Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Fritz' blundercheck bug?

Author: Steve Coladonato

Date: 05:38:01 03/14/00

Go up one level in this thread


Boris,

Blundercheck analyzes a game and looks for "blunders" based on the threshold
setting you supply.  That setting is used as the delta in the evaluation to
identify a "blunder".

Is your threshold setting 0.15 or 15?  The threshold is measured in hundredths
of a pawn so that a threshold setting of 15 would cause Fritz to make a comment
if the move it picked was 15/100 better than the move played.  The output does
look like you had it set at 15.  Also, 10s/move is fairly quick although you are
getting to 10 ply in the evaluation. (What's your CPU speed?)  The game you
included didn't show the variation Fritz picked for 26.Ba4.  I would suggest
that you load the game and turn analysis on and open up the analysis window to
show multiple variations for this position.  I have encountered some positions
where the engine changed its evaluation of a move for a given position and the
variation reported was truncated.

Steve

On March 14, 2000 at 05:58:29, Boris Burrakowski wrote:

>Hi!
>Can someone please explain to me what goes on in Fritz blundercheck.
>This is a game I played at ICC the other day. Timelimit 3 0.
>Blundercheck settings 10 s/move, threshold 0.15. Side to analyze white
>
>
>[Fritz 6a]
>

<snip>



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.