deep.09

[Dann Corbit]

I can easily see getting your king to that position through something like this:

[Event "?"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "?"]
[Black "?"]
[Result "*"]
[FEN "6r1/6r1/2p1k1pp/p1pbPq2/Pp1p1PpP/1P1P2NR/1KPQ3R/8 w - - 0 1"]
[SETUP "1"]

1. Nxf5 gxf5 2. Rg3 h5 3. Rg1 Bf3 4. Qe1 Bd5 5. Qg3 Rc7 6. Rf2 Re8 7. Kc1 Rd8 8. Rff1 Re7 9. Kd2 Rg7 10. Re1 Re7 11. Ke2 Rde8 12. Ra1 Rg7 13. Kf2 Reg8 14. Rae1 Bf3 15. Rb1 Bd5 16. Rbc1 Kd7 17. Ra1 Kc7 18. Rae1 Re8 19. Qh2 Bf3 20. Kg3 Re6 21. Qf2 *

Giving this position

[d]8/2k3r1/2p1r3/p1p1Pp1p/Pp1p1PpP/1P1P1bK1/2P2Q2/4R1R1 b - -

But this position is definitely drawn. So I guess what I am asking is to see all of the actual steps from the initial puzzle position. I have three chessmen that can be moved to counter things (one of the rooks is obviously nailed to the board).

[jhellis3]

Why do you keep playing bad moves for white? I already showed you 1 improvement in my previous post.

I really don't like having to repeat things, but I will do it once more (please not again).....

[d]4r3/4r3/2p1k3/p1pbPp1p/Pp1p1PpP/1P1PR3/1KPQ4/4R3 b - -

Qf5 Nxf5 gxf5 Rg3 h5 Re2 Re7 Re1 Rge8 Re3

The point is that the black king can not move from e6 because it allows white to play e6.


If you really must play it out, we can do that.... but with a little work on your end, I think you find the study is very likely flawed.

[peter]

[d]4r3/4r3/2p1k3/p1pbPp1p/Pp1p1PpP/1P1PR3/1KPQ4/4R3 b - -

Qf5 Nxf5 gxf5 Rg3 h5 Re2 Re7 Re1 Rge8 Re3

The point is that the black king can not move from e6 because it allows white to play e6.

Why and where does the black King have to move from e6 at all? Maybe you're right, that the author's solution unnecessarily gives such an example, maybe to prevent the Bishop from being castled by white c4, I don't think that to be the real danger.
You saw your first idea c3 fail, if a black Rook is on b-line at the moment of white b-Pawn to move, ...bxc3 makes another fortress or even a white disadvantage. Did you see the threats of the black b- and or g-pawn to break through themselves after White's b- pawn is gone?
We don't need to play it our move by move, if you show a single winning line for White, that cannot be denied.
Target- postions won't help without lines toward and forward

[peter]

You saw your first idea c3 fail, if a black Rook is on b-line at the moment of white b-Pawn to move, ...bxc3 makes another fortress or even a white disadvantage.

Of course I meant, "at the moment of the white c-Pawn to move", not b-Pawn as mistyped

[Sven]

I think White wins by moving a rook to c1 and then playing c2-c3

You mean e.g. like this?

6r1/6r1/2p1k1pp/p1pbP2q/Pp1p1PpP/1P1P2NR/1KPQ3R/8 b - - 0 1

1... Qf5 2. Nxf5 gxf5 3. Rg3 h5 4. Rg1 Rc7 5. Rc1 Rg6 6. c3 bxc3+

Don't see a sure White win.
Rather think the fortress is stable [...]

Hi Peter,

quite a lot has been written since my post above, and I did not have the time to check everything, so I just want to explain my findings. Please apologize if someone else has already found something better for White in this interesting fortress-style position where engine-based analysis is difficult since engines always display "+-" even in those positions where there is actually a fortress for Black ...

After 1...Qf5 2.Nxf5 gxf5 3. Rg3 h5 I want to play:

4.Rf2 Rd7 (or Rg6 or another "do nothing" move) 5.Rf1 Rdg7 6.Rc1 Rd7 (this time the rook might be placed optimally on d7 to defend d6 against the queen possibly arriving at c5, and to eventually create threats against d3) 7.c3 dxc3+ (or bxc3+ - ignoring the rook sac would obviously be bad for black due to 8.c4! which wins the bishop and the game) 8.Rxc3 bxc3+ 9.Qxc3, and now I see no defense for Black, e.g.:

9...Rb8 10.Qxa5 Rxb3+ 11.Kc2 +- (the white king is safe on e3, White wins c5, can advance his passed pawn, can activate his rook, and Black needs to defend his king against the queen so he can't attack with both rooks - I hope everyone agrees this is a won position for White?)

9...Ra8 10.Qxc5 Kf7 (10...Rb8 11.d4) 11.Re3 Ke6 12.d4 +- (I hope everyone agrees ...)

9...Kf7 10.Qxc5 (10.e6+? +/= is tempting but gives away too much of the white advantage, even though it kind of removes the fortress condition) 10...Rb8 11.d4 +-

Any suggestions where this can be improved for Black (or even for White?)

[peter]

After 1...Qf5 2.Nxf5 gxf5 3. Rg3 h5 I want to play:

4.Rf2 Rd7 (or Rg6 or another "do nothing" move) 5.Rf1 Rdg7 6.Rc1 Rd7 (this time the rook might be placed optimally on d7

I don't think so, Sven, as I told Joseph already in another posting, I don't think white c3 nor c4 does work with a black Rook on b-line already.
So how do you get progress after 6...Rb7 instead of Rd7 (?) in your line?



1... Qf5 2. Nxf5 gxf5 3. Rg3 h5 4. Rf2 Rd7 5. Rf1 Rdg7 6. Rc1 Rb7

[peter]

I don't think white c3 nor c4 does work with a black Rook on b-line already.

Which doesn't mean of course, that one black Rook has to stay on b-line all the time, but if a white Rook appears on c- line, I would want a black one on a b-square

[jhellis3]

I don't think so, Sven, as I told Joseph already in another posting, I don't think white c3 nor c4 does work with a black Rook on b-line already.

Why and where does the black King have to move from e6 at all?

You are correct that c3 and c4 do not directly work. However, if the Black King does not move from e6, White can achieve a winning position similar to this:
[d]1r6/6r1/2p1k3/p1pbPp1p/Pp1pRPpP/1P1P2K1/2P2Q2/4R3 b - -

The winning plan for White is to sacrifice one of his rooks for Black's d and c pawns. The only way for Black to effectively stop this is for Black to achieve the following set up:
[d]1r6/2kr4/2p1b3/p1p1Pp1p/Pp1pRPpP/1P1P2K1/2P2Q2/4R3 b - -

Without the King on C7, the Bishop on E6, and a Rook on the D-File, Black very likely loses.

So the question of the original "fortress" becomes can White prevent the Black king from reaching C7 before it sacrifices its Rook on D4 for the D and C Pawns. And the answer appears to be yes, because White can move 1. Re2 from this position:

[d]6r1/6r1/2p1k3/p1pbPp1p/Pp1p1PpP/1P1P2R1/1KPQ3R/8 w - - 0 4

The Rook locks the Black King to E6, because if it moves White will play E6 and game over. So the Black King is stuck on E6, which means the Black Bishop is stuck on D5, which means White gets to sac a Rook on D4 and win the game.

Here is the position that will happen if Black does not move the king from e6:
[d]r2r4/8/2p1k3/p2bPp1p/Pp1Q1PpP/1P1P4/1KP5/4R3 b - -

[peter]

What's your winning line, Joseph?

[jhellis3]

What's your winning line, Joseph?

I didn't say it is winning with 100% certainty. I said it is very likely a win, and that the fortress (shown) does not hold. I do not know if the position I posted is the absolute best White can do, but it can certainly be treated as a lower bound.

It is not clear to me whether White needs to be concerned with the advancement of the g-pawn or not. If that turns out to be the case, then White will probably want to do the king walk to blockade the g-pawn first (it has time to do this as Black can only shuffle). If it doesn't matter then White can just proceed as I have already outlined, and/or possibly do the king walk later. It is not clear whether it is necessary for White.

Otherwise, just study the last position I posted, it shouldn't take too much effort to see that the situation is pretty dire for Black. I'm not really in the business of hand-holding though.

In some of my analysis, White could end up with a Queen and 3 isolated pawns vs 2 Rooks, which were just able to hold a blockade on the 6th rank. But letting the analysis sit for a while in antecedent positions the eval climbed ever higher. Black may have drawing chances, but they certainly are not very good ones, IMHO.