|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |
Nov-21-19
 | | MissScarlett: Discrepancies in the sources for this game are addressed in C.N. 7040 and C.N. 10042. Curious that Winter should claim: <The worst treatment is on page 308 of The Games of Alekhine by Rogelio Caparrós and Peter Lahde (Brentwood, 1992). Alekhine’s brilliant conclusion beginning with 22 Qh5+ was not even mentioned...> when <Alexander Alekhine’s Chess Games, 1902-1946 by L.M. Skinner and R.G.P. Verhoeven (Jefferson, 1998)> doesn't mention it either. |
|
| Aug-13-22 | | Brenin: I would have played 17 fxe6+ Bxe6 18 Bxe6+ Bxe6 and only then 19 e5, but I guess that's why I'm not world champion, and AA was. |
|
| Aug-13-22 | | mayankk06: I was hoping 17 fxe6+ Bxe6 18 Bxe6 Kxe6 19 Qb3+ Ned5 20 exd5+ cxd5 21 Re1+ may work. Evidently I had no clue. |
|
| Aug-13-22 | | jrredfield: I nearly chose 17.Bxe6 but switched to 17.e5, seeing the wisdom of chasing the black N away. Still didn't see much of the full continuation leading to White dominating. Getting the King stuck toward the middle was the key. |
|
| Aug-13-22 | | mel gibson: I chose 17. fxe6+.
Stockfish 15 followed the text line 17. e5
for 3 minutes and then changed its mind to 17. fxe6+
This evaluation is for over 9 minutes on an 8 core CPU: 17. fxe6+
(17. fxe6+ (f5xe6+ ♗d7xe6 ♗c4xe6+
♔f7xe6 e4-e5 ♘f6-g4 ♕c2-e4 h6-h5 ♗a3xe7 ♔e6xe7 ♕e4-g6 ♔e7-d7 ♖f1-f7+ ♔d7-c8
♖f7xc7+ ♔c8xc7 ♖d1-e1 ♖e8-e7 ♕g6xh5 ♘g4-h6 ♕h5-h4 ♘h6-g8 ♕h4-g5 b7-b5 a2-a4
b5xa4 e5-e6 ♔c7-b8 ♕g5-a5 ♖d8-c8 ♕a5xa4 ♖e7-b7 h2-h3 ♔b8-a8 c3-c4 ♘g8-e7
♕a4-a5 g7-g5 d4-d5 c6xd5 c4xd5 g5-g4 d5-d6 g4xh3 d6-d7 ♖b7-c7 d7xc8♖+
♖c7xc8 ♕a5-a3 ♘e7-c6 e6-e7 ♖c8-e8 ♕a3-f3) +8.88/43 556) score for White +8.88 depth 43. |
|
| Aug-13-22 | | Amarande: Yeah, Black is just THAT busted.
Stockfish, now that I've a chance to run it on this game, even confirms that 20 Qh5+ at once does win, too (although it still prefers 20 Bb3). Ultimately White seems to wind up two exchanges up or with the lone Bd6 remaining after all other pieces are exchanged, for an eval of about +7. (A curious computer analysis moment as well: 17 ... Nfd5, which most books give claiming to be basically leading to mate in a few moves, actually isn't that bad, at least as far as moves in a position where all moves are hopeless is concerned. Stockfish seems to waffle between this or 17 ... Ned5 as best, with evaluations also in the +7 range in White's favor.) |
|
| Aug-13-22 | | jrredfield: <mel gibson Stockfish 15 followed the text line 17. e5 for 3 minutes and then changed its mind to 17. fxe6+> Interesting. I ran Stockfish 15 for a long time, all the way to depth 48. It never wavered from 17.e5. When I stopped it, it showed +8.11 for White. |
|
| Aug-13-22 | | mel gibson: < Interesting. I ran Stockfish 15 for a long time, all the way to depth 48. It never wavered from 17.e5. When I stopped it, it showed +8.11 for White.> OK -
I am running this version:
Stockfish_15_x64_avx2
What are you running? |
|
| Aug-13-22 | | newzild: I also ran the position through Stockfish 15, and it took just one minute to switch from 17. e5 to 17. fxe6+. That was on my clunky 8-year-old MacBook Pro. |
|
| Aug-13-22 | | mel gibson: < newzild: I also ran the position through Stockfish 15, and it took just one minute to switch from 17. e5 to 17. fxe6+. That was on my clunky 8-year-old MacBook Pro.> Very strange.
I only run 8 cores so the CPU goes to a bit over 50%.
I am using the Arena interface and I allocate 22 Gig of RAM
out of the 32 Gig I have for the Stockfish engine. What are you doing? |
|
Aug-13-22
 | | agb2002: I know this game.
The queen sac 22.Qh5+ is a bit famous. For example, position 31 in Reinfeld's "Brilliant Ways to Checkmate". |
|
| Aug-13-22 | | newzild: <mel gibson>
I don't really know what I'm doing, but my Mac is only quad-core with 16gb ram.
Also, it's showing a different line from you at the same depth: 17. fxe6+ Bxe6 18. Bxe6+ Kxe6 19. e5 Ng4 20. Qe4 h5 21. Bxe7 Kxe7 22. Qg6 Kd7 23. Rf7+ Kc8 24. Rxc7+ Kxc7 25. Rb1. Your computer gives 25. Re1. |
|
| Aug-13-22 | | mel gibson: <Your computer gives 25. Re1.> You're correct -
I get:
17. f5xe6+ Bd7xe6
18. Bc4xe6+ Kf7xe6
19. e4-e5 Nf6-g4
20. Qc2-e4 h6-h5
21. Ba3xe7 Ke6xe7
22. Qe4-g6 Ke7-d7
23. Rf1-f7+ Kd7-c8
24.Rf7xc7+ Kc8xc7
25. Rd1-e1 Re8-e7
26. Qg6xh5 Ng4-h6 |
|
| Aug-13-22 | | CaliWest: The engine says: "better is 15. g4"?? I never would play that. Why is it better? |
|
| Aug-13-22 | | CaliWest: I suppose because black's position is so cramped that it really doesn't create any weakness for white's king |
|
Aug-13-22
 | | chrisowen: I o pow bark eunuch fxe6+ arrive ogle pe5 fend affix principle baggy its ear oary boat take down fxe6+ egatz pe5 buff; |
|
| Aug-13-22 | | jrredfield: <mel gibson> I'm running the same Stockfish version as you. Maybe it has to do with some settings, such as hash table sizes or something else. |
|
| Aug-13-22 | | saturn2: 17 e5 wins back the piece under hopefully good circumstances. So far I came. As to 7...Bxc3 I watched a banter blitz by Radjabov when confronted by a Nimzoindian defence. His opponent played Bxc3 without⁸ necessity that is a3 and Radjabov blamed this as a loss of time. |
|
| Aug-13-22 | | saturn2: <played a number of grandmasters, and it is surprising how much more they apprehend in a position than we ordinary players, as well as how much more rapidly they get there.>
They apprehend the necessary things much quicker. Me I make cumbersome calculations of useless lines. Also I mix up different lines. I could never be a Top player. |
|
Aug-13-22
 | | perfidious: <saturn2....They apprehend the necessary things much quicker. Me I make cumbersome calculations of useless lines....> As Larsen noted in a 1972 interview with Hugh Alexander: <....Analytic power and positional insight vary very much for different Grandmasters. Smyslov and Petrosian never liked to analyze complex positions very much, Tal obviously does it very well and Korchnoi's whole play is based on analysis. If you are Smyslov or Petrosian then you have such positional understanding that you can avoid many complications and dangers and the need for so much analysis. Analytical power improves very much with practice; you learn to find what is critical in a position, what is worth analysing - you see which pieces are active, where the weak points are. When you don't know what to analyse, you sit there for a long time, picking variations almost at random. Tal and Korchnoi were probably born with a greater gift for analysis than I was and I was born with a greater gift than Smyslov and Petrosian; and in positional insight it is the other way round. Korchnoi is fantastic at calculating complex variations, especially when he is hard pressed; but he must analyze because his judgment when he doesn't calculate is very bad - he has to get through a lot of variations before he knows what's happening. Of course many things that worry the ordinary player are not problems at all for a Grandmaster; but in the difficult positions Spassky has said "Korchnoi is always wrong". Maybe a strong point of Fischer is that he is good both at analysis and in judgment - though Tal can calculate better and possibly Korchnoi also.'> Alexander, <A Book of Chess>, p 88 |
|
| Aug-13-22 | | Stanco: I found the move in 5 seconds.
e5!
Qf5+, f file, e6, Bd6 etc... |
|
| Aug-13-22 | | 2071 S Milwaukee: Ah but were you playing music? |
|
| Aug-13-22 | | stst: make use of the X-ray of RxN on f-file, rather than the long process of fxe+:
17.Bd6 (harass Q as prom-fake) QxB
18.e5 Qc7
19.fxe+ BxP
20.Qh7 (pinning N by R & Q) Rh8
21.RxN+ Ke8
22.QxR+ Kd7
23.BxB#
see how Alekhine did differently... |
|
| Aug-14-22 | | Stanco: depends on what band of frequency you consider awry |
|
| Sep-20-22 | | TheBish: <Amarande: After some analysis, it looks like 20 Qh5+ actually forces mate, without the need to wait two extra moves.
I'm not surprised Alekhine chose to prepare the sacrifice, however: there are MANY variations in this line, and when plumbing it with Ruffian 1 and WinBoard I am noticing horizon effects are ENORMOUS here. One move Black seems OK or just a bit down, next move it's duly evaluated at 327+. Here is the "primary" variation: 20 Qh5+ Nxh5 21 fxe6+ Kg6 22 Bd3+ Kg5 23 Rf5+ Kh4 24 g3+ Nxg3 25 hxg3+ Kxg3 26 Be2! g6 27 Rf3+ Kg4 28 Rf2+ Kg5 29 Ba3 Rf8 30 Bc1+ Rf4 31 Rxf4 Rf8 32 Rxf8+ Kh4 33 Rf4+ Kg5 34 Rf2+ Kh4 35 Rh2+ Kg3 36 Rd3#.> It seems that Black has a big improvement with 25...Kg4! (instead of 25...Kxg3). Then, according to Stockfish 14+ (lichess' engine), with best play White has about a pawn advantage after 26. Rf4+ Kg5 27. exd7 Qxd7 28. Rdf1 h5 29. Rf7 Nh6 (29...Qh3 30. Rxg7+ Kh6 31. Rff7 mates soon) 30. Rxd7 Rxd7 31. Be4! (only move to keep a slight edge) as Black can't play 31...Rxd6 32. exd6 Rxe4 33. d7 and the pawn queens. Alekhine did have the same winning queen sac one move earlier, but it takes a few moves more to mate than the way the game ended: 21. Qh5+!! Nxh5 22. fxe6+ Kg6 23. Bc2+ Kg5 24. Rf5+ Kh4 (24...Kg6 25. Rf6+ Kg5 26. Rg6+ Kh4 27. g3+ Nxg3 28. hxg3+ Kh5 29. g4+! Kh4 30. Kg2 followed by 31. Rh1#; or 24....Kg4 25. Rdf1 g6 26. Bd1+ Kh4 27. g3+ Nxg3 [or 27...Kh3 28. Rxh5+! gxh5 29. Rf4 and 30. Rh4#] 28. R5f4+! Kh3 [28...Kg5 29. h4#] 29. R1f3 and 30. Rxg3#) Position after 24...Kh4
click for larger view25. d5! (threat is 26. Rd4+) 25...c5 26. Re1 Ngf6 27. exf6 Rxe6 28. dxe6 Bc6 29. g3+ Kg4 (29...Kh3 30. Rxh5+ Kg4 31. Rh4+ Kg5 32. Re5+ Kxf6 33. Rf4#) 30. h3+ Kxh3 31. Rxh5+ Kg4 32. Rh4+ Kg5 33. Re5+ Kxf6 34. Rf4#. |
|
 |
|
|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |
Kibitzing
sac: 16.Nxf7