Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: petition request for Andrew Williams

Author: Duncan Roberts

Date: 12:46:37 01/08/05

Go up one level in this thread


On January 07, 2005 at 10:37:47, Vincent Diepeveen wrote:

>On January 07, 2005 at 10:33:39, Duncan Roberts wrote:
>
>I'm under the impression you still do not realize how stupid a material only
>search is.

a material only search is not designed to find good moves, and it is also not
designed to tell you if it does find this forced material loss whether   it is a
win for white, as black could find compensation (either material or positional.)
it IS designed to find (if it exists) material loss.


it has according to dann I believe a more than likely chance even with null move
of doing this.

duncan


>
>>On January 07, 2005 at 10:02:15, Vincent Diepeveen wrote:
>>
>>>On January 07, 2005 at 08:12:52, Duncan Roberts wrote:
>>>
>>>If there would be material win for white within say 60 plies of search from
>>>openings position then:
>>>  a) it would have been found already if winning for white by GM's and
>>>especially by corr players.
>>
>>as only a small percentage of the positions possible withion 60 plies have been
>>analysed by gms' it is by no means impossible that they have missed something
>>
>>
>> Hell, even Uri Blass with his autistic logics could
>>>figure it out.
>>>  b) when not winning for white black has some real great compensation
>>>     perhaps black has a mate in 2 or something which my evaluation sees
>>>     tens of plies in advance but your 60 ply beancounter will miss it
>>>  c) a 60 ply material only search will reveal nothing significant as the
>>>position is only evaluated by material.
>>
>>well if there is a pawn to be taken in 60 plys, there is a good chance that the
>>60 ply search will reveal it. It is useful to know that. black may or may not
>>have compensation, but if you do not try you will not know.
>>
>>>Who knows, perhaps white has a passer on
>>>the 7th row already promoting in 1 step.
>>
>>yes a 60 ply material search will not reveal a material gain on 61 ply.
>>
>>and similarly a 34 ply material search will not reveal a material gain on 35
>>ply.
>>
>>the solution is NOT to not bother trying to go for a 34 ply search, but to first
>>try to get to the 34 ply search and then later as resources allow to get to the
>>35 ply search.
>>
>>
>>Rome was not built in  a day
>>
>>duncan
>>
>>
>>
>>
>>>
>>>Are you sure that you didn't realize all this yet and the other 100 reasons why
>>>it will contribute nothing to science.
>>>
>>>Oh i realize again why i hardly read any posting on this forum... ...it's just
>>>such a waste of time. People at kids level...
>>
>>I appreciate you taking the time to share your vast computer chess knowledge
>>with this forum.
>>
>>
>>duncan



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.