Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: What is Ferret doing?

Author: Dave Gomboc

Date: 13:50:56 06/21/99

Go up one level in this thread


On June 21, 1999 at 16:40:52, Stefan Meyer-Kahlen wrote:

>On June 20, 1999 at 12:52:07, Robert Hyatt wrote:
>
>>On June 20, 1999 at 11:52:01, James T. Walker wrote:
>>
>>>On June 19, 1999 at 21:07:56, Robert Hyatt wrote:
>>>
>>>>On June 19, 1999 at 14:33:47, Ernst Walet wrote:
>>>>
>>>>>If I didn't know better I would say that Ferret has cracked under the pressure
>>>>>after move 55!  Maybe later Bruce can explain what is going on right now or has
>>>>>someone else any idea?.
>>>>>
>>>>>Ernst-J.
>>>>
>>>>
>>>>Here's what I saw:
>>>>
>>>>by move 25 or so white was not just lost, but dead lost.  I was talking with
>>>>Roman at the time and he said something like "I have _never_ seen white get
>>>>into such a dead lost position so quickly..."  But then Ferret did what many
>>>>computer programs do when they start an attack... they go for the first pawn
>>>>they can see, rather than keeping the pressure up.  I suspect that a good GM
>>>>would have taken black and ended the game within 25 more moves.  But as ferret
>>>>went for that pawn it dissipated the attack it had going and turned the
>>>>initiative over to shredder.  Shredder slowly caught up and at one point we
>>>>all had it as over +1.5 or so...  but then Ferret began to free up its pieces
>>>>and almost pulled even again...  then it let the e-pawn advance too far and
>>>>sacked the exchange to win it...  and then Shredder began to make progress
>>>>again.  But at the end, Shredder apparently overlooked the draw that most
>>>>programs saw nearly instantly... and the game ended a draw...  a neat see-saw
>>>>game if there ever was one...  both sides had multiple winning chances...
>>>
>>>In the "Shredder apparently overlooked the draw that most programs saw nearly
>>>instantly... " scenario isn't it possible that Shredder had it's contempt factor
>>>set to accept a draw sooner than normal since it still wins the Championship?
>>>Jim Walker
>>
>>I suspect not because there were other places where it could have simply
>>repeated had it wanted to...  but your explanation is definitely possible...
>>although it would have been mighty high to be over a pawn ahead and go for
>>the draw...
>
>My contempt factor was 0 because I have never tested it in a real game.
>This almost backfired as Shredder was trying to avoid a repetition in this game
>when Ferret would have happily taken it.

Uh... yikes. :-)

>Stefan

Dave



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.