Author: Lanny DiBartolomeo
Date: 12:40:51 08/10/98
Go up one level in this thread
On August 09, 1998 at 23:04:23, Shaun Graham wrote:
>
>>>from whence:
>>>
>>>"the so-called "Walter-Pilz-settings" [W-P] (Att./Def. -4, Select. 10, Mat.Pos.
>>>4, Mob. 103, King Saf. 170, Pawn Weakn. 125), plus my special piece values
>>>[SPV](Queen 10, Rook 5.5, Bishop 3.5, Knight 3.3 - similar to MChess 7 which has
>>>Knight=3.25). Standard opening book. [Supposedly stronger than CM 5500.]"
>>>
>>>1. What statistical evidence is there to support the contention that these
>>>settings produce better results than the default? While I realise that it is
>>>play against a variety of programs run on other computers that is the key here,
>>>I report my own findings of W-P + SPV played in auto-tournaments against the
>>>standard personality... The standard wins, both at fast games and at less fast
>>>games.
>>>
>>>2. How were these settings arrived at? :-)
>>>
>>>Kind regards
>>>
>>>fca
>>
>>I cannot figure how the settings didnt win against the default setting of
>>chessmaster, just the selective search=10 should have better chances than the
>>default of 6. The chess genius5 selective search default is 12 if set to
>>11,10,9, 8.. this makes it weaker. Mabey try just selective search=10 .
>>Thanks,
>> Lanny :)
>
>
>I have never heard of anyone altering the selective search i think that's a
>definite no no. In my post what i stated is that the chessmaster 5555 / walter
>pilz settings only slightly increases the strength of the default chessmaster.
>The real strength increase that chessmaster has, has primarily come from it's
>faster hardware. In fact that's were this whole chessmaster thread started.
>Freddie mentioned fritz and rebel on a PII 400, and i said chessmaster would
>beat either of them on that hardware. The ssdf only has tested chessmaster
>5000, on a P90. The p90 is not sufficient to get the full strength of the
>program. What is being ascerted here is that chessmaster benefits far more and
>to a much greater extent from faster hardware than other programs. Indeed the
>statistics of all tournaments i have seen has beared this to be true. Example
>Lubke's web page, Komputer Kup, some other tourneys listed on the gambitsoft
>page, and also my own personal experiences playing chessmaster against rebel 9,
>fritz 5, and genius. The pilz settings etc., the only thing at present to be
>said about them, is that they are neither significantly stronger or weaker than
>the default. My experience does dictate that they do increase it's positional
>abilty though slightly(especially concerning king safety, and the bishop pair).
in the walter/pilz settings the selective does get set to 10. When you play
40moves n 2hrs the selective search=10 plays stronger im not sure about blitz or
10min games.
Thanks
Lanny :)
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.