Author: Robert Hyatt
Date: 15:21:08 07/03/03
Go up one level in this thread
On July 02, 2003 at 12:47:00, Russell Reagan wrote: >On July 02, 2003 at 08:33:36, Michel Langeveld wrote: > >>Does a (bugfree) program need 26 ply, 18 ply, or different to see that this >>position is > 2. (normal plies, no pawn, singulair extensions). >> >>I think a normal ply searcher needs more as 18 ply to see in all variations that >>the f5 or b6 goes off the board. >> >>How much ply does your program need? >> >>[D]8/k7/3p4/p2P1p2/P2P1P2/8/8/K7 w - - > >I believe you need 26 ply without a hash table, and 18 ply with a correctly >working hash table. This position is usually used to test whether or not your >hash table is working correctly. > >Also, this position is Fine 70 (not 60), IIRC. :) There is no right number of plies for hashing. It is too sensitive to table size, replacement strategy, random numbers used, and good old lady luck. 26 plies is all you need to win a pawn. If you find it in less than 26 plies, hashing is the explanation. Many will find it at 18-19 plies, but doing so shows a weakness in the program's move ordering (mine included.)
This page took 0.01 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.