Author: Laurence Chen
Date: 11:46:24 08/19/99
Go up one level in this thread
Hi Terry,
After much thought and reading of all previous posting about "bad" opening
book lines, it made me realize why such lines exists at all in the chess engine
book. I believe the problem lies with the user expectations versus the
programmers expectations of how the chess engine would be used. A lot of us are
not professional chess players who plays chess for a living, and our time to
study chess is very limited. Consider this, if you were to sit down and play
your chess engine, and the chess engine played a bad line would you be able to
refute it. I believe most of us would not be able to refute the bad line even if
our booked up opponent played an inferior line against us in an OTB game. So the
inferior lines in the chess opening book in the engine was designed against
human players not against computers. That's why one should design an opening
book for chess engines which play against other chess engines. They are a
different type of beast. Instead of complaining how bad the chess engine opening
book is, just remember that it was designed with the human player in mind, and
if the chess engine played only good lines then how would you, a human player,
learn opening theory. Would you know why such and such line is bad if the chess
engine never played? Remember that most chess engines, Fritz 5.32/5.16, Junior
5, Hiarcs 7.32, Hiarcs 7, MChess 8, and so on, have learning capabilities, and
will learn from the loses and eventually avoid playing these losing lines once
enough games have been played. As a human player I would like that the chess
engine to play some inferior lines so that I can beat the chess beast once in a
while for a change.... So be happy and don't worry about inferior lines. :)
Laurence
On August 18, 1999 at 17:48:32, Terry Ripple wrote:
> This is one "Huge" Opening mistake and i here that there are more in Hiarcs7.32
>opening book! So what does one do? Maybe the Power Books 99 is a solution but it
>states that it will weaken your program somewhat over the long run because
>it will have many opening lines not suited for chess programs because of lines
>that lead to closed positions that need long term planning that programs
>aren't good at doing, but it got to be better than losing lines already in the
>book of Hiarcs, don't you think? If we only knew how many lines in Hiarcs book
>were at fault because it might not matter if only a few were bad and we were
>told which ones to deleat out.
> Any other opening books that would be worth a try?
>
>Thankyou for any information,
>Terry
>
>PS. Here is the game, the first move out of book at move #21. Nb6+ (-5.81/14)
>
>Event "60'/30+0'/0+30'/Game"]
>[Site "?"]
>[Date "1999.07.21"]
>[Round "7"]
>[White "Hiarcs 7.32"]
>[Black "Comet B02"]
>[Result "0-1"]
>[ECO "C56"]
>[PlyCount "50"]
>
>{W=13.8 ply, B=10.8ply} 1. e4 {0} 1... e5 {0} 2. Nf3 {0} 2... Nc6 {0} 3. d4 {0}
>3... exd4 {0} 4. Bc4 {0} 4... Nf6 {0} 5. O-O {0} 5... Nxe4 {0} 6. Re1 {0} 6...
>d5 {0} 7. Bxd5 {0} 7... Qxd5 {0} 8. Nc3 {0} 8... Qa5 {0} 9. Nxe4 {0} 9... Be6 {
>0} 10. Bd2 {0} 10... Bb4 {0} 11. Bxb4 {0} 11... Qxb4 {0} 12. Neg5 {0} 12...
>O-O-O {0} 13. Nxe6 {0} 13... fxe6 {0} 14. Ng5 {0} 14... Rd7 {0} 15. Nxe6 {0}
>15... Re8 {0} 16. Qg4 {0} 16... d3 {0} 17. Qxb4 {0} 17... Nxb4 {0} 18. Nc5 {0}
>18... Rxe1+ {0} 19. Rxe1 {0} 19... dxc2 {0} 20. Nxd7 {0} 20... Nd3 {
>Both last book move 0} 21. Nb6+ {-5.81/14 418} 21... axb6 {-5.14/14 470} 22.
>Rf1 {(Ra1) -5.81/14 219} 22... c5 {(c1Q) -5.22/13 296} 23. a4 {
>(f3) -6.26/14 586} 23... c1=R {(c1Q) -5.39/11 27} 24. Rxc1 {-6.51/14 1358}
>24... Nxc1 {-5.39/9 1} 25. Kf1 {-6.51/13 209} 25... Kd7 {-5.57/10 34} 0-1
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.