Computer Chess Club Archives


Search

Terms

Messages

Subject: Request, and also some RYBKA results to make people jump?

Author: Ed Murak

Date: 22:21:34 12/12/05

Go up one level in this thread


On December 12, 2005 at 14:34:30, Daniel Mehrmannn wrote:

>Hello Vasik,
>
>after some review work from some experts here,
>i must send you my apology ! Sorry about my posting about Rybka
>and congrats to a so strong engine.
>Its seems to be i was dazzled from some moves and outputs.
>
>Sorry!
>
>Best,
>Daniel


Daniel,

Always best to apologize when wrong, well done, IMO. If only all people could
follow your example.  Usually they do not.

Your kind message above has been sent to Mr Rajlich, who I am sure could be
relieved he didn't do anything wrong.  He knew that already, of course.  :)

But could I make a little suggestion, purely voluntarily on your part, which
could even be viewed as a favor or request just from me (a total nobody) and no
one else?

The original suggestion, theory or even accusation as some could have read it,
by you appeared in-
http://www.talkchess.com/forums/1/message.html?469130
The best way to retract, IMO, is for you to post a _reply_ to that i.e. to your
own original message, with a SHOUTING subject:, something like:
 "SORRY, I MADE A MISTAKE! PLEASE IGNORE THIS THREAD"
Daniel, this is only my suggestion... you can do it or not. As a programmer
yourself, you can imagine that it can be hurtful to read incorrect things about
one's work. :)

I have got very good news for you and everyone.  In my play of the latest Rybka
test, it is leading (at 40/40+40/40+40, opening book Fritz9.ctg for both, modern
PC) against Fritz 9 by a huge margin!  Right now, game 12 is being played (each
opening is played twice with the usual reversal), and the score to RYBKA is
+8 =2 -1 which is almost unbelievable.  That is 81.8%, +261 ELO T.P.
The RYBKA is Beta version latest, released versions will be much better because
of endgame knowledge being added and EGTB support and other things.  Fritz 9 was
playing with full 3/4/5 EGTBs, even many 6 ones but I forget whether it looks at
those.
Of course the match is too short, but 81.8%... eh?
I may post in CCC the interim results and of course all the PGN when more games
are played.

Thanks for your time.



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.