Modular opening book SF analysed 87417 pos., beta-1

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

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Frank Quisinsky
Posts: 5098
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: Modular opening book SF analysed 87417 pos., beta-1

Post by Frank Quisinsky » Tue Mar 14, 2017 5:08 am

Hi Dann,

do you know the written theory book from John Shaw (The King's Gambit)? One of the best books to theory I know.

For an example: To find to C38 balanced lines isn't in reality possible. But with such books John Shaw have written you can added lines ... engines will give evaluation from +1 because to many of lines engines never able to give us a good evaluation because engines have often not understandings to complicated lines.

I have a lot very good written opening books by grandmasters. If my database in six months are ready I will looking what I have as final results. To rarely lines I will add by hand again important lines from such written books.

Best
Frank

I do it often for my FCP Live Book ... but most of that material I added from hand now the engines reject :-(

I must do all work again ... need also time because I don't have it in a database ... made again and again the same mistake. Better is to added newer book material in additional databases.
I like computer chess!

Dann Corbit
Posts: 11488
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: Modular opening book SF analysed 87417 pos., beta-1

Post by Dann Corbit » Tue Mar 14, 2017 7:03 am

The link to my data has lots of analysis to 36 plies or more.
Maybe some of it will be helpful.

20 plies just weeds out the truly bonehead moves and nothing more
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

Frank Quisinsky
Posts: 5098
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: *** If programmers / users have interest to help here!

Post by Frank Quisinsky » Tue Mar 14, 2017 8:59 am

Hi Dann,

The file contains beta-v01 with the udpate I made without move transpositions with some additional information to the positions. So, all the bad lines I added from newer theory I found are inside ... 80% from FCP Rating List games and 20% newer theory I found and ECO codes I haven't enough. Bad lines comes the newer 20% of material I added before I start the project.

Very interesting material!
But which engines are used for it?
It seems Stockfish ... think so!

I can added in my download files such files by users / programmers. But more interesting is beta-v03. So I can collect in a user-file directory engine analysis to beta-v03 (most important file). Better to use beta-v03_endpos_n.epd. So we can compare it much more easy with Excel other programs.

All the time I am thinking ... others can help here ... one reason to create an open project about it.

Thanks for your hint ... I am not looking in Rybka forum. TalkChess and CSS forum is enough for me.

Best
Frank
I like computer chess!

Dann Corbit
Posts: 11488
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: *** If programmers / users have interest to help here!

Post by Dann Corbit » Tue Mar 14, 2017 9:11 am

Frank Quisinsky wrote:Hi Dann,

The file contains beta-v01 with the udpate I made without move transpositions with some additional information to the positions. So, all the bad lines I added from newer theory I found are inside ... 80% from FCP Rating List games and 20% newer theory I found and ECO codes I haven't enough. Bad lines comes the newer 20% of material I added before I start the project.

Very interesting material!
But which engines are used for it?
It seems Stockfish ... think so!
Mostly SF but sometimes Komodo or Houdini.
I go by depth.
So if I have 35 ply SF data and Houdini analysis goes to 36 plies I will store the Houdini data and move the SF row to my "alternates" table.
Then if Komodo goes to 39 plies for the same position, it will replace and displace the old row again.

I can added in my download files such files by users / programmers. But more interesting is beta-v03. So I can collect in a user-file directory engine analysis to beta-v03 (most important file). Better to use beta-v03_endpos_n.epd. So we can compare it much more easy with Excel other programs.

All the time I am thinking ... others can help here ... one reason to create an open project about it.

Thanks for your hint ... I am not looking in Rybka forum. TalkChess and CSS forum is enough for me.

Best
Frank
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

Post Reply