Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A fact about table base usage

Author: Robert Hyatt

Date: 12:32:57 08/16/99

Go up one level in this thread


On August 14, 1999 at 14:05:10, James T. Walker wrote:

>On August 14, 1999 at 11:28:59, Marc van Hal wrote:
>
>>I actualy did think about the previous post i made and Roberts answer where he
>>showed me where the table bases where used
>>Later on i though but in fact the only programs who realy take advance of this
>>is actualy the faster engines cause more is to be found in the tablebases
>>So hiarc7.32 and Chess system Tal do not have as much advantage of it as Crafty
>>is this treu or am i totaly wrong here.
>
>Hello Marc,
>Maybe you can explain what you mean by "faster engines".  I have seen a position
>posted here where Hiarcs 7.32 found the tablebase move when Crafty did not.  I
>think the "faster engines" are the one that find the best move in the least
>amount of time and it has very little to do with nodes per second.
>Jim Walker

Could you post the position?  I'll try it.  The only way I can see that this
might happen is in a position where Hiarcs extends a lot more than I do (and it
does this a lot, good or bad).  As if it goes deep enough to see a trade to a
5 piece file before crafty can, yes it will find the right move.  But in
endgames, extensions are hard to come by (few checks, etc.)

I've had a whole lot longer to tune the tb probing algorithm and deciding when/
where/how-deep to do it...  I shouldn't get 'skunked' by anybody I'd hope...




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.