Qc2 is absolutely stronger, no question about it.

Note that after 30.Qh2, black can gum up the works after Qxg6+ by playing Rf7 followed by Rd6.

1: Evgeny Vladimirov - Paul Van der Sterren, Ostend op 1990

click for larger view

Analysis by Rybka 2.3.2a mp : 20-ply

1. (2.62): 29.Nxg6 hxg6 30.Qc2 Rg7 31.fxg7 Bxg7 32.Qh2 Kf8 33.Ne2 Qe6 34.Ka1 a3 35.Rdf1+ Ke8

2. (1.13): 29.e4 dxe4 30.Nxe4 b3 31.a3 Bd7 32.f7+ Kxf7 33.Rxh7+ Bg7 34.Ka1 Bf5 35.Nf6 Kf8

OTB I'd not hesitate to play 29.Nxg6 and follow the breadcrumbs home. But it seems rather double-edged... Black may be able to return the piece for counterplay. There conceivably may be dozens of continuations - any one of which could go either way.

This kind of position definitely qualifies as "insane" - so I give up since I'm still stuck on (Friday's) Jaurgui-Mendes 1959 which must surely be more forcing - yet which continues to elude me.

I was surprised at how one-sided the attack went...I get a feeling there are some spoilers - involving not capturing the knight ...e.g. g3 or h3 attacking on the a-file or even Bc8

<dzechiel> Analysis after <33...Bxf7>:

1: Evgeny Vladimirov - Paul Van der Sterren, Ostend op 1990

click for larger view

Analysis by Rybka 2.3 mp 32-bit (20-ply):

1. (5.19): 34.Rdf1 Bd6 35.Rf6+ Kd7 36.Rxf7+ Kc8 37.Rxa7 Kb8 38.Rxa4 Bxg3 39.Rc1 Bd6 40.Qxg6 Qb7

2. (5.06): 34.Rhf1 Bd6 35.Rf6+ Kd7 36.Rxf7+ Kc8 37.Rxa7 Kb8 38.Rxa4 Bxg3 39.Rc1 Bd6 40.Qxg6 Qb7

3. (2.80): 34.Rc1 Ke7

4. (2.80): 34.Rd2 Ke7

I loved 30 Qc2 as well, but not for the reason above. To me, this move does two things. First, it threatens Qxg6+!, with the idea that after black follows with Bg7, then f7+ wins black’s queen.

It only really took a few seconds to figure out Nxg6. What else, really? Other than the direct line-opening sac (which enables the necessary additional leverage of any h-file reinforcement hitting h8 as well as h7), what could White do?

The only things that even suggest themselves are an h-file pileup (Qh2, maybe Rh4 and Rdh1), further line-opening with e4, and perhaps f7+. But what of them?

h-file pileup: 29 Qh2 or 29 Rh4 and Black will just guard with Be8. No more sac at g6, and h7 is not going to fall, especially not with both Black's other major pieces ready to guard the second rank if needed ...

Line-opening ... No dice here. We have Knights, he has Bishops. Superiority is definitely localized, in fact Black has the better of it on the rest of the board. 29 e4?? would give us a weak pawn at d4 and basically surrender the game eventually even if not now.

Finally, as for f7+ ... the pawn sac might have been something to consider in other circumstances (especially as this was a Sunday puzzle so we look for the non-obvious), in order to open piece access to the weak square at f6, but in this particular game it seems likely that it wouldn't ever be. Mainly this is because Black has the two Bishops and thus any weak square exploitation will be difficult, unless we devote both Knights to it so as to play Nxf6 after Bxf6 (however, this deprives us of other things like the chance to sac a Knight) ...

Meanwhile, as we saw the simple h-file reinforcement fails, it's clear we have to do something now; it is also clear that any attempt other than Nxg6 fails. But 29 Nxg6 looks pretty enterprising, especially since Black has no really good way to cover h8 so an invasion after hxg6 30 Qh2 is assured.

Material: 2Ns + P for 2Bs. White has a strong K-side attack against a weak dark-square complex on the Black K-side, with Pf6 and Pg5 forming a permanent wedge. Both Nf4 and Ng3 are near the Black K-side, covering the light squares, with the move 29.Nxg6 possible to open the h-file for Rh1. The White Qd2 and Rd1 can both join an attack along the h-file, so White has maximized his development. The main defenders of the Black K-side are Ra7, which protects Ph7 laterally, and Bf8, which covers the weak dark squares. The Bb5 can reinforce the K-side with 29…Be8, so 29.Nxg6 is a one-time offer.

Candidates (29.): Nxg6

29.Nxg6

Black might refuse the sacrifice, but then White has won another P and furthered his strategic aim of demolishing the Black K-side. Detailed analysis is unnecessary, therefore, because White can neither make a more aggressive move nor improve his defenses.

29…hxg6

Candidates (30.): Qh2, Qc2

30.Qh2, threatening 31.Qh8+ Kf7 32.Qh7+

I burned all my time on 32.Rh7+ Ke8 (instead of 32.Qh7+), but could not make it work. Time to peek and check the kibitzing. I will post with computer aid later, <dzechiel>.

I looked hard at 31.Qc2 Be8 32.Qh2, but I thought that 32...Bc6 reduces Black's liabilities by removing Qg6's discovered attack on the unprotected Qb6 with f7+. I missed the importance of being able to move Rf1 to back up Pf6, however. Unfortunately, I could not convince myself 31.Qc2 was better than 31.Qh2, because I was so hung up on 31.Qh8+ Kf7 32.Rh7+, stranding Rh7 in enemy territory. There are a number of pretty themes in that line, none of which quite work (although my imagination got a good work-out ;>)

The saying is not exactly as I remember it, but your version is even more wonderfully obtuse than I remember! Many thanks, <Samagonka>.

<Those who can't use their head must use their back.>

Dynamic Evaluation: Moving the ph7 would allow entry into the King's domain. g6 is a critical square. Moving the Q to g2 would attack d5 twice but it can easily be defended twice. Moving the Queen to h2 for an assault through the h file could be easily defended. Nf4 taking g6 should weaken the guard. But what of another way, a direct assualt.

29.Rxh7 ...Rxh7 30.Rd-h1 ...Rxh1 31.Nxh1 ...Qd6

32.Qc2

PM= That line goes down in flames with Rc8.

30. Qh2 Qa6 31. Qh8+ Kf7 32. Qh7+ Ke6 33. Qxg6 Kd7

click for larger view

The resulting position evaluates at (+1.75, 20-ply) and the expected continuation is: