#26: Feller - Williams, Novi Sad 2009
22 Rxb1 !?
(!!) At many levels, this is undoubtedly a great sacrifice. However, a more human line would be 22 Qe1 Bc2, 23 Bxc6 Qxc6, 24 Qxe5 [White is clearly better]
Viktor Moskaleno, The Diamond Dutch (New in Chess, 2014)
An innocent enough comment. Until, that is, you see that the guy playing White was Sebastien Feller and remember that he was one of the Frenchies who was up to no good three years ago.
Turn off the stupid computer and analyse for yourself
A O CP - comments to TISE Confusion
Did Feller play this game without computer assistance? Or did his silicon friend whisper 22 Rxb1 in his ear? It doesn't really matter does it? Just the thought that he might have had some help is enough to ruin this game and all his others too.
Anyhoo, whether you consider the exchange sac unhumanlike or not, it's not too difficult to understand. The rook on a1 is not contributing to White's attack in the slightest, the bishop defends h7. Why not give up one for the other? I suppose the thinking is that it's not so much giving up five for three as much as it's trading nothing for something.
My own theory is that ISE's can be loosely divided into two types, dynamic and static ... [the former ] ... are much more difficult, because the compensation will or won't turn out to be enough according to various tactical details.
JC - comments to TISE Confusion
Clearly we have a Type One here. As before, understanding the nature of the compensation is one thing, knowing whether White has enough is quite another.
As it happens after 22 Rxb1 Qxb1+, 23 Kg2 Moskalenko thinks the GingerGM's ... Qg6 is a losing move and he suggests 23 ... Qd1 as an improvement.
I'll leave the resulting position, the calculation and the judgement as to whether there's sufficient compo or not, to you. Well, to you or to you and your electronic friend if you prefer.
2014 Exchange Sac count: 8