Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rebel Century users: can anyone reproduce this error (logfiles included)

Author: Ed Schröder

Date: 14:05:13 12/16/99

Go up one level in this thread


>Posted by Rafael Vasquez on December 16, 1999 at 15:40:58:
>
>I'm seriously thinking that AMD K6-3 450mhz is producing some strange errors
>with REBEL CENTURY.Want to confirm it by reproducing it in a Pentium machine.
>I put Rebel Century in player-player mode and went thru the position shown in
>the logfile. Then I turned the books off and set autoplay. The time control was
>40 in 2:30. Suddently RC declared chekmate? by two different moves Be3+ and
>Bd3+(move 10)

>--------------------------------------------------------------------------------
>I cannot believe this and put the position in analisys mode. This time it will
>not declare mate and I decided to move forward according to the PV. Now it will
>show a -4128768 eval! for position in black's 10th move.
>logfile2:
>--------------------------------------------------------------------------------

Looks like your PC is having some problems. Overheated, dust,
broken RAM who can tell. Here is a system to test your PC.

Download the following EPD file:

www.rebel.nl/rg4.epd

Run Rebel Century using "analyze EPD" (menu EXTRA)

When finished the NPS counter for every position (see the logfile
"rg4.txt") should always be 100% the same. If there are differences
(no matter how small) consider your PC as (very) suspect if not
broken.

The system is simple: "rg4.epd" is nothing more than 500 x running
the same position (on 10 ply) over and over. You can cancel the
process any time (hitting ESC) and then check the created logfile
"rg4.txt".

Ed




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.