Author: Vincent Diepeveen
Date: 17:17:58 09/21/00
Go up one level in this thread
On September 21, 2000 at 17:08:30, Chessfun wrote: >On September 20, 2000 at 17:37:33, Vincent Diepeveen wrote: > >>On September 20, 2000 at 15:59:28, Joshua Lee wrote: >> >>>I was wondering which programs beat Gandalf in 40/2? and how it compares with >>>Hiarcs 7.32. Lastly if it has such a high LCT2 and BS results were the test ran >>>properly?? was each position looked at for 10 or 15Minutes?? I have run tests >>>where others said the rating was one thing example Ikarus 0.18 was supposed to >>>be 2700 but in actuality on my 880Mhz Athlon it equaled Hiarcs at 2670.... >> >>sorry i think you are confusing 2 things >> - solving positions from which one knows the solution in advance and >> only tactical power is needed to solve it or optimistic eval >> - scoring points in games. Auto232 games where learning rules? >> Or winboard games? > >I don't think he is. Seems to me he is asking two separate questions. > >Thanks. >Sarah. Note that gs2930 is heaven for programs who love to put opponents under pressure using open files, like Tiger and Gandalf, so i'm not surprised both programs do very well on this testset. In both cases i'm dead sure the evaluations find the positions, not the search. gs2930 testset just tests : "how quick do you want to give a pawn for a bit of vague attack?". Both of the above programs get a nobel prize for that obviously.
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.