This is hot. Thank you.

Best deal for the money anywhere...

Bob

Vas, I've only heard very good things about 2.3.2 beta 2..People in the forum(testers) seemed to be very impressed with this engines ability to play very elegant chess..congrads for all ur patience and work..

dareapa

Hi Vas,
very much thanks (to you and your team)for this further steps. This should bring Rybka again to a higher level and sure will be a success.
But one additional question, please:
Is there any progress in dealing with more than two cores of one processor or even with more processors (for example 8 cores or more) cause I read about of poor performancegain for more than 2 cores.
Thanks in advance for giving a statement about this stuff.
Wolfram

Vas,

Was that endgame Knowledge entered before or after today's game with Zappa? :-)  Yeah! There goes that winning personality again! (Of, course, I'm only kidding (but, did you?).:-)

Thanks for the ABSOLUTELY GREAT chess engine-no kidding!!!

Robert

P.S.
Soon That great Chess Engine Will Have A Great GUI For A Home!!!

Great work, Vas. You and LK have become a fantastic team.

Human creativity is an amazing thing. Sometimes it happens when just the right individuals appear at just the right time in human intellectual endeavor, that they are able to make enormous strides, creating new value where it did not exist before.
  You and LK should not plan to retire from the scene as long as the magic is happening.

Vas, 
 
Absolutely thanks to you and all your team for this magnificent 2.3.2 engine.  
 
I strongly believe that our cooperation here inside the Forum it is also very important to continue helping a little in the Rybka improvement. I believe that the sentence: The union makes the force, it is quite certain.

I believe that on behalf of all the members here in this Forum, our thanks, truly, also for your patience and do not stop, continue with your excellent work in Rybka!  

Regards, 
Gambito.

Thanks for all of the nice comments. To answer the various questions:

1) Rybka 2.3.2 is being distributed by Convekta this morning. This can take a couple of days - if you don't get your copy within a couple of days, please send them an email (info@chessok.com).

2) Multiprocessor efficiency - increasing this is a high priority for Rybka 3. There are no improvements from 2.3.1 to 2.3.2 which are specific to the MP version.

And yes - we have a really excellent team, I could not ask for a better one.

Vas

The results in the CCRL list are:

http://www.computerchess.org.uk/ccrl/404.live/rating_list_all.html

  Glaurung 2-epsilon/4 64-bit 2CPU 3126 +211 −201 58.3% −42.0 50.0% 6  
66.4%
1 Rybka 2.3.2 Beta2 64-bit 2CPU 3084 +37 −35 77.4% −200.1 30.2% 305
60.2%
  Rybka 2.3.1 64-bit 2CPU 3078 +23 −23 80.6% −233.4 23.5% 862
78.9%

I wonder if beta2 rating is going to be improved.
I expected more information in this list after I saw in the chess computer club that  there is already new list with new movei
but it seems that when more games for movei were added no more games for rybka were added.

I decided to delete irrelevant part of the post that was based on wrong link because I clicked from the ccc on wrong link.

> - separate endgame module which attempts to estimate winning chances in king and pawn endgames

I don't find any separate endgame modules with my download, nor any mention of it in the engine options. So is it integrated into the engine automatically?

Hi,
Thanks and congratulations.
She has been an excellent programm up to now. The improvement make her more attractive.;-)
Regards
Hetman

Thanks Vas! I'm sure it's great. I'll try it out in Parallels. :)

BTW: Convekta was on time! WOW!

Hi Vas

I anwriting you because until now I have not received the link to download this new version. Yesterday I wrote to info@chessok.com, and today in the morning I have not yet an answer. I have been registrated with the mane DAVID AMIEL and my e-mail is damielp@hotmail.com. And I received without problems the former links of the Rybka 2 series

I hope you can help me because I an very anxious into test this new giant

best rgards

David Amiel

Application joins options set parameters

I have noticed that often my Rybka, currency the position in center game "draw"
+/- 0,5 p.c. and then when it enters on end, (when it begins to read the tablebases) his
evaluation clearly passes to the sudden one negative -1,5/-2,5 c.p.
I think that this is due to the fact, that begins to consult the tablebases
of the ends, too late.
According to me, to obviate to this, would be interesting to have the possibility
to manually plan the point in which the engine begins to consult the tablebases.
I have seen that in Nimzo 8, in Crafty and other engine, is possible to plan
the moment of departure of the tablebases in base to the number of present pieces on the chessboard. Or if I for example plan the parameter to 14, Nimzo it begins
to consult the tablebases from when they remain on the chessboard 14 pieces.
This would be an useful function to plan the reading of the tablebases to,
according to of the speed and characteristics of the mine PC.
In Rybka I can choose only among: rarely, normal and frequently.
But I was perhaps able choose the moment in base to the remained pieces, it would be better.
Another question:
in Nimzo it is possible to write some instructions in the file of configuration (*. config)
immediately to load in ram some tablebases.
If this were also possible with Rybka, it would surely be a more speed
of the game in final. It is not possible to insert in the file Rybka.config, (where I have written the command for the randomiz) to insert some commands to load the tablebases more used?  (type pKppK and RpKRK).
You greet to everybody

I mostly use Rybka to analyze my OTB tournament games post mortem, and this program has proven to be a tremendous resource. Rybka seems to find the best move more often and much quicker than Fritz.

Excellent product, thanks Vasik!