A Chess forum. ChessBanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » ChessBanter forum » Chess Newsgroups » rec.games.chess.computer (Computer Chess)
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

Tags: , ,

Nalimov TBs problems



 
 
Thread Tools Display Modes
  #1  
Old July 12th 04, 10:59 PM
Pistarino
external usenet poster
 
Posts: n/a
Default Nalimov TBs problems

I downloaded some six pieces Nalimov's tablebases from Hyatt's FTP
site. I know I have to remove the middle dot to use them in Fritz 8.
But what about files with a number in their names, like

kqbnkq.0.nbb.emd File EMD 810.430 KB 05/06/2003 0.00.00
kqbnkq.0.nbw.emd File EMD 685.176 KB 05/06/2003 0.00.00
kqbnkq.1.nbb.emd File EMD 824.312 KB 05/06/2003 0.00.00
kqbnkq.1.nbw.emd File EMD 656.936 KB 05/06/2003 0.00.00
kqbnkq.2.nbb.emd File EMD 499.917 KB 05/06/2003 0.00.00
kqbnkq.2.nbw.emd

How can I handle them?

Moreover, do I need to download both nbb AND nbw to work with Fritz 8?

Thanks.
Ads
  #3  
Old July 14th 04, 12:29 PM
Pistarino
external usenet poster
 
Posts: n/a
Default Nalimov TBs problems

On 13 Jul 2004 00:50:03 GMT, CeeBee
wrote:

(Pistarino) wrote in rec.games.chess.computer:

I downloaded some six pieces Nalimov's tablebases from Hyatt's FTP
site. I know I have to remove the middle dot to use them in Fritz 8.
But what about files with a number in their names, like

kqbnkq.0.nbb.emd File EMD 810.430 KB 05/06/2003 0.00.00
kqbnkq.0.nbw.emd File EMD 685.176 KB 05/06/2003 0.00.00
kqbnkq.1.nbb.emd File EMD 824.312 KB 05/06/2003 0.00.00
kqbnkq.1.nbw.emd File EMD 656.936 KB 05/06/2003 0.00.00
kqbnkq.2.nbb.emd File EMD 499.917 KB 05/06/2003 0.00.00
kqbnkq.2.nbw.emd

How can I handle them?

Moreover, do I need to download both nbb AND nbw to work with Fritz 8?



You don't need to remove dots for use with the Fritz interface (or the
latets to be able to read 6 man TB's for that matter). The numbers
indicate large files split in parts. IIRC this is done because some OS's
can't read files larger than a certain size.

If you download tablebase files you need to consider dependencies. If
you download one tb and leave out a tb which could be the result of the
other you won't get proper results.


I have all 3-4-5 man tablebases. How can I check which tablebases are
needed for a specific 6 man tbs (knppkr for example). Moreover, do you
know how use splitted files in Deep Fritz 8? (I posted another message
with this question, don't kill me, please!

  #4  
Old July 27th 04, 04:43 PM
David Richerby
external usenet poster
 
Posts: n/a
Default Nalimov TBs problems

Pistarino wrote:
I have all 3-4-5 man tablebases. How can I check which tablebases are
needed for a specific 6 man tbs (knppkr for example).


You're playing a game with knpp vs kr. What combinations could arise?
Any of the pieces could be captured and either of the pawns could promote.
So, you need any filename formed by deleting one of the characters that
isn't a k and any filename formed by changing one of the p's into a q, r,
n or b.


Dave.

--
David Richerby Pickled Happy Puzzle (TM): it's like
www.chiark.greenend.org.uk/~davidr/ an intriguing conundrum that makes
your troubles melt away but it's
preserved in vinegar!
  #5  
Old July 28th 04, 07:57 AM
SethB
external usenet poster
 
Posts: n/a
Default Nalimov TBs problems

Pistarino wrote:
I have all 3-4-5 man tablebases. How can I check which tablebases are
needed for a specific 6 man tbs (knppkr for example).


You're playing a game with knpp vs kr. What combinations could arise?
Any of the pieces could be captured and either of the pawns could promote.
So, you need any filename formed by deleting one of the characters that
isn't a k and any filename formed by changing one of the p's into a q, r,
n or b.


Dave.


A succinct description. Which translates into ~30 tablebases (~15 pairs
for B,W) to cover 6pc tablebases for only knppkr. (You already have all
the necessary 3,4,5 man tablebases this translates to by piece capture).
You need every one of the knppkr transposition tablebases or it will not
function at all. You may want to go to Hyatt's site and add up the file
sizes of these ~30 tablebases so you can plan where to put them. It's a
large number of gigabytes for just this one set. The complete set of 6pc
tablebases is over 1000 gigabytes.

The combined size of all 3,4,5 man table bases is only about 7 gigabytes,
by way of comparison.
  #6  
Old July 28th 04, 05:15 PM
Robert Hyatt
external usenet poster
 
Posts: n/a
Default Nalimov TBs problems

SethB wrote:
Pistarino wrote:
I have all 3-4-5 man tablebases. How can I check which tablebases are
needed for a specific 6 man tbs (knppkr for example).


You're playing a game with knpp vs kr. What combinations could arise?
Any of the pieces could be captured and either of the pawns could promote.
So, you need any filename formed by deleting one of the characters that
isn't a k and any filename formed by changing one of the p's into a q, r,
n or b.


Dave.


A succinct description. Which translates into ~30 tablebases (~15 pairs
for B,W) to cover 6pc tablebases for only knppkr. (You already have all
the necessary 3,4,5 man tablebases this translates to by piece capture).
You need every one of the knppkr transposition tablebases or it will not
function at all. You may want to go to Hyatt's site and add up the file
sizes of these ~30 tablebases so you can plan where to put them. It's a
large number of gigabytes for just this one set. The complete set of 6pc
tablebases is over 1000 gigabytes.


The combined size of all 3,4,5 man table bases is only about 7 gigabytes,
by way of comparison.



I will add that for Crafty (I can't speak for other programs) you can get
by without having the promotion tables. IE KRPPKR will work just fine
without the other 10 promotion cases. Having them is better, as you can
reach KQRPKR via normal captures as well as by a pawn promotion, so you will
have more info to use in the search. But crafty will play just fine without
the promotion tables, and it won't get hung up as some programs do and avoid
promoting since it can't see the forced mate. It has special code to avoid
this particularly ugly problem.


--
Robert M. Hyatt, Ph.D. Computer and Information Sciences
University of Alabama at Birmingham
(205) 934-2213 136A Campbell Hall
(205) 934-5473 FAX Birmingham, AL 35294-1170
  #7  
Old August 1st 04, 10:15 PM
Simon Krahnke
external usenet poster
 
Posts: n/a
Default Nalimov TBs problems

* Robert Hyatt (2004-07-28) schrieb:

But crafty will play just fine without the promotion tables, and it
won't get hung up as some programs do and avoid promoting since it
can't see the forced mate. It has special code to avoid this
particularly ugly problem.


Is this special to promotion, or does is work for other missing tables
as well?

Does it just start searching again?

mfg, simon .... l
  #8  
Old August 1st 04, 10:49 PM
Robert Hyatt
external usenet poster
 
Posts: n/a
Default Nalimov TBs problems

Simon Krahnke wrote:
* Robert Hyatt (2004-07-28) schrieb:


But crafty will play just fine without the promotion tables, and it
won't get hung up as some programs do and avoid promoting since it
can't see the forced mate. It has special code to avoid this
particularly ugly problem.


Is this special to promotion, or does is work for other missing tables
as well?


Does it just start searching again?


mfg, simon .... l


If you have a missing table, the basic idea is to take all the moves at the
root (assuming the table for the root position is available and that it says
this game is won) and for each root move you must either get (after making
the move) win in N-1, draw, or lost for the score, except for those with
no tables. If there are no win in N-1 moves there, then you must promote
or capture to continue the win.

Ie the most common ugly case is a pawn on the 7th, but you don't have the
promotion table. The program would rather keep the pawn which leads to a
mate even though it is no closer than the last move's mate score, rather
that to promote the thing to a queen and get no mate score at all. Crafty
will promote just fine in that case...


--
Robert M. Hyatt, Ph.D. Computer and Information Sciences
University of Alabama at Birmingham
(205) 934-2213 136A Campbell Hall
(205) 934-5473 FAX Birmingham, AL 35294-1170
 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Nalimov endgames Wayne Martin rec.games.chess.computer (Computer Chess) 3 March 3rd 04 06:51 AM
problems with CB 8 Henk van Weersel rec.games.chess.computer (Computer Chess) 0 February 26th 04 12:52 AM
CB update causing CD problems??! HD rec.games.chess.computer (Computer Chess) 2 January 11th 04 03:02 PM
Positional problems books kempelen rec.games.chess.analysis (Chess Analysis) 15 September 27th 03 08:51 PM
Collection of my Diary Problems Fernando rec.games.chess.analysis (Chess Analysis) 4 September 10th 03 01:26 AM


All times are GMT +1. The time now is 02:58 PM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2008, Jelsoft Enterprises Ltd.Content Relevant URLs by vBSEO 2.4.0
Copyright ©2004-2008 ChessBanter, part of the NewsgroupBanter project.
The comments are property of their posters.
Mortgage Calculator - Repair Bad Credit - T Shirt Quilts - Loans - Mortgage Calculator