Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: CCT4 Tao/Quark

Author: David Rasmussen

Date: 01:50:11 01/27/02

Go up one level in this thread


On January 26, 2002 at 20:40:54, Thomas Mayer wrote:

>
>will since the last release version there were several changes... I think
>current Quark should be about 50 ELO better then v1.50... not in blitz - seems
>to be weaker in blitz then v1.50... but not tested well enough - Quark was not
>playing much games lately, just testpositions and some games from several
>starting positions... To be honest, I think I myself am the most amazed person
>about it's performance so far... But I am sure tomorrow evening I will be back
>to reality with Quark somewhere in the middle field... remember: this are only
>one game all the time... Any engine that is above 2300 can win against Shredder
>a game... not a series, but a game... and besides - Quark did loose badly
>against Yace, was already lost against Monsoon but survives and lost the hard
>way against Zarkov... So not everything is shining... but today this was the day
>of my live, you can be sure... don't know if this will happen again once... (But
>okay, there was Leiden 2001 - Quark crushes at the same day "The King" and
>"SOS", so the potential is maybe there since a long time... and today seems to
>be eastern and christmas at the same day for Quark... :)
>
>>Have a nice day!
>oh yeah, I had it ... :)
>
>Greets, Thomas

:) Congratulations, Thomas!

I think people here have suddenly forgotten about statistical significance
somehow. I am sure that Quark is a very strong program, and my intuition tells
me that it is probably stronger than the lot of ~2300 amateur engines. But
almost nothing conclusive can be said from any programs CCT4 results. If we look
at the standings now, then there is some sense to it. The 1800~2100 rated
engines are at the bottom, and at the top there are strong commercials and
strong amateurs. But there are also many many programs that are "out of place".
9 games as statistical material, is virtually nothing.
That said, it is an achievement to beat such strong programs. That means you
must be doing something not totally wrong :) It tells me more than anything,
that of Quark isn't as strong as it's CCT4 result so far hints, it can be with
time. I am just saying it is amazing what conclusions people are willing to draw
on the basis of 9 games.

Congratulations again, Thomas! I hope Quark and Chezzz don't get paired today :)

/David



This page took 0.01 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.