Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: BS 2830 Problem #5 is wrong.

Author: Uri Blass

Date: 01:29:06 03/18/04

Go up one level in this thread


On March 18, 2004 at 03:57:49, Bas Hamstra wrote:

>>I do not think that it is confirmed that white wins.
>>Maybe white wins but Vincent is the last person that I can trust based on
>>other things that he says.
>
>Hi Uri,
>
>Since at least 1 person confirmed it, it is confirmed.
>
>But seriously, if a FM master analysed the position intensively, it's good
>enough for me. I don't trust the computers here, they can easily give +1.50 pure
>speculative.

I did test it but the way that I suggest to test is generating a tree of logical
alternatives(the tree may have something like 20 leaves when it is clear by
computer analysis that every move out of the tree is bad) and try comp-comp
games at long time control(like 1 hour per move) from every leave and after
looking at the games trying to improve the moves of black in case that black is
losing or to improve the moves of white in case that white draws and play again
comp-comp games if they are needed.


Note that I did not claim which move is better but only that Vincent's words
prove nothing and we have no convincing evidence which move is better.

Claiming that he looked at it for hours without giving specific lines to beat
programs with white after Rxe6 is no proof even if kasparov says it.

Posting some games when he beated top chess programs after Rxe6 at 1 hour per
move is more convincing evidence.

It is not important for me to spend computer time about investigating the truth
so I do not analyze it but if other care about the truth they should spend time
about it by playing comp-comp games at long time control.

Uri



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.