 |
| The tower looks impressive when first seen looming above you, but it was built as recently as 1895, albeit on a site that had been repeatedly fortified since pre-Roman times |
French Irregular
1.e4 e6 2.d4 d5 3.Nc3 Ne7!?
There are just 251 examples of this move in ChessBase's 2026 Mega database, but it has been tried by strong players.
 |
| How should White respond? |
*****
*****
*****
*****
4.Bd3
It is hardly surprising there is no out-and-out refutation, but the commonest continuation is 4.Nf3, which is also the top choice of Stockfish18 and Dragon1.
4...a6!?
The engines do not mind this.
5.Nf3 Nbc6!? 6.e5 Ng6?!
This seems to be a novelty - known moves are 6...Nb4 and 6...g6!? - and is strongly disliked by the engines.
7.h4 Bb4 8.h5 Nf8!?
The engines prefer this over 8...Nge7.
9.h6 g6 10.Rh4!? Be7!?
This may be best, but Black has lost time.
11.Rg4!? Qd7 12.Ng5
The engines prefer 12.a3 or 12.Kf1!?
12...f5 13.exf6 Bxf6 14.Qf3?!
This wastes much of White's advantage, whereas 14.Be3 leaves White with a positionally won game, according to the engines.
14...Qe7 15.Be3
 |
| How should Black proceed? |
*****
*****
*****
*****
15...Bd7?
Better is 15...Nxd4! 16.Rxd4! Bxg5, although 17.0-0-0 gives White more than enough for a pawn, according to the engines.
16.0-0-0
The engines point out the strength of 16.Nge4!. when 16...dxe4 can be met by 17.Nxe4, and if 17...Bxd4?!, then 18.Bxd4 Nxd4 19.Nf6+.
16...0-0-0
 |
| Both sides have castled long, but who stands better, and by how much? |
*****
*****
*****
*****
White's activity and pressure is worth a little more than 1.5 pawns, according to the engines, but only if White finds a sharp continuation.
17.a3?!
17.Kb1 keeps an advantage, but the engines give as best 17.Nf7! Qxf7 18.Rf4, one line running 18...Nxd4 19.Rxf6! Nxf3 20.Rxf7 Ne5 21.Rg7!?, when they reckon White has much more than enough for a pawn.
17...Re8?!
The engines reckon Black is fine after 17...Kb8!?, but note that 17...Nxd4?! 18.Bxd4 Bxd4 can be met by 19.Bxa6!? or 19.Nf7!?, in both cases with advantage to White.
18.Nh3
Best seems to be 18.Nge4!? dxe4 19.Nxe4 Nxd4 20.Qxf6, when the engines reckon White has a winning advantage.
18...Rg8 19.Rf4 Bh8 20.Rf7 Qd6
 |
| White to play and gain at least the upper hand |
*****
*****
*****
*****
21.Ng5?
Promising is 21.Bf4 e5 22.dxe5 Nxe5 23.Bxe5 Qxe5 24.Nf4 c6 25.Na4!?, when White's better development is probably the most important factor, but the engines also point out 21.Bb5!?, when 21...axb5? 22.Nxb5 traps the black queen, so best may be 21...g5!? 22.Nxg5 axb5 23.Nxb5 Qh2!? 24.g3 Bxd4!? 25.Nxd4 Ne5 26.Qf6!, when 26...Ng4!? 27.Ndf3!? Qxh6 28.Qd4 e5! (in order that Qa7 can be answered by ...Qa6) 29.Qxd5 Qc6 (29...Bc6? 30.Qa5) 30.Qa5 leaves White up a pawn and with a continuing attack.
21...Bxd4?
The wrong capture. After 21...Nxd4 22.Bxd4 Bxd4 23.Bxa6!? Qe5! 24.Rxd4! Qxd4 25.Be2 White has good compensation for the exchange, but, at best, a slight edge, according to the engines.
22.Bxd4 Nxd4 23.Qe3
Even stronger seems to be 23.Qf6!?, eg 23...Nf5 24.Bxf5 gxf5 25.Rg7!, when 25...Ng6 can be met by 26.Rxh7, and 25...Rxg7 26.hxg7 Ng6 by 27.Nxh7, in each case with a large advantage for White, according to the engines.
23...Nf5?
Better is 23...e5, but White has several promising replies, perhaps headed by 24.f4. Note that 23...Nc6?, which prevents White's game continuation, allows 24.Rg7 Ne7 25.Nce4! dxe4 26.Nxe4, when White has all the chances, eg 26...Qb6 27.Qxb6 cxb6 28.Nf6 Rxg7 29.hxg7 Rd8 30.Be4!
24.Qa7
 |
| After fiendish complications, at least in the notes, White has a winning queen sortie |
*****
*****
*****
*****
24...c6
If 24...c5, then 25.Bxf5 gxf5 26.Nxe6! Rxe6 27.Nxd5 wins, eg 27...b6 28.Rg7!? Rh8 29.Ne7+ Qxe7 30.Rxe7 Rxe7 31.Qxa6+ Kd8 32.Qa8+! Kc7 33.Qa7+ Kc6 34.b4! cxb4 35.axb4, when Black has rook, bishop and knight for queen and two pawns, but the hopelessly weak black king and scattered black pieces decide matters, one engine-line running 35...Re4 36.Qb8 Re6 37.Qg3, after which Black cannot reasonably save the king's rook.
 |
| How should White proceed? |
*****
*****
*****
*****
25.Bxa6?
Best is 25.Na4! I rejected it because of 25...Qf4+ 26.Kb1 Qxa4, missing that after 27.Qa8+ Kc7 White has 28.Qxe8! Also strong is 25.Bxf5 gxf5 26.Rg7, with ideas similar to those already covered.
25...Qb8 26.Qc5
The engines give 26.Bxb7+!? Qxb7 27.Qc5, when 27...Nxh6? runs into 28.Rd4!, so perhaps best is 27...Qc7, but 28.g4! Nxh6 29.Rd4! Nxf7 30.Nxf7 seems to give White a winning attack, despite being a rook down, eg 30...Rg7 31.Ra4! Rxf7 32.Nb5! Qb8!? 33.Nd6+ Kc7 34.Nxf7 Qb5 35.Qa7+ Qb7 36.Qe3!? c5!? (otherwise 37.Ra7) 37.Qxc5+ Qc6 38.Qa5+ Qb6 39.Qc3+ Qc6 40.Qe5+ Kb7 41.Nd6+ Qxd6 42.Qxd6 Bxa4 43.Qb4+ - another fiendishly complicated engine-line with many alternatives along the way.
26...Qd6
Dragon1 reckons 26...bxa6!? saves the game, but Stockfish18 continues 27.g4 Nxh6 28.Rf3, which Dragon1 comes to agree is winning for White, eg 28...Qc7 29.Na4 Qe5 30.Nb6+ Kc7 31.Qa5! Qxg5+ 32.Kb1 Kd6 33.Nc4+! Ke7 34.Qb4+ Kd8 35.Qb8+ Bc8 36.Na5 Re7 37.Nxc6+ Kd7 38.Rc3 Re7 39.Ne5+ Ke7 40.Qb4+! Kf6 41.f4!? The engines also point out that if, right at the start, Black tries to save a tempo with 28...Qe5!?, White has 29.Nce4!
[I doubt if either of us saw even a small faction of most of these lines.]
27.Qb6 Qc7 ½–½
