I was so close to playing 16.Bb5, and I still think the idea would have been quite good. Instead, 16.Rf1 was probably playable, but I hardly bothered to look at it - did not want to give up time or ground if I could help it. But returning to the exchange sac idea, this seemed to offer an obscure tactical fight of the kind I used to love in my younger days...

I was shocked that Bb5 wasn't played. Certainly was just the move I was thinking of.

This was always the planned follow-through. I half-expected Black might 'desperado' the bishop here and play 17...Bxf2ch, but thought a queen move probably more circumspect. Sure enough.

What a sweet move and a very powerful intermezzo. There is now long term pressure against the b7 pawn, the f7 pawn, the e7 pawn and black has weak dark squares around the king. Not only that but black still has two pieces that haven't moved yet and aren't likely to. Most likely queen will move, white will take one of the bishops and then the other bishop will find a better square. That is a long list of pro's just for an exchange sac. Sweet idea by white.

Now this game just got fun 😎

I was having a ball, here, as this is the type of game I like to play. That this Queen move attacks the knight didn't trouble me as the next move had already been decided.

Qd8 seems fine. I wonder about Qc8 as well keeping an eye on the light squared bishop. It seems like 17..Qc8 18. Bxe7 Re8 19. d6 is just not something that black is interested in at all with pressure on f7 a great pawn on d6 with a bishop on e7. maybe 19..Be6 holds but with 20. Qxb7 Qxb7 21. Rxb7 white is holding all the trumps.

All part of the plan. It was the dark squares around the Black King that had drawn my attention. Had the White Queen been standing at b2 right now, White would have been threatening mate in one.

Could the bishop not go to c3? I did think Black might opt for 18...Bc3 19.Rc1 Nd7 20.Qxc3 Qxc3 21.Rxc3 and try his luck in the end game. Probably White would have had an edge, but those connected Q-side passers would have been a worry. But the actual move to a5 was no surprise. +++++++++++++ Response to charleshiggie: I agree: I would most probably have kept the queens on, but I wanted to make clear that the notion of ...Bc3 wasn't totally out of court. But ...Ba5 was obviously best anyway.

I think after 18..... Bc3 19. Rc1 Nd7, 20.Rxc3 is an improvement on that analysis?

Line clearance.

You are very welcome!

Compare the activity of the pieces!

My kind of game 😎😎😎 Good going archduke_piccolo

After 20...e5, I was looking at 21.Rd1, or 21.Nh6ch Kg7 22.Rd1. But there was another defence that might have made things interesting: 20...h5!? 21.Bh6 e5 and then a choice between 22.Nxe5 and 22.Rd1. I agree with Charles, though: ...f6 was not the best, and was very pleased when I saw it.

It isn't often you get to see this sort of bayonet attack in the centre. On the wings, yes.

Yes the game is over here. It seems silly to resign with the extra material but...black looks close to dead.

This move wasn't as hard to find as some others were. But the problem with the sort of attack in this game, and it was not over yet, was choosing from a range of promising options. For instance, 23.Rc1 looks fairly attractive here. The purpose of this offer was to decoy the queen from guarding e5: 23.Bc4 Qxc4 24.Nh6ch! ... (forcing the King onto a dark square so that the next move is a check) 24... Kg7 25.Qxe5ch Rf6 26.Qxe7ch ... and now [A] 26....Rf7 27.Bf6ch Kxh6 28.Qe3ch ... and mates in a couple more moves; or [B] 26...Qf7 27.Bxf6ch Kxh6 28.Qe3ch ... etc.

The mark of the better players is having the ability to make the most of a losing position. Here, white does an excellent job of standing tall under immense pressure and finding the correct response to extend the game.