Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Anyone for an 'increasing expectation' heuristic?

Author: Guido Schimmels

Date: 07:57:45 05/04/99

Go up one level in this thread


On May 03, 1999 at 17:16:41, Will Singleton wrote:

>Hmm.. by "untouched", do you mean +- inf?  Or rather, do you keep the window
>narrow around the expected result?  Because if, for example, there's only one
>good move, then moves 2-n will fail low, and you won't get the score back.

About those implementation details only Chessbase can know.
I can't answer how they do it, but ...

...setting the window to +-inf is the easiest way to do.
You could have an individual aspiration window for each of the first n moves,
that will make things a bit tricky, though - I don't think they do it this way.

>And is it pvs, mtd, or ?
>Will

I guess Fritz uses normal aspiration search or pvs.

PS:
There was an exhibition match where Fritz ran on a dual processor machine
provided by the sponsor (Siemens ?). Especially for that purpose a mtd version
was programmed - but it was only 30% faster than the single cpu version.

- Guido -



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.