Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Goliath Light 1.5

Author: Mike Hood

Date: 04:40:31 02/22/02

Go up one level in this thread


On February 22, 2002 at 07:21:01, Bertil Eklund wrote:

>On February 22, 2002 at 07:15:14, Mike Hood wrote:
>
>>Has anybody done any serious testing with Goliath Light 1.5? I assume the
>>program tested by the SSDF (rated 32nd in the current list) is version 1.2.
>>After running a series of engine tournaments it performs better than any of the
>>other free Chessbase engines, and it's not far behind the likes of Fritz 7,
>>Shredder 6 and the Tigers. Of course, I haven't played enough games to have any
>>statistical relevance -- and I'm only playing on a single computer -- but I'm
>>very impressed. I hope the SSDF will test this version (or the newer Winboard
>>version?)
>>
>>P.S. It's downloadable at
>>http://www.chessbase.com/download/index.asp?cat=Engines
>
>Hi!
>
>I have played a few houndred games with it and I believe the improvement vs
>Goliath Light is very small, in example (clearly) weaker then Crafty, Yace or
>SOS.
>I think it must take a clear step forward before it is time to incorporate it in
>the list.
>
>Bertil

Maybe this is an example of a program which performs better in a single-computer
environment (with no pondering) than in a two-computer environment. An opposite
example is Nimzo (both versions 7.32 and 8.0) which does so badly in engine
tournaments that I have difficulty believing its high position in the SSDF list.
In my typical setting of 40'/40+40'/40+40' Nimzo usually uses too much time in
the early moves, and by the end of its time controls it has about a minute left
for 10 moves. I presume this doesn't happen when it has the opponent's thinking
time available.




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.