Re: [Bug-gnubg] Bearoff dB position, few questions
From:
Massimiliano Maini
Subject:
Re: [Bug-gnubg] Bearoff dB position, few questions
Date:
Mon, 29 Jun 2009 12:26:52 +0200
Robert-Jan Veldhuizen <address@hidden>
wrote on 28/06/2009 23:39:32:
> For money, the 2-sided database is perfect, so there's no lookahead
> needed. 0-ply gives the final answer already.
>
> For matchplay, the 2-sided database is not perfect. Obviously
> cubeful equities from the db are of no use, therefore cube actions
> also need to be determined using normal lookahead. But moves can be
> different too, and GnuBG plays according to score (and cube!). So,
> 0-ply uses the db cubeless figures, applies "match-janowski"to
get
> cubeful equities, and then uses the best CUBEFUL move. This is NOT
> always the best move for cubeless equity, nor is it always the same
> move that the database says is best for money cubeful.
That's the explanation I gave to Misja, written in
a much more understandable manner :) Christian has since explained the meaning of cubeless
figures (which are in fact not really cubeless): 2ply cubeless is computed
using cubeful selection of best moves and cube actions (if cubeful
eval is enabled). Calling these GWCs "cubeless" is probably
a bit misleading but as Misja said, once you know ...