Author: Keith Ian Price
Date: 13:01:50 08/22/98
Go up one level in this thread
On August 22, 1998 at 13:42:35, blass uri wrote: > >On August 22, 1998 at 12:36:06, Thorsten Czub wrote: > >>the game is in the 23rd move... moves so far: >> >>[Event "summer-tournament"] >>[Site "k6/200 40/120"] >>[Date "0098.07.22"] >>[Round "5"] >>[White "ChessSystemTal win95"] >>[Black "ChessTiger 11.5"] >>[Result "*"] >> >>1. e4 c6 2. d4 d5 3. exd5 cxd5 4. c4 Nf6 5. Nc3 g6 6. Qb3 e6 7. Nf3 dxc4 >>8. Bxc4 a6 9. Bg5 b5 10. Bd3 Be7 11. a4 Nc6 12. axb5 Nxd4 13. Qd1 Nb3 >>14. Qxb3 Qxd3 15. Ra4 Qd7 16. Rf4 Qd8 17. O-O O-O 18. Rd1 Nd7 19. Bh6 Re8 >>20. Qa2 Qc7 21. Nd5 Qb7 22.Nxe7+ Rxe7 23.* >> >>looks like take is in it's element... >> >>CStal says +4.08 and wants to play Qa5 so far, tiger expects this and said -1.12 >>with playing Rxe7... >> >>23.Qa5 came. ... >> >>to be continued... >chess system tal is winning >I see no defence for black after 23.Qa5 Re8 24.Ng5 and if f6 25.Ne4 >and if 24...f5 25.Qc3 > >chess system tal and crafty are the surprising programs in this tournament. > >I do not understand the results because I believed mark young that >chess system tal is a weak program. I've always liked CSTal, even though the released version on CD has a bug, and the Paris version is not real strong against computers. I like the way it plays, and I've always thought that it will improve with a faster processor more than the tactical fast searchers. I think that the entire program, plus most of its hash-tables could fit within the 2MB L2 cache of a Xeon, and thus operate at a 450-Mhz clip in virtually all operations. I don't know if the Win95 version will be able to do this, as there is a lot of overhead in interfacing to an API which is poorly written for space conservation. I'd love to see a CSTBlack (which I don't have, so I can't speak for sure, but which is supposed to be better against computers) running on a 450 Mhz 2MB of L2 cache Xeon in a tourney against all the best of the rest, just to see. One of these guys here is bound to have one soon. (Tom, can you convince Bill Bader that this would be a good way to test the new Xeons?) >Is it a new version of chess system tal that mark young does not have? >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.