Computer Chess Club Archives


Search

Terms

Messages

Subject: The nature of the move from DTM EGTs

Author: GuyHaworth

Date: 08:51:04 06/16/03

Go up one level in this thread



As I've suggested before on CCC, the DTZ metric has its merits:

- DTZ == "Depth to (Move-Count) Zeroing (move)" i.e. to P-push/capture/mate

- helps guard the length of the current phase re the 50-move rule

- has a 'close relative' in the DTZ50 metric which is occasionally useful
    [ 'wins' are not wins under DTZ50 if a 50-move-draw claim is possible ]


However, you cannot just point your chess-engine at non-DTM EGTs without telling
it that they are not DTM EGTs ... or else (e.g.) they walk up to captures and
then defer the captures for one move.

So, 'moving' from DTM EGTs to DTZ EGTs requires:

a)  the temporary use of more disc-space for DTZ EGTs as well as DTM EGTs
b)  new access-code in chess-engines
c)  potentially, more sophisticated EGT-server-code to
        deal with EGTs of different metrics in an intelligent way

i.e. moving to DTZ EGTs is not just a straight swapout of DTM EGTs, hence my
comments about non-forward-compatible.

The 'move' could be backward-compatible in that new EGT-file-formats and
EGT-server-code could still deal with the current DTM EGTs in their current
format.


[ A colleague and I hope to publish on all this later this year. ]

Guy




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.