Author: Marc Lacrosse
Date: 00:29:46 11/19/05
Go up one level in this thread
Although I must confess that Jeroen 's comment is very well written and pleasant to read, it is a bit incredible to see such a masterpiece of self-satisfaction by a book-cooker whose engine is dead-lost with white ONE MOVE after leaving its book! Moreover, what is said about current theory is purely and simply erroneous. So here are a few comments of mine (ML) with the corresponding JN comments : Fruit - Diep Dutch open 2005, round 3 ------------------------------------------------------------- 1.e4 c5 2.Nc3!? 2... d6! 3.Nge2 Nf6 4.d4 cxd4 5.Nxd4 a6 6.Be3 6... e5 7.Nb3 Be6 8.f3 Nbd7 9.g4 b5!? JN : "Diep repeats the line it played against Fritz in the WCCC some years ago. That could turn out to be a risky business." ML : 9...b5 used to be the main line a few years ago but most top-class players switched to 9..Nb6 which has had better results recently. 10.g5 b4 11.Nd5 JN :"An interesting moment. Databases show a higher score for 11.Ne2, but nowadays GM's agree this move is worse than 11.Nd5". ML : I do not agree. I have 33 master games 2003-2005 with the position after 10...b4. In 16 games white played 11.Nd5 and had only 41% success whereas 11.Ne2 was played in 17 games with a much better 62% success rate. The trend did not change in 2005 : Six master games so far . In two games white played Nd5 and lost both. In four games white played Ne2 and got 1.5/4 So where is the so-called grandmaster agreement for 11.Nd5? In fact the whole variation is extremely dubious and performed badly in recent master games. So I understand why you say that "11.Ne2 is worse than 11.Nd5" instead of saying that "Nd5 is better than Ne2" : but is it good book-cooking to be with white in an inferior position at move 11 ? 11... Nxd5 12.exd5 Bf5 13.Bd3 Bxd3 14.Qxd3 Be7 15.Rg1 O-O 16.O-O-O a5 17.Nd2 JN:" Of course the white knight is doing nothing on b3 and merely 'asks for' a a6-a5-a4 attack, so it hastens to approach the wonderful e4 square." ML: In the one and only master game that reached that position in 2005, super-GM Ivanchuk played 17.Kb1 here. Most interestingly his opponent who went on to win the game never played a5-a4 so it is not clear at all that leaving the b3 square is requested before Black plays a5-a4. In quite a few variations where black's e5 pawn moves or is exchanged, moving the b3 knight to d4 is a good option so it is not evident that moving it prophylactically is good. 17... a4?! JN: "Interesting: Diep avoids the known 17...f5. But this turns out to be a highly risky decision. Because it didn't put me out of book, on the contrary!" ML: "Interesting: Diep avoids the known 17...f5. It could well be a master decision not to go for the well-known f5 plan" 18.Ne4! JN:"A very clever move and of course a very natural one. The square e4 is a wonderful centralised post for the white knight, that can assist in defence and attack at the same time. Fruit played it from my book and I had still some variations in it. Now Diep was out of book and immediately Vincent started to look worried. I explained him that black needs f7-f5 in this line, but Diep wanted to play otherwise and showed a clear advantage for White here (I believe something like +0,5 for White, Vincent will correct me if I'm wrong). ML: most engines already evaluate this position as =+. The exclamation mark for 18.Ne4 is far from clear. 18... Qc7? JN : "A clearly second rate move. 18... f5! is the right move. But this was the clever side of 18.Ne4! When thrown out of book, no program wants to play 18... f5! as after 19.gxf6 the g-file is opened, with the rook on g1 'viciously' looking at g8. So progs refrain from the best move and now they all go for one plan: attack on the queen's side with queen, 2 rooks and the stormram a4-b4. In many test games I have seen this plan to be inadequate. And there is a simple reason for that: Black has only 3 heavy pieces to attack, but all white's pieces can assist in the defence. Furthermore, there is no way for Black to open the files, as b4-b3 can always be answered by cxb3!,axb3 a3! shutting the door and leaving black with no targets. Take a look at black's minor pieces: the bishop is passive at e7, while white's counterpart on e3 is very active. The knight on d7 would like to assist, but Nc5 or Nb6 will be answered by Bxc5 and Bxb6 respectively. In the remaining position the wonderful knight on e4 would be superior to the helpless bishop on e7. So the verdict is clear: without the move f7-f5 the attack initiated by black is doomed to fail. White has a clear advantage. Black will go for b4-b3 (they all play like that), white closes the position with cxb3!,axb3 a3! and calmly picks up the pawn at b3. After that Black is dead. I recall Pro Deo won many games in this line, against various progs like Fritz and Shredder. There is one 'but'. Fruit now HAS to play the move 19.Kb1! It is the only good move and the move that will preserve White's clear lead. Diep expected that move and showed a clear white plus." 19.h4? JN : " O, horror! A very natural move, but a bad one! The king had to leave c1, to make cxb3 possible after b4-b3." ML: How can you have Fruit making what you call a blunder at move 19 whereas you said after your 18th (!) move :"Fruit played it from my book and I had still some variations in it." That's the problem : you did not have 18...Qc7 in your preparation, and you did not see that Fruit woud react inappropriately against it. 19... b3! JN : "At the last moment, after calculating for several minutes, Diep comes up with this hammer blow. Believe it or not, but I think that now White is lost.... " ML : with a well-deserved crushing victory ! All this game is a perfect illustration of inadequate book-cooking : An interesting line but: - incomplete knowledge of the most recent theory and master games (see comments after move 11) - totally inadequate choice for the supported engine : Fruit blunders at the very first move after leaving the book! Marc Lacrosse
This page took 0.01 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.