruffian, etc... in such positions? Thanks. >>>>> >>>>>Will >>>> >>>> >>>>Actually Crafty seems fairly good in these types of positions. Especially where >>>>a pawn break can occur. >>>> >>>> >>>>Ed >>> >>>No >>> >>>I know that Crafty has anti symmetric evaluation and it means that it is >>>relatively bad in these positions because it may try to open the position also >>>when it is wrong. >>> >>>I prefer to use an objective program to analyze games and not a program that >>>tries to avoid closed positions. >>> >>>Uri >> >>Fine. Which program gives the right answer in such positions then? None I >>know of. > > >Which human gives the right answer in such positions? Nobody's perfect. The point was, however, that if asymmetry for such problems is _bad_ then which program without asymmetry does it well? I'm not aware of any that don't have problems with blocked positions. It's the way the "draw masters" do their evil on ICC. :) > >As an aside, it is worth pointing out that DB2 did not do too badly in that >closed Ruy game against Kasparov, where Kaspy resigned in a drawn position. Correct, but who knows how they did it? The idea for "asymmetry" is well- known. > >I think that in *general*, for the purpose of *analysis*, either type of program >will do more or less equally well. Computers usually suck at such positions >regardless.
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.