In this game my opponent was Hans-Dietrich Rohde of Kirchheim, Germany.

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Na5 6.Bb5+ c6 7.dxc6 bxc6 8.Be2 h6 9.Nf3 e4 10.Ne5 Bd6 11.f4

11.d4 exd3 12.Nxd3 Qc7 13.b3 0-0 14.Bb2 is a major alternative for White.

11...exf3

11...Qc7 is playable, for example 12.0-0 0-0 13.Nc3 Bf5 14.a3 Nd5! Timman-Gligoric, Bad Lauterberg 1977.

12.Nxf3 0-0 13.d4 Qc7

13...c5!?

14.0-0 c5 15.Nc3 a6 16.d5 Bb7

With this move, I was following a couple of games by Lengyel.

16...Re8 17.Kh1 Rb8 18.a3 Ng4 19.h3 Ne3 20.Bxe3 Rxe3 21.Rb1 Qe7 was Estrin-Levinfish, USSR 1949. The judgment of Gligoric in ECO-2 is that Black has plenty for his pawn.

17.Kh1 Rad8

Possibly better is 17...Rfe8

A. 18.Bd2 Ng4! and White's 18th move appears to have been a waste of time (18...Rad8 see note to Black's 18th in the game; 18...Rab8 19.b3! with advantage to White);

B. 18.Nh4 Be5 19.Nf5 Rad8 20.Bxh6 Nxd5 21.Bd3 Nxc3 22.Qg4 Qc6 23.Bxg7 Qxg2+! was satisfactory for Black in Estrin-Letic, postal 1964.

18.Bd2!

This is new, so far as I know, and it is a very strong challenge to 16...Bb7.

Game position after 18. Bd2

18.Nh4 Be5 19.Nf5 Nxd5 20.Nxd5 Bxd5 21.Qe1 Rfe8 22.Qh4 Re6 was equal in Bagirov-Lengyel, Beverwijk 1965;

18.Be3 1/2-1/2 was Estrin-Lengyel, Lublin 1971.

18...Bf4

I played this move after coming to the conclusion that it was very important to prevent White from forming the battery Qe1, Bd2 against Black's rim-knight.

More natural, but apparently no better, was 18...Rfe8 and now:

A. 19.a3

A1) 19...Bf4 20.Re1 (20.Bxf4 Qxf4 is even) 20...Bxd2 21.Qxd2 Bxd5 (21...Nxd5 22.Bxa6 Nxc3 23.Rxe8+ Rxe8 24.Bxb7 Na4 25.Re1 favors White) 22.Nxd5 Rxd5 with an even game;

A2) 19...Nxd5? 20.Nxd5 Bxd5 21.Bxa5 Qxa5 22.Qxd5 Rxe2 23.Nd4;

B. 19.b3 Bf4! (19...Nxd5? 20.Nxd5 Bxd5 21.Bxa5 Qxa5 22.Qxd5 Rxe2 23.Nd4; 19...Bf8? 20.Qe1 and the rim-knight is quite vulnerable) 20.Bxf4 (20.Bc4? Nxc4 21.bxc4 Ng4!) 20...Qxf4 21.Bd3 Nxd5 with an even game.

C. But White can simply go ahead and play 19.Qe1! in view of 19...Nc4?! (19...Bf4 20.Bxf4 Qxf4 transposes back into the game) 20.Bxh6! Nxd5

Analysis position after 20...Nxd5

C1) 21.Nxd5 Bxd5 22.Qc3 (22.Bg5! f6 23.Rd1 see B2) 22...Be5! (22...Ne5 23.Bg5 f6 24.Bf4 Ng4 25.Bxd6 Qxd6 26.Bc4 favors White) 23.Nxe5 Qxe5 24.Qxe5 Rxe5 25.Bxc4 Bxc4 26.Bf4 Re2 with equality;

C2) 21.Bg5! f6 22.Nxd5 Bxd5 23.Rd1 fxg5 (23...Bf7 24.Bc1 and Black has nothing to compensate for his pawn) 24.Rxd5 Ne3 25.Rxg5 Bf4 (25...Nxf1? 26.Bc4+) 26.Qh4 c4 27.Rh5! Nxf1 28.Ng5! Kf8 29.Bxf1 and White has a very powerful initiative;

19.Bxf4 Qxf4 20.Qe1 Rfe8 21.Nd2

This is innacurate, but it makes no difference, since Black has nothing better than to allow the previous position to be repeated.

21...Qg5

21...Qd6 22.Qf2 Bxd5 23.Rad1 Qe5 (23...Re5 24.Nc4 Nxc4 25.Bxc4 also favors White) 24.Nf3 Bxf3 25.Qxf3 and Black doesn't have enough play to make up for his inferior pawns.

21...Qc7? 22.Rxf6 gxf6 23.Nde4

22.Nf3 Qf4

Black has nothing better than to repeat the position.

22...Qe3

A. 23.Rd1! Nxd5 (23...c4 24.d6) 24.Nxd5 Bxd5 25.Bxa6 is significantly better for White;

B. Less efficient would be 23.Bd3 c4 24.Qxe3 Rxe3 25.Bf5 g6 26.Bh3 Nxd5 and the game is approximately even.

23.Nh4!

This time White finds the best move.

23.Rd1 Qb4 24.b3 Nxd5 25.Nxd5 Rxd5 with approximate equality.

23...Qg5

23...Qc7 24.Rxf6! gxf6 25.Nf5 Qf4 26.Qf1! and the exchange-down ending is nevertheless very good for White.

Game position after 23...Qg5

24.Rf5

I believe White can obtain the advantage with 24.Qf2!

A. 24...Bxd5 25.Nxd5 Rxd5 26.Bxa6 Rd2 27.Qf4 Qxf4 28.Rxf4 Rxc2 29.Ra4 Nc6 (29...Re4? 30.Rxa5 Rxh4 31.Bd3) 30.Bb5 Rc8 31.Nf5 looks good for White;

B. 24...c4 25.Rad1 Bxd5 26.Nxd5 Rxd5 27.Rxd5 Qxd5 28.Nf5. Although the material is even, Black has nothing to compensate for his inferior pawn structure. Also, White's Knight is beautifully posted.

24...Qe3 25.Qf1

25.Rd1 Nc4 and the rim-knight gets back into the game.

25...Qd4

25...Nxd5? 26.Rxf7 Nxc3 27.Rxg7+! is a disaster.

26.Rf4

26.Nf3 Qb4 27.Rb1 Nxd5 and the game is even.

26...Qe5 27.Ra4 Nxd5 28.Nf3

28.Rxa5 Nxc3 29.Bxa6 (29.Bc4 Nd5 30.Rd1 Qf6 is also even) 29...Bxa6 30.Rxa6 Nd1 31.Nf3 Qe2. Black's more active pieces compensate for his material disadvantage;

28.Nxd5 Rxd5 29.Bd3 c4 30.Bxc4 Nxc4 31.Rxc4 (31.Qxc4? Bc6!) 31...Rd2 is at least equal for Black.

28...Qe6 29.Rxa5

29.Nxd5 Qxe2 30.Qxe2 Rxe2 31.Nc3 Rxc2 32.Rxa5 Bxf3 33.gxf3 Rdd2

Analysis position after 33...Rdd2

This was a key position in my reckonings, and I expected that the game might well go this way. White's only active rook is too far afield to prevent ...Rxh2+, after which repeated checks or threats of two-rook mates along the first and second ranks ensure the draw for Black, in spite of his missing piece.

29...Nxc3 30.Bxa6 Bxf3 31.bxc3

31.gxf3? Qb6;

31.Qxf3 Nd1 32.Bf1 Qe1 33.Ra7 Rf8 34.Kg1 Qd2 35.Bd3 (35.Kh1 Nf2+ 36.Kg1 Nd1=) 35...Qe1+ 36.Bf1 Qd2=

31...Ba8

Less clear but seemingly adequate is 31...Qb6 32.Ra3 Be4 and now:

A. 33.Bc4 Re7 34.Rb3 (34.Rb1? Bxg2+ 35.Kxg2 Rd2+ 36.Kh1 Qc6+ 37.Kg1 Qg6+ 38.Kh1 Qe4+ 39.Kg1 Qe3+ 40.Kh1 Qe5 41.Qh3 Qe4+ 42.Kg1 Qxc4) 34...Qg6 35.Qf2 Re5 and it appears that Black's initiative is worth more than White's pawn.;

B. 33.Bd3 Bxd3 34.cxd3 Rxd3 35.Qxd3 Qb2 with an approximately equal game.;

C. 33.Bb5? Bxg2+

32.Bd3

The game now finishes neatly with a piece sacrifice and perpetual check.

32.Bc4 Bxg2+ is similar to the game continuation.

32...c4 33.Bxc4 Bxg2+ 34.Kxg2 Rd2+ 35.Kg1 Qg4+ ½-½