In a position like this one, there of course are numerous ways of going about a win. The real crux is to get into positions like this first.

--

Going back to a different part of my life, this was something that always bothered me about self-defense classes. I saw much time spent on nuances of how to control an oponent that was already in a great disadvantage. But students were jumping for each other when they practiced throws.

Not a complete analogy to chess combat, true, as there is not realy an equivalent of "jumping" for practice oponent in chess, but our frequent discussions if a check-mate is more accurate because it comes a move earlier reminds me of it.

Kxd3 would string out the game more moves, because Black can eliminate the bishop with Ng1, which supports the b7-Ra8+ mating pattern against Black's Kb8.

I myself hadn't recognized the hidden mating pattern with b7, so there is some value in analysing these already decided sequences.

When computers evaluate a position, they assign important features of the game a numerical pawn equivalent. In addition to N=B=3P, R=5P, and Q=9P of the textbooks, computers assign a pawn equivalent value to open lines, rooks on 7-th rank, centralized position of Ns and Bs, passed pawns, bishop pairs, and so on. This is then all tallied up and used to evaluate the position over all. Eg., if White sacrificed two pieces for a rook (-1P), but has compensation in that his extra rook is on 7-th rank and Black king is under some sort of an attack, the overall evaluation may come out something like +0.75. This means that Fritz or Junior thinks that White is de-facto up 3/4 of a pawn.

I do not use computer engines, but these evaluations are a handy way of communicating position assesments. Positive numbers mean that White is up, negative that advantage goes to Black. I think 0.15P is more or less noise. Opening advantage comes out somewhere in the neighborhood of 0.25P. Between 0.5P and 0.75P, White has already obtained a serious advantage. In endgame, such an evaluation usually means that White is fightting for win, Black for a draw. At 1.0P, the stronger side is realy on the verge of winning, but possibly is not quite there yet. Over 1.2P, position is probably objectively won/lost with the best play for both sides. When an evaluation jumps from, say 0.6 to 2.75, you know that Black just blundered a piece or an exchange. And so it goes.

When I first went through this game I was wondering why Akiba went with 28.Qxf2! instead of 28.Kb3. It turns out that 28.Kb3 is a very inferior move to the one played.

Also, I couldn't help but laugh when I saw the final move of 29.Bc5! A simpler win for White would have been 29.Bxg8 but Akiba chose the very artistic Bc5! instead which leaves Black's Rook and Knight both hanging.

And the brilliant 25.Qb6! is reminiscent of Marshall's infamous Qg3! move.

Yea, Ive always found it strange that when Akiba opened with e4 and was meet by ...e5, he invariably replied with 2.f4

I thought a player like him would prefer something like the Lopez or maybe even the Giouco.

<Man i really like Rubinstein's great gems like this.. how cool is this game??? i daydream of playing like this> LOL. I wish I could play a game as marvellous as this too. So many moves in this game deserve exclamation marks IMO.

Moves 10, 12, 19, 20, 23, 25, 26, 28, & 29 all deserve an exclamation mark IMO.

He occasionally played other 1. e4 e5 openings -- this is a very fine Ruy Lopez, highly praised by Nimzowitsch in the tournament book.

19.Kc2! is a very nice move which gives the King the security of the pawn chain and also connects the Rooks.

20.Rf3! which sets up White for a double attack on the f7 and a7 squares.

23.g3! a nice little pawn sac which removes the Knight of the f-file.

25.Qb6! which is just brilliant and devastating.

28.Qxf2! is the simplest and most accurate way to continue.

29.Bc5! which hangs Blacks Rook and Knight.