Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: DCCC 25th : a few comments from a Fruit's friend.

Author: Ryan B.

Date: 02:57:57 11/13/05

Go up one level in this thread


On November 13, 2005 at 05:44:58, Uri Blass wrote:

>On November 13, 2005 at 04:57:19, Marc Lacrosse wrote:
>
>>
>>>                  Score       SB   1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6
>>>-------------------------------------------------------------------
>>> 1: Zappa        5.0 /  6   17.00  X = = . . 1 1 1 . . . 1 . . . .
>>> 2: Diep         4.5 /  6   16.00  = X 1 0 . . 1 1 . . 1 . . . . .
>>> 3: The King     4.5 /  6   12.00  = 0 X . 1 . . 1 1 . . . . . . 1
>>> 4: Ktulu        4.0 /  6   11.75  . 1 . X = . . 0 = 1 . . . . 1 .
>>> 5: Deep Sjeng   3.5 /  6   10.50  . . 0 = X = . . . 1 = 1 . . . .
>>> 6: Gandalf      3.5 /  6    9.25  0 . . . = X . . = . 1 . 1 = . .
>>> 7: Fruit        3.5 /  6    9.25  0 0 . . . . X . 1 1 . = . 1 . .
>>> 8: Tao          3.0 /  6    8.50  0 0 0 1 . . . X . . . 1 . 1 . .
>>> 9: Pro Deo      3.0 /  6    6.75  . . 0 = . = 0 . X . . . . 1 1 .
>>>10: Homer        3.0 /  6    6.00  . . . 0 0 . 0 . . X 1 . 1 . 1 .
>>>11: Neurosis     2.5 /  6    4.25  . 0 . . = 0 . . . 0 X . 1 . . 1
>>>12: XiniX        2.5 /  6    2.75  0 . . . 0 . = 0 . . . X . . 1 1
>>>13: Zzzzzz       2.5 /  6    2.00  . . . . . 0 . . . 0 0 . X = 1 1
>>>14: IsiChess     2.0 /  6    3.00  . . . . . = 0 0 0 . . . = X . 1
>>>15: Ayito        1.0 /  6    0.00  . . . 0 . . . . 0 0 . 0 0 . X 1
>>>16: Usurpator    0.0 /  6    0.00  . . 0 . . . . . . . 0 0 0 0 0 X
>>
>>A few comments on the present state of affairs at the 25th Dutch computer chess
>>open championship, mainly from a fruit-friend point-of-view.
>>
>>1.There is only one surprise in this tournament so far : the relatively
>>disastrous results achieved by Fruit.
>
>I disagree
>I do not think that it is a surprise.
>
> Other results seem quite normal with a few
>>exceptions (ProDeo could be a bit higher and Tao lower in the list).
>>But since Fruit 2.1 has been released, never did Fruit (2.1, wccc or 2.2)
>>achieve such a disappointing result in any official public appearance.
>
>
>It means that fruit was extremely lucky in other tournaments or had a clearly
>better book that the opponents.
>
>With all respect to fruit
>Fruit is not better than Deep Shredder with 4 processors but it scored better in
>WCCC.
>
>It can be or because of luck or because you had better book than sandro.
>
>>At WCCC Fruit had finished 2nd with 77.3% against a stronger field.
>>At CCTblitz Fruit finished equal first with 83.3%.
>>Even playing with a non-tuned book, Fruit is presently leading the field at WBEC
>>riederkerk with 68.3 %.
>>Here Fruit has a miserable 58.3%¨performance so far and it has lost as many
>>half-points in 6 games than in the whole stronger 11-games event in Reijkjavik.
>
>Fruit lost against the same opponents that it lost in WCCC.
>
>>OK Fruit has already played (and lost) against the two strongest opponents but
>>it still has to meet The King, Ktulu, Deep Sjeng and/or Gandalf. If Fruit cannot
>>manage to win its three remaining games this tournament will be a real black
>>stone in Fruit's garden.
>
>I do not expect fruit to win the remaining games and I think 2/3 or 2.5/3 is not
>a disaster for fruit.
>
>I see no reason to expect fruit to win the tournament when it supports only one
>processor.
>

I agree, even if fruit is 50 elo stronger than Zappa or Shredder on one cpu good
use of the extra 3 cpus would overcome the advantage (maybe +25 elo per cpu)

>If we talk about the opponents they were able to learn from fruit open source so
>they may be even stronger than the opponents in WCCC.
>
>Uri

I agree again, it does not take long to learn by reading Fruits eval code.  If
people could read Zappas or Shredders eval functions we would have a better
chance of beating them or tuning a program to beat them.



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.