Hi friends,

I have always been fascinated by Fortresses in chess, where one side deficit in material builts a position with the reamining pieces that is impossible to penetrate by the stronger side. Should probably be the most aesthetic side of chess!.

Was just thinking wouldnt it be nice to have a list of fortresses that we know or have played or seen or even constructed in our minds. I am sure there must be simple fortresses with 3 pieces to complex fortresses with many pieces on board. But the aim is it shouldnt be penetratable at any cost.

One that comes immediately to mind.

You are welcome to share your fortress ideas.

Cheers,

Arun

Actually, the first one is a win for White. This fortress idea has been busted.

According to Nalimov, White has a win in 55 starting with Qg2.

Move black's pieces up the board one rank (Pf7, Re6, Kf8), and we have a draw. Amazing how that extra space actually hurts black in original position.

The concept of fortresses is an amazing place for creative defensive positions which may be new to some readers.

Thanks, Ahhca, for opening this discussion.

Busted by rybka 4.0!

Another one. This should also qualify as a fortress right?. This is a fortress with a deficit of 4 points (with 1,3,3,5,9 convention). Anyone to better :).

Cheers,

Arun

Was just wondering if 55 moves without a pawn move considered as a draw from this position?

I haven't played it through, but you provbably force a pawn move somewhere.

One of my recent over the board tournament games -- I was white and managed to reach this position in the end.  My opponent tried to break it down for another 50 moves or so.

I disagree with Nalimov. Whites king is on the wrong side of that rook, so as it is out of play entirely. It must stay in the cage between  d4 & a1 it can not effect the game at all.

That leaves the white queen with the job of seperating the black king from the f6 pawn all by herself and this is impossible if black plays good.

Black is in no danger in this position if black plays solid, white is not a threat,

Post the entire line, it must have a blunder by black

I know it's old, and you're all getting tired of seeing it, but it fits the topic too well to be omitted:

That one is a classic worth reposting. Thanks!

My Bongcloud Fortress Variation L01

You are welcome to disagree with Nalimov, but rather than someone posting the 55  moves (and many, many sub variations), I suggest it would be more instructive for you or anyone to play out the position on http://www.k4it.de/index.php?topic=egtb&lang=en if you do not possess the tablebase.

On initial review, it seems as if the white king is trapped away from activity, but that extra rank provides white queen to force black to make concessions, either moving the rook to an unguarded square, the pawn to advance, the king to move to less comfortable squares. Not many of the moves are natural hence the tablebase and a good endgame book to assist.

Amazingly, in Dvoretsky's Endgame Manual, the author covers these positions, and notes with the pawn moved one square (as in the original posting of this thread):

• with a pawn on a6 or h6, the position is drawn.
• with a pawn on b6 or g6, a draw can be reached, and the black king should remain behind the pawn.
• with a pawn on c6, d6, e6, or f6, the position is lost.
  

For breaking the fortress in the last case, Dvoretsky cites the following plan for white;

1. Occupy squares behind the black pawn via zugzwang, and force the black king in front of his pawn, thereby obstructing his rook,
2. Cross the 5th rank with the white king.
3. Breakthrough with the with king to attack the pawn.

After winning the pawn, white next must solve the infamous Q vs. R endgame which is also problematic.

Enjoy!

"The tablebase contains the game-theoretical value (win, loss, or draw) of each possible move in each possible position, and how many moves it would take to achieve that result with perfect play. Thus, the tablebase acts as an oracle, always providing the optimal moves. Typically the database records each possible position with certain pieces remaining on the board, and the best moves with White to move and with Black to move."

I guess it isn't obvious why this fortress can be broken.  Yes the Rook in front of a pawn like that is a type of fortress, but the specific position you give is not a true fortress.  A bishop's pawn cannot be on the 6th rank in this type of fortress, if it is the fortress can be broken down.  Philidor showed a similar position back in 1777.

As Dvoretsky states: In the following position White's plan is

1)To occupy squares behind the pawn, with the help of zugzwang, and drive the black king out to d5 where he obstructs his own rook.

2)To cross the 5th rank with his king.

3) To break through with the king to the e-file and the pawn.

I am glad to see nobody disagrees with my fortress.

Well, I'll tell you, I wont resign as Black! White will have to show me, even if he's a GM :)

I was wrong, but seriously 99.9% of Humans w/o looking directly at a table base are not going to be able to to win that game against solid play. A queen vs rook endgame is hard enough as is if u have the queen, adding that pawn into the mix seemed like enough to hold back the queen but after about an hour of analyzing the table base for it, it is fairly clear the queen has a win.

However if the black pieces were all back 1 rank it is a draw.