[Event "TCEC 20 Premier 2020"]
[Site "tcec-chess.com INT"]
[Date "2021.01.06"]
[Round "33.2"]
[White "KomodoDragon 2647.00"]
[Black "ScorpioNN 3.0.11"]
[Result "1-0"]
[ECO "B54"]
[WhiteElo "3522"]
[BlackElo "3496"]
[Annotator "Monokroussos,Dennis"]
[PlyCount "57"]
[EventDate "2020.12.27"]
[EventType "tourn"]
[EventRounds "56"]
[SourceTitle "The Week in Chess 1366"]
[Source "Mark Crowther"]
[SourceDate "2021.01.11"]
[SourceVersion "1"]
[SourceVersionDate "2021.01.11"]
[SourceQuality "2"]
{[%evp 0,57,28,34,52,48,90,62,64,33,27,10,31,31,58,56,66,65,72,53,59,39,72,20,
40,38,48,31,44,33,43,17,52,59,85,28,36,11,56,61,61,-108,-20,-98,-48,-98,-55,
-12,45,-184,52,0,0,0,0,-10,492,382,29999,-30000]} 1. e4 c5 2. Nf3 d6 3. d4 cxd4
4. Nxd4 Nf6 5. f3 a6 $6 {Not a particularly good choice, but maybe it was part
of the organizer's book rather than the move chosen by the engine.} (5... e5) (
{and} 5... Nc6 {are standard here.}) 6. c4 e6 7. Nc3 Qc7 $6 {While this is a
common square for the queen in hedgehog-type positions, Black should have
developed its minor pieces first.} 8. Be3 Nbd7 9. Rc1 b6 10. g4 h6 {Now it's
time for something great.} 11. Kd2 $3 {I can't recall seeing this idea before,
but maybe those of you with the relevant repertoires can let me know if it's a
standard idea. White "castles" by hand, tucking the king away on b1, and it
looks absolutely fine.} g6 12. Kc2 Bg7 13. Be2 (13. Kb1) 13... Bb7 14. Kb1 {
The king is perfectly safe on b1, and now White is ready to resume kingside
play with h4 and g5.} h5 {Forcing the immediate g5 for some reason.} (14... Rc8
{seems the wiser choice, not that Black is in great shape here.}) 15. g5 Nh7
16. f4 {White's advantage is overwhelming, and even a 3496-rated engine has
little chance to escape to even a lousy ending.} Nc5 17. f5 Be5 18. Ka1 {
All the time in the world.} Rc8 19. Rf1 O-O {The king wasn't safe in the
center, either, but now that White knows the king's permanent address it's
time to pay him a visit.} 20. Bxh5 $1 {Of course.} gxh5 21. Qxh5 Nxe4 {This is
the only thing that makes 20.Bxh5 something less than completely automatic.}
22. Nce2 $1 Qd7 $2 (22... Nc5 {may have been better, preventing the idea we're
about to see in the game.}) 23. Rf3 exf5 24. Rg1 $1 (24. Rh3 $2 f6 {is Black's
defensive idea, and is what White's 24.Rg1 is aimed to prevent (or at least
punish).}) 24... f6 (24... Kh8 25. Rxf5 $1 {(Not 25.Rh3)} Rce8 26. Nf4 Bxf4 27.
Rxf4 Nc5 28. Bd2 $1 $18 {followed by some combination of Bc3 and Rh4, with a
winning attack.}) 25. g6 Nhg5 26. Nxf5 $1 Nxf3 27. g7 Nfg5 28. Qg6 Qf7 {
Odd - I can't recall seeing an engine allowing itself to get mated when it can
delay it. I don't know if it's a bug, but Scorpio has been mated repeatedly in
the tournament.} ({Just to be clear, there's no saving the game for Black.
After the "better"} 28... Qxf5 {White won't have any trouble winning after} 29.
gxf8=Q+ Kxf8 30. Qxf5 Rxc4 31. Nf4 {. The material situation isn't *that* bad
for Black, but with a permanently weak king there is no chance to survive
against an elite engine. White will play h4 soon, opening more avenues for the
attack.}) 29. Nh6# 1-0
