Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: The "same threat extension" as effective way to resolve horizon prob

Author: Tord Romstad

Date: 10:03:16 10/01/03

Go up one level in this thread


On October 01, 2003 at 10:02:25, Sergei S. Markoff wrote:

>Hello!
>
>>I just posted a message almost identical to my current one on the Winboard
>>forum, but I guess CCC is a better place to discuss this.  Is the pseudo-code
>>below a correct implementation of your idea?
>
>I have two corrections.

OK, thanks for the corrections!

I am experimenting with the idea right now.  In tactical positions, my
program needs 10-20% more time to reach the same search depth when the
Markoff-Botvinnik extension is enabled.  On the other hand, some tactics
are found one ply earlier.  Whether it is effective in real games remains
to be seen; I am playing a match between a version with the extension and
a version without at the moment.

In my implementation, I only extend when the threat move represents a
rather serious threat (at least two pawns).  In positions where the
extension applies, I add half a play to the search depth for all moves.
This is in addition to all other extensions, but I take care that no
move is extended by more than one ply.

Tord



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.