On Tue 29 Mar 2005 (08:27 +0100), Ian Shaw wrote:
From: Jim Segrave [mailto:address@hidden
On Mon 28 Mar 2005 (16:47 +0100), Ian Shaw wrote:
Could we reduce the size of the two-sided database by omitting gin
positions?
Gnubg could assume that if the position was inside the
database space
but the result was not stored, then the position is gin. A simple
0-ply evaluation could be used to verify who has won, and
as a sanity check.
I don't know how much effect this would have, but it would have
greater benefits for the larger databases because more
positions would be gin.
I've not looked at the code, but I think it calculates where
to look as a fixed function of the current position. If
that's the case, then you need different indexing of the
database, which might not be a gain.
I was thinking the index number would remain the same. If there is no
entry for that index, then the position is gin. Or is the index number
not actually part of the database? Does gnubg just count CRs or commas
or something?
I'm not at all sure what the structure of the bearoff files is. You
might be right that there's a separate index with pointers, in which
case it would be possible to have a value indicating no data (or
pointing to a common entry). I think it might be time for a bit of
scripting to see how many duplicate entries there are (gin being just
one possibility).