Author: Drexel,Michael
Date: 07:12:29 11/12/05
Go up one level in this thread
On November 12, 2005 at 09:51:19, Günther Simon wrote:
>On November 12, 2005 at 09:06:12, Drexel,Michael wrote:
>
>>On November 12, 2005 at 07:58:21, Günther Simon wrote:
>>
>>>On November 12, 2005 at 07:49:04, Uri Blass wrote:
>>>
>>>>On November 12, 2005 at 07:39:38, Günther Simon wrote:
>>>>
>>>>>On November 12, 2005 at 06:43:52, Drexel,Michael wrote:
>>>>>
>>>>>>On November 12, 2005 at 06:27:24, Kurt Utzinger wrote:
>>>>>>
>>>>>>>On November 12, 2005 at 06:17:10, Drexel,Michael wrote:
>>>>>>>
>>>>>>>>On November 12, 2005 at 03:59:54, Kurt Utzinger wrote:
>>>>>>>>
>>>>>>>>>In the running Open Dutch Computerchess Championship
>>>>>>>>>it was the program "Usurpator" to choose with White
>>>>>>>>>twice the queen move Qh5 in the second move. Of course,
>>>>>>>>>both games endet with a disaster. What the hell is the
>>>>>>>>>idea of the programmer to choose such a bad opening line?
>>>>>>>>>Kurt
>>>>>>>>
>>>>>>>>Both games ended in a disaster because Ursupator is a very weak program.
>>>>>>>>
>>>>>>>>I´m sure this would be a great opening move to play against all engines because
>>>>>>>>you quickly get them out of book and the chances to get some nonsense moves are
>>>>>>>>very high.
>>>>>>>>
>>>>>>>>Michael
>>>>>>>
>>>>>>>
>>>>>>> Hi Michael
>>>>>>> I very much doubt ... you can say what you want,
>>>>>>> but an opening where Black is slightly better
>>>>>>> after the second move should be avoided at any
>>>>>>> cost ... at least in matches between computer
>>>>>>> programs.
>>>>>>> Regards
>>>>>>> Kurt
>>>>>>
>>>>>>Nope, this should only be avoided if you don´t have some further book moves.
>>>>>
>>>>>What further book moves can repair this? ;)
>>>>>Tell me a few further book moves and I set up a match
>>>>>between 2 decent programs both with your super 2.Qh5? book.
>>>>>I bet White will reach > 65% at least.
>>>>
>>>>If white will reach >65% then it proves Michael's claim.
>>>>I guess that you mean black.
>>>>
>>>>Uri
>>>
>>>Yes I mixed the colours, see my other post.
>>>Assuming average distribution is around 53.5: 46.5,
>>>I bet White will be < 42% with that opening.
>>>
>>>Guenther
>>
>>[Event "Blitz:10'"]
>>[Site "?"]
>>[Date "2005.11.12"]
>>[Round "?"]
>>[White "Drexel"]
>>[Black "Fritz 9"]
>>[Result "1-0"]
>>[ECO "C20"]
>>[PlyCount "85"]
>>[TimeControl "600"]
>>
>>{64MB, Fritz.ctg, ATHLONXP 2.2Ghz} 1. e4 {0} e5 {0} 2. Qh5 {(Sf3) 6} Nc6 {0} 3.
>>Bc4{2} g6 {0} 4. Qe2 {(Df3) 2} Nd4 {-0.09/12 23} 5. Qd1 {2} Qg5 {-0.03/11 14} 6.
>>g3 {6} d5 {-0.19/12 14} 7. Bxd5 {8} Qf6 {-0.27/12 14} 8. c3 {12} Bg4 {
>>-0.06/13 2} 9. Qxg4 {(f3) 24} Nc2+ {-0.66/12 13} 10. Kf1 {12} c6 {-0.43/12 10}
>>11. Bc4 {8} b5 {-0.32/12 11} 12. Qd1 {48} Nxa1 {0.28/13 0} 13. Be2 {3} h5 {
>>0.39/11 14} 14. b3 {45} h4 {0.82/14 0} 15. g4 {25} a5 {1.25/13 0} 16. Bb2 {5}
>>Nxb3 {1.33/11 6} 17. Qxb3 {11} a4 {1.34/11 0} 18. Qc2 {6} Qe6 {1.50/12 18} 19.
>>c4 {21} a3 {1.65/13 0} 20. Ba1 {5} b4 {1.73/12 9} 21. d4 {7} Bg7 {1.70/11 3}
>>22. d5 {3} Qe7 {1.69/12 13} 23. Nh3 {7} Nf6 {1.75/11 5} 24. g5 {9} Nd7 {
>>2.07/12 7} 25. Kg2 {7} c5 {2.06/11 7} 26. Nd2 {4} Kf8 {2.33/13 8} 27. Rf1 {5}
>>Kg8 {2.60/13 14} 28. Bg4 {23} Rh7 {2.99/13 0} 29. f4 {15} Qd6 {3.25/12 0} 30.
>>Nb3 {10} Nb6 {3.23/12 3} 31. Nf2 {16} Qe7 {3.23/12 0} 32. Nd3 {2} exf4 {
>>3.27/12 11} 33. e5 {7} Qxg5 {3.44/12 11} 34. Kh3 {2} Na4 {3.69/12 16} 35. Qe2 {
>>9} Nc3 {3.42/11 6} 36. Bxc3 {3} bxc3 {4.36/13 9} 37. Nbxc5 {10} Rb8 {5.12/12 0}
>>38. Ne4 {9} Qd8 {5.83/12 6} 39. Nxc3 {5} f5 {5.98/12 8} 40. exf6 {19} Qxf6 {
>>7.27/12 2} 41. Ne4 {11} Qe7 {8.10/12 0} 42. Be6+ {3} Kh8 {9.95/14 10} 43. Nxf4
>>{1} 1-0
>>
>>Note that I had access to Fritz 9 evaluation during the game.
>>Of course Fritz will play different moves in a tournament game but I am sure it
>>would be a matter of some hours to figure out how to beat top engines almost
>>always in tournament games with 2.Qh5 for me.
>>
>>Michael
>
>Michael, why did you suddenly change the topic back to Human vs. program,
>when you already knew and even stated youself we were talking about
>program vs. program?
>http://www.talkchess.com/forums/1/message.html?461244
>Thus I don't see what your last posting has to do with all we were
>talking about before?
>
>Guenther
In my opinion it is possible to create an engine book which always plays 1.e4 e5
2.Qh5 and is very sucessful, since the other books do not contain the
"refutation".
The game might be already a good start.
Many engines like to play Qg5 or/and sacrifice a pawn for development with d5.
You are certainly right that 1.e4 e5 2.Qh5 is a bad opening for computer vs.
computer _unless_ you have a lot of lines in the book which cover all important
variations. In this case 2.Qh5 might be a good idea.
Michael
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.