The Old Firm

Even in the online format, matches between Aberystwyth and Cardigan have something of the feel of an Old Firm rivalry, and the match in the Dyfed Online League on Tuesday 23rd February was no exception. Top boards Rudy van Kemenade and Howard Williams have had many fierce fights over the years, so it was only to be expected that this latest meeting would be a complex struggle. In a sharp King's Indian, Rudy went a pawn down and had to seek piece activity in compensation. Eventually the only way out was to give up a piece with the aim of eliminating all White's pawns, and, with Howard short of time and having to defend against mate threats, this was enough to earn Rudy the draw. In contrast, Adam Watkin-Jones's game against Ben Brewer was over quickly, as he overlooked a threat to his h pawn in a Slav, and resigned in disgust, though there was still plenty of play left. Once again I had some positional problems with my Dutch Stonewall, this time against Gareth Williams. When he offered a piece for a couple of pawns and an unbalanced position, I should probably have taken it, but I hadn't seen very far into the sequence and passed it up, after which the game settled into equality. Tom Gunn's Greek gift sacrifice in his Colle System wasn't sound, but Joshua Brewer didn't find the best defence, and the White attack broke though for the win. The match was drawn, 2-2.

[Event "Dyfed Online League"] [Site "Lichess"] [Date "2021.02.23"] [Round "6.1"] [White "Williams, A Howard"] [Black "Van Kemenade, R."] [Result "1/2-1/2"] [ECO "E97"] [WhiteElo "2340"] [BlackElo "1999"] [Annotator "kemen"] [PlyCount "108"] [EventDate "2021.??.??"] [WhiteTeam "Cardigan"] [BlackTeam "Aberystwyth"] {[%evp 0,108,19,31,14,30,45,39,20,40,37,44,33,41,42,39,73,35,1,18,4,-25,101,60, 56,62,62,61,80,76,108,52,55,41,45,5,24,24,24,10,91,126,204,204,197,217,208,208, 208,90,80,80,86,61,92,74,77,86,79,83,87,92,127,101,99,59,47,70,85,7,22,73,72, 76,40,64,17,27,52,28,5,0,0,0,0,0,0,0,0,0,41,49,109,126,194,197,190,132,112,91, 91,91,109,109,94,94,74,74,69,69,69]} 1. d4 Nf6 2. Nf3 g6 3. c4 Bg7 4. Nc3 O-O 5. e4 (5. Bg5 c5 6. e3 cxd4 7. exd4 d6 8. Be2 Nc6 9. d5 Nb4 10. O-O Na6 11. Nd4 Nc5 12. Be3 {Williams-Pleasants, Dyfed Open 2020, 1-0}) (5. Bg5 d6 6. e3 h6 7. Bh4 Nc6 8. d5 Ne5 9. Nxe5 dxe5 10. Be2 Nd7 11. e4 f5 12. f3 c5 13. d6 Bf6 14. Bf2 exd6 15. Qxd6 Qe7 {Williams-Van Kemenade,Dyfed League 2015,1-0}) 5... d6 6. Be2 e5 7. O-O (7. d5 Na6 8. O-O Qe8 9. Nd2 Nd7 10. Rb1 f5 11. b4 Nf6 12. c5 f4 13. Ba3 Rf7 14. Nc4 Bf8 15. Na5 g5 {Williams-Van Kemenade,Dyfed League 2008,1/ 2-1/2}) 7... Nc6 (7... exd4 8. Nxd4 Nc6 9. Nc2 Re8 10. f3 Be6 11. Bg5 { Williams-Davis,WCPL 2016,1-0}) 8. Be3 {Gligoric variation} (8. d5 Ne7 9. Ne1 Nd7 10. Nd3 f5 11. Bd2 fxe4 12. Nxe4 Nf6 13. Nxf6+ Bxf6 14. Bf3 Bf5 15. Qe2 c6 16. dxc6 Nxc6 17. Bc3 Qb6 18. Rad1 {Williams-Botterill, Wales 1980,1-0}) 8... Ng4 9. Bg5 f6 10. Bh4 h5 (10... g5 {main}) 11. d5 (11. h3 {main} Nh6 12. dxe5 dxe5 13. Rb1 a5 14. Qxd8 Nxd8 15. Nd5 Ne6 16. Rfd1 Rf7 17. Rbc1 Nf4 18. Bf1 Be6 {Vickery-Van Kemenade, Leeds-Huddersfield 1988,0-1}) 11... Ne7 12. Nd2 (12. h3 Nh6 13. Nd2 f5 14. f3 f4 15. b4 Nf7 16. c5 g5 17. Bf2 Nh6 { Tolatito-Hearthstones9, Playchess.com 2013,0-1}) 12... Nh6 13. f3 g5 {main} ( 13... f5 $5 {1.5-2,5}) 14. Bf2 f5 (14... Ng6 {alternative, 2 Wh wins, 2 draws, 1 Bl win}) 15. exf5 {1 win each} (15. c5 {3 Wh wins, 1 draw,1 Bl win}) 15... Nhxf5 16. Nde4 Ng6 17. c5 Nd4 (17... Nf4 18. cxd6 cxd6 19. Rc1 a6 20. a4 (20. Re1 Nd4 21. Bf1 g4 22. fxg4 Bxg4 23. Qd2 Bh6 24. Kh1 Nf3 25. Qe3 Nxe1 26. Qxe1 Rc8 {Saigin-Kalinichenko, Spartakiad Leningrad 1967,0-1}) 20... Kh8 { Yakovlev-Kochurov,RUS ch tm U18, Orsk 2002,1-0}) (17... Nf4 18. Kh1 Nd4 { LC0 & Houdini3}) 18. cxd6 cxd6 19. Nb5 Nxe2+ (19... Nxb5 {is what Bl meant to play, but found himself taking the B} 20. Bxb5 Nf4 $11 {Stockfish13}) 20. Qxe2 a6 {Stockfish13 doesn.t like this move} (20... Nf4 21. Qd2 Rf7 22. Nbxd6 Rd7 23. Nxc8 Rxd5 24. Qc2 Rxc8 25. Qb3 Kh8 26. Qxb7 g4 $16 {Stockfish13}) 21. Nbxd6 $18 Nf4 22. Qd2 g4 {Bl just has to keep going keeping pieces active in hope of some compensation} 23. fxg4 (23. Nxc8 {is worth consideration to deprive Bl of the B pair}) 23... Bxg4 24. Nxb7 {not liked by lichess Stockfish} (24. h3 Nxh3+ (24... Bd7 25. Kh2 (25. Nxb7 Qe8 26. Bc5 Qg6 27. Bxf8 Rxf8 28. Nbc5 Bxh3 29. Rf3 Bg4 30. Raf1 $18 {Stockfish13})) 25. gxh3 Bxh3 26. Rfe1 $18 {Stockfish13}) 24... Qxd5 25. Qxd5+ Nxd5 26. h3 Bf5 (26... Rab8 27. Nbc5 Be2 28. Rfb1 Nf4 29. Nd7 Bd3 30. Re1 Rxb2 31. Nxf8 Bxf8 $16 {Stockfish13}) 27. Nbd6 Bg6 {the Ns are a bit clumsy protecting one another, yes, but they both need the same squares. Bl was hoping to get a N to d4, while keeping the two Bs} 28. Rad1 Nf4 (28... Nb4 29. a3 Nc2 30. Rc1 Nd4 {Stockfish13- is a quicker route there}) 29. Kh2 Rab8 (29... Ne6 {original idea, but the K on h2 gave Bl some ideas of getting to the 7th}) 30. b3 Rb4 {another Engine dislike} (30... Ne6 31. Be3 Nd4 $16 { still possible, but Wh retains the pawn advantage, Bl's g7 B is trying to get out, but for the moment it is the usual staunch defenfer of the K}) 31. Rfe1 ( 31. Nc5 $1 a5 32. Be3 $18 {Stockfish13}) 31... Ne6 32. Bc5 {looks forcing, but Bl now has 2 Bs, which threaten to become active} (32. Kg1 $16 {Stockfish13}) 32... Nxc5 33. Nxc5 Rf2 {going for practical chances- the hyper-correct Engines dislike it, and Wh's reply} (33... Rd4 {was looked at by Bl , not thought forcing, but it is the Engine preference} 34. Ne6 Rxd1 35. Rxd1 Rf2 36. Nxg7 Kxg7 37. a3 Rb2 38. b4 Rb3 39. Nc4 e4 $11 {Stockfish13- the Wh pieces are in defensive positions}) 34. Nd3 (34. Nc4 Rxa2 35. Rd8+ Kh7 36. Rd7 Kh6 37. Ne6 $16 {Stockfish13-keeps Wh's advantage}) 34... Bxd3 35. Rxd3 Rxa2 (35... Rd4 36. Ne4 Rxd3 37. Nxf2 Rd2 38. Ne4 Rxa2 $11 {Stockfish13- most accurate sequence}) 36. Nf5 (36. Rg3 {Stockfish13-best execution of Wh idea} Kf8 37. Nc4 $14) 36... Bf6 $11 (36... e4 $2 {trying to get that ch in, fails} 37. Rg3 Rb7 38. Rxe4 $18 ) 37. Rd6 Kf7 38. Rc1 Rf4 {Bl is holding on- the B a key protector, & still hoping to get e4 in} 39. Rc7+ Kg6 40. Ne7+ Kg5 41. Nd5 (41. Rd8 Bxe7 42. Rxe7 Rf5 $11 {Stockfish13}) 41... Rff2 42. h4+ (42. Rxf6 Rxf6 43. h4+ Kf5 44. Nxf6 Kxf6 45. Rh7 Kg6 $11) 42... Kxh4 {other moves lose, Bl just has to go for getting rid of all Wh's pawns (with the odd mating possibility for either side} (42... Kg6 43. Kh3 $18 (43. Rcc6 Rxg2+ 44. Kh1 Rh2+ 45. Kg1 Rhg2+ 46. Kf1 Rgf2+ 47. Ke1 Rfe2+ 48. Kd1 Red2+ 49. Kc1 Rg2 50. Rxf6+ Kg7 {and it seems Bl still has a draw!}) 43... Rab2 44. Rb7 Rbd2 45. Rbb6 $18 {Stockfish13- careful to the end}) 43. Rxf6 (43. Nxf6 Rxg2+ 44. Kh1 Rh2+ 45. Kg1 Rag2+ 46. Kf1 Rf2+ 47. Ke1 Re2+ 48. Kd1 {running too far into the middle will lose, as then both Rs have extra space at both ends} (48. Kf1 {draws}) 48... Ra2 49. Rc4+ e4 $1 ( 49... Kh3 {draws} 50. Rd3+ Kg2 51. Rc2+ Rxc2 52. Kxc2 Rh1 {and the h pawn saves Bl}) 50. Rxe4+ Kg5 51. Rc6 Ra1+ 52. Rc1 Rh1+ 53. Kd2 Raxc1) 43... Rxg2+ 44. Kh1 Rgd2 (44... Rh2+ 45. Kg1 Rag2+ 46. Kf1 Rb2 {simplest- Wh must return to g1 immediately or lose, according to Stockfish13}) 45. Ne3 Rd3 $6 (45... Ra1+ {is the most accurate way to equality- it was considered by Bl} 46. Nf1 ( 46. Rf1 Rxf1+ 47. Nxf1 Rd3 48. Rc6 (48. Rb7 {Bl thought this might be fine for Wh, but then the Wh R is out of action} Kg4 49. Kg2 h4) 48... Rxb3 49. Rxa6 $11 ) 46... Rb1 47. Rc3 Re2 48. Kg1 Kg4 49. Rxa6 {looks tricky, but Stockfish13 is content that this is a drawn position}) 46. Rc4+ Kg5 (46... Rd4 $5) 47. Rf5+ Kg6 48. Rxe5 $6 (48. Rc6+ $1 Kg7 49. Rxe5 {gives Wh the winning edge} Kf7 ( 49... Rxb3 50. Nf5+ {in the game Bl is able to avoid this} Kf8 51. Rf6+ Kg8 52. Re8+ Kh7 53. Rh6#) 50. Rc7+ Kf6 51. Rce7 Rd6 52. Re8 Kf7 53. R5e7+ Kf6 54. Re4 $16 {but this still looks far from over}) 48... Rxb3 $11 49. Rc6+ Kf7 50. Rf5+ Kg7 51. Rc7+ Kg6 52. Rc6+ Kg7 53. Rc7+ Kg6 54. Rc6+ Kg7 {Wh had 3 mins left to Bl's 27} 1/2-1/2 [Event "Dyfed Online League"] [Site "Lichess"] [Date "2021.02.23"] [Round "6.2"] [White "Watkin-Jones, Adam"] [Black "Brewer, Ben"] [Result "0-1"] [ECO "D12"] [WhiteElo "1918"] [BlackElo "1679"] [Annotator "kemen"] [PlyCount "20"] [EventDate "2021.??.??"] [WhiteTeam "Aberystwyth"] [BlackTeam "Cardigan"] {[%evp 0,20,80,29,30,20,30,13,36,35,47,-2,10,18,19,19,19,15,15,0,4,-83,-78]} 1. Nf3 d5 2. d4 Bf5 {new to both players} 3. c4 c6 {transposing to a Slav looks safest} (3... e6 {has been played at elite GM level, but Bl needs tactics if Wh then goes for b7 with the Q} 4. Nc3 (4. Qb3 Nc6) 4... Nf6 5. Qb3 Nc6) 4. Nc3 e6 5. e3 (5. Qb3 Qb6 6. c5 {main, when Bl can swap Qs or go to c7}) 5... Nf6 6. Nh4 Bg6 7. Nxg6 hxg6 {2550 games} 8. Be2 (8. Bd2 {main}) (8. g3 Nbd7 9. Bg2 dxc4 10. Qe2 Nb6 11. O-O {81 games, 55% for Wh, with such as Shirov, Aronian & Kramnik et al}) (8. Bd3) 8... Nbd7 9. Qc2 Qc7 (9... Bd6 10. h3 (10. g3) (10. Bd2 {42 games})) 10. Bd2 (10. h3) (10. g3) 10... Rxh2 {having missed this Wh decided resignation was in order before worse things happened. However.} (10... Rxh2 11. Rxh2 (11. Rg1 $15 {Stockfish13}) (11. O-O $15 {Stockfish13-Wh is only half a pawn down}) 11... Qxh2 12. Bf1 Ng4 $19 13. cxd5 cxd5 14. Qd3 a6 15. Qe2 (15. e4 $11 {Stockfish13}) 15... Ngf6 16. O-O-O Qh5 17. Qd3 Ng4 18. Qe2 Qf5 ( 18... Ngf6 $15) 19. f3 Ngf6 20. g4 Qg5 21. e4 $18 {BaljKhaev-Filippov, RUS ch U10 Serpukhov 2001,1-0}) 0-1 [Event "Dyfed Online League"] [Site "Lichess"] [Date "2021.02.23"] [Round "6.3"] [White "Williams, R Gareth"] [Black "Francis, Matthew"] [Result "1/2-1/2"] [ECO "A90"] [WhiteElo "1643"] [BlackElo "1648"] [Annotator "kemen"] [PlyCount "82"] [EventDate "2021.??.??"] [WhiteTeam "Cardigan"] [BlackTeam "Aberystwyth"] {[%evp 0,82,22,22,29,-14,44,23,68,74,75,45,54,55,59,54,68,-5,21,12,19,6,51,64, 73,80,65,64,64,65,75,68,83,63,71,50,42,50,75,52,71,68,96,28,31,4,5,9,27,24,14, 16,14,12,12,20,24,9,36,44,39,40,44,44,49,46,60,65,67,58,58,60,64,17,6,2,1,0,0, 0,0,0,0,-14,0]} 1. d4 e6 2. c4 f5 3. g3 c6 (3... Nf6 {main} 4. Bg2 d5 (4... c6 5. Nh3 Be7 6. O-O O-O 7. Nc3 d6 8. e4 e5 {M Jones-Francis, Dyfed League previous rd, 2021,1-0}) 5. Nc3 c6 6. Nh3 Be7 7. O-O O-O 8. Qc2 Qe8 { Pinch-Francis, Dyfed Closed,2021,0-1}) 4. Bg2 d5 5. Qc2 Bd6 6. Nh3 Nf6 7. O-O O-O 8. cxd5 $6 $146 {this looks a premature exchange, as Bl is able to retake with the e pawn, getting the e file & a less obstructed B,} (8. Bf4 {main}) (8. Nd2 Nbd7 9. cxd5 cxd5 {now forced , 3 Wh wins, 2 draws, 2 Bl wins}) (8. b3) 8... exd5 9. Nd2 (9. Bf4) 9... Na6 {well, the N can find a good square on e6} ( 9... Re8 {looks simplest} 10. Nf3 Ne4 $11 (10... Nbd7 $11 11. Qxf5 Re4 12. Qg5 Rxe2 $15 {Stockfish13})) 10. a3 Qc7 $6 {obstructs the N, & falls in with Whs plan anyway} (10... Nc7 11. Nf3 Ne6 $11) 11. Nf3 $16 Qe7 12. Bf4 Bd7 (12... Ne4 $5 13. Bxd6 Nxd6 14. e3 Nb8 15. b4 Nd7 16. Nf4 Nf6 17. a4 $16 { Stockfish13-looks a logical sequence for both sides}) 13. Bxd6 Qxd6 14. Nf4 Nc7 15. Nd3 (15. Rfc1) 15... Ne6 16. e3 Rac8 {feels a little strange, the pawn structure suggests Bl needs to play on the K side} (16... g5) (16... Be8 { both look worth a try}) 17. Nfe5 (17. b4 {should be ok as well}) 17... c5 $2 { positionally it leaves the centre very vulnerable} 18. Nxd7 Nxd7 19. Nxc5 { but this way gives Bl a chance- Wh had seen this, but 'was desperate for imbalance to try and win'} (19. dxc5 {gets to game continuation safely}) 19... Ndxc5 (19... b6 {is possible} 20. b4 {is what Wh intended, gettings two pawn for the piece} (20. Nb7 {requires a careful look as well, Rudy wondered about this at the time} Qe7 21. Qb3 Nf6 22. Bxd5 Nxd5 23. Qxd5 Rc7 24. Nd6 Rd8 25. Nxf5 Rxd5 26. Nxe7+ Rxe7 27. Rac1 $15 {Wh has 3 pawns for the piece}) 20... bxc5 21. dxc5 {Bl has a choice- return the piece for 2 pawns; or, Rudy barricade the pawns & go for K side attack} Qc6 22. Rad1 {Rudy had been looking to play g5 quickly , but there is a lot of pressure on Bl's d5 pawn} Nc7 (22... Nf6 23. Qxf5 $11) 23. Qd3 Nf6 24. Qxf5 Kh8 $15 {Stockfish13-but it is difficult finding forcing moves for Bl In summary, it looks like the piece sac does give Wh chances , if taken up}) 20. dxc5 Nxc5 21. Qd1 $16 (21. Rfd1 $18 {Stockfish13- the Queen pawn is a fatal weakness} Ne4 22. Qb3 Rfd8 23. Qxb7 a5 24. Bxe4 fxe4 25. Rd2 Rc7 26. Qb3 Rc5 27. Rad1 h5 28. Qa4 h4 29. b4 Rcc8 30. Rxd5 Qxd5 31. Rxd5 Rxd5 32. Qb3 Rcd8 33. bxa5 (33. Kg2) 33... h3 34. Kf1 $18 { Stockfish13- humans would feel more comfortable with the K on g2}) 21... Ne4 22. f3 $2 {ends the B activity on the long diagonal and weakens K side} (22. Qb3 Rc7 23. Rad1 Rd8 24. Rd4 Qe6 25. Rfd1 Rcd7 26. h4 b6 27. Qc2 $16 { Stockfish13, Bl remains under pressure; though no clear breakthrough showing}) 22... Nf6 $11 23. Rc1 Kh8 (23... Qb6 24. Qd4 Qxd4 25. exd4 Rc4 $11 {Stockfish13 }) 24. Qd2 Qb6 25. Qd4 (25. Kf2 $5 $11) 25... Qxd4 26. exd4 $11 {if anything, the d4 pawn is now more vulnerable than the d5 one} Rfe8 27. Kf2 g6 28. Rfd1 b6 29. Bf1 Rxc1 30. Rxc1 Re7 31. Bb5 $14 Kg7 32. Rc2 g5 (32... f4 33. g4 (33. gxf4 Nh5) 33... g5 34. Bd3 $11 {Stockfish13}) 33. Bd3 f4 34. gxf4 gxf4 35. a4 Ne8 { nothing much now for either side} 36. a5 bxa5 37. Rc5 Rb7 38. Rxa5 (38. Rxd5 Rxb2+ 39. Ke1 (39. Kg1 Nf6 40. Rxa5 Rd2 41. Rxa7+ Kh6 42. Ra3 Nd5 43. Be4 Rxd4 $11) 39... Rxh2 40. Rxa5 h5 41. Rxa7+ Kf6 42. Rh7 {remains equal as well}) 38... Rxb2+ 39. Be2 Nf6 40. Rxa7+ Kg6 41. Ra4 h6 $11 1/2-1/2 [Event "Dyfed Online League"] [Site "Lichess"] [Date "2021.02.23"] [Round "6.4"] [White "Gunn, Tom"] [Black "Brewer, Joshua"] [Result "1-0"] [ECO "D04"] [WhiteElo "1398"] [BlackElo "1481"] [Annotator "kemen"] [PlyCount "55"] [EventDate "2021.??.??"] [WhiteTeam "Aberystwyth"] [BlackTeam "Cardigan"] 1. d4 d5 2. Nf3 Nf6 3. e3 Bd7 {an usual idea; Bl it seems is aiming for rapid Q side pawn advances. There have been a couple of games Gunn-Osinga , Dyfed Major 2020 1/2-1/2 & Dyfed League 2020, 0-1, where a similar Q side advance happened, though with Bd7 a liitle later in these games} (3... Bf5 {as played in elite GM games eg Giri, is a good alternative to the main}) (3... e6) (3... Bg4 {another possibility, among lesser Gms, with just Firouzja among the top}) 4. c4 (4. Bd3 {main, carry on the Colle}) 4... e6 5. Bd3 (5. Nc3 {50 games,66 % for Wh}) 5... c5 6. cxd5 (6. O-O {1 Wh win, 2 Bl wins}) 6... exd5 7. O-O $14 c4 8. Bc2 Bd6 9. Nc3 O-O 10. Re1 (10. e4 $14 dxe4 11. Nxe4 Nxe4 12. Bxe4 Qc8 ( 12... Nc6 13. Qc2 Nb4 14. Qxc4 {wins a pawn, though Stockfish13 thinks only slight Wh plus since Bl gets a lot of piece play- Normally this kind of pawn sac is done by Wh, But here it is Bl with the lead in development.}) 13. Qc2 f5 14. Bd5+ Be6 15. Bxe6+ Qxe6 16. d5 Qe4 17. Qxe4 fxe4 18. Nd2 e3 19. fxe3 Rxf1+ 20. Kxf1 Nd7 $11 {Wh is behind in development with pawn weaknesses}) 10... a6 ( 10... Nc6) 11. e4 dxe4 12. Nxe4 Nxe4 13. Bxe4 Ra7 {Bl wants that Q side pawn advance, but, it never gets there} (13... Nc6 14. Qc2 h6 15. Qxc4 Rc8 16. Qd3 $16) 14. Bxh7+ $2 {But the excellent position goes to Wh's head} (14. Qc2 h6 15. Qxc4 $18 {just picks off Bl's pride & joy}) (14. Ne5 {is considered even stronger by Stockfish13} Bc6 (14... Bxe5 15. Qh5 (15. dxe5 Bc6 16. Be3 Ra8 17. Bf5 $18) 15... g6 16. Qxe5 Re8 17. Bg5 Rxe5 18. Bxd8 Rb5 19. d5 $18 {The Bl pieces are tangled up on the K side while Wh threatens a rapid K side invasion} ) 15. Nxc4) 14... Kxh7 15. Ng5+ Kg8 16. Qh5 Bf5 {only move, clearly missed by Wh, but Bl should now be winning.} 17. Re3 (17. Bd2 $5) 17... Bg6 18. Qh4 f6 ( 18... Re8 {removes Wh's threat because of the backrank}) 19. Rh3 (19. Ne6 Qa5 20. Nxf8 Bxf8 21. b3 Nc6 (21... cxb3 22. Qg3 Qh5 23. Ba3 Nd7 $19) 22. bxc4 Qa4 $19 {the 2 pieces are better than the R}) 19... fxg5 {forced but should be winning} 20. Bxg5 Be7 $2 {miscounting or hoping that b5 has been played already?} (20... Qd7) (20... Qe8 21. Qh8+ Kf7 22. Rf3+ Ke6 23. Re1+ (23. Re3+ Kd5 24. Rxe8 Rxh8 25. Rxh8 b5 {Stockfish13-Bl's 2 pieces and Q side majority are winning assets}) 23... Kd5 24. Rxe8 Rxe8 $19) (20... Qa5 {all look fine for Bl}) 21. Bxe7 $18 Qe8 22. Qh8+ Kf7 23. Bxf8 Qe4 24. Qxg7+ Ke6 25. Re3 (25. Qe7+ {Stockfish13- the engine has worked out the mate- humans happily settle for winning the Q & thinking later} Kf5 (25... Kd5 26. Qd6#) 26. Rf3+ Qxf3 27. Qe5+ Kg4 28. h3+ Qxh3 29. gxh3+ Kh4 30. Be7+ Kxh3 31. Qg3#) 25... Qxe3 26. fxe3 Bd3 27. Qe5+ (27. Rf1 {the Engine ignores material & goes for mate- a far cry from the obsessively materialistic Sinclair Specrum chess engine from some 40 years ago} Kd5 (27... Bxf1 28. e4 Nd7 29. d5# {very elegant}) 28. Qe5+ Kc6 29. Qd6+ Kb5 30. a4+ Kxa4 31. Qb4#) 27... Kd7 28. Qxb8 1-0