MYG

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Rebel, chrisw

IWB
Posts: 1539
Joined: Thu Mar 09, 2006 2:02 pm

Re: MYG

Post by IWB »

Laskos wrote:One possible issue is that Contempt in Stockfish is broken or at least too rudimentary. Komodo has much more involved Contempt, maybe that's why it is gaining significantly ELO in rating lists such as yours.
i can agree to a brocken SF contempt implementation but that would be something someone would have discovered already - I hope

And Komodo "gaining significantly" of a contempt? I wrote about the test on the 20.6.2016 of K10 without contempt (please check). It lost less than 1 SD or 7 Elo. Is that significant? But K10 without contempt performed much better against SF - I am not a fan of contempts at all, mostly their influence is overrated (no pun intended :-)
Jouni
Posts: 3283
Joined: Wed Mar 08, 2006 8:15 pm

Re: MYG

Post by Jouni »

Houdini 4 was also tested by Ingo :wink: . No other possibility as clones are not tested.See

http://www.talkchess.com/forum/viewtopic.php?t=50304
Jouni
JJJ
Posts: 1346
Joined: Sat Apr 19, 2014 1:47 pm

Re: MYG

Post by JJJ »

Okey, now I m sure it is not Stockfish dev or Stockfish clone.
So I m still betting on Houdini.
JJJ
Posts: 1346
Joined: Sat Apr 19, 2014 1:47 pm

Re: MYG

Post by JJJ »

Jouni wrote:Houdini 4 was also tested by Ingo :wink: . No other possibility as clones are not tested.See

http://www.talkchess.com/forum/viewtopic.php?t=50304
december 2013

1 Houdini 4 3117
2 Komodo 5.1 3025
3 Stockfish 4 3019

Stockfish was third. And now again ? :)
IWB
Posts: 1539
Joined: Thu Mar 09, 2006 2:02 pm

Re: MYG

Post by IWB »

+23, =70, -19

there was a typo, sorry.
JJJ
Posts: 1346
Joined: Sat Apr 19, 2014 1:47 pm

Re: MYG

Post by JJJ »

Now barely winning against Stockfish. When I see a strong program like this doing so bad against Stockfish, I m sure there is a problem with Stockfish rating. It is indeed deflated because Stockfish draw too many game and is doing much better against close opponent.
IWB
Posts: 1539
Joined: Thu Mar 09, 2006 2:02 pm

Re: MYG

Post by IWB »

JJJ wrote:Now barely winning against Stockfish. When I see a strong program like this doing so bad against Stockfish, I m sure there is a problem with Stockfish rating. It is indeed deflated because Stockfish draw too many game and is doing much better against close opponent.
Do not forget that K performes much better against SF without contempt ...

What I think is that it is much more complex the the "allegedly" high draw rate!
User avatar
Guenther
Posts: 4605
Joined: Wed Oct 01, 2008 6:33 am
Location: Regensburg, Germany
Full name: Guenther Simon

Re: MYG

Post by Guenther »

Jouni wrote:Houdini 4 was also tested by Ingo :wink: . No other possibility as clones are not tested.See

http://www.talkchess.com/forum/viewtopic.php?t=50304
Only if he knew before it was an SF derivate in a closed source program.
Last edited by Guenther on Sat Sep 09, 2017 1:35 pm, edited 1 time in total.
https://rwbc-chess.de

trollwatch:
Chessqueen + chessica + AlexChess + Eduard + Sylwy
Modern Times
Posts: 3546
Joined: Thu Jun 07, 2012 11:02 pm

Re: MYG

Post by Modern Times »

As I said at the beginning, I think it is either latest Komodo or Shredder. Not sure which. But actually I don't recall Stefan releasing this way before ? So that raises doubts. Komodo team on the other hand are a fan of IPON, so....
JJJ
Posts: 1346
Joined: Sat Apr 19, 2014 1:47 pm

Re: MYG

Post by JJJ »

Looks like it wil stick to 3318.