Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rebel8 and Rebel9 on SSDF (explained)

Author: Albert Silver

Date: 06:20:29 12/16/97

Go up one level in this thread


On December 16, 1997 at 05:07:08, Jouni Uski wrote:

>>
>>Thoralf reported that after removing the Rebel-Rebel matches in
>>question R9-200MMX would go down 15 points and the R9-P90 would go
>>up 17 points.
>>
>Heh, heh! Correction is now not so important, when it actually LOWERS
>R9 200 MMX rating. IMHO games with wrong book should immediately
>removed.
>
>But so after one year work Rebel team hasn't beaten amateur (Hiarcs6
>team)!

Agreed, but Ed correctly pointed out that one also doesn't know what the
result would have been had they played against each other with the
proper books. For my part, I find both programs to be extremely
impressive. I have a few tests of my own, and one of them consists of a
proposed rook exchange that would lead to a won pawn endgame. Rebel and
Hiarcs are the only programs (not using any endgame tablebase) that go
for the exchange AND see that it is completely won BEFORE actually
making the exchange. Frankly it's hard to say which is better as they do
play somewhat differently.



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.