user923005 Position

General discussion about computer chess...
Post Reply
zullil
Posts: 82
Joined: Thu Jun 10, 2010 10:17 am
Real Name: Louis Zulli
Location: Pennsylvania, USA

user923005 Position

Post by zullil » Sun Feb 16, 2014 5:09 pm

Saw this at rybkaforum.net

r1b1k2r/pp1p1ppp/2n1p3/qN2P3/5P2/2bBB3/P1P2KPP/R2Q3R b kq -

Decided to let Stockfish take a long look, using MultiPV=3:

Code: Select all

info depth 48 seldepth 71 score cp 0 nodes 1283832057191 nps 16873327 time 76086477 multipv 1 pv c3b4 a2a3 b4e7 h1e1 e8g8 f2g1 a7a6 b5d6 e7d6 e5d6 b7b5 f4f5 a5c3 d1d2 c3d2 e3d2 a8b8 a3a4 b5b4 a1b1 f8d8 c2c3 b4b3 b1b2 f7f6 d3c4 c6a5 c4d3 a5c6

info depth 48 seldepth 71 score cp -19 nodes 1283832057191 nps 16873327 time 76086477 multipv 2 pv e8g8 d3h7 g8h7 d1d3 h7g8 b5c3 b7b6 c3e4 c6b4 d3b3 c8b7 e3d2 a5d5 b3d5 b4d5 e4d6 b7c6 c2c4 d5e7 h1e1 e7f5 d2b4 f5d6 b4d6 f8c8 h2h4 c6b7 e1c1 f7f5 a2a4 b7e4 g2g3 c8c6 f2e3 a8c8 e3d4 e4d5 c4c5 g8f7 c1c3 c8a8 a1c1 c6c8 c5b6 c8c3

info depth 48 seldepth 71 score cp -21 nodes 1283832057191 nps 16873327 time 76086477 multipv 3 pv c3a1 b5d6 e8e7 d1a1 e7f8 h1b1 h7h5 f2g1 f8g8 a2a4 h5h4 c2c4 a5d8 d3e4 h4h3 g2g3 d8f8 c4c5 g7g6 a1c3 g8g7 c3d2 a8b8 d2d3 h8h7 d3e2 g7g8 e2d2

User923005
Posts: 616
Joined: Thu May 19, 2011 1:35 am

Re: user923005 Position

Post by User923005 » Tue Feb 18, 2014 4:45 am

I had the castle move as best, but I had only analyzed to 37 plies.
r1b1k2r/pp1p1ppp/2n1p3/qN2P3/5P2/2bBB3/P1P2KPP/R2Q3R b kq - acd 37; acs 795; bm O-O; ce 16; pm O-O; pv O-O Bxh7+ Kxh7 Qd3+ Kg8 Nxc3 b6 Bd2 Ba6 Qxd7 Rfd8;

ChessDrone
Posts: 78
Joined: Mon Jan 28, 2013 10:56 am

Re: user923005 Position

Post by ChessDrone » Thu Feb 20, 2014 1:29 am

User923005 wrote:I had the castle move as best, but I had only analyzed to 37 plies.
r1b1k2r/pp1p1ppp/2n1p3/qN2P3/5P2/2bBB3/P1P2KPP/R2Q3R b kq - acd 37; acs 795; bm O-O; ce 16; pm O-O; pv O-O Bxh7+ Kxh7 Qd3+ Kg8 Nxc3 b6 Bd2 Ba6 Qxd7 Rfd8;
Hi user923005 :)

How do you edit/read and hold the database of EPD with spreadsheet? I ask the question because I know you hold a large database EPD. But mostly because I did not find available tool for this. Currently I'm trying EPD with libreoffice calc.

User923005
Posts: 616
Joined: Thu May 19, 2011 1:35 am

Re: user923005 Position

Post by User923005 » Thu Feb 20, 2014 3:11 am

My database is in SQL Server format.
I can easily change it to any other database, but I have dozens of stored procedures that would not port so easily as the structures and data.

For instance, this is the table Epd:


USE [Chess]
GO

/****** Object: Table [dbo].[Epd] Script Date: 2/19/2014 6:10:58 PM ******/
SET ANSI_NULLS ON
GO

SET QUOTED_IDENTIFIER ON
GO

SET ANSI_PADDING ON
GO

CREATE TABLE [dbo].[Epd](
[Epd] [varchar](255) NOT NULL,
[acd] [smallint] NULL,
[acn] [int] NULL,
[ce] [smallint] NULL,
[pv] [varchar](255) NULL,
[am] [varchar](255) NULL,
[dm] [smallint] NULL,
[id] [varchar](255) NULL,
[CheckoutCount] [smallint] NULL,
[c0] [varchar](255) NULL,
[c1] [varchar](255) NULL,
[c2] [varchar](255) NULL,
[c3] [varchar](255) NULL,
[c4] [varchar](255) NULL,
[acs] [int] NULL,
[EpdID] [uniqueidentifier] NOT NULL,
[tag] [char](10) NULL,
[white_wins] [int] NULL,
[black_wins] [int] NULL,
[draws] [int] NULL,
[coef] [float] NULL,
[bm] [varchar](1000) NULL,
[pm] [varchar](1000) NULL,
[Opening] [varchar](400) NULL,
[games] AS (([white_wins]+[black_wins])+[draws]),
CONSTRAINT [PK_Epd_1] PRIMARY KEY CLUSTERED
(
[EpdID] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY]

GO

SET ANSI_PADDING OFF
GO

ALTER TABLE [dbo].[Epd] ADD CONSTRAINT [DF_Epd_EpdID] DEFAULT (newid()) FOR [EpdID]
GO

ALTER TABLE [dbo].[Epd] ADD CONSTRAINT [DF_Epd_white_wins] DEFAULT ((0)) FOR [white_wins]
GO

ALTER TABLE [dbo].[Epd] ADD CONSTRAINT [DF_Epd_black_wins] DEFAULT ((0)) FOR [black_wins]
GO

ALTER TABLE [dbo].[Epd] ADD CONSTRAINT [DF_Epd_draws] DEFAULT ((0)) FOR [draws]
GO

User923005
Posts: 616
Joined: Thu May 19, 2011 1:35 am

Re: user923005 Position

Post by User923005 » Thu Feb 20, 2014 3:15 am

ChessDrone wrote:
User923005 wrote:I had the castle move as best, but I had only analyzed to 37 plies.
r1b1k2r/pp1p1ppp/2n1p3/qN2P3/5P2/2bBB3/P1P2KPP/R2Q3R b kq - acd 37; acs 795; bm O-O; ce 16; pm O-O; pv O-O Bxh7+ Kxh7 Qd3+ Kg8 Nxc3 b6 Bd2 Ba6 Qxd7 Rfd8;
Hi user923005 :)

How do you edit/read and hold the database of EPD with spreadsheet? I ask the question because I know you hold a large database EPD. But mostly because I did not find available tool for this. Currently I'm trying EPD with libreoffice calc.
I recommend against spreadsheet format, because it is slow and limited. Many spreadsheets will only hold 64000 rows and besides, you cannot index the spreadsheet columns.

PostgreSQL is a nice format. I can send you PostgreSQL schema for my database if you like.

ChessDrone
Posts: 78
Joined: Mon Jan 28, 2013 10:56 am

Re: user923005 Position

Post by ChessDrone » Thu Feb 20, 2014 4:27 am

Thanks for the offer :) , but I fear not know how use it. I was more referring to a tool all in one as for PGN files.

User923005
Posts: 616
Joined: Thu May 19, 2011 1:35 am

Re: user923005 Position

Post by User923005 » Thu Feb 20, 2014 4:38 pm

From here:
http://www.7sun.com/chess/
there is a tool called pgn2fen by Tim Foden.

It does a real nice job of turning PGN files into Epd records.
I often use it like this:
pgn2fen file.epd -e -l
which will turn a PGN file called file.epd into EPD records (the -e flag) and also includes the last move (-l for leaf nodes).

Rémi Coulom has a similar utility at his chess site that even comes with source code.

I usually use some sort of stream editor to pull off the full move number and the half move clock value. I trust the chess engines to track that since the basic concepts of the position are mostly independent of those two values when using many tools.

PGN-Extract by Barnes is extremely useful.

Norm Pollock has a mighty army of PGN tools.

ChessDrone
Posts: 78
Joined: Mon Jan 28, 2013 10:56 am

Re: user923005 Position

Post by ChessDrone » Fri Feb 21, 2014 11:57 am

Thank you for the useful information :)

Post Reply