It is called the Beverwijk Variation of the Berlin Defence in the Spanish opening.

It was favoured by Boris Gulko (as you can see if you click the first link). In recent years, it has been played on a few occasions by Anand, Salov and Leko.

It was previously thought that 5.Nxe5 was the best reply, but eventhough it has never been refuted, many GMs today prefer 5.c3 because it is less unclear.

A. After 5.Nxe5:

I

5..Nxe5 6. d4

black can play 6..c6 or 6..a6 .

(a) 6..a6 became popular after J J Ady vs Spassky, 1984 and is the most common move today. White oftens plays 7.Ba4 to steer it into a variation of the Ruy Lopez (C78) , but better is 7.Be2 as in Fressinet vs C Ionescu, 2001 with a good game for white.

(b) 6..c6
After the older (but perhaps better) continuation 6..c6 the best play for white is 7. dxe5 Nxe4 8. Bd3 d5 9. exd6 Nf6 and here he can chose between 10.Re1+ (as in Fedorowicz vs Kaidanov, 1993), 10. Bg5 (as in J Benjamin vs B Finegold, 1993) or 10.Qe2+ (as in Krogius vs Spassky, 1959).

II

5..Nxe4

6. Qe2 Nxe5

(a) 7.d4 Be7 8. dxe5 Nc5 and being rather drawish.

(b) 7.Qxe4 Qe2 8.d4 Nc6 (after ..Ng6 white should play Bxd7+! , see Savon vs Alburt, 1970 ) and there are some unclear complications (see some as Gulko's games, for example Huebner vs Gulko, 1987 ).

B. After 5.c3

The most common continuation is 5..O-O 6. d4 Bb6 7. Bg5 h6 8. Bh4 d6 This wasn't played much by white earlier as the game was thought to be close to equal. However, it turned out that 9.Qd3! now gives white excellent chances. There are many examples in the database , e.g. Grischuk vs D Lima, 2000 , Anand vs Leko, 2000 and Bacrot vs I Sokolov, 2004 .

Here's why I ask. Black can play ...a6 and ...b5 (maybe not right away, but at least eventually) and force the bishop to b3. I consider those pawn moves to be to black's advantage. He gains space on the queenside and is now ready to develop his queen's bishop (though he reserves the option of developing it to the kingside later instead). And white's bishop is on the same diagonal when on b3 or c4.

I'm not really sure that the bishop is better off on b3 than on c4. So I don't see white's advantage in playing 3.Bb5.

Italian:
Opening Explorer

Ruy L.:
Opening Explorer

I would say one of the main advantages for white having the B on b3 as opposed to c4 is that the sting of a d5 push by black is greatly reduced. Also the pawn on b5 is not necessarily a strength, and can become a target after an eventual a4 by white. White will simply focus on the usual c3/d4 plan, and black's options are somewhat reduced. In some lines there are also tactics with Nxe5/d4

<If white doesn't intend to play the exchange variation, why play the Ruy at all instead of the Italian game?> It's all about sustained pressure against black's center (e5 pawn) which can lead to white maintaining the initiative longer than he/she can in the Italian game. Along with threatening to destroy black's center, white plans to play c3 and d4 to dominate the central territory. In the Italian game 1. e4 e5 2. Nf3 Nc6 3. Bc4 Bc5 4. c3 (<I used to play this quite a lot. White plans to maintain pawns on d4 and e4>) 4..Nf6 5. d4 exd4 6. cxd4 (<white has them>) 6...Bb4+ 7. Bd2 Bxd2+ 8. Nxd2 d5! (<but white can't maintain them and black should enjoy even chances in the game>)

<Here's why I ask. Black can play ...a6 and ...b5. I consider those pawn moves to be to black's advantage. He gains space on the queenside> Those moves gain space but do nothing to prevent white's best positional plan of playing c3 and d4 to dominate in the center.

<I'm not really sure that the bishop is better off on b3 than on c4. So I don't see white's advantage in playing 3.Bb5.> I'm sure you know that with 3. Bb5 white threatens to eventually capture the Nc6 (currently the only defender of e5) Black has to respond to that threat if he/she wants to maintain his share of the center (e5) and not give white a permanent positional advantage. Tarrasch showed in the 19th century (a long time ago :-)) that playing 3...d6 to guard the pawn will not allow black to hold onto his share of the center (e5) for much longer 1. e4 e5 2. Nf3 Nc6 3. Bb5 d6 4. d4 Bd7 (<breaking the pin on the knight and after 5. Bxc6 Bxc6 attacks the e4 pawn>) 5.Nc3 (<now the threat is Bxc6 and after black recaptures with the bishop the e4 pawn is guarded and e5 falls>) Nf6 (<developing and attacking e4 again to stop Bxc6>) 6. O-O Be7 7. Re1 (<now black has to play the somewhat anti-positional 7...exd4 (all of his previous moves were played according to the plan of maintaing e5) because of a forcing variation we all should memorize>) 7...O-O 8. Bxc6 Bxc6 9. dxe5 dxe5 10. Qxd8 Raxd8 11. Nxe5 (<e5 is gone>) Bxe4 (<black in turn captures e4>) 12. Nxe4 Nxe4 (<13. Rxe4 loses to 13...Rd1+>) 13. Nd3! (<blocking the d-file check possibility and exposing a rook skewer on the Ne4 and Be7>) 13...f5 (<protecting the knight>) 14. f3 Bc5+! 15. Nxc5 Nxc5 16. Bg5! (<wins material because black has no good answer to 17. Be7>) Tarrasch vs G Marco, 1892

<Also, why does black not play it this way often? 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 b5 5. Bb3 Bc5 -- a standard Italian game, except black has in two free moves for queenside expansion, and white can't play the Evans gambit (bishop is blocking the b-pawn).> After that sequence I believe that White will just play c3 and d4 with more space in the center and a strong position or he/she might transpose the game to the line below after 6. O-O Nf6

Recently I played a quick game against a low level computer; it was Ruy Lopez opening, and in the game, I offerred a pawn sac on the queenside, which turned out good. Fischer, in a game against Spassky in their 1972 match, let Spassky grab a pawn on the queenside for an attack Fischer vs Spassky, 1972

[Event "Playchess"]
[Site "Watford"]
[Date "2007.02.15"]
[Round "?"]
[White "Haeron"]
[Black "Guest"]
[Result "1-0"]
[ECO "C62"]
[PlyCount "43"]

1. Nf3 Nc6 2. e4 e5 3. Bb5 d6 4. d4 exd4 5. Nxd4 Bd7 6. Nxc6 Bxc6 7. Bxc6+ bxc6 8. O-O d5 9. Re1 dxe4 10. Rxe4+ Be7 11. Rd4 Qb8 12. Nc3 Bc5 13. Re4+ Ne7 14. Bg5 f6 15. Bxf6 gxf6 16. Qh5+ Kf8 17. Qxc5 Qd8 18. Rd1 Qe8 19. Rde1 Rd8 20. Rxe7 Qg6 21. Re8+ Kf7 22. Qe7# 1-0

1. e4 e5 2. Nf3 Nc6 3. Bb5 Nge7 4. Nc3 g6 5. d4 exd4 6. Nxd4 Bg7 7. Be3 a6 8. Be2 O-O 9. O-O d6 10. f4 f5 11. Bc4+ Kh8 12. Ne6 Bxe6 13. Bxe6 fxe4 14. g4 Ng8 15. Nd5 Nf6 16. c4 Nxd5 17. cxd5 Qe7 18. g5 Nd8 19. f5 gxf5 20. Bxf5 Bxb2 21. Rb1 Be5 22. Qh5 Rg8 23. Kh1 c6 24. g6 cxd5 25. Bg5 Qc7 26. Rg1 Nc6 27. Bh6 Nd4 28. Rbc1 Nc6 29. gxh7 Rxg1+ 30. Rxg1 Ne7 31. Qf7 Nxf5 32. Rg8+ 1-0

What is the variation called (Nge7->g6) ?

I had no idea how to reply to this, but Dvoretzky's ideas helped me. 7...a6 put me in a dilemma whether to exchange the Bishop or not. I decided not to exchange so the obvious move was Be2 because if Ba4 then b5, Bb3 Na5 and my is out. My idea was that if black plays b5->Bb7 then I can play f4->Bf3 challenging my opponent .

My aggressive intentions paid off as I was able to cause enough trouble to my opponent with my double s.

One example:

<To avoid the narrative flow being clogged up, or bogged down, with explanations of technicalities, the book concludes with a seven-page glossary of chess terms, such as the following (from page 166):

‘Ruy Lopez: ... This attack involves a penetration of the center by the king and queen’s pawns with the subsequent rapid opening of the bishop file, from which squares the knights may control the center. Direct and powerful attack meant to march down the board; it is easily defeated by a Sicilian defense in which the advance of the pieces is blocked. Now in partial disfavor, the Ruy Lopez has been displaced by the modernists. It is still often seen in amateur games. The Ruy Lopez is capable of producing a Fool’s Mate against an inexperienced or nervous player who might fail to note positioning of the queen and king’s bishop.’>

After 3...Nge7 it is easy to play because the next several moves are fairly automatic.