Steve Berry RIP

ANYONE who plays chess regularly in the London area will probably have been as shocked as I was to learn of the death of Steve Berry at the age of 67.
The Wimbledon stalwart was taken into hospital shortly before Christmas, I understand, and died of an infection on January 5.
Steve was a Fide master with a peak rating, as far as I can discover, of 2379. In July 1999 he had an ECF grade of 232, and at the time of his death was ranked 88th among active players in England, and just outside the world's top 5,000.
In many years Steve was often the only credible challenger to grandmaster Keith Arkell's dominance of the annual Paignton congress in September.
Keith is among those who have paid tribute to Steve at the English Chess Forum, writing: "Because we shared similar tastes in tournaments, I spent many many evenings simply chatting with Steve over a pint, and on each and every occasion I found his company very agreeable."
My best memories of Steve are when he made an occasional visit to the Bad Wörishofen pre-Easter tournament in Bavaria.
As I recall it, he was not overly fond of the longish morning walks some of us went on, but was always happy to chip in with analysis when we were postmorteming in the evening. His good conversational German came in handy too.
I only played Steve once, at one of George Goodwin's monthly rapidplays in north London.
Berry (222) - Spanton (164)
Highbury Rapidplay 1993
Double e-Pawn Opening (Konstantinopolsky)
1.e4 e5 2.Nf3 Nc6 3.g3!?
In his Forum tribute, GM Arkell notes that "Steve had a very natural positional style," and this slow build-up move, which has been played by Topalov, reflects that.
3...Nf6 4.d3 d5
White's opening sequence is sometimes called the Konstantinopolsky Opening. According to Wikipedia(!), Black gets "an easy game with the natural and strong" moves I played. The online encyclopedia's analysis ends at this point.
5.exd5 Qxd5
More popular is 5...Nxd5, as played by Kramnik.
6.Bg2 e4?!
Black should probably just get on with development.
7.Nfd2?
But this is premature and helps justify Black's last move.
Best, according to Stockfish10, is the natural 7.Nc3 with the continuation 7...Bb4 8.Nd2 Bxc3 9.bxc3 Bg4 10.Nxe4! when the engine reckons White has a slight edge.
7...Bg4
Even better, according to Stockfish10, is 7...Qa5!, the point being that 8.dxe4 is met by 8...Bg4 9.f3 Be6, when Black's lead in development apparently gives more than enough compensation for the pawn.

How should White meet the threat to his queen?

8.Nxe4! Bxd1?
I should have castled long or moved the queen.
9.Nxf6+ gxf6 10.Bxd5 Bxc2 11.Be4?
Best was 11.Bxc6 bxc6 12.Kd2, when Black's bishop-pair is not sufficient compensation for having six isolated pawns, four of which are doubled!
11...0-0-0
This wins a pawn.
12.Nc3
No improvement is 12.Kd2 Bxd3! 13.Bxd3 Nb4 etc.
12...Bxd3 13.Bxd3 Rxd3 14.Be3 Bb4 15.Ke2 Rhd8 16.Rhc1 Bxc3 17.Rxc3 Rxc3 18.bxc3 Ne5 19.Rb1 Nc4?
Stockfish10 gives 19...a5 to keep White's rook off the fourth rank, from where it can harass Black's weak kingside.
20.Bd4?!
20.Rb4 Nxe3 21.Kxe3 gives excellent drawing chances. The text may have been an objectively misguided attempt to keep winning chances against weaker opposition.
20...Re8+ 21.Kd3 Ne5+ 22.Ke4
Stockfish10 gives White's best as 22.Bxe5 Rxe5, which may look similar to the previous note but here Black's rook is more active and White's less so.
22...Nc6+?
It seems that best was 22...Nd7, so that in the event of the game continuation of 23.Kf5, White cannot capture on f6 with his king and thus threaten f7.
23.Kf5 Nxd4+ 24.cxd4 Re2 25.f4 Rxa2 26.Kxf6 Rxh2 27.Kxf7

White is a pawn down but his active king and dangerous passed pawn make the position difficult for Black

27...a5?
I can no longer remember but I presume I did not appreciate the danger Black was in. Correct is 27...Rf2, significantly slowing down the passer.
28.f5 Rg2 29.Rb3?
The position is tricky, but Stockfish10 and Komodo9 reckon White is close to winning after 29.Ra1 Rxg3 30.Rxa5.
29...a4?
This should have lost, whereas the engines believe Black is hanging on with 29...Rf2.
30.Ra3?
Going passive. 30.Rf3 wins, eg 30...Rd2 31.f6 Rxd4 32.Ke6 Rd6+ 33.Ke5 etc.
30...b5
Now Black should be fine.
31.f6 Kd7?
But not any longer. Correct, as usual, was 31...Rf2.
32.Kg7
Winning was 32.Kg8! The difference between this and the text will become apparent at White's 35th move.
32...Rf2 33.f7 b4 34.Rxa4 Ke7
I offered a draw.
35.Rxb4
If White's king were on g8, White would win here with 35.Ra7 as 35...Rxf7?? would lose trivially to 36.Rxc7+.
35...Rxf7+ 36.Kh6 Kd6 37.Rb5 Ke6 38.Rc5 Rd7 39.Re5+ Kf6 40.Re4 Kf5
My scoresheet ended here, but the game was drawn.