Author: Dave Gomboc
Date: 23:27:59 10/20/98
Go up one level in this thread
On October 20, 1998 at 13:34:51, Ulrich Tuerke wrote: >On October 20, 1998 at 13:23:50, Ulrich Tuerke wrote: > >>On October 20, 1998 at 13:21:56, Amir Ban wrote: >> >>>On October 20, 1998 at 12:36:49, Ulrich Tuerke wrote: >>> >>>>On October 20, 1998 at 10:08:33, Amir Ban wrote: >>>> >>>>>On October 20, 1998 at 01:16:06, blass uri wrote: >>>>> >>>>> >>>>>> >>>>>>The question is if hiarcs6 claimed a draw with making the last move. >>>>>>If it is not the case it is a bug in hiarcs6(hiarcs6 knew that it was a draw >>>>>>otherwise there was no chance for fritz3 to do a capture only in the 101 ply. >>>>>> >>>>>>Uri >>>>> >>>>> >>>>>In auto232 games there is no agreed result, so some confusion about it may >>>>>occur. In the DOS autoplayer, practically the only way to end a game is to stop >>>>>playing, and let the game be terminated by a timeout. This means that when one >>>>>program resigns or claims a draw, the other program only sees a timeout and can >>>>>only guess what happened. >>>>> >>>>>In this case, it's possible that Hiarcs printed on the screen "It's a draw !", >>>>>but continued playing anyway. Another possibility is that both programs realized >>>>>this was a draw, but the tester who looked at the final position did not know >>>>>this and thought Fritz won. >>>>> >>>>>Amir >>>> >>>>Confusion even grows cause some programs themselves decide to terminate games, >>>>when auto232 is used. When some threshold score is reached, they refuse to send >>>>moves to the lpt (or whatever) device. No chance for the tester to continue. I >>>>learned this when I was puzzled about a prematurely (IMO) terminated game which >>>>was lost by Comet and asked for this at SSDF. >>>>In general such option may be useful but it should really be an option. >>>> >>>>Uli >>> >>>This depends: It's perfectly ok to terminate a game if you want to resign, and >>>you are free to resign whenever you want. If you mean that a program refused to >>>continue because it claimed a win, this should not be allowed at any score. >> >>Yes, I meant that. > >[Event "40Z/2h"] >[Site "Autoplayer"] >[Date "1998.06.27"] >[Round "12"] >[White "Shredder 2.0 P200 MMX"] >[Black "Comet-A.90 P200 MMX"] >[Result "1-0"] >[PlyCount "46"] > >1. d4 d5 2. c4 e6 3. Nf3 Nf6 4. cxd5 exd5 5. e3 c6 6. Bd3 Bd6 7. h3 O-O >8. O-O >Na6 9. Bxa6 bxa6 10. b3 Qe7 11. Qd3 Re8 12. a4 Rb8 13. Ba3 Bxa3 14. Rxa3 >Ne4 >15. Ra2 Qb7 16. Rb2 Bf5 17. Nh4 Bd7 18. Rc1 Rec8 19. Rcc2 c5 20. f3 g5 >21. fxe4 >dxe4 22. Qf1 cxd4 23. Nf5 Rxc2 1-0 > >How do you think about this game which - according to SSDF - was claimed as a >win by Shredder. It was also counted that way by SSDF. >May be it's a very obvious win for a chess expert. But I am not one and I would >have liked to see a few more moves. > >Uli > >> >>> >>>Amir If the final position is not checkmate, and Shredder claimed a win and refused to move, it should have lost the game by time forfeiture. Of course, this presupposes that the operator did not preconfigure Shredder to terminate winning games early when its default is to play until checkmate. Dave Gomboc
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.