doefmat wrote on 07/04/21 at 07:18:59:

I didn't find the argument weird. Your counter-arguments are worth considering, but consider also this scenario: Tomorrow Giri starts playing the Dragon against other top players. Would this new fact increase your confidence in the chessable repertoire, decrease your confidence, or make no difference?

A suspicious mind, verging on paranoia, is a positive trait for a chessplayer. The reason why a Dragon player might be suspicious that some blithe Dragon analysis will not hold up in practice is because they have seen it before. I am sure Giri knows a lot, and is a good analyst. What trips up an author in an opening they don't play is the lines they don't know and haven't analysed. In a sharp opening such as the Dragon, you want the author to be a fellow practitioner because they will have an eye for alternatives that need checking. There is also the classification of opening problems into those which can be figured out at the board, and those which can't. Here again, the practitioner of an opening is much more likely to know which is which.

Quote:


I have been putting some dedication to the line with 9.Bc4 Bd7 10.0-0-0 Rc8 11.Bb3 Ne5 12.Kb1 and 12...Re8 seems to hold it. But it is super complicated and you have to be ready to play variations such as 13.h4 h5 14.g4 hxg4 15.h5 Nxh5 16.Bh6 Kh7 17.Bxg7 Kxg7 18.fxg5 Nxg4 19.Nf5+ Bxf5 20.exf5 Rh8 21.fxg6 Nxg6 and here one of the lines goes 22.Nd5 Nf6!23.Rxh8 Qxh8 24.Qg2 Qd8! 25.Nf4 Qe8 26.Rg1 Qc6 27.Ne6+ Kg8 28.Qg3 d5 29.Nf4 Qd6!

Also there is a deviation with 16.Rdg1 Qa5 where after 17.fxg4 Bxg4 18.Rh4 Rxc3 19.bxc3 Nf6 20.Bh6 Nxe4 21.Qe3 Nxc3+ 22.Ka1 Bf6 23.Rhxg4 Nxg4 24.Bxf7+ Kh8 25.Qh3 Qh5 26.Qxc3 Nxh6 27.Bxe8 Qc5 28.Qxc5 dxc5 where with precise play black gets a draw.
Here 17.Bh6 Bf6 18.fxg4 Bxg4 19.Bg5 Bg7 20.Rh4 Rc4! seems to do the trick! Alternatively 20...Rc5 could be an option. But I didn't study it yet.

On 14.Bh6 Nc4! 15.Bxc4 Rxc4 16.Bxg7 Kxg7 17.Qe3 Qb6! seems to hold well after 18.Rd3 Rec8 19.Nd5 Nxd5 20.exd5 R4c5 21.Qe4 Qc7 22.Rc1 Re8 23.a3 e5! 24.dxe6 Bxe6 25.b3 b5 with equality. In this variation 17.Nd5 e5 18.Nxf6 Qxf6 19.Nb3! Be6! 20.Qxd6 Rec8 21.c3 R4c6 with compensation due to the bad knight and the activity of the black pieces.

I barely saw Giri's course. I wouldn't trust it. On 9.0-0-0 d5 10.exd5 Nxd5 11.Nxc6 bxc6 12.Bd4 e5 13.Bc5 Re8 14.Ne4 Qc7 he talks about 15.h4 f5. But what about 15.g4?
I do not recall to have seen Giri playing one single Dragon game. So you need to be careful with his material.

kylemeister wrote on 06/07/21 at 20:00:04:


Kruppa-Tiviakov, 1991 is basically given although he stops short of the end. 14.Ne4 is where the original analysis starts.

A bit later comment on that opposite coloured bishops endgame was "And this position is of course a draw, although the computer rightly points out that Black still has to be precise. Some key lines:"  with some analysis to move 35-40.  The indication is that the current computers are able to analyse the position to a draw.  How easy it is to hold OTB is another matter, but it doesn't look impossible to me at first glance.

gewgaw wrote on 06/07/21 at 19:15:44:


I'm not sure I'm really qualified but I would say they are far from final.  I believe that Giri has both deviated from his Najdorf course and beaten players who played his recommendations against him as white.  These are interesting new lines generated by a top playing pushing the current strongest engines on strong hardware, I'm sure in 5-10 years there will be some different ideas. Chess is not yet dead.

bragesjo wrote on 06/07/21 at 19:26:59:


Giri thinks 9.Bc4 Bd7 10.0-0-0 Rc8 11.Bb3 Ne5 12.Kb1! is just better for white so 11...Nxd4 is mandatory, but rather than learn all the Soltis stuff as well (due to 10.h4 h5 move order) it is better to just play Nxd4 earlier.

For the Kaufman line, Giri basically follows it to the end and says black is fine.  Kaufman was claiming +0.14 of an advantage so I would say that is OK...

Lifetime repertoar and in Dragon looks like an odd combo but since Bc4 recommendation is a side line it might work.

About 2 Nc3 g6 there are some lines you must know after 3 d4. For example Kaufmanns book went for something that looks like a standard  yugoslav attack but with the Knight at e2 instead of d4 thus all theory are gone.

Quote:

Sounds like += to me. Karpov in his heyday would have made it seem almost hopeless for black.

Quote:

That does seem interesting.  In my awareness, 13...Re8 has been considered as leading to += since the '90s (including in the recent ECO-B2).