| Jul-27-20 | | MordimerChess: Anti-Moscow Gambit in Semi Slav Defense. An interesting lesson about chess opening principles. Sometimes it's good to move your king into the safety ;) Ding Liren could go for very promising line 14...O-O-O with 15.Qd2 b4! attractive side line ... 13 games in DB, 6 won by black, 7 draws...
Full video analysis of this game and variations:
https://youtu.be/LZ48jRyp22s
Enjoy and learn! |
|
| Jul-27-20 | | savage sanctuary: Does black have better than <25...Nxa1?> Acc to the Chessgames engine (SF9), Sure does! (I say that sarcastically)... 25...g4 26. Re1+ Re7 27. Bd6 Nxa1 and now it gives the hilarious 28. Qxg4 Rhh7 (instead of blk mates 28...Rxe1#) and now it recommends 29.Qg6+ by white, not finishing the game immediately with Qc8#.
Man, that engine cannot be trusted! |
|
Jul-27-20
 | | beatgiant: <savage sanctuary>
I just checked it, and it says <27. Be6> and not <27. Bd6> as you posted above. |
|
| Jul-27-20 | | savage sanctuary: <bg> You are correct...
25...g4 26.♖e1+ ♖e7 <27.♗e6> ♘xa1 28.♕xg4 ♖hh7 29.♕g6+ ♔d8 30.♖xa1 ♖hg7 31.♖d1+ ♖d7 32.♕f6+ ♗e7 33.♗xd7 ♗xf6 34.♗xb5+ +6.82 Damn my eyesight! |
|
Jul-27-20
 | | AylerKupp: <savage sanctuary> You didn't indicate the search depth for the analysis. So I ran it and it reached d=20. A d=20 is pretty much useless for any engine in order to have any confidence in its evaluation, and for Stockfish even more so. Because of its aggressive search tree pruning Stockfish needs to be run to a depth of at least 35-ply and preferably over 40-ply in order to have any confidence in its evaluation. So you're right, at d=20 or d=25 that engine cannot be trusted. Fortunately, again because of its aggressive search tree pruning, it doesn't take Stockfish much time to get to these search depths. I let my Stockfish 11 analyze the position after 25.Bd5 and even on my relatively slow 32-bit computer it reached d=30 in less than 7 minutes. At that point it considered Black's "best" move to be 25...g4 (Which doesn't make any sense to me either; how does that help protect Black's king) with an evaluation of [+9.91], clearly lost. And it evaluated 25...Nxa1 as Black's 3rd "best" move with an evaluation of [+15.76]. But it doesn't matter much at this point, Black is clearly lost with its exposed king and White's 2 bishops. Did you get a chance to check out <MordimerChess>'s video listed in the first post on this page? Do so, it's enjoyable. According to <MordimerChess> Black's fatal mistake was 18...Nc5. He gives several variations which show that Black can sacrifice the knight and get sufficient compensation due to his extra pawns. |
|
| Jul-28-20 | | SChesshevsky: Seems like getting into this anti-moscow gambit against Carlsen is risky. Looks like, for a pawn, black is stuck with a weakened king side, especially light squares and with the king often stuck in the middle. For a guy like Carlsen who apparently loves to go after f7 and does it very successfully, opening seems tailor made. Carlsen vs A Groenn, 2005 Carlsen vs Van Wely, 2008 Carlsen vs M Matlakov, 2019 Think I mentioned in a previous post somewhere, might want to always keep that f7 square in mind when black against Carlsen. |
|
| Jul-28-20 | | Atking: <MordimerChess> Thanks for your analysis. Maybe on 14...0-0-0 15.Bh5 Rh7 16.Qb1!? Rg7 17.Rd1 then 18.b3 Obviously to open widely Black castle but also with an hidden attack on light square. For example 17...Be7 18.b3 b4 19.a5 Qa7 20.Na4 c3 21.Bxf7 RxB 22.Qg6 Rf8 23.Qxe6 White attack seems serious. |
|
| Jul-29-20 | | MordimerChess: <AylerKupp> Huge thanks for the recommendation and checking out my channel. Indeed 18...Nc5 is losing but it's understandable that 18.Be7 wasn't the first choice in rapid time control game. The best option is to sacrifice the piece for... almost nothing :D <Atking> Yes, 14...0-0-0 is the main line and with following 15. Qd2 b4 it's extremely attractive for black. Your 15. Bh5 looks interesting but check some insane lines like 15...Bf4 16. Bxf7 Nxe5 with attack on the bishop on f7, that would be worth to see in real game! :D Guys, sorry for the late answer, if you post your comments under the youtube video, I react much quicker as I can do it from my phone easier ;) |
|
| Jul-29-20 | | SChesshevsky: <... Maybe on 14...0-0-0 > I would've thought that long castle would be more common in this line after the ...h6 ...g5 weakening. But it seems that ...0-0-0 isn't a major or most common thought for black when trying to hold the pawn in this anti-moscow gambit. Wondering why as it does seem a logical move? Guessing it's just practically too unclear for black strategically. Firstly, black has taken on the weaknesses just to get the big pawn majority on the queenside. After ...0-0-0, advancing that pawn majority is trickier and potentially more dangerous with the king in the line of fire. Another consideration is if the king is actually safer after ...0-0-0. After a possible Bf3, you have both white B's directed at the king with a rook still on the a-file. At a minimum, it probably needs a lot of calculation for black unless he's done all the work before. For instance, I don't know if it's good but one of many possibilities Black has to consider is something simple like 14...O-O-O 15. Qc2 c5 16. Nxd5 Bxd5 17. axb5 axb5 18. dxc5 Bxc5 19. Bf3 and might have sufficient compensation for the pawn. Of course, it's easy for a computer to recommend ...0-0-0 as it sees that it's safe so quickly and accurately. But a main reason ...0-0-0 isn't as common as one would think is that it probably takes an awful lot of human calculation to evaluate and the result likely still ends up as unclear. |
|
| Jul-29-20 | | savage sanctuary: <AylerKupp>
Sorry for the late response, and thanks for your post. Clearly you are more knowledgeable at understanding AI language when running games/checking moves. But I'm learning. And yes, I did get a chance today at watching <Mordimer's> instructive vid. His vids are a blessing for us amateurs. And yes, 25...g4 doesn't make any sense to the human eye. |
|
| Jul-29-20 | | savage sanctuary: Adams vs R Edouard, 2020 This one is pleasing too. From the "hey black, leave your king in the center" collection. |
|
| Jul-30-20 | | MordimerChess: <savage sanctuary> Thanks. If you have any ideas what would you like to see or improve in my vids, let me know in the comments... I am still learning what is useful for players :) |
|
| Jul-30-20 | | Atking: Thanks <MordimerChess> to note to my attention the game of Gelfand! It's incredibly complex and aesthetic. A great player. I enjoy it. By the way I was impressed how the "veterans" played during during the first phase of this tournament. Except for Carlsen who dominates and Anand who wasn't at his best here, the veterans lost control under time pressure but strategically they were on the side teaching some lessons to the young. |
|
| Jul-31-20 | | MordimerChess: <Atking> Thank you for the support. I think it's also about computer things... if you were born with mouse in your hand, it's obvious that your natural environment favors you. Old masters were raised with OTB training. We could see that even Anand did some mouse slips and lost a couple of games in equal positions... |
|
Aug-03-20
| | AylerKupp: <MordimerChess> Thank you. I now finally have an excuse for all those games that I lose playing online. I just say "That wasn't a blunder, it was a mouse slip." Too bad that we can't say that about Fischer's 29...Bxh2 in Spassky vs Fischer, 1972. Or maybe we can? Pretty much everything else has been said about Fischer's reasons for that move. |
|
Aug-03-20
 | | moronovich: 29...Bxh2 bode for a wealth of research and interesting speculations.
Much better than if they had just taken the draw
in the completely equal position. |
|
| Aug-04-20 | | savage sanctuary: Top engines nowadays find nothing wrong with 29...Bxh2 |
|
Aug-04-20
| | AylerKupp: <savage sanctuary> We must be looking at a different page and have a different definition of what the "top engines" are. In Spassky vs Fischer, 1972 all of these top engines, particularly Stockfish, indicate that Black falls into what the engines consider to be a totally lost position evaluation-wise but they greatly overestimate White's winning chances because they don't recognize the fortress that Black can establish in the a8-b7 corner. And then Fischer throws the draw away by 39...f5. I guess it must have been a mouse slip. If you have any analyses that show different results, please post them at Spassky vs Fischer, 1972 . |
|
| Aug-05-20 | | savage sanctuary: <Ayler>
I had analysis that was done by, I believe Larry Evans in which in the end he found "nothing wrong with the Bishop capture." It is decades-old analysis examined by engines of today, unfortunately it's locked away in my storage unit, along with my other competition equipment from my tournament days, some 25 miles from my new address.
As soon as it's feasible to retrieve, I'll gladly look it over and post it. |
|
Aug-05-20
| | AylerKupp: <savage sanctuary> I'm personally not interested in decades-old analysis done by Larry Evans or anyone else, although others might be. What I would be interested in, as I'm sure others would be, is one or more top-engine analyses that corroborate that decades-old analysis that indicate that, as you said, there is nothing wrong with 29...Bxh2. And when you get the chance, please post those engine analyses in the Spassky vs Fischer, 1972 page, not this one. |
|
Kibitzing
sac: 25.Bxd5