Author: GuyHaworth
Date: 04:23:38 02/09/03
Go up one level in this thread
Leaving aside 7-man EGTs, let's suppose you are generating 6-man EGTs.
'Seeding' the whole process requires two steps after basic setup of the
workspaces.
1) Find 'the first set' of positions that are 'win in 0' in DTC terms.
These are the stm-mated and stm-forced-to-capture-to-loss positions.
2) Find 'the second set' of positions which 'win in 1' in DTC terms.
These are the 'mate in 1' and/or 'capture to win in 1' positions
Your idea is performing step 2 which, as above, has to be done anyway - but only
identifies DTC = 1 positions (which are captures).
'Unmoving' from the 5-man positions once rather than 5 times is a small economy
across 5 6-man EGTs but calculating the index of the resultant 6-man position
still has to be done 5 times.
I don't think you are saving a great deal in %-terms overall - probably not
enough to justify the extra complexity of part-seeding 5 EGTs ... and you still
have to scan for 'mates in 1' so you might as well pick up 'captures in 1' at
the same time.
g
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.