Computer Chess Club Archives


Search

Terms

Messages

Subject: I Agree

Author: Terry McCracken

Date: 12:14:56 12/04/05

Go up one level in this thread


On December 04, 2005 at 14:37:13, Robert Hyatt wrote:

>On December 04, 2005 at 03:00:01, Terry McCracken wrote:
>
>>On December 04, 2005 at 00:21:59, chandler yergin wrote:
>>
>>>On December 04, 2005 at 00:05:12, Terry McCracken wrote:
>>>
>>>>On December 03, 2005 at 23:48:36, chandler yergin wrote:
>>>>
>>>>>Ever Scroll a PV and see it change it's Eval over time?
>>>>>If yes, then that ends the discussion.
>>>>
>>>>Irrelevant.
>>>Tell that to Hyatt!
>>
>>Why? Your post is jibberish. Did Dr. Robert Hyatt spew nonsense, as you seem to
>>suggest? No!
>>
>>Chan wake up, you're arguing with the pros, you're not even a novice in chess
>>progamming.
>>
>>Why don't you listen to the best in the field? The best post here, with a few
>>exceptions. You're an orderly whose disorderly, telling neurologists how to do a
>>brain transplant!
>>
>>Terry
>
>I don't know why this is continuing.  Here is a simple example.
>
>Let's do _just_ a 5 ply search, no more.  And in that search, we find a forced
>mate in 3.  That is, a move for white, a move for black, a move for white, a
>move for black, a move for white, and now white has no legal moves and is in
>check.  We return "mate in 3".
>
>Now exactly _how_ can we do a 6 ply search and find that the mate in 3 was
>wrong, that there is really a mate in 4 or more moves and not 3?  If the mate in
>3 is there at depth=5, it is there at depth=6, 7, 8, 9, ... N, and will _never_
>go away, if it is a real forced mate.  If it isn't a real forced mate, the
>search has a bug because returning mate in 3 is an absolute, not an
>approximation.
>
>Surely this has been explained enough times that we can move on to something
>more interesting?

More than enough, I was trying to convey this to Chan.



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.