Author: Dann Corbit
Date: 00:08:59 07/29/03
Go up one level in this thread
On July 29, 2003 at 03:07:18, Dann Corbit wrote:
>On July 29, 2003 at 02:52:43, Russell Reagan wrote:
>
>>On July 29, 2003 at 00:28:46, Dann Corbit wrote:
>>
>>>If the position is not in the hash table anywhere (ignoring depth), no need to
>>>search -- you have not seen it yet.
>>
>>Hi Dann,
>>
>>I'm confused :-/
>>
>>What do you mean by "ignoring depth"?
>
>Well, you are right. For a repetition hash table no need to bother storing any
>depth information. It's irrelevant.
>
>>
>>Isn't it possible for a position to not be in the hash table and still have
>>occured in the game (ex. if the position was overwritten)?
>
>Sure. But it's rare. If you are going to use a hash table to detect
>repetitions you will have to:
>1. Live with an occasional blunder due to overwrite
{aside}
Since there are only 100 of them, max, you should definitely store the whole 64
bits as the key.
>OR
>2. Make some rule like "NEVER overwrite a position tagged as pv"[*]
>OR
>3. Chain the darn thing
>OR some such.
>
>I suspect most people do #1.
>
>[*] implies that the hash table stores at least one bit to say a position is in
>the pv {and pv's do change...}
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.