Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Ferret vs Gerbil

Author: Brian Richardson

Date: 19:35:56 03/25/02

Go up one level in this thread


On March 25, 2002 at 20:02:24, Dann Corbit wrote:

>On March 25, 2002 at 19:49:40, Tom Kerrigan wrote:
>
>>On March 25, 2002 at 12:50:57, Uri Blass wrote:
>>
>>>Gerbil is clearly more complicated than tscp and gerbil
>>>is also clearly better than tscp.
>>>Gerbil is using hash tables and ponder when tscp does not.
>>
>>Searching faster/deeper via hash tables and pondering does not necessarily make
>>it better. Last I checked, Gerbil relied entirely on piece/square tables.

Gerbil v2 has some additional eval terms per the eval.c comments below:
//	This eval function evaluates:
//
//	1.	Pawn structure, which is hashed and handled in another module.
//	2.	Piece location, which is usually related to central occupation or
//		in the case of pawns, simple advancement.
//	3.	Minors blocking d2/e2 pawns.
//	4.	King on f1/g1 with rook on g1/h1.
//	5.	Pawn shelter.
//	6.	Endgame king position values versus middlegame values.



>>It wouldn't surprise me if TSCP does enough evaluation to beat piece/square table
>>programs (unless it's getting ridiculously outsearched).
>
>From A. Herrmann's list (http://www.wbholmes.de/elo/eloceleron.txt):
>place engine       rating   games   +    =    -     pts     %  oppo  diff
>=========================================================================
>...
> 62. Gerbil R02    2141     496   167   57  272   195.5  39.4  2218   -77
>...
> 74. Tscp 1.73 *   2034     276    68   19  189    77.5  28.1  2199  -165
>...
>
>TSCP is fairly close to Gerbil.  Gerbil is a very fast searcher, so TSCP's close
>performance is rather surprising, I think.



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.