Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Is this a correct fail-soft?

Author: Zach Wegner

Date: 20:45:34 09/21/03

Go up one level in this thread


On September 21, 2003 at 21:23:09, Mikael Bäckman wrote:

>On September 21, 2003 at 20:02:11, Zach Wegner wrote:
>
>>On September 21, 2003 at 18:35:48, Dieter Buerssner wrote:
>>
>>>On September 21, 2003 at 18:21:11, Dieter Buerssner wrote:
>>>
>>>>>	nLegal = 0;
>>>>
>>>>nlegal is not declared here.
>>>
>>>Same for other variables, you are using. Either your code snippets are
>>>incomplete, or all those vars are global. I would guess the first (otherwise, I
>>>would suspect total crazy search or crashes). When you give complete code
>>>snippets, it will be easier to answer (even when the code is longer). In
>>>principle, your code looks like proper fail soft.
>>
>>all the 'junk' is stripped.
>>
>>>What is the intention of your eval func (which always returns mate scores ...)?
>>
>>thats material.
>>
>>I'm not quite sure whats going on, but i would see if it still happens if you
>>seperate
>>
>>if(score>alpha)
>>{
>>     if(score>=beta)return score;
>>     alpha=score;
>>}
>>
>>to
>>
>>if(score>=beta)return score;
>>if(score>alpha)alpha=score;
>
>Same thing isn't it? My way saves an if when score <= alpha. :)

Yes, but I can't see anything wrong with your posted code, so I think something
in your real code is causing some scores to be >= beta but <= alpha. Just a
guess. I'd like to hear if that is what's happening.

>
>
>>Regards,
>>Zach
>
>
>/Mikael



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.