@ rimy
I would like a firm quote on this, rather than an opinion
A final visit to Hollis-Florian I already stated that an important part of my findings isnot new. The russian master I.Yavarev came to the same conclusions and published an article about his analysis in 2008. A fact I was not aware of.
The article in question is actually an internet article on the Russian website Chesspro:
http://www.chesspro.ru/_events/2008/tn.html Earlier I wrote: “The article is mainly concerned with variation A and B1, and contains less detail. Also the winning method in variation B.1.y diverges at move 11. Variations B.1.x and B2 thru D are not considered in Yanvarev's article.”
Looking at this again this is not exactly true.
There actually is a connection between the article and my line D.
Besides Yanvarev's article gave me an idea for an improvement over my previous analysis 
My article centred here around 5.Rc7 to show a win.
However Yanvarev gave the following sideline to
5.Kc2, Rf5? (5....Re3 draws)
6.Rc7, Rb5
7.b7, Ke6 8.Kc3, Kd5!? (In our posts we considered 8....f6 as the main line)
9.Rxf7,Rb6 Diagram 19 With a position very similar
but not exactly equal to the positions in line D.
Diagram 20 This is a position from my line D.
The only difference is,
that in the above diagram, White has the move, while in
the diagram from Yanvarevs analysis, it is Blacks move (after the continuation Rg7).
The above diagram provided the starting point for my variation D1. And I showed you in the previous posts that White wins with
[b][11]Kd3, Rb3+
[12]Kd2![/b]
One of my lines ran
[12] Kd4 Now Yanvarev considered the position after 12...Kd4 by transposition. And he mistakenly evaluated it as a draw. In my original analysis I showed that White actually does win after
[13]Rd7+!, Kc4
[14]Ke2!! +- The key move to make progress.
14.Rg7?, Kd4 Repeats the position after Blacks 12th move.
14.Rc7?,Kd5 15.Rg7 Repeats the position after White's 11th move.
----------------------------------------------------
Diagram 20 (repeated) Now Yanvarevs analysis gave me an idea to streamline the winning process and cut out some of the more complex variations, especially line D1.y.3. [11]Kd2! This is the improvement over my old main line (iso 11.Kd3)
It cuts out the complicated win in var. D1.y.3.
After [11]Kd2 we have
Variation E1 [11] Rb3
Variation E2 [11] Kc4
Variation E3 [11] Ke4!
Variation E1 [11] Rb3
With this move we return to Yanvarevs lines.
[12]Rxg6! (Yanvarev) I also analysed this position but only considered that White could play the triangulation 12.Kc2,Rb6 13.Kd3 transposing to my old main line (position after 11.Kd3 iso 11.Kd2)
[12] Rxb7 [12]....Rb2+?! [13]Ke3,Rxb7 only helps White, who now is a tempo up over the main line.
'
[13]Ke3! In fact we have now reached a position that I gave as winning in line D1.y.1. But the honours for first discovery of course go to Yanvarev.
Diagram 21 
[13] Rh7
[14]Kf4 Rh8
[15]Kf5 Rf8+!? Here Yanvarevs line diverges from mine:
I gave here (in the pgn) [15]....Rh7 [16]Rg8!+-
[16]Kg5 Rh8
[17]Rh6 Rg8+! Setting a nasty trap.
Diagram 22 [18]Kf6! Note that White could have spoiled everything with
[18]Kxh5??,Rxg3= draw [18] Rxg3
[19]Rxh5+ +- A well known technical ending has appeared on the board. And although the Black king is no longer cut off along the rank, there is just too much distance between this monarch and the h-pawn to keep the draw. [19] Kd6!?
[20]Rf5! Strangely enough the only winning move
After for instance
[20]Rh8, Rf3+ [21]Kg6,Ke7!= Blacks king is close enough to ensure a draw.
[20] Kd7
[21]Kf7! Rh3 [22]h5 Kd7 Note how White with his 20th to 22th move kept the Black king on the d-line. [22]...Rh1 [23]Rf6+! followed by h5-h6.
[23]Rd5+
White puts as much distance between the Black king and the pawn as possible. [23] Kc6
[24]Rg5+
The main line is the winning line of Yanvarev's analysis of the sideline indicated at the top of the article. Variation E2 diagram 20 repeated [11]Kd2
[11] Kc4 This is a position not considered in Yanvarevs analysis. [12]Ke3 [12]Rxg6, Rxb7 [13]Ke3 wins as well. We have reached support position nr.4.
[12] Kd5 [13]Kf4! Kd4
[14]Rxg6 Rxb7
[15]Rg5! Rh7 (Black last move is forced, otherwise White transposes to won endgame R+2p vs. R)
Diagram 23 [16]g4! Variation E3 [11] Ke4! This is the most stubborn defence.
Again this position was not covered in Yanvarevs analysis. [12]Rd7!
And now we have a further split
Variation E.3.x. [12]....Ke5
Variation E.3.y [12]....Kf3
[u]Variation E3.x[/u] [12] Ke5
Diagram 24 [13]Kd3, Ke6 On other moves White will continue Kc4!
[14]Rg7 Kd5
[15]Rxg6! I showed in my previous analysis that 15.Ke3 wins as well (See line D.1.y.1). But 15.Rxg6 was in my original analysis as well. [15] Rxb7
[16]Ke3! And we have transposed to variation E1 after white's 13th move (and are in fact also back in the side variation I gave in line D1.y.1.)
Variation E3.y [12] Kf3 This line has a connection with an allready previously presented piece of analysis: The refutation of Dvoretzky's drawing line against 5.Rc7. diagram 25 [13]Kc2!! The key move. White uses his rook on d7 for the indirect defence of the g-pawn.
We have seen this method in line B.
(the refutation of Dvoretzky's drawing line see diagram 6 thru 8): White threatens a side check, followed by posting his rook behind his own b-pawn.
Preferable when he can put his rook between the Black rook and the pawn. [13] Kg2 [13]...Kxg3? [14]Rd3+,Kxh4 [15]Rb3,Rxb7 ([15]...Rc6+ [16]Kb2 +-)
[16]Rxb7 is a technical win.
[14]Rg7! Now the g-pawn is truly immune. White gains time to move his king up the board.
[14] Kh3 We have in fact transposed to line B. [14]...Kxg3 [15]Rxg6+ with an attack on the b6-rook.
[14]...Rb5!? (to evade that Rxg6 will attack Blacks rook)
[15]Kc3!,Kxg3!? [16]Kc4!! wins ([16]Rxg6?+,Kxh4=)
This position was actually examined in line B.
See my correction post on Dvoretzky's drawing line. [15]Kc3! Rb1 [16]Kc4 +- Diagram 6 repeated [16] Kxg3
[17]Rd7!!+- Now white wins thx to the thread Rd3+ followed by Rb3.
[17]....Kxh4 [18]Rd4+,K? [19]Rb4 +- Note that
[17]Rxg6+?? (iso [17]Rd7)
,Kxh4= again leads to a draw.
And that is about all I have to say about this endgame.
I will post a pgn containing the final variations shortly.
Pages: [1] 
RookEndgame Hollis-Florian (Read 39587 times)
