If Stockfish can't figure it out, than I certainly can't. If I was white I would just try to exchange rooks after Rc1, doubling the back rank.
Two extra pawns - enough for a win?!
Sort:
Dec 28, 2014
0
#3
...Possibly give up f2, exhange rooks, and see if the king can do more damage than your opponent's rook. To make progress you can't stay paralysised by the black rooks so that makes sense. Stockfish is better than all of us though.
Why not put in the suggestions from the computer as moves, that might help to clarify the situation.
This may be one of those endgames you look at and say "well how complicated could it be" that turns out to be insanely complicated, involving zugzwang and corresponding squares etc. My guess is it's a draw, white is merely one pawn up unless he wants to continue in perpetual check.
I'd bet huge it is a draw.
Wallacabayka je napisao/la:
If Stockfish can't figure it out, than I certainly can't. If I was white I would just try to exchange rooks after Rc1, doubling the back rank.
What do you mean? How to exchange rooks when Rc1 is met by rxf2# ?
J-Star-Roar je napisao/la:
...Possibly give up f2, exhange rooks, and see if the king can do more damage than your opponent's rook. To make progress you can't stay paralysised by the black rooks so that makes sense. Stockfish is better than all of us though.
That was my idea as well, but I wasn't able to find anything satisfactory for White, for example:
and Black will keep simuntaneously checking the white king and attacking his pawn.
HueyWilliams je napisao/la:
Maybe Black can draw this, but I doubt it.
The obvious question is, what did the comps suggest as continuations? Were they just shifting pieces back and forth pointlessly? The corollary to your remark about computers not being able to recognize fortress situations is that they will also act sometimes as though it is a fortress when it isn't.
It looks to me like White has two ideas here. Either he can immediately play Kf4 or he can try Rc5 and d5 with the idea of playing Kf4 then (with the d-pawn more advanced). That gets a bit complicated...so back to the immediate king move: 1 Kf4 Rxf2+ 2 Rxf2 Rxf2+ 3 Ke5 Rf3. And now 4 Rc7+ Kg6 5 Rc3 and it looks like Black will have a tough time dealing with the d-pawn.
Stockfish 5, for instance, showed +1.79 in all variations, but it just moved pieces aimlessly, as previously mentioned.
In the previous post I gave an analysis of the variation you have in mind and it ends in a draw. 4. ... Kg6 is probably possible as well, but safer is Kf8! so that the king doesn't get cut off.
And in the other case, :
and if Rc7, then ...Kf6
The beautiful complexity of R & P endings. If white could manage to trade off a pair of rooks, he would have some chances, but not gonna happen in this position.
Roby-Boby wrote:
Wallacabayka je napisao/la:
If Stockfish can't figure it out, than I certainly can't. If I was white I would just try to exchange rooks after Rc1, doubling the back rank.
What do you mean? How to exchange rooks when Rc1 is met by rxf2# ?
I assume he means Rcc1, otherwise notated as Rc1.
Anyway, this shows that the two pawns are balanced out by the rook pair on the 7th rank. Similar situation with a bishop pair, or a protected passed pawn,
It is necessary to understand how to assess imbalances in a position. You can't just look at the material differences.
Can I find a winning plan for white? Not for free....
It is a headache to look at, but if white can keep from repeating moves and force g7-g6 with the rook placed on H file... Maybe he can sac the f pawn & work the king up to pick off the backward forementioned g pawn... a real headache position... But if you want a winning plan, that is it...
"Can I find a winning plan for white? Not for free.... "
Stockfish works for free and it can't find one. But maybe with your awesome chess abilities you should be paid to find it. Give me draw odds, name your bet. This position is really clearly drawn and the insight here ought to be the power of doubled rooks on the 7th.
@Roby-Boby The final position of your line is a clear win for White after 13. Rg5 followed by capturing the h pawn.
I let Deep Junior & Deep Shredder look at it (they aint free) and both programs win by a timely sac of the f2 pawn along with the white king securing a post at g5.. The insight from here is, that you have two pawns to give to negotiate an edge with and still be equal in material... that has to be worth something.... The care to be taken is not to allow repetition...
Y'all ought stop restricting your brain cells to the shackles of the position...
Doesn't look difficult at all. Put the rook "away" first to have more liberties (say the a-file) and then let the f2 pawn go.
Black is dead lost in all cases.
Do not rely on engines to find a way out of such situations- their top suggestions would be nonsensical moves which keep the two-pawn advantage.
Roby-Boby wrote:
J-Star-Roar je napisao/la:
...Possibly give up f2, exhange rooks, and see if the king can do more damage than your opponent's rook. To make progress you can't stay paralysised by the black rooks so that makes sense. Stockfish is better than all of us though.
That was my idea as well, but I wasn't able to find anything satisfactory for White, for example:
and Black will keep simuntaneously checking the white king and attacking his pawn.
That ending position is a win for white...
I achieved this position as Black and offered a draw, which my opponent accepted a few moves later. I've been analysing it with help of Houdini 1.5a and Stockfish 5, and they both gave White a prevalent advantage. However, I wasn't able to find a win for White and I think this position is a "fortress" and that's why computers can't evaluate it properly. What do you think of it? Can you find a winning plan for White?
It's White's move