Search found 390 matches

by ozziejoe
Sun Jun 03, 2007 10:03 pm
Forum: Computer Chess Club: Tournaments and Matches
Topic: rating list...strategic openign repertoire
Replies: 1
Views: 972

Re: rating list...strategic openign repertoire

Interestingly, in two of the standard caro kann lines (see below), the computers just do not seem to be any good at the black side. When white plays bc4, black only wins 27% When white plays h4, black only wins 37% In contrast, based on the huge base tree (chessassistant), humans playing black win 4...
by ozziejoe
Sun Jun 03, 2007 9:42 pm
Forum: Computer Chess Club: Tournaments and Matches
Topic: rating list...strategic openign repertoire
Replies: 1
Views: 972

rating list...strategic openign repertoire

I've added many more games. details of tourni are at bottom. It looks like toga 1.3x4 is an absolute beast in this kinds of positions. spike 1.2 was a little disappointing but it is a strong group, and strelka is probably as strong as shredder 10 single cpu Program Elo + - Games Score Av.Op. Draws 1...
by ozziejoe
Sat Jun 02, 2007 10:15 pm
Forum: Computer Chess Club: General Topics
Topic: chess assistant 9 question
Replies: 0
Views: 747

chess assistant 9 question

hi folks, I am trying to analyze a position using chess assistant and rybka, and I want the engine to foucs its analysis only on certain lines if i recall correctly, this is done under "analyze current position". For some reason when I view the position (whether in the tree or an actual game), the "...
by ozziejoe
Fri Jun 01, 2007 6:17 am
Forum: Computer Chess Club: General Topics
Topic: Open Letter & $100,000 challenge to FIDE
Replies: 85
Views: 41531

Re: Open Letter & $100,000 challenge to FIDE

>>>i will say this much about Rajlich's calling of the "party faithful" to arms >>>>on his financial behalf Steve, I am not sure this is entirely fair. Why can't people be enthusiastic about an engine, and not be labeled as "party faithful" (which implies a kind of brainwashed obedience). I think ta...
by ozziejoe
Sun May 27, 2007 11:56 am
Forum: Computer Chess Club: Tournaments and Matches
Topic: strategic opening rating list
Replies: 3
Views: 1886

Re: strategic opening rating list

rats ,

can someone show me how to convert epd notation to an image
by ozziejoe
Sun May 27, 2007 11:54 am
Forum: Computer Chess Club: Tournaments and Matches
Topic: strategic opening rating list
Replies: 3
Views: 1886

Re: strategic opening rating list

Interestingly, in the below position, chess assistant ratings suggest c5 followed by b4 (reverse benko gambit) is =, whereas nc6 is advantage for black Yet...black scores 58 % when playing c5 (and faces reverse benko), and only 44% when playing nc6 (based on 42 games). I know this is small numbers a...
by ozziejoe
Sun May 27, 2007 11:39 am
Forum: Computer Chess Club: Tournaments and Matches
Topic: strategic opening rating list
Replies: 3
Views: 1886

strategic opening rating list

Hi folks, I have decided to form a rating list based on a strategic opening test set. Opening test set consists of 31 openings taken from ----white strategic opening repertoire by im John Donaldson. Based on nf3, c4, g3 setup for white) -----Roman Dzindzi. 's nimzo and bogo indian recomendations for...
by ozziejoe
Sun May 20, 2007 2:27 am
Forum: Computer Chess Club: General Topics
Topic: estimating error bars in engine tournis-a problem sometimes?
Replies: 0
Views: 717

estimating error bars in engine tournis-a problem sometimes?

According to my understanding, the error bars are based on number of games and draw rate. Engines are said to differ if their error bars do not overlap I am wondering if this is an excessively conservative way of doign things, in some circumstances. Consider the following two scenarios (Assume draw ...
by ozziejoe
Thu May 17, 2007 1:26 am
Forum: Computer Chess Club: General Topics
Topic: can we estimate strength of an engine given time handicap?
Replies: 11
Views: 5169

Re: can we estimate strength of an engine given time handica

Thanks Christofhe, I took the liberty of applying your formula to various speed increases new_elo = old_elo + 70 * log ( new_speed / old_speed ) / log ( 2 ) I used log 10, if that is ok? Assumptions: STart ELO of 3000 at 180 minutes against same oponent also taking 180 minutes, 70 pnt decrement esti...