Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Crafty and single-computer winboard matches

Author: Robert Hyatt

Date: 19:26:29 10/05/99

Go up one level in this thread


On October 05, 1999 at 20:26:59, Peter Kappler wrote:

>
>Bob,
>
>This is really a shame, because Crafty is the best known of all the winboard
>engines, and is often used as *the* standard for assessing the strength of other
>programs.  Anytime I hear that "program X" is a really strong amateur engine, my
>first thought is "well, lets just see how it stacks up against Crafty".
>
>As much as possible, I think you should try to address some of these
>single-computer winboard issues - especially the no-ponder time management
>problem, which just doesn't seem that difficult.  Problems that can't be solved
>should at least be quantified, performance-wise, and this information could be
>published.  As Christopher said in his previous post - if the net effect is 5 or
>10 ELO points, who really cares?  On the other hand, if you can prove that for
>Crafty the difference is 50 or 75 ELO points, then at least people will take
>this into account when interpreting results.
>
>--Peter


The problem is not as easy to address as you think.  The current time allocation
code was developed over _many_ months of testing, going thru log files, checking
how it was doing at various points in the game, etc.  I don't have the time to
do that again with ponder=off.  The only place to do that would be on ICC.  And
would take a _lot_ of time, while weakening crafty substantially while the
testing is going on.

In short, why test in a mode it is not designed to run in?  I added ponder=off
solely for testing, so that I can get reproducible results over several moves
when adding something new.  It wasn't designed to allow single computer games
for anything other than testing specific things that are modified...  The
easiest thing to do would be to remove the ponder=off option, as that would
definitely solve the 'problem' and would only take a few seconds of testing to
do...  to modify the timing code takes a lot of work.  Mike Byrne and I spent
months looking at this stuff and tweaking it for the various modes like the
icc inc/no inc games, normal primary/secondary time control games, etc.  Doing
it all again for ponder=off is a big waste of time, overall, when there is al-
ready way too little time to work on the things that need work...




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.