TCEC groupe stage

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
JJJ
Posts: 1285
Joined: Sat Apr 19, 2014 11:47 am

TCEC groupe stage

Post by JJJ » Wed Apr 27, 2016 3:56 pm

http://www.chessdom.com/tcec-season-9-g ... ticipants/

It seems that Stockfish will play the "easy" group.

On facebook, they says to stay tuned because of particular case of Houdini and Rybka, any idea about what they mean ?

User avatar
Graham Banks
Posts: 32997
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: TCEC groupe stage

Post by Graham Banks » Wed Apr 27, 2016 9:29 pm

JJJ wrote:http://www.chessdom.com/tcec-season-9-g ... ticipants/

It seems that Stockfish will play the "easy" group.

On facebook, they says to stay tuned because of particular case of Houdini and Rybka, any idea about what they mean ?
Houdini 5 is expected to replace Houdini 4 at some stage.

Fritz 15 could replace Rybka 4.1 at some stage.
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz

JJJ
Posts: 1285
Joined: Sat Apr 19, 2014 11:47 am

Re: TCEC groupe stage

Post by JJJ » Thu Apr 28, 2016 7:08 am

Ah, but nothing new with Houdini.
Ok for Rybka, nice !

User avatar
Ajedrecista
Posts: 1397
Joined: Wed Jul 13, 2011 7:04 pm
Location: Madrid, Spain.
Contact:

Re: TCEC Season 9, group 1 stage.

Post by Ajedrecista » Thu Apr 28, 2016 8:25 am

Hello Jean:
JJJ wrote:http://www.chessdom.com/tcec-season-9-g ... ticipants/

It seems that Stockfish will play the "easy" group.
I do not think so. I would have done the pairings the same way: given the full rating list of participants, group 1A has all the odd ranking places (1st, 3rd, 5th, ..., 29th and 31st) and group 1B has all the even ranking places (2nd, 4th, 6th, ..., 30th and 32th).

Regarding ratings and averages, everyone must expect a higher rating average in group 1A due to the chosen pairing method. But how much? If I am not wrong:

Code: Select all

<Rating &#40;1A&#41;> = 2978.75   Elo.
<Rating &#40;1B&#41;> = 2940.0625 Elo.
Without including the leaders of both groups (Komodo and Stockfish):

Code: Select all

<Rating &#40;1A&#41; | without Komodo>    ~ 2953.47 Elo.
<Rating &#40;1B&#41; | without Stockfish> = 2913.4  Elo.
Furthermore, if you look at the a priori most difficult engines for the group ranking leaders (let's say less than 400 Elo of difference), then there are nine engines in each group (number 2 to number 10), and their average ratings are:

Code: Select all

R_A = <Rating &#40;1A&#41; | &#123;without Komodo,    rating&#40;i&#41; > 2958&#125; ~ 3115.3 Elo.
R_B = <Rating &#40;1B&#41; | &#123;without Stockfish, rating&#40;i&#41; > 2940&#125; ~ 3093.7 Elo.

Rating &#40;Komodo&#41;    - R_A ~ 242.7 Elo.
Rating &#40;Stockfish&#41; - R_B ~ 246.3 Elo.
I would not call that a significant difference.

Regards from Spain.

Ajedrecista.

User avatar
cdani
Posts: 2104
Joined: Sat Jan 18, 2014 9:24 am
Location: Andorra
Contact:

Re: TCEC Season 9, group 1 stage.

Post by cdani » Thu Apr 28, 2016 12:08 pm

I hope the version of Andscacs running on TCEC will be like 40 elo better or more than the 3150 announced; that was of 0.85. Also I know that Nirvana will be stronger, and probably other engines in every group. So is really difficult to tell which group is stronger.

JJJ
Posts: 1285
Joined: Sat Apr 19, 2014 11:47 am

Re: TCEC Season 9, group 1 stage.

Post by JJJ » Thu Apr 28, 2016 1:33 pm

Fire could be stronger that last time where it played less good than usual. But I don't have any news about it.
A strong update of Fire would make it close to Houdini 4.

Werewolf
Posts: 1189
Joined: Thu Sep 18, 2008 8:24 pm

Re: TCEC groupe stage

Post by Werewolf » Thu Apr 28, 2016 8:24 pm

Graham Banks wrote:
Fritz 15 could replace Rybka 4.1 at some stage.
Why couldn't Rybka just be the tournament name for what is actually Fritz 15.1?

User avatar
lucasart
Posts: 3037
Joined: Mon May 31, 2010 11:29 am
Full name: lucasart
Contact:

Re: TCEC groupe stage

Post by lucasart » Sat Apr 30, 2016 1:36 am

JJJ wrote:http://www.chessdom.com/tcec-season-9-g ... ticipants/

It seems that Stockfish will play the "easy" group.

On facebook, they says to stay tuned because of particular case of Houdini and Rybka, any idea about what they mean ?
How did FireFly and Myrddin get selected for TCEC ?

They are about 1000 elo behind many good engines that were not selected. Why not a random mover while we're at it ?

And do not have SMP. I thought one of the pre-requisite of TCEC was SMP.
Theory and practice sometimes clash. And when that happens, theory loses. Every single time.

Jesse Gersenson
Posts: 575
Joined: Sat Aug 20, 2011 7:43 am
Contact:

Re: TCEC groupe stage

Post by Jesse Gersenson » Sat Apr 30, 2016 6:20 am

Graham Banks wrote:Fritz 15 could replace Rybka 4.1 at some stage.
Fritz 15 could finish a respectable top-3 in any of the stages.

User avatar
Guenther
Posts: 3012
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: TCEC groupe stage

Post by Guenther » Sat Apr 30, 2016 7:09 am

lucasart wrote:
JJJ wrote:http://www.chessdom.com/tcec-season-9-g ... ticipants/

It seems that Stockfish will play the "easy" group.

On facebook, they says to stay tuned because of particular case of Houdini and Rybka, any idea about what they mean ?
How did FireFly and Myrddin get selected for TCEC ?

They are about 1000 elo behind many good engines that were not selected. Why not a random mover while we're at it ?

And do not have SMP. I thought one of the pre-requisite of TCEC was SMP.
For the computerchess community it's much better to have a few wildcards,
which are not selected by strength, but by other criteria like stamina,
integrity and whatever, besides hunting for the biggest E.. (fill in),
which is a main reason for numerous clones and useless pseudo derivates.

Guenther

Post Reply