Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Why use opening books in machine-machine competitions?

Author: Claude Le Page

Date: 02:33:17 11/25/03

Go up one level in this thread


Hello Uri!
I think you are right about Nunn-style tournaments ; but why do you request that
positions are not known in advance? the only difference is that with known
positions it becomes a thematic tournament; why not? with thematic matches or
tournaments you can make very exciting experiments : so when you see that an
engine has difficulties with some type of positions you can make a thematic
tournament to know whether it is the same with others; you can do it , either
with a panel of prépared variations or with a specialized opening book:I have
dozens of them!thus , I could make some discoveries about engines:
Here is an example : in the Fegatello variation (known since 400 years)
1 e4 e5 2 Nf3 Nc6 3 Bc4 Nf6 4 Ng5 d5 5 exd5  out of book most of engines
play 5...Nxd5 and still fewer can play 6 Nxf7 (Polerio) or 6 d4 (Jaenisch);
only 2 of the CB engines pass the twofold test: shredder7 and junior7;
junior8 passes only the first ; fritz8 and hiarcs9 pass none
this learns very much about functioning of engines , as depth of analysis was
sufficient  (at least for fritz8) to avoid this trap:there is something else
More (I come back to initial purpose) it gives a justification to the use of
opening books: it avoids the engines to fall in these sort of traps:
one could say : a chess engine is worth 1000 ELO in opening, 2000 in middlegame,
 3000 in the endgame
Friendly Yours
Claude Le Page



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.