Author: Uri Blass
Date: 06:26:46 12/27/01
Go up one level in this thread
On December 27, 2001 at 09:14:32, Uri Blass wrote: >On December 27, 2001 at 08:54:19, Gian-Carlo Pascutto wrote: > >>On December 27, 2001 at 08:32:14, José Carlos wrote: >> >>>On December 27, 2001 at 06:43:23, Gian-Carlo Pascutto wrote: >>> >>> Mine needs 6:40 (PIII-900) to show a negative score here, in ply 16. There's >>>so much to do in eval... >> >>I get a negative score in 18 seconds, but let me show you what I mean >>with 'nullmoves itself to death': >> >>[...] >> >>10 37 413 2523987 g5 ! >>10 37 476 2889129 Kf8 ! >>10 37 500 3027747 e5 ! >>10 37 518 3134806 h5 ! >>10 37 545 3306723 a5 ! >>10 37 565 3411045 a6 ! >>10 37 596 3591841 g6 ! >>10 37 663 4009756 h6 ! >>10 37 675 4087799 Kh8 ! >>10 37 688 4153091 Kf7 Nc2 Ke7 Kg1 a5 b4 a4 b5 e5 Ne3 >>11 31 793 4902264 Kf7 Nc2 a5 Kg1 g5 b4 a4 b5 g4 Ne3 h5 >>11 32 797 4935549 g5 ! >>11 32 1003 6147838 h5 ! >>11 32 1057 6480470 Kf8 ! >>11 32 1105 6769974 g6 ! >>11 32 1180 7247897 a5 ! >>11 32 1215 7450589 e5 ! >>11 32 1231 7546691 a6 ! >>11 32 1258 7694831 h6 ! >>11 32 1303 7980135 Kh8 ! >>11 32 1335 8172728 Kf7 Nc2 a5 Kg1 g5 b4 a4 b5 g4 Ne3 h5 >>12 -18 1810 11465842 Kf7 ?? >>12 -17 1921 12091128 g5 ! >>12 -17 2922 18491468 g6 ! >>12 -17 3306 21019493 h5 ! >>12 -5 3486 22062248 h6 ! >>12 -5 3701 23193981 Kf8 ! >>12 -5 3776 23600593 a5 ! >> >>[...] >> >>-- >>GCP > >I know what is null move but I do not understand this output. >If 2 moves have the same score then >what is the reason that your program changes it's mind. > >If a move fail high then it means that the score is clearly >bigger than the previous score so if the previous score was >37 the new score should be at least 38 >I assume that you use integers for scores. > >Uri I understand from other posts that it was fail high that was not confirmed but I still do not understand how fail high can be wrong. I read in bruce pages that it happens only to strong programs and I also remember from previous discussion about MTD that it is related with hash tables. Is the reason for the fail high that were not confirmed only null moves or also hash tables? I am also interested to see the tree that cause to this problem chessbase interface can understand tree so if it is possible to post a practical tree as small as possible that cause the problem with comments after every node (like number of node,alpha value,beta value) then it may be productive for me to look at it I think that a program that translate nodes of a program to a tree that chessbase interface can understand may be productive for chess programming because I believe that looking at the tree node after node may be productive to get ideas. 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.