Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Kill this auto232-players

Author: Ulrich Tuerke

Date: 04:09:55 04/30/98

Go up one level in this thread


On April 30, 1998 at 05:49:49, Peter Herttrich wrote:
>
>Ok, then why not use the Fisher-System and add 3 seconds per move?
>

I am afraid that will not help either. If you add 3 seconds per move,
the programs will use them (unless you don't tell them to use the
Fischer clock).

>
>ooooooops, You will tell me, such programs exist after so much
>experience and time?

For sure. Not long ago, a game between HIARCS-6 and COMET was mailed to
me, where HIARCS announced a draw by repetition, and COMET didn't. I
checked that Comet was right. This is in fact a little bit tricky,
because most programs terminate the search inside the tree when
detecting the 1st repetition and return a value of 0 (just to save a lot
of nodes) to the parent node. But at the root, you must handle the
situation differently, and go on until the second repetition occurs. I
guess that something like this confused HIARCS.
You see that even top programs are not free of bugs.

>
>I think, if all programs are implemented well and the 3-seconds-per-move
>add-on is used, the time-difference will be really small.
>
>
>To use an extra-machine (can FICS this do?) sounds for me, to shoot
>with cannons for sparrows.

In principle I agree. I could imagine solutions not being perfect, but
very well suited to work. But then there is still some space for debates
and doubts.

>
>
>
>BTW: Of course I know the xboard, I use crafty with xboard under LINUX !
>
>Use UNIX or die    :-)
>
>Where is the comet-version for LINUX ?  ;-)

I have a well working version for DIGITAL UNIX. Roland Pfister has just
generated a version for WIN-95 (winboard). In case you are interested, I
would email the files to you.

Regards, Uli



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.