The Official “Dragon Refutation” Forum
Sort:
NM
FizzyBand
Jul 29, 2020
0
#41
Chinese is decent but objectively White gets a nice advantage. Most people don't know the theory necessary to avoid the pitfalls though
Jul 29, 2020
0
#46
well I just love comparing engine strength, it's almost as if we humans forgot how to analyze games by ourselves.
FizzyBand έγραψε:
@pfren Yes most modern Dragon manuals give 11...Nxd4 because it is the sound move. 11...Ne5 is not terrible but a bit unsound. GM Gawain Jones wrote a chapter on it after 12.Kb1 (if u want to play 11...Ne5) and recommends 12...Nc4. He admits that it might not all be sound but gives many lines showing how Black can get decent practical chances.
Might well be more unsound than what Jones is assuming.
19.Rd3! TN (main idea is using the rook on the third rank as king shield) appeared at Correspondence games 5 years ago, and so far white has won every single game, but one OTB game, in that line.
Not very encouraging, right?
pfren wrote:
FizzyBand έγραψε:
@pfren Yes most modern Dragon manuals give 11...Nxd4 because it is the sound move. 11...Ne5 is not terrible but a bit unsound. GM Gawain Jones wrote a chapter on it after 12.Kb1 (if u want to play 11...Ne5) and recommends 12...Nc4. He admits that it might not all be sound but gives many lines showing how Black can get decent practical chances.
Might well be more unsound than what Jones is assuming.
19.Rd3! TN (main idea is using the rook on the third rank as king shield) appeared at Correspondence games 5 years ago, and so far white has won every single game, but one OTB game, in that line.
Not very encouraging, right?
OK, I’m not surprised that engine-aided White players are doing well in correspondence. My point about this line is not so much in regards to correspondence, where White has time, resources, and engines to help him defend accurately, but rather in a practical OTB game, where White has to find all the correct moves (often not just to keep the edge, but to not lose.) Anyways, is the position that horrible for Black after 19.Rd3...Rc8? Jones gives 20.Qd2...Qxc4 21. Rb3N (Nb3?!...Bc3!) and says that Black has good compensation after either d5 or a5
I agree that objectively 11...Nxd4 is best as it is fully sound, but I think that 11...Ne5 is a decent practical weapon as it is hard to find all the right moves without engine assistance, and even if White finds all the moves Black still has chances to hold.
Game stats can be very misleading, but that 100% score is clearly indicative of something being very wrong in this line.
Jul 31, 2020
0
#51
I think 33... Ng7 is an improvement in one of those correspondence games (21...e5).
The engine keeps jumping between +1 and +2 and I am not sure Black can hold it. Some of the lines lead to endgames where White is a full piece up for only one pawn but only with an h-pawn left. I have no idea if such endgames are holdable and no one would be surprised if White is easily winning but somehow the engine eval is not as high as one would expect in a piece up endgame.
Anyway for humans it's anybody's game. It's impossible for White to play perfect with such an open king.