Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
corres
Posts: 689
Joined: Wed Nov 18, 2015 10:41 am
Location: hungary

Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by corres » Mon Jun 11, 2018 8:38 am

TCEC proves the difference between developer version of Stockfish and the other top engines decreased to now.
I think it is the time to change the basic engine (Stockfish 7) to a stronger one (Stockfish 8).
I suppose it may show a more real picture about the power of Stockfish dev. than Stockfish 7 does.
Moreover it may stimulate the developers to make changes to gain Elo instead of making cosmetics code modifications with noisy results in Elo.

Vinvin
Posts: 4118
Joined: Thu Mar 09, 2006 8:40 am

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by Vinvin » Mon Jun 11, 2018 11:18 am

Where do you see "Stockfish 7" ?

Uri Blass
Posts: 8221
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by Uri Blass » Mon Jun 11, 2018 12:09 pm

corres wrote:
Mon Jun 11, 2018 8:38 am
TCEC proves the difference between developer version of Stockfish and the other top engines decreased to now.
I think it is the time to change the basic engine (Stockfish 7) to a stronger one (Stockfish 8).
I suppose it may show a more real picture about the power of Stockfish dev. than Stockfish 7 does.
Moreover it may stimulate the developers to make changes to gain Elo instead of making cosmetics code modifications with noisy results in Elo.
TCEC proves nothing because of not having enough games

corres
Posts: 689
Joined: Wed Nov 18, 2015 10:41 am
Location: hungary

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by corres » Mon Jun 11, 2018 4:23 pm

Uri Blass wrote:
Mon Jun 11, 2018 12:09 pm

TCEC proves nothing because of not having enough games
Only in the case of small Elo difference is important the number of games.
If the difference is enough big even two games played from the same starting position with reversed colors may show the power relation of the two engine.

Uri Blass
Posts: 8221
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by Uri Blass » Mon Jun 11, 2018 8:53 pm

corres wrote:
Mon Jun 11, 2018 4:23 pm
Uri Blass wrote:
Mon Jun 11, 2018 12:09 pm

TCEC proves nothing because of not having enough games
Only in the case of small Elo difference is important the number of games.
If the difference is enough big even two games played from the same starting position with reversed colors may show the power relation of the two engine.

My point is that there is no proof that the difference between stockfish and the rest of the programs got smaller.

Stockfish has 12 wins and 32 draws in the premier of season 12 total result of 63.636%
Stockfish had 39 wins and one loss and 44 draws in the premier of season 11 total result of 72.619%

I think that the expected result may be 69% when stockfish was lucky in season 11 and unlucky in season 12

corres
Posts: 689
Joined: Wed Nov 18, 2015 10:41 am
Location: hungary

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by corres » Mon Jun 11, 2018 10:09 pm

In my opinion the top chess engines are: Stockfish, Komodo, Houdini.
The recent standing is:
Stockfish 160518... -110 Elo
Komodo 12............+59 Elo
Houdini 6.03..........-27 Elo
No comment...

AndrewGrant
Posts: 421
Joined: Tue Apr 19, 2016 4:08 am
Location: U.S.A
Full name: Andrew Grant
Contact:

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by AndrewGrant » Mon Jun 11, 2018 11:35 pm

No comment...
Edgy closers don't replace statistics.

corres
Posts: 689
Joined: Wed Nov 18, 2015 10:41 am
Location: hungary

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by corres » Tue Jun 12, 2018 11:36 am

Uri Blass wrote:
Mon Jun 11, 2018 8:53 pm

I think that the expected result may be 69% when stockfish was lucky in season 11 and unlucky in season 12
You would be right if the participants are the same.
But it is not the situation.
Now Stockfish is relative weaker than it was.

corres
Posts: 689
Joined: Wed Nov 18, 2015 10:41 am
Location: hungary

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by corres » Tue Jun 12, 2018 11:40 am

AndrewGrant wrote:
Mon Jun 11, 2018 11:35 pm
No comment...
Edgy closers don't replace statistics.
If you are interested in statistics please, make those statistics.
The data are public.
To me it is obvious that to loose 110 Elo during 44 games this is not a statistical issue.

sovaz1997
Posts: 150
Joined: Sun Nov 13, 2016 9:37 am

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by sovaz1997 » Tue Jun 12, 2018 11:59 am

TC you don't understand, why are you talking about this?
Zevra chess engine
Binary, source and description here: https://gitlab.com/sovaz1997/zevra2/tags

Post Reply