Author: Brian Katz
Date: 03:17:39 04/17/02
Go up one level in this thread
On April 17, 2002 at 05:42:50, Brian Katz wrote:
>On April 17, 2002 at 01:25:02, Leen Ammeraal wrote:
>
>>On April 16, 2002 at 23:15:47, Tom Likens wrote:
>>b>>situation as a draw (essentially a case of missing the 2-man tablebase le).
>
>
>
>
>>>
>>>>>
>>>I recently had what I considered an interesting endgame tablebase bug which
>>>cost my program 1/2 a point on ICS. The problem occurred when a KR vs. KR
>>>endgame was reached and the opponent captured the program's rook. Instead of
>>>grabbing the rook with its king it moved the king away and immediately lost the
>>>game. Looking at the log file revealed that it had been in the endgame
>>>tablebases for a while. In fact, the way I structure this is that if I'm not
>>>in swindle mode and the root position is a tablebase draw then it simply prints
>>>out the PV to the maximum search depth and returns (skipping the search
>>>entirely).
>>>
>>>The problem was that the EGTB probe code did not recognize the K vs. K
>
>
>I have had a similar problem which occurs in the following position if you are
>using the 4 disc Nalimov set (Endgame Turbo)
>[D]1r6/6P1/5K2/6R1/8/4k3/8/8 w - - 0 1
>
> Of course the correct move should be to immediately promote the pawn with g8=Q
>However, you will find if you are using Fritz 6 and before, and quite probably
>Fritz 7 (I have not tried this with the Fritz 7 engine), that rather than
>promote the pawn, Rg3 check will be played instead.
> The reason for this is that the Nalimov set is missing some important tablebase
>files, such as KRR KR and others. This being the case, the program does NOT then
>know the KRR KR endgame (and others) perfectly, so it avoids immediately
>promoting the pawn. There are others that are also missing which also cause this
>problem.
>Make sure you have all the KRP vs KR possibilies such as:
>KRPKR, KRQKR, KRRKR, KRNKR, KRBKR.
>The remedy for this is to go into Bob Hyatts wonderful site and download the
>Tablebase files that are missing. It is as simple as that. Make sure that you
>download both files for each endgame situation you choose to download.
>Here is an interesting note:
>The Crafty program is has the intelligence to play the correct move and
>immediately promote the pawn even with the missing Tablebase files.
>
>I am not sure which other programs out there will have problems with the above
>mentioned program. Another interensting note: If you only used the tablebases
>that were included on the Fritz6 disc then this endgame will be played properly.
>
>Bob Hyatts site also gives an explanation of the missing tablebase problem.
>
>I hope this helps some of you.
>Regards
>Brian
ftp://ftp.cis.uab.edu/pub/hyatt/TB/five/
Above is the url for the 5 man tablebase files
http://www.chesskit.com/aarontay/Winboard/Winboard3.html#[C.7]
Above is the url for the Tablebase problem
http://www.chesskit.com/aarontay/Winboard/Winboard.html
Above is the url for the Winboard/Engine FAQ's
BELOW IS THE HYATT TABLEBASE ARTICLE FROM THE FAQ
C.7] How many tablebases do I need? Can I download only some of them?
The more tablebase files you install, the stronger the program will be. However,
a full set of 3,4 and 5 Tablebase files takes about 8 GB of Hard disk space!
Most people download a full set of 4 Piece databases and select only a few of
the 5 pieces.
Generally the 5 piece endgames with rooks are reached most frequently and should
be downloaded . However, there are some pitfalls that you should be aware of.
If you want to use the KRPKR tablebase [and assuming you have all the 3 and 4
tablebases], make sure that you have the following endgame tablebases as well,
KQRKR, KRBKR, KRRKR, KRNKR, . This is to ensure that the promotion cases are
included.
If you lack say the KQRKR tablebase, some programs refuse to queen the pawn in
the KRPKR even if that would lead to a win, because such a move, would cause the
program to drop from a position flagged as "win", to a position that is
uncertain since they lack the relevant endgame table.
A similar problem can result if you download only the KQPKQ tablebase without
KQQKQ , KQRKQ etc..
Some programs like Crafty and Yace are "smart" enough to avoid this problem, but
most like Amy , or The Crazy Bishop cannot handle this.
Also take note that not all programs support all the Tablebases. Yace for
example, currently does not use the 4+1 tablebases [ King and 3 pieces/pawns on
a side versus a alone King] , because such positions are easily won most of the
time barring rare cases [Like double rook pawns, wrong coloured bishop and King
vers King]. Also I think only Crafty currently, supports 6 piece tablebases.
Szots Gabor also reports that Wildcat , Capture and perhaps Inimichess cannot
handle incomplete 5 men tablebase.
I hope this helps some of you out there.
Regards
Brian
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>>>So instead of leaving the tablebases it played a losing move that kept it in
>>>the tablebases, aarrgghh!! Simple to fix, but it begs the larger question of
>>>what happens in general if a program has access to an N piece tablebase but
>>>not all the N-1 piece tablebases that the position could become if a piece or
>>>pieces were traded off. ...
>>
>>I cannot remember having had problems with this situation. But
>>there is a similar case, which was more serious for me:
>>the program may be dealing with N-piece tablebases in which
>>one of the N pieces is a pawn, which is about to promote,
>>while the corresponding N-piece tablebases with that pawn
>>replaced with a queen is lacking. The effect was that the
>>pawn refused to promote. For this reason, I now
>>temporarily switch off tablebase probing when a pawn is
>>about to promote.
>>Leen
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.