Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Book/No-book more data from a longer run

Author: Dann Corbit

Date: 11:54:22 02/21/05

Go up one level in this thread


On February 21, 2005 at 13:58:15, Robert Hyatt wrote:

>On February 21, 2005 at 13:40:37, Dann Corbit wrote:
>
>>This is on the 2.2 GHz 64 bit computer (which had fewer problems than the slower
>>machine).  The first set is the raw data, and the second set has had all
>>duplicate games removed.
>>
>>  Program                    Elo    +   -   Games   Score   Av.Op.  Draws
>>1 Polyglot fruit-book      : 2553   31  35   372    59.9 %   2483   18.3 %
>>2 Polyglot glaurung-book   : 2540   32  35   372    57.5 %   2487   16.7 %
>>3 Polyglot glaurung-no-book: 2481   32  34   372    46.2 %   2507   16.7 %
>>4 Polyglot fruit-no-book   : 2427   40  29   372    36.3 %   2525   14.0 %
>>
>>  Program                    Elo    +   -   Games   Score   Av.Op.  Draws
>>1 Polyglot fruit-book      : 2550   32  37   348    59.3 %   2484   17.5 %
>>2 Polyglot glaurung-book   : 2536   33  35   363    56.7 %   2489   16.5 %
>>3 Polyglot glaurung-no-book: 2483   33  36   343    46.9 %   2505   16.9 %
>>4 Polyglot fruit-no-book   : 2430   41  30   354    36.9 %   2524   12.7 %
>>
>>It seems that Fruit may benefit from a book slightly more than glaurung.  At
>>least, in the case of fruit, there is now enough data to say for sure that a
>>book does help.
>
>
>I don't believe there was ever any question about whether a book helped or not.
>But whether or not it helps by a 700+ rating point advantage is the question at
>hand.  The answer is "no" IMHO.
>
>:)

I think it very clear that a 2700+ Elo program like Shredder cannot possibly be
helped by 700 points to reach 3400.  A 1500 Elo program with a 50 ply deep book
with no bad lines in it might get that much improvement.

At any rate, Vincent was surely engaging in hyperbole.

Also, I think at issue was TPR and not actually Elo, which are quite a bit
different.

For instance, I might have 50 machines playing matches against Shredder 9.x to
look for possible weaknesses around the clock for a month.  With this data, I
could surely produce a book to give Shredder problems.

So, with special preparation, you can potentially create a "killer book" good
for one tournament that might give a huge TPR boost.

That was (I believe) what Vincent meant.



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.