Author: Kurt Utzinger
Date: 05:41:34 11/11/05
Go up one level in this thread
On November 11, 2005 at 07:50:24, Uri Blass wrote:
>On November 11, 2005 at 07:40:42, Kurt Utzinger wrote:
>
>>On November 11, 2005 at 07:33:15, Jouni Uski wrote:
>>
>>>This is stunning, when other testers have reported normally Fruit - Junior
>>>to about 70 - 30.
>>>
>>>Jouni
>>
>> Hi Jouni
>> Not necessarily: you know SSDF are testing
>> the whole package with own books, learning on,
>> ponder=on and thus partly not the strength of
>> the engines but more the goodness of the opening
>> books ... and Junior 9 is one of the engines to
>> benefit more than other programs from its rather
>> limited opening book repeating won lines many
>> times in a match thus sometimes producing doubles.
>> Regards
>> Kurt [http://www.utzingerk.com]
>
>I did not know that opening book help Junior so much.
>
>I knew that fruit performed clearly better than Junior also when own books were
>used
>
>see http://www.talkchess.com/forums/1/message.html?460698
>
>I did not follow the games to check for doubles but I remember matches when
>Fruit beated Fritz9 or Shredder9 convincingly when original opening book were
>used and the result of the ssdf are surprising.
>
>Maybe Fruit does not like the athlon1200 and it may be interesting to know if it
>earns more speed from faster hardware relative to Fritz or Junior or Shredder.
>
>Uri
Hi Uri
Some further explanations: still too less
games / Junior 9 benefits more than Fruit
from ponder=on / pure accident
Regards
Kurt
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.