Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: About debugging HashTables implementation .

Author: Ralf Elvsén

Date: 23:32:27 04/07/02

Go up one level in this thread


On April 07, 2002 at 15:09:21, Sune Fischer wrote:

>On April 07, 2002 at 14:09:04, John Marountas wrote:
>
>>Hello !
>>
>>I have implemented HashTables (at Pyotr Club Edition) using the Zobrist key but
>>Pyotr randomly makes unexplained blunders.
>>
>>If I disable HashTables then Pyotr plays other moves.
>>
>>It seems that I have made  a wrong implementation. But how can I prove it ?
>>I cannot find a bug in my code. Is there any test suite which can help me?
>>
>>Any suggestion welcome !
>
>I have made a debug routine that generates the key from the ground up at every
>node, just to verify that is matches the incrementally constructed one.

I subscribe to the same philosophy. It's no fun to debug your hash
probing code if you're uncertain whether your hashkey is correct or not.

It is worth the "trouble".

Ralf

>
>Also be sure you mask it all into the key, castle rights, ep and side to move.
>
>-S.
>
>
>>This is the problem which delays the release of the Club Edition ! :-(
>>
>>Thanks,
>>Marountas John



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.