Author: Len Eisner
Date: 18:59:57 12/20/99
Go up one level in this thread
On December 20, 1999 at 21:48:29, Len Eisner wrote: >On December 20, 1999 at 21:14:48, Gustavo Pereira wrote: > >>On December 20, 1999 at 17:53:06, Len Eisner wrote: >> >>>On December 20, 1999 at 16:52:07, Torstein Hall wrote: >>> >>>>On December 19, 1999 at 20:10:59, Len Eisner wrote: >>>> >>>>>In the following endgame position, Fritz 6 plays f7?? I know this is the >>>>>tablebase at work but it seems strange to me. And after …a2, Fritz 6 plays >>>>>Kg2?? which also look weird. Then after Black queens the pawn with a1-Q, Fritz >>>>>6 goes out of the table base and plays f8=q which is only good for a draw >>>>>according the Thompson database on the Fritz 5 CD. If I had the Nalimov QPvsQ >>>>>tablebase, Fritz 6 would have played g8=q and won. But that table base does not >>>>>come with the program. >>>>> >>>>>I created this position to test the table base. It did not arise in an actual >>>>>game. >>>>> >>>>>wKg1,Pf6,g7/bKb5,Pa3 >>>>> >>>>>Len >>>> >>>>Strange, My F6 plays g8Q, says that its check mate in 7 and do not hit the >>>>tablebase at all. Are you sure you not are playing in sparring mode, friend mode >>>>or other weak mode? >>>> >>>>Torstein >>> >>>My Fritz 6 plays f7 in analysis mode and announces mate in 13! If I set it for >>>the long game level 40/40 it moves immediately and announces mate in 25! >>> >>>I have the tablebases from Hiarcs 7.32 loaded on my hard drive. I'm not >>>actually using the ones that came on the F6 CD. I wonder if that could account >>>for the difference. >>> >>>Len >> >>Are you sure all the tablebases for 5 pieces are present in your drive? >>If they are not that may explain the problem. >>If you use tb's it is much better if you have all the 5 pieces tb's, not just a >>set of them. > >I only have the 5 piece tablebases that are included on the Hiarcs 7.32 and >Fritz 6 CDs. The other 5 piece tablebase files are too large to download over a >dial up connection. I ordered the Turbo Endgame 4 CD set from Chessbase, but >even that only has the "most important" 5 piece endings. I should have it in >about a week according to Chessbase. > >In any case, the position I mentioned should be covered by the KPP vs KP 5 piece >tablebase which I do have loaded. So I don't know why Fritz 6 (and H7.32 for >that matter) plays f7 since it is obviously not the fastest winning line. > >I'm also curious why Torstein's Fritz 6 does not even attempt to look up the >position in the tablebase since he has it loaded. > >Once again, here's the position: > >wKg1,Pf6,g7/bKb5,Pa3 > >Len Here's the position in the other format: 8/6P1/5P2/1k6/8/p7/8/6K1 w - - 0 1 Len
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.