Author: Uri Blass
Date: 05:41:19 04/08/01
Go up one level in this thread
On April 08, 2001 at 06:47:56, Aaron Tay wrote: >On April 08, 2001 at 06:14:44, José Carlos wrote: > >>On April 07, 2001 at 11:18:34, Urban Koistinen wrote: >> >>>I have written down a algorithm for computing endgame tablebases that should be >>>about 10 times quicker than the Nalimov algorithm and requires much less ram. >>>It is similar to the Arlazarov&Futer algorithm of 1979 but is more general as it >>>does not require a pawn. >>>It might be too technical for most here, >> >> Thank you. Too technical algorithms are not for a crowded-of-programmers forum like this. We wouldn't understand anything. >> >> José C. > >Hey , be nice. He's new here. > >He is just feeling a little fustrated after posting at r.g.c.c and getting only >one response. Knowing the high level of general technicial expertise here, I >suggested to him that he should post here. > >So what do the experts think? Is the algorithim as good as he claims? Layman >like me want to know! 1)I tried to explain the paper in more words in order to understand it and the first part that I did not understand was the sentence: "t4 depends on t3 and d" It seems to me that t4 is dependent also on t1. Here is an example of position in t4 [D]2k5/8/2K1Q3/8/8/8/8/8 b - - 0 1 The position after Kb8 is in t3 but the position after Kd8 is in t1. I prefer not to try to understand the rest of the paper before I understand this point. 2)Here is my explanation of the part that I did understand: I call capturing or pawn move in the same word conversion. tables ti are clear They include all the positions when it is a win for white but white need i plies to get the conversion. Table d include only legal positions that are not stalemate and black cannot save itself by conversion. We want to find positions that are win for white. other positions are not intereting because it is obvious that they are not win for white. It is clear that t1 depends on t0 and previous tables that can be reached by a conversion(a position is in t1 only if white wins by conversion or by checkmate in 1 ply). t2 are positions when black is to move. The position is legal and black cannot save itself by conversion and it means that t2 is included in d. If black try to play a move that is not conversion it gets to position in t1 so it depends also on t1. t3 are positions when white is to move. White cannot win by conversion otherwise the position is in t1. White can get by a move that is not conversion to position in t2. This is the reason that t3 is dependent on t2 and t1. Uri
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.