Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Feng-Hsiung Hsu's talk at Microsoft

Author: Vincent Diepeveen

Date: 04:39:07 10/09/02

Go up one level in this thread


On October 09, 2002 at 02:30:04, Gian-Carlo Pascutto wrote:

>On October 08, 2002 at 14:42:33, Robert Hyatt wrote:
>
>>Vincent, they are the _same_.
>
>In theory, perhaps yes. In practise, you'd have hard time getting one
>to work as the other.

But we all know bob can make perfectly fine evaluations to
select which moves to try, don't we? :)))))

>>>Apart from that, SE simply doesn't catch all those moves at all. instead
>>>it sees things already when it's too late. the only reason i use SE is
>>>to solve tactical tricks from the testsets quicker and because i already
>>>search not much deeper than 10 to 11 plies. Above that SE is too expensive.
>>
>>If you don't believe it helps in games, why do it?
>>
>>wait, don't answer that...
>
>He already did in the paragraph you quoted.

yes for testsets, i don't want to look too weak at testsets.
But after having it turned on for 1.5 years now and experimented with
several singular extensions i can assure you that it gives positional
shit.

>--
>GCP



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.