What did we decide on this position? win or draw?

[peter]

I always believed that this position was draw but maybe now I begin to have some doubt !
I am not sure here 46 .. h4 is the better move for black
We can have instead 46 .. Kb7 !

Did let SF play out after 46.Qe2 Kb7 (just because that's your proposal) but SF came along with ...h4 already at 47th after LC0's 47.Kb1, so it won't prove a thing to you neither, if LC0 would win that again, would it?

So after this one more full point LC0-SF of the kind we already had one above here, I restarted the game after your 46.Qe2 Kb7 with LC0 against Crystal, 3070ti for LC0 and 30 threads of 16x3.5GHz CPU for Crystal again:

[pgn][Event "AMD, Schnellschach 30.0min+10.0sek"]
[Site "Dort"]
[Date "2023.04.15"]
[Round "1"]
[White "Lc0 v0.30.0-dag+git.1842e13d"]
[Black "Crystal 5 KWK"]
[Result "1-0"]
[ECO "E87"]
[Annotator "0.90;3.24"]
[PlyCount "127"]
[TimeControl "1800+10"]

{AMD Ryzen 9 5950X 16-Core Processor 3400 MHz W=28.8 plies; 19kN/s; 71 TBAs B=37.8 plies; 31.210kN/s; 4.619.938 TBAs} 1. c4 g6 2. Nc3 Bg7 3. d4 Nf6 4. e4 d6 5. f3 e5 6. d5 Na6 7. Be3 Nh5 8. Qd2 Qh4+ 9. Bf2 Qe7 10. O-O-O O-O 11. Be3 f5 12. Bd3 f4 13. Bf2 Bf6 14. Nge2 Bh4 15. Bg1 Bd7 16. Kb1 b6 17. Nc1 Nc5 18. Bc2 a5 19. Bxc5 bxc5 20. Ba4 Bxa4 21. Nxa4 Rfb8 22. Qc2 Kh8 23. Rd3 Rb4 24. Ra3 Qe8 25. Nd3 g5 26. h3 Nf6 27. Rc1 Kg7 28. b3 Nd7 29. Qd2 Rb7 30. Ndb2 Nb6 31. Ka1 h5 32. Nd1 Nxa4 33. Rxa4 Rb4 34. Nc3 Qc8 35. Nb5 Bg3 36. Qe2 Kf6 37. Rb1 Ke7 38. a3 Rxa4 39. bxa4 Kd8 40. Nc3 Rb8 41. Rb5 Rxb5 42. axb5 Qa8 43. Na4 Qa7 44. Ka2 Kc8 45. Qb2 Qb6 46. Qe2 Kb7 {Beide letzter Buchzug} 47. Kb1 {0.90/20 448} Kc8 {3.24/44 863 (h4)} 48. Qd2 {1.10/33 394 (Ka2)} Qa7 {3.73/42 618 (h4)} 49. Qb2 {1.10/34 240} Qb6 {3.77/42 144} 50. Kc2 {1.15/33 1 (Kc1)} Kd8 {4.32/44 356 (h4)} 51. Kc3 {1.18/32 0 (Kb3)} Qa7 {2.86/39 263 (Kd7)} 52. Kd3 {1.55/30 199} Qb6 {4.37/38 0} 53. Qb1 {1.69/33 64} h4 {4.55/34 5} 54. Ke2 {1.74/35 109} Kc8 {5.11/35 0} 55. Kf1 {1.82/34 82 (Db3)} Kb8 {5.91/33 81} 56. Qc2 {1.95/33 20} Kb7 {5.46/30 4} 57. Nb2 {2.22/32 10 (Db3)} Qa7 {6.03/31 28} 58. Qb3 {2.28/30 31} Bh2 {6.33/31 0 (Kb8)} 59. g4 {2.51/29 167 (Sa4)} fxg3 {6.83/29 5} 60. Kg2 {2.65/27 31} g4 {7.66/33 0 (Da8)} 61. fxg4 {4.85/17 162} Qb8 {11.01/36 0 (Da8)} 62. Qf3 {5.12/18 84 (b6)} Qh8 {24.35/29 42 (a4)} 63. g5 {6.48/15 92} a4 {#23/59 0} 64. Nxa4 {7.54/15 62 (Sd1) adjud.} 1-0
[/pgn]
As you see, Crystal adjourned ...h4 this time some further on, when it yet was played at 53rd, Crystal's evals were already at 4.55, so you'd probably not make this single one move responsible for the loss at this point anymore, would you, regards

[fishpov]

You're right ! Thanks.
Evaluation climb up with time.
So this is win for white.
Very interesting. Sting here, is not able to see winning position.
Without top engine and fortress detect, this is practically impossible for a human to find.

Analysis by Crystal 5 KWK bmi2:
....
1.Kb3 Qa7
+- (3.37 --) Depth: 36/26 00:00:26 265MN
....
1.Kb3 Kb8 2.Kc2 Kc8 3.Kd3 Kb7 4.Qb1 h4 5.Ke2 Kc8 6.Qg1 Kb8 7.Qd1 Kb7 8.Kf1 Bh2 9.Qb1 Bg3 10.Qb3 Bh2 11.g4 Bg3 12.Nb2 Qa7 13.Qc2 Kb8 14.a4 Kc8 15.Qd2 Kb7 16.Qc3 Qb6 17.Nd3 Bh2 18.Qb3 Bg3 19.Qb2 Kb8 20.Qc3 Kb7 21.Nc1 Qa7 22.Kg2 Kb8 23.Nb3 Kb7 24.Qxa5 Be1 25.Qxe1 Qxa4 26.Qc3 Qa2+ 27.Nd2
+- (3.36 --) Depth: 41/60 00:03:09 1686MN
....
1.Qb1 h4 2.Qb2 Kb7 3.Kb1 Kb8 4.Qd2 Kc8 5.Kc2 Qa7 6.Kd3 Qb6 7.Ke2 Kb8 8.Kf1 Kb7 9.Qc2 Kb8 10.Nb2 Kb7 11.Nd3 Qa7 12.a4 Qb8 13.Qc3 Qa8 14.Kg1 Kb6 15.Nxe5 dxe5 16.Qxe5 Qd8 17.Qg7 Be1 18.Kf1 Bb4 19.Ke2 Qc8 20.Kd3 Ba3 21.Qc3 Qf8 22.Qxa3 g4 23.hxg4 h3 24.gxh3 Qh8 25.Kc2 Qd4 26.Qd3 Qa1 27.Kb3 Qe1 28.e5 Qb4+ 29.Kc2 Qxa4+ 30.Kd2
+- (3.95 --) Depth: 48/88 00:26:17 12860MN
....
1.Kb3 Kb7
+- (3.65 ++) Depth: 51/87 01:43:22 50889MN
1.Kb3 Kb7
+- (4.09 ++) Depth: 51/97 01:47:37 52993MN
....
1.Qd2
+- (4.36 ++) Depth: 52/88 02:24:07 70975MN
1.Kb3
+- (4.65 ++) Depth: 52/101 03:07:22 91854MN

[fishpov]

Sorry last Eval from this position:
Analysis by Crystal 5 KWK bmi2:
..
1.Kb3
+- (4.65 ++) Depth: 52/101 03:07:22 91854MN
[pgn][Event "Junior Tournament"]
[Site "Schilde 1970"]
[Date "????.??.??"]
[Round ""]
[White "Crystal 5 KWK"]
[Black "Hasaj"]
[Result "1/2-1/2"]
[SetUp "1"]
[FEN "2k5/2p5/1q1p4/pPpPp1pp/N1P1Pp2/P4PbP/K2Q2P1/8 w - - 0 1"]
1.Kb3 Kb7
[/pgn]

[Jouni]

https://en.chessbase.com/post/computer-challenge-01 says easy draw!?

[peter]

https://en.chessbase.com/post/computer-challenge-01 says easy draw!?

Not taking the Queen at once isn't analysed at all, I tried two times with analyse- engine at site offered and won each time easily with LC0 running on home- PC at 3070ti and short ponder- time. Here's one of the lines I did let play online and sent to chessbase per email:

[pgn][Event "White to play and win"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "Computer challenge 01"]
[Black "?"]
[Result "1-0"]
[FEN "2k5/q1p5/3p4/pPpPp1pp/N1P1Pp2/P4PbP/KQ4P1/8 b - - 0 45"]
[SetUp "1"]
[Annotator "Karsten Müller"]
[Setup "1"]

{[%evp 0,21,36,671,671,671,671,671,671,671,671,671,671,671,671,671,671,671,671,671,671,671,671,671]} 45...Qb6 $1 46.Kb3 ( 46.Nxb6+ {and draw, as any human will quickly recognise.} 46...cxb6 47.h4 {White would like to use the h3 square for his queen to penetrate the other side of the board.} 47...gxh4 48.Qd2 h3 $1 {With this sacrifice Black will lock up the position.} 49.gxh3 h4 {and now, as you can see, there is no way the white queen can move to the black half of the board.} 50.Kb3 Kb7 51.Ka4 {White probably considered sacrificing the queen on a5, but Black can easily block the passed b-pawn and hold the draw.} 51...Ka7 52.Qg2 {He worked out that the sacrifice doesn't work.} 52...Kb7 53.Qb2 Ka7 54.Qc2 Kb7 55.Qc3 Ka7 ) 46...Qb7 47.Nxc5 dxc5 48.Ka4 Qb6 49.Qxe5 Bh4 50.Qe8+ Kb7 51.e5 g4 52.fxg4 hxg4 53.hxg4 Bg5 54.Qh5 Qh6 55.Qxh6 Bxh6 56.Kxa5 1-0[/pgn]
Of course weak resistance of online- engine, just to show the line it played, regards