Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rybka blitz results slightly disturbing

Author: Mike Byrne

Date: 16:30:51 12/31/05

Go up one level in this thread


On December 31, 2005 at 13:53:12, James T. Walker wrote:

>On December 31, 2005 at 12:05:01, Mike Byrne wrote:
>
>>On December 31, 2005 at 11:38:01, Norm Pollock wrote:
>>
>>>On December 31, 2005 at 10:03:42, Torstein Hall wrote:
>> The
>>>"Beta" version of Rybka gets better results, and the "proof" is that the vast
>>>majority of engines on playchess use the "Beta".
>>
>>haha  -- That is proof.   If you want to see how people feel about something -
>>see which side they take they put their money on the line ...or in this case
>>rating points!
>
>Sorry but I don't follow your comments.  I suspect that more people use the Beta
>version on Playchess because it was obtained free and more people have it.  But
>with more comments like mine above it is a disturbing trend to see the Elo
>dropping with "newer" versions.  I guess if it's only affecting the blitz
>ratings that's probably acceptable to the author.  He seems more interested in
>creating an analysis tool which works better at long time controls.  If this
>trend continues, I will be stuck with a beta version that has no opening book
>and very poor endgame play since I'm only interested in blitz games.

Ignore my comment -- I was just joking about the rating points.  I personally
see little difference between the two versions -- althought there is a
difference.  I don't think preview is weaker than beta.



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.