|
< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> |
Aug-07-20
 | | NM JRousselle: This game reminds me of the very famous Pillsbury-Lasker game from 1896 where Lasker sac-ed each rook on a3. |
|
| Aug-07-20 | | saturn2: .I looked at 25 Rg6
25... hxg6 26. Qh6+ Kg8 27. Qxg6+ Kh8 28. Qh6+ K-
g8 29. Rg1
25...Bxd4 26. Qxd4+ Rf6 27. Qxf6+ Qxf6 28. Rxf6 25...Qe7 26. Re1 in this line did not see the quick win but after Rxf6 white would at least get back the exchange |
|
| Aug-07-20 | | Brenin: <King.Arthur.Brazil>: In line 10 of your post "After trying ...", after 29 Rg1 Bc8 (did you mean Be8?) White can play 30 Rg6 since 30 ... Qxg6 is answered with 31 Qxf8+ Qg8 32 Bxf6#. Black has no escape from mate or serious loss of material. |
|
Aug-07-20
| | Predrag3141: <King.Arthur.Brazi: However, for the sequence 25...♕e7 26. ♖xf6 ♖xf6 27. ♖e1 I suppose that Black would answer 27...♕f7! 28. ♕h6 ♖f8. Then, I can't see a clear win for white> Above I suggested 27 ... Qf7 28 Bh5, and I appear to have ruled out the two most obvious replies, 28 ... Qxd5 and 28 ... Qxh5. All I can see that's left is to lose a tempo, e.g. 28 ... Qg7 and the game continues much like actual play. |
|
| Aug-07-20 | | Everett: Really attractive finish. Interesting that Black had <24...f4!> at his disposal, delaying White’s Q’s access time the K-side while freeing his own LSB along the c8-h3 diagonal. |
|
| Aug-07-20 | | RandomVisitor: After 16.Bh6 black was fine, 16...gxf5 holds well
click for larger viewStockfish_20073110_x64_modern:
<52/38 03:20 0.00 16...gxf5 17.Bxf8 Qxf8> 18.Rab1 fxg4 19.Nd2 Bg5 20.Ne4 Nxe4 21.Bxe4 Qh6 22.Qa4 Kf8 23.Rxb7 Bxb7 24.Rb1 Bh4 25.Rxb7 Qc1+ 26.Kg2 Qf4 27.Qc2 Rc8 28.c4 h5 29.Rh7 Rxc4 30.Qxc4 Qxf2+ 31.Kh1 Qe1+ 32.Kg2 |
|
| Aug-07-20 | | mel gibson: No - didn't see this one at all.
Stockfish 11 says mate in 23:
25. Rg6
(25. Rg6
(♖g1-g6 ♗d7-e8 ♖g6xf6 ♕d8xf6 ♕d2-h6 ♗e8-g6 ♗d4xf6+ ♖f8xf6 h2-h4 ♔h8-g8
h4-h5 ♗g6xh5 ♕h6xh5 ♖f6-g6 ♖a1-e1 ♖a8-f8 ♖e1-e7 ♖g6-g7 ♖e7xg7+ ♔g8xg7
♕h5-g5+ ♔g7-h8 ♕g5-h6 ♔h8-g8 ♕h6xd6 ♔g8-g7 ♕d6-e7+ ♖f8-f7 ♕e7-g5+ ♔g7-h8
♕g5-h6 ♖f7-g7 d5-d6 ♖g7-g6 ♕h6-f8+ ♖g6-g8 ♕f8-f6+ ♖g8-g7 d6-d7 h7-h5
♕f6-h6+ ♔h8-g8 d7-d8♕+ ♔g8-f7 ♗f3-d5+) +M23/70 324) |
|
| Aug-07-20 | | Everett: <RV> it often seems that “winning” the exchange while giving your opponent the bishop pair is a mixed bag. There will at least be some compensation. |
|
| Aug-07-20 | | RandomVisitor: <Everett>Yes, in the above line black's pieces become active. Other moves do not work so well for black and in fact after 16...Rf6 white has the advantage. A longer run indicates that 16...gxf5 is likely best and holds for black: 
click for larger viewStockfish_20080713_x64_modern:
<73/54 3:45:51 0.00 17.Bxf8 Qxf8 18.Rab1 fxg4> 19.Nd2 Qh6 20.Ne4 Nxe4 21.Bxe4 Bg5 22.Rfe1 Qh5 23.c4 Bf4 24.f3 gxf3 25.Qf2 Bf5 26.Bxf5 Qxf5 27.Qxf3 Bxh2+ 28.Kg2 Bf4 29.Qh3 Qxh3+ 30.Kxh3 b6 31.a4 Kf7 32.a5 bxa5 33.Rb5 Kg6 34.c5 dxc5 35.Rxc5 Kf5 36.Rxa5 Rd8 |
|
| Aug-07-20 | | RandomVisitor: By the way, the Stockfish evaluation posted above is the new NNUE version, which is a neural net which has been 'trained' to play chess, the text below explains a little bit, translated from Japanese: "What is the NNUE evaluation function?
Chess and shogi have similar gameplay characteristics. If you take Stockfish’s search engine, which is the thought engine of chess, and bring it to Shogi, you will find that it works well. For the last five years or so, the developers of the shogi AI have borrowed a lot of ideas from Stockfish, with almost nothing to return to the Stockfish community. However, one of the things that the Shogi AI was unable to borrow from Stockfish was an evaluation function for that [game]. The evaluation function has evolved independently in Shogi AI community, and now the mainstream of Shogi AI evaluation function is called NNUE, using a neural network consisting of all the connections of a lightweight, shallow layer." As a summary, Stockfish has created a new version which is stronger, and uses this NNUE 'technology' borrowed from the Shogi AI community. Let's see how it performs on these chessgames puzzles. |
|
| Aug-07-20 | | jith1207: So, Shogi AI has offered something back to Chess, after all. |
|
| Aug-07-20 | | RandomVisitor: <jith1207>Yes, and what a gift it was. You have to download a *.nnue file and check a box in the Stockfish configuration to run the NNUE version, which I forgot to do, so moving forward over the next few days I will be posting Stockfish NNUE evaluations. |
|
| Aug-07-20 | | jith1207: <Random Visitor> excited to see your findings going forward. It would be great to know if it differs a lot from the latest Stockfish evaluations without NNUE. |
|
| Aug-07-20 | | RandomVisitor: Stockfish NNUE quickly finds the drawing line after 16.Bh6:
click for larger viewStockfish_20080713_x64_modern:
Found 510 tablebases
<NNUE evaluation using nn-5a3983417e4f.nnue enabled.> <47/37 02:52 527,043k 3,052k 0.00 16...gxf5 17.Bxf8 Qxf8 18.Rab1 fxg4> 19.Nd2 Bg5 20.Ne4 Nxe4 21.Bxe4 Qh6 22.Qa4 Kf8 23.Rxb7 Bxb7 24.Rb1 Bh4 25.Rxb7 Qc1+ 26.Kg2 Qf4 27.Qc2 Rc8 28.c4 Rxc4 29.Rb8+ Kg7 30.Rb7+ Kg8 31.Qxc4 Qxf2+ 32.Kh1 Qe1+ 33.Kg2 |
|
| Aug-08-20 | | RandomVisitor: Stockfish NNUE quickly finds the drawing line after 16.Bh6:
click for larger viewStockfish_20080713_x64_modern:
Found 510 tablebases
<NNUE evaluation using nn-5a3983417e4f.nnue enabled.> <55/50 19:33 4,284,379k 3,652k 0.00 16...gxf5 17.Bxf8 Qxf8 18.Rab1 fxg4> 19.Nd2 Bg5 20.Ne4 Nxe4 21.Bxe4 Qh6 22.f3 g3 23.Rfe1 Bf4 24.hxg3 Bxg3 25.Re2 Kh8 26.Rg2 Qe3+ 27.Kh1 Qf4 28.Bxh7 Bh3 29.Qg6 Qxf3 30.Qxg3 Bxg2+ 31.Qxg2 Qh5+ 32.Qh2 Qxh2+ 33.Kxh2 Kxh7 34.Rxb7+ Kg6 35.Rd7 Rc8 |
|
| Aug-08-20 | | TheBish: After 25. Rg6! Qe7, my first impulse was to play 26. Rxf6, but didn't immediately see the follow-up (I went back later to confirm that 26. Rxf6 wins as in the game). So I switched to 26. Re1 (also the engine's choice) and soon found that it wins in all lines. I believe only <agb2002> gave 26. Re1, but not the defense 26...Be5:
click for larger viewI'll leave this as a bonus puzzle for now. White to play (27.?) |
|
| Aug-08-20 | | saturn2: <The Bish>
The white queen on d2 is missing in your diagram.27. Qh6 looks good because black cannot defend the pawn d6. |
|
| Aug-08-20 | | 5hrsolver: 25. Rg6 Qe7 26. Rxf6 Rxf6 27. Re1 Qf7 28. Qh6 Rf8 29. Bh5 Qxd5+ 30. f3 |
|
| Aug-08-20 | | 5hrsolver: <TheBish>27. Rxd6 |
|
| Aug-08-20 | | jith1207: I read it as <5hrslover> ;=) |
|
| Aug-08-20 | | 5hrsolver: <Rg6 Qe7 26. Rxf6 Rxf6 27. Re1 Qf7 28. Qh6 Rf8 29. Bh5 Qxd5+ 30. f3> I just realized that my line is wrong because of 30... Qxd4!! 31. Bg6!! R8f7 32. cxd4 Rxg6. White may still have some advantage. Oh well. I should probably start now with the Saturday puzzle. |
|
| Aug-08-20 | | jith1207: 6 hours to go for Sunday puzzle to be up.
So you should probably start now on the Saturday puzzle :) |
|
Aug-08-20
 | | OhioChessFan: < I read it as <5hrslover> ;=) > Puts a 60 Minute Man to shame... |
|
| Aug-08-20 | | 5hrsolver: Smiles |
|
| Aug-08-20 | | TheBish: Whoops! Thanks <saturn2>, I'm usually pretty good at proofing, but posted that very late and missed it even after glancing again at my posted diagram. Here is the corrected diagram, with White to play (27.?)
click for larger view<5hrsolver> got the key move: 27. Rxd6!, followed by winning the queen after 27...Qxd6 28. Bxe5+, or even 28. Rxe5 with a deadly discovery coming. |
|
 |
|
< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> |
Kibitzing
sac: 26.Rxf6