Introducing the *.EBF project

Discussion of chess software programming and technical issues.

Moderators: hgm, Rebel, chrisw

flok

Re: Introducing the *.EBF project

Post by flok »

hgm wrote:
Personally, I think it wise to store every pv node an engine ever finds in a file. Load the file on startup. Whenever a new pv node is found or any pv node is improved, update the file.

Let's call it permanent pv hash.
That is another kind of book information.
Pv nodes are rare, and therefore valuable.
Funny you bring that up. I was just considering implementing my mini-Shogi book in Shokidoki that way.

What is a PV node? What I think a pv is, is the list of moves from root to the leave of the ideal exchange of moves. Are those the PV nodes? Or?
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Introducing the *.EBF project

Post by Dann Corbit »

flok wrote:
hgm wrote:
Personally, I think it wise to store every pv node an engine ever finds in a file. Load the file on startup. Whenever a new pv node is found or any pv node is improved, update the file.

Let's call it permanent pv hash.
That is another kind of book information.
Pv nodes are rare, and therefore valuable.
Funny you bring that up. I was just considering implementing my mini-Shogi book in Shokidoki that way.

What is a PV node? What I think a pv is, is the list of moves from root to the leave of the ideal exchange of moves. Are those the PV nodes? Or?
A pv node is a principal variation node.
It is one of the moves in the sequence that the engine intends to play.
https://chessprogramming.wikispaces.com ... +variation
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
flok

Re: Introducing the *.EBF project

Post by flok »

Dann Corbit wrote:A pv node is a principal variation node.
It is one of the moves in the sequence that the engine intends to play.
https://chessprogramming.wikispaces.com ... +variation
Ok but why can we have multiple?
PK
Posts: 893
Joined: Mon Jan 15, 2007 11:23 am
Location: Warsza

Re: Introducing the *.EBF project

Post by PK »

In Rodent there is an option to record scores and depths for root positions, to retrieve them deeper in the search. I do not use it much, because every change renders these data useless, but toying with it for a few days shown some convergence with human book lines. Once it learned to avoid Alechine and to play Petroff, the other time I wanted it to learn to avoid some dubious line in QGD, but it kept shuffling between 1.d4 d5, 1.d4 Nf6 and 1.d4 e6 instead.
User avatar
Evert
Posts: 2929
Joined: Sat Jan 22, 2011 12:42 am
Location: NL

Re: Introducing the *.EBF project

Post by Evert »

flok wrote: What is a PV node? What I think a pv is, is the list of moves from root to the leave of the ideal exchange of moves. Are those the PV nodes? Or?
Typically, a "node" is a position that is reached after a move. A PV-node is a node that is reached after making a move that is in the PV. The term is also commonly used for moves that are searched with an open (alpha != beta-1) window in a PVS search (technically, those would be "expected PV nodes").
User avatar
hgm
Posts: 27790
Joined: Fri Mar 10, 2006 10:06 am
Location: Amsterdam
Full name: H G Muller

Re: Introducing the *.EBF project

Post by hgm »

In this context a PV node is just a node for which you have an exact score, rather than a bound.
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Introducing the *.EBF project

Post by Dann Corbit »

I suggest storing them in sets according to material signature.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
User avatar
Rebel
Posts: 6991
Joined: Thu Aug 18, 2011 12:04 pm

Re: Introducing the *.EBF project

Post by Rebel »

First contribution is in by our member Folkert, embla.ebf

http://rebel13.nl/misc/ebf/
Robert Pope
Posts: 558
Joined: Sat Mar 25, 2006 8:27 pm

Re: Introducing the *.EBF project

Post by Robert Pope »

Rebel wrote: The goal of this project is to create an extended opening book (or multiple books) for all chess engines that roughly will ensure:

Book hits for the first 10 plies with a 90-95% security
Book hits for the first 15 plies with a 75-80% security
Book hits for the first 20 plies with a 50-55% security
Book hits for the first 25 plies with a 25-30% security
Book hits for the first 30 plies with a 10-15% security.

If these numbers are feasible has to be seen, we are going to try it anyway.
What do you mean by "security"? Just that the book is wide enough that it will find a move X% of the time at that ply?
User avatar
Rebel
Posts: 6991
Joined: Thu Aug 18, 2011 12:04 pm

Re: Introducing the *.EBF project

Post by Rebel »

Yep.