We are reaching the Max Odds that a top engine can give the top 3 GM's

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

lkaufman
Posts: 6284
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA
Full name: Larry Kaufman

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by lkaufman »

chrisw wrote: Mon May 31, 2021 9:05 pm
[Event "knight odds match"]
[Site "TalkChess"]
[Date "2021.05.31"]
[Round "1"]
[White "KomodoDragon2, MCTS."]
[Black "Whittington, Chris"]
[Result "*"]
[Annotator "Kaufman,Larry"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[PlyCount "97"]
[EventDate "2021.05.20"]
[EventType "match"]

1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 Bg4 5. Be2 Nc6 6. c4 Nb6 7. e6 fxe6 8. O-O
g6 9. a4 a5 10. d5 Bxf3 11. Bxf3 exd5 12. cxd5 Ne5 13. Be4 Bg7 14. f4 Ned7 15.
f5 O-O 16. Ra3 Ne5 17. Rg3 Qe8 18. b3 c6 19. dxc6 bxc6 20. Rf2 e6 21. fxg6 Rxf2
22. gxh7+ Kh8 23. Kxf2 Qf7+ 24. Kg1 Rf8 25. h3 d5 26. Bf3 c5 27. Qe2 Nc6 28.
Bb2 Nd4 29. Qe1 Nd7 30. Bd1 e5 31. Rg5 e4 32. Qg3 Nf3+ 33. Bxf3 Bxb2 34. Rg8+
Kxh7 35. Rg5 Bg7 36. Bh5 Qf6 37. Bg6+ Qxg6 38. Qh4+ Qh6 39. Rh5 Nf6 40. Rxh6+
Bxh6 41. g4 Kg6 42. Qf2 Nd7 43. Qh4 Be3+ 44. Kh1 Rf1+ 45. Kg2 Rf2+ 46. Kh1 Nf6
47. Qh8 Bd4 48. Qf8 e3 49. Qc8 e2 [/pgn]
e2
[/quote]

[pgn]
[Event "knight odds match"]
[Site "TalkChess"]
[Date "2021.05.31"]
[Round "1"]
[White "KomodoDragon2, MCTS."]
[Black "Whittington, Chris"]
[Result "0-1"]
[Annotator "Kaufman,Larry"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[PlyCount "98"]
[EventDate "2021.05.20"]
[EventType "match"]

1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 Bg4 5. Be2 Nc6 6. c4 Nb6 7. e6 fxe6 8. O-O
g6 9. a4 a5 10. d5 Bxf3 11. Bxf3 exd5 12. cxd5 Ne5 13. Be4 Bg7 14. f4 Ned7 15.
f5 O-O 16. Ra3 Ne5 17. Rg3 Qe8 18. b3 c6 19. dxc6 bxc6 20. Rf2 e6 21. fxg6 Rxf2
22. gxh7+ Kh8 23. Kxf2 Qf7+ 24. Kg1 Rf8 25. h3 d5 26. Bf3 c5 (26... Nc8!) 27.
Qe2 Nc6 28. Bb2 Nd4 29. Qe1 Nd7 30. Bd1 e5 31. Rg5 e4 32. Qg3 Nf3+ ?! (32...
Bh6!) 33. Bxf3 Bxb2 34. Rg8+ Kxh7 35. Rg5 Bg7 36. Bh5 Qf6 37. Bg6+ Qxg6!
38. Qh4+ Qh6 39. Rh5 Nf6 40. Rxh6+ Bxh6 41. g4 Kg6 42. Qf2 Nd7 43. Qh4 Be3+ 44.
Kh1 Rf1+ 45. Kg2 Rf2+ 46. Kh1 Nf6 47. Qh8 Bd4 48. Qf8 e3 49. Qc8 e2 0-1

Dragon resigns. Congratulations! According to Dragon, Black only made two significantly non-optimal moves in the whole game, neither of which put the win in jeopardy. Did you actually average under a minute per move (it's okay if you took longer, it wasn't a rule, just your estimate)? That's a very high level of play for just one minute per move. Anyway if you still want to play the game with g1 off, I'm willing.[/pgn]
Komodo rules!
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by Chessqueen »

Probably two pawns was the most that you could have given Chrisw to start with, but now we now that time is the essence of Blunder and Chrisw showed us that if it is NOT game in 15'+10" he could manage to play without making more than two inferior moves at 1 minute per move, but how can we make sure of that. Well I believe that you or Modern time did some Centipawn analysis of this game to determine that only 2 inferior moves were made.

Note: If you take Snowy which is only rated around 1977 by CCRL it shows that at that level it is imposible for Komodod Dragon 2 MCTS or Stockfish 13 to beat it with a Knight Odds. I started it from position 29 to see how a low rated 1977 engine playing as Black handle stockfish13.

[pgn]
[Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.05.31"]
[Round "1"]
[White "Stockfish_21052415_x64_bmi2"]
[Black "Snowy_0_2_x64"]
[Result "0-1"]
[BlackElo "1977"]
[Time "16:39:36"]
[WhiteElo "3500"]
[TimeControl "900+1"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[Termination "unterminated"]
[PlyCount "107"]


1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 Bg4 5. Be2 Nc6 6. c4 Nb6 7. e6 fxe6 8.
O-O g6 9. a4 a5 10. d5 Bxf3 11. Bxf3 exd5 12. cxd5 Ne5 13. Be4 Bg7 14. f4
Ned7 15. f5 O-O 16. Ra3 Ne5 17. Rg3 Qe8 18. b3 c6 19. dxc6 bxc6 20. Rf2 e6
21. fxg6 Rxf2 22. gxh7+ Kh8 23. Kxf2 Qf7+ 24. Kg1 Rf8 25. h3 d5 26. Bf3 c5
27. Qe2 Nc6 28. Bb2 Nd4 29. Qe1 Nd7 30. Bd1 e5 31. Rg5 e4 32. Qg3 Nf3+ 33.
Bxf3 Bxb2 34. Rg8+ Kxh7 35. Rg5 Bg7 36. Bh5 Qf6 37. Bg6+ Qxg6 38. Qh4+ Qh6
39. Rh5 Nf6 40. Rxh6+ Bxh6 41. g4 Kg6 42. Qf2 Nd7 43. Qh4 Be3+ 44. Kh1 Rf1+
45. Kg2 Rf2+ 46. Kh1 Nf6 47. Qh8 Bd4 48. Qf8 e3 49. Qc8 e2 50. Qe6 Kg5 51.
b4 cxb4 52. Kg1 Kf4 53. Qf5+ Kg3 54. Qd3+ {White resigns} *[/pgn]
Last edited by Chessqueen on Mon May 31, 2021 11:53 pm, edited 1 time in total.
lkaufman
Posts: 6284
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA
Full name: Larry Kaufman

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by lkaufman »

Chessqueen wrote: Mon May 31, 2021 11:42 pm Probably two pawns was the most that you could have given Chrisw to start with, but now we now that time is the essence of Blunder and Chrisw showed us that if it is NOT game in 15'+10" he could manage to play without making more than two inferior moves at 1 minute per move, but how can we make sure of that. Well I believe that you or Modern time did some Centipawn analysis of this game to determine that only 2 inferior moves were made.

Note: If you take Snowy which is only rated around 1977 by CCRL it shows that at that level it is imposible for Komodod Dragon 2 MCTS or Stockfish 13 to beat it with a Knight Odds. I started it from position 28 to see how a low rated 1977 engine playing as Black handle stockfish13.

[pgn]
[Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.05.31"]
[Round "1"]
[White "Stockfish_21052415_x64_bmi2"]
[Black "Snowy_0_2_x64"]
[Result "0-1"]
[BlackElo "1977"]
[Time "16:39:36"]
[WhiteElo "3500"]
[TimeControl "900+1"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[Termination "unterminated"]
[PlyCount "107"]


1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 Bg4 5. Be2 Nc6 6. c4 Nb6 7. e6 fxe6 8.
O-O g6 9. a4 a5 10. d5 Bxf3 11. Bxf3 exd5 12. cxd5 Ne5 13. Be4 Bg7 14. f4
Ned7 15. f5 O-O 16. Ra3 Ne5 17. Rg3 Qe8 18. b3 c6 19. dxc6 bxc6 20. Rf2 e6
21. fxg6 Rxf2 22. gxh7+ Kh8 23. Kxf2 Qf7+ 24. Kg1 Rf8 25. h3 d5 26. Bf3 c5
27. Qe2 Nc6 28. Bb2 Nd4 29. Qe1 Nd7 30. Bd1 e5 31. Rg5 e4 32. Qg3 Nf3+ 33.
Bxf3 Bxb2 34. Rg8+ Kxh7 35. Rg5 Bg7 36. Bh5 Qf6 37. Bg6+ Qxg6 38. Qh4+ Qh6
39. Rh5 Nf6 40. Rxh6+ Bxh6 41. g4 Kg6 42. Qf2 Nd7 43. Qh4 Be3+ 44. Kh1 Rf1+
45. Kg2 Rf2+ 46. Kh1 Nf6 47. Qh8 Bd4 48. Qf8 e3 49. Qc8 e2 50. Qe6 Kg5 51.
b4 cxb4 52. Kg1 Kf4 53. Qf5+ Kg3 54. Qd3+ {White resigns} *[/pgn]
You started this game from the position near the end of the game after it was already hopeless. That has nothing to do with whether Snowy could beat Dragon with knight odds from the start. Also note that I said that Chris only made two SIGNIFICANTLY inferior moves, I don't mean that every other move was optimal except in the sense that he always preserved his winning advantage. It is not so clear how to define inferior moves if they keep the win intact.
Komodo rules!
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by Chessqueen »

lkaufman wrote: Mon May 31, 2021 11:53 pm
Chessqueen wrote: Mon May 31, 2021 11:42 pm Probably two pawns was the most that you could have given Chrisw to start with, but now we now that time is the essence of Blunder and Chrisw showed us that if it is NOT game in 15'+10" he could manage to play without making more than two inferior moves at 1 minute per move, but how can we make sure of that. Well I believe that you or Modern time did some Centipawn analysis of this game to determine that only 2 inferior moves were made.

Note: If you take Snowy which is only rated around 1977 by CCRL it shows that at that level it is imposible for Komodod Dragon 2 MCTS or Stockfish 13 to beat it with a Knight Odds. I started it from position 28 to see how a low rated 1977 engine playing as Black handle stockfish13.

[pgn]
[Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.05.31"]
[Round "1"]
[White "Stockfish_21052415_x64_bmi2"]
[Black "Snowy_0_2_x64"]
[Result "0-1"]
[BlackElo "1977"]
[Time "16:39:36"]
[WhiteElo "3500"]
[TimeControl "900+1"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[Termination "unterminated"]
[PlyCount "107"]


1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 Bg4 5. Be2 Nc6 6. c4 Nb6 7. e6 fxe6 8.
O-O g6 9. a4 a5 10. d5 Bxf3 11. Bxf3 exd5 12. cxd5 Ne5 13. Be4 Bg7 14. f4
Ned7 15. f5 O-O 16. Ra3 Ne5 17. Rg3 Qe8 18. b3 c6 19. dxc6 bxc6 20. Rf2 e6
21. fxg6 Rxf2 22. gxh7+ Kh8 23. Kxf2 Qf7+ 24. Kg1 Rf8 25. h3 d5 26. Bf3 c5
27. Qe2 Nc6 28. Bb2 Nd4 29. Qe1 Nd7 30. Bd1 e5 31. Rg5 e4 32. Qg3 Nf3+ 33.
Bxf3 Bxb2 34. Rg8+ Kxh7 35. Rg5 Bg7 36. Bh5 Qf6 37. Bg6+ Qxg6 38. Qh4+ Qh6
39. Rh5 Nf6 40. Rxh6+ Bxh6 41. g4 Kg6 42. Qf2 Nd7 43. Qh4 Be3+ 44. Kh1 Rf1+
45. Kg2 Rf2+ 46. Kh1 Nf6 47. Qh8 Bd4 48. Qf8 e3 49. Qc8 e2 50. Qe6 Kg5 51.
b4 cxb4 52. Kg1 Kf4 53. Qf5+ Kg3 54. Qd3+ {White resigns} *[/pgn]
You started this game from the position near the end of the game after it was already hopeless. That has nothing to do with whether Snowy could beat Dragon with knight odds from the start. Also note that I said that Chris only made two SIGNIFICANTLY inferior moves, I don't mean that every other move was optimal except in the sense that he always preserved his winning advantage. It is not so clear how to define inferior moves if they keep the win intact.
I will restart it from move 8. 0-0 and post it in 5 minutes :shock:
chrisw
Posts: 4773
Joined: Tue Apr 03, 2012 4:28 pm
Location: Midi-Pyrénées
Full name: Christopher Whittington

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by chrisw »

lkaufman wrote: Mon May 31, 2021 10:56 pm
chrisw wrote: Mon May 31, 2021 9:05 pm
[Event "knight odds match"]
[Site "TalkChess"]
[Date "2021.05.31"]
[Round "1"]
[White "KomodoDragon2, MCTS."]
[Black "Whittington, Chris"]
[Result "*"]
[Annotator "Kaufman,Larry"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[PlyCount "97"]
[EventDate "2021.05.20"]
[EventType "match"]

1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 Bg4 5. Be2 Nc6 6. c4 Nb6 7. e6 fxe6 8. O-O
g6 9. a4 a5 10. d5 Bxf3 11. Bxf3 exd5 12. cxd5 Ne5 13. Be4 Bg7 14. f4 Ned7 15.
f5 O-O 16. Ra3 Ne5 17. Rg3 Qe8 18. b3 c6 19. dxc6 bxc6 20. Rf2 e6 21. fxg6 Rxf2
22. gxh7+ Kh8 23. Kxf2 Qf7+ 24. Kg1 Rf8 25. h3 d5 26. Bf3 c5 27. Qe2 Nc6 28.
Bb2 Nd4 29. Qe1 Nd7 30. Bd1 e5 31. Rg5 e4 32. Qg3 Nf3+ 33. Bxf3 Bxb2 34. Rg8+
Kxh7 35. Rg5 Bg7 36. Bh5 Qf6 37. Bg6+ Qxg6 38. Qh4+ Qh6 39. Rh5 Nf6 40. Rxh6+
Bxh6 41. g4 Kg6 42. Qf2 Nd7 43. Qh4 Be3+ 44. Kh1 Rf1+ 45. Kg2 Rf2+ 46. Kh1 Nf6
47. Qh8 Bd4 48. Qf8 e3 49. Qc8 e2 [/pgn]
e2
[pgn]
[Event "knight odds match"]
[Site "TalkChess"]
[Date "2021.05.31"]
[Round "1"]
[White "KomodoDragon2, MCTS."]
[Black "Whittington, Chris"]
[Result "0-1"]
[Annotator "Kaufman,Larry"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[PlyCount "98"]
[EventDate "2021.05.20"]
[EventType "match"]

1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 Bg4 5. Be2 Nc6 6. c4 Nb6 7. e6 fxe6 8. O-O
g6 9. a4 a5 10. d5 Bxf3 11. Bxf3 exd5 12. cxd5 Ne5 13. Be4 Bg7 14. f4 Ned7 15.
f5 O-O 16. Ra3 Ne5 17. Rg3 Qe8 18. b3 c6 19. dxc6 bxc6 20. Rf2 e6 21. fxg6 Rxf2
22. gxh7+ Kh8 23. Kxf2 Qf7+ 24. Kg1 Rf8 25. h3 d5 26. Bf3 c5 (26... Nc8!) 27.
Qe2 Nc6 28. Bb2 Nd4 29. Qe1 Nd7 30. Bd1 e5 31. Rg5 e4 32. Qg3 Nf3+ ?! (32...
Bh6!) 33. Bxf3 Bxb2 34. Rg8+ Kxh7 35. Rg5 Bg7 36. Bh5 Qf6 37. Bg6+ Qxg6!
38. Qh4+ Qh6 39. Rh5 Nf6 40. Rxh6+ Bxh6 41. g4 Kg6 42. Qf2 Nd7 43. Qh4 Be3+ 44.
Kh1 Rf1+ 45. Kg2 Rf2+ 46. Kh1 Nf6 47. Qh8 Bd4 48. Qf8 e3 49. Qc8 e2 0-1

Dragon resigns. Congratulations! According to Dragon, Black only made two significantly non-optimal moves in the whole game, neither of which put the win in jeopardy. Did you actually average under a minute per move (it's okay if you took longer, it wasn't a rule, just your estimate)? That's a very high level of play for just one minute per move. Anyway if you still want to play the game with g1 off, I'm willing.
[/quote]

Okay, we can play mirror Alekhine. I need a couple of days to recover, that was quite the stressy session.

When did we begin? Ten days ago? Fifty moves, five a day, so, in theory that’s two hours of awake time per move. In practice moves were in surges, where I’ld worked out where the game was going, then there’s 18 hours or so. Obviously I’m not thinking position all that time, but I can chew over general principles, in this case: defence principles. I liked Ne5, Qe8, light squares. I think e6 was a mistake. Pawn could perfectly well have stayed on e7, and I basically had a tactical defence of g6, namely the f-file.
After that I figured as long as Q stayed connected to Bg7, I ought to be able to manoeuvre around the strong centre with the nuisance knights and you’ld just run out of possibilities. Er, wrong.
Nf3+ was a shot at being okay by getting rid of your Bb2, but again wrong, and against defensive principles, I got careless and started mixing it tactically, which is absolutely dumb thing to do. However, again I figured B back to g7 and all would be defendable again. But.
Well, but. Fortunately, by temptation or not, and possibly just running out of possibilities, you went for the Bg6+ line and Q vs RBN-centre pawns which I was very confident would be a win.
On reflection, especially after e6, much of the game was quite predictable and much of it was forced. White had to press. Black had to try and make fortresses where everything defends everything and nothing is hanging and human doesn’t have to keeping thinking tactical. Then we come down to space and mobility (white had in spades) against one more defensive piece (and a centre). QRBB is a very formidable combination of pieces of course. Curious how e6 played such a key part for both sides, then finally becoming the decisive pawn.

So, yes, mirror Alekhine, couple of days break.[/pgn]
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by Chessqueen »

Chessqueen wrote: Mon May 31, 2021 11:55 pm
lkaufman wrote: Mon May 31, 2021 11:53 pm
Chessqueen wrote: Mon May 31, 2021 11:42 pm Probably two pawns was the most that you could have given Chrisw to start with, but now we now that time is the essence of Blunder and Chrisw showed us that if it is NOT game in 15'+10" he could manage to play without making more than two inferior moves at 1 minute per move, but how can we make sure of that. Well I believe that you or Modern time did some Centipawn analysis of this game to determine that only 2 inferior moves were made.

Note: If you take Snowy which is only rated around 1977 by CCRL it shows that at that level it is imposible for Komodod Dragon 2 MCTS or Stockfish 13 to beat it with a Knight Odds. I started it from position 28 to see how a low rated 1977 engine playing as Black handle stockfish13.

[pgn]
[Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.05.31"]
[Round "1"]
[White "Stockfish_21052415_x64_bmi2"]
[Black "Snowy_0_2_x64"]
[Result "0-1"]
[BlackElo "1977"]
[Time "16:39:36"]
[WhiteElo "3500"]
[TimeControl "900+1"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[Termination "unterminated"]
[PlyCount "107"]


1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 Bg4 5. Be2 Nc6 6. c4 Nb6 7. e6 fxe6 8.
O-O g6 9. a4 a5 10. d5 Bxf3 11. Bxf3 exd5 12. cxd5 Ne5 13. Be4 Bg7 14. f4
Ned7 15. f5 O-O 16. Ra3 Ne5 17. Rg3 Qe8 18. b3 c6 19. dxc6 bxc6 20. Rf2 e6
21. fxg6 Rxf2 22. gxh7+ Kh8 23. Kxf2 Qf7+ 24. Kg1 Rf8 25. h3 d5 26. Bf3 c5
27. Qe2 Nc6 28. Bb2 Nd4 29. Qe1 Nd7 30. Bd1 e5 31. Rg5 e4 32. Qg3 Nf3+ 33.
Bxf3 Bxb2 34. Rg8+ Kxh7 35. Rg5 Bg7 36. Bh5 Qf6 37. Bg6+ Qxg6 38. Qh4+ Qh6
39. Rh5 Nf6 40. Rxh6+ Bxh6 41. g4 Kg6 42. Qf2 Nd7 43. Qh4 Be3+ 44. Kh1 Rf1+
45. Kg2 Rf2+ 46. Kh1 Nf6 47. Qh8 Bd4 48. Qf8 e3 49. Qc8 e2 50. Qe6 Kg5 51.
b4 cxb4 52. Kg1 Kf4 53. Qf5+ Kg3 54. Qd3+ {White resigns} *[/pgn]
You started this game from the position near the end of the game after it was already hopeless. That has nothing to do with whether Snowy could beat Dragon with knight odds from the start. Also note that I said that Chris only made two SIGNIFICANTLY inferior moves, I don't mean that every other move was optimal except in the sense that he always preserved his winning advantage. It is not so clear how to define inferior moves if they keep the win intact.
I will restart it from move 8. 0-0 and post it in 5 minutes :shock:
You are correct it requires at least an engine rate around 2275 by CCRL to beat Stockfish 13 NNUE from where you started the game against Chrisw. He deserve to be congratulated by all of us Humans. :lol:
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by Chessqueen »

Chessqueen wrote: Tue Jun 01, 2021 1:17 am
Chessqueen wrote: Mon May 31, 2021 11:55 pm
lkaufman wrote: Mon May 31, 2021 11:53 pm
Chessqueen wrote: Mon May 31, 2021 11:42 pm Probably two pawns was the most that you could have given Chrisw to start with, but now we now that time is the essence of Blunder and Chrisw showed us that if it is NOT game in 15'+10" he could manage to play without making more than two inferior moves at 1 minute per move, but how can we make sure of that. Well I believe that you or Modern time did some Centipawn analysis of this game to determine that only 2 inferior moves were made.

Note: If you take Snowy which is only rated around 1977 by CCRL it shows that at that level it is imposible for Komodod Dragon 2 MCTS or Stockfish 13 to beat it with a Knight Odds. I started it from position 28 to see how a low rated 1977 engine playing as Black handle stockfish13.

[pgn]
[Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.05.31"]
[Round "1"]
[White "Stockfish_21052415_x64_bmi2"]
[Black "Snowy_0_2_x64"]
[Result "0-1"]
[BlackElo "1977"]
[Time "16:39:36"]
[WhiteElo "3500"]
[TimeControl "900+1"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"]
[Termination "unterminated"]
[PlyCount "107"]


1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 Bg4 5. Be2 Nc6 6. c4 Nb6 7. e6 fxe6 8.
O-O g6 9. a4 a5 10. d5 Bxf3 11. Bxf3 exd5 12. cxd5 Ne5 13. Be4 Bg7 14. f4
Ned7 15. f5 O-O 16. Ra3 Ne5 17. Rg3 Qe8 18. b3 c6 19. dxc6 bxc6 20. Rf2 e6
21. fxg6 Rxf2 22. gxh7+ Kh8 23. Kxf2 Qf7+ 24. Kg1 Rf8 25. h3 d5 26. Bf3 c5
27. Qe2 Nc6 28. Bb2 Nd4 29. Qe1 Nd7 30. Bd1 e5 31. Rg5 e4 32. Qg3 Nf3+ 33.
Bxf3 Bxb2 34. Rg8+ Kxh7 35. Rg5 Bg7 36. Bh5 Qf6 37. Bg6+ Qxg6 38. Qh4+ Qh6
39. Rh5 Nf6 40. Rxh6+ Bxh6 41. g4 Kg6 42. Qf2 Nd7 43. Qh4 Be3+ 44. Kh1 Rf1+
45. Kg2 Rf2+ 46. Kh1 Nf6 47. Qh8 Bd4 48. Qf8 e3 49. Qc8 e2 50. Qe6 Kg5 51.
b4 cxb4 52. Kg1 Kf4 53. Qf5+ Kg3 54. Qd3+ {White resigns} *[/pgn]
You started this game from the position near the end of the game after it was already hopeless. That has nothing to do with whether Snowy could beat Dragon with knight odds from the start. Also note that I said that Chris only made two SIGNIFICANTLY inferior moves, I don't mean that every other move was optimal except in the sense that he always preserved his winning advantage. It is not so clear how to define inferior moves if they keep the win intact.
I will restart it from move 8. 0-0 and post it in 5 minutes :shock:
You are correct it requires at least an engine rate around 2275 by CCRL to beat Stockfish 13 NNUE from where you started the game against Chrisw. He deserve to be congratulated by all of us Humans. :lol:
I do NOT know exactly what is the minimum rating for an engine to beat either Stockfish13 NNUE or Komodo Dragon2 MCTS with a Knight odds, but it could be in the range of 2300 CCRL to 2400, anyway I tried Dragon 4.6 rated 2417 and it beat Stockfish easily probably other people will test engine lower than 2417 with Komodo Dragon 2 MCTS comtempt =150
[pgn]
[Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.05.31"]
[Round "?"]
[White "Stockfish_21052415_x64_bmi2"]
[Black "Dragon 4.6"]
[Result "1-0"]
[BlackElo "2417"]
[Time "19:21:59"]
[WhiteElo "3500"]
[TimeControl "900+1"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[Termination "normal"]
[PlyCount "83"]
[WhiteType "program"]
[BlackType "program"]

1. e4 e5 2. d4 Qh4 3. Nc3 Bb4 4. dxe5 Qxe4+ 5. Qe2 Bxc3+ 6. bxc3 Qxe2+ 7.
Bxe2 Nc6 8. Bb2 Nxe5 9. c4 f6 10. O-O-O Ne7 11. Rhe1 O-O 12. f4 N5c6 13.
Bf3 d6 14. g4 Ng6 15. c5 dxc5 16. Bxc6 bxc6 17. f5 Nf4 18. Re4 Nd5 19. c4
Nb6 20. Ba3 Ba6 21. Bxc5 Rfe8 22. Rde1 Rxe4 23. Rxe4 Bxc4 24. a3 Bf1 25. h4
Nd7 26. Bb4 c5 27. Ba5 Nb6 28. Re7 Nc4 29. Bxc7 Nxa3 30. g5 Nb5 31. Ba5 Bh3
32. gxf6 gxf6 33. Kd1 Bxf5 34. Rb7 a6 35. Rb6 Rd8+ 36. Ke2 Rd6 37. Rb8+ Kf7
38. h5 c4 39. Kf2 Rc6 40. Rb7+ Ke6 41. Rb8 c3 42. Bb6 {White resigns} *[/pgn]
lkaufman
Posts: 6284
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA
Full name: Larry Kaufman

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by lkaufman »

Chessqueen wrote: Tue Jun 01, 2021 3:03 am
I do NOT know exactly what is the minimum rating for an engine to beat either Stockfish13 NNUE or Komodo Dragon2 MCTS with a Knight odds, but it could be in the range of 2300 CCRL to 2400, anyway I tried Dragon 4.6 rated 2417 and it beat Stockfish easily probably other people will test engine lower than 2417 with Komodo Dragon 2 MCTS comtempt =150
[pgn]
[Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.05.31"]
[Round "?"]
[White "Stockfish_21052415_x64_bmi2"]
[Black "Dragon 4.6"]
[Result "1-0"]
[BlackElo "2417"]
[Time "19:21:59"]
[WhiteElo "3500"]
[TimeControl "900+1"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[Termination "normal"]
[PlyCount "83"]
[WhiteType "program"]
[BlackType "program"]

1. e4 e5 2. d4 Qh4 3. Nc3 Bb4 4. dxe5 Qxe4+ 5. Qe2 Bxc3+ 6. bxc3 Qxe2+ 7.
Bxe2 Nc6 8. Bb2 Nxe5 9. c4 f6 10. O-O-O Ne7 11. Rhe1 O-O 12. f4 N5c6 13.
Bf3 d6 14. g4 Ng6 15. c5 dxc5 16. Bxc6 bxc6 17. f5 Nf4 18. Re4 Nd5 19. c4
Nb6 20. Ba3 Ba6 21. Bxc5 Rfe8 22. Rde1 Rxe4 23. Rxe4 Bxc4 24. a3 Bf1 25. h4
Nd7 26. Bb4 c5 27. Ba5 Nb6 28. Re7 Nc4 29. Bxc7 Nxa3 30. g5 Nb5 31. Ba5 Bh3
32. gxf6 gxf6 33. Kd1 Bxf5 34. Rb7 a6 35. Rb6 Rd8+ 36. Ke2 Rd6 37. Rb8+ Kf7
38. h5 c4 39. Kf2 Rc6 40. Rb7+ Ke6 41. Rb8 c3 42. Bb6 {White resigns} *[/pgn]
When you post an engine game, you should always include the time limit at least, that's vital info. I'm guessing 15' + 10", but just a guess. I suppose it is single thread, but this is also important to state.
Since you have so many of the near or under 2000 ccrl engines and post games with them, perhaps you can tell what engine, if any, you are able to score about evenly against yourself in blitz, say 3' + 2"? I think you estimated your rating to be in the 1900 to 2000 range, so I'm guess it would be an engine a few hundred elo below that, but "a few" is not a number. Anyone else with a known blitz rating (server or FIDE) who can name an engine which they are evenly matched with, no matter how weak, please post here (and specify hardware and threads). Or it can be a Skill level on a strong engine.
Komodo rules!
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by Chessqueen »

TC was 15'+1" for both and single thread for both, but I strongly believe that with either Knight b1 or Knight g1 odds the 1. e4 Opening is bad for White :roll:

Here it is with the 1. d4 opening for White :roll:

[pgn][Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.05.31"]
[Round "2"]
[White "Stockfish_21052415_x64_bmi2"]
[Black "Dragon 4.6"]
[Result "1/2-1/2"]
[BlackElo "2417"]
[Time "20:26:09"]
[WhiteElo "3500"]
[TimeControl "900+1"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[Termination "adjudication"]
[PlyCount "110"]
[WhiteType "program"]
[BlackType "program"]

1. d4 d5 2. c4 dxc4 3. e3 Be6 4. e4 c6 5. Nc3 Nf6 6. Be2 b5 7. Bf3 Na6 8.
O-O Nb4 9. a4 a6 10. axb5 axb5 11. Rxa8 Qxa8 12. b3 cxb3 13. d5 cxd5 14.
Qxb3 dxe4 15. Qxb4 exf3 16. Nxb5 Nd5 17. Rd1 Qc6 18. Qa5 f6 19. Nd4 Qc7 20.
Qxc7 Nxc7 21. Bf4 Bd5 22. Bxc7 fxg2 23. Ne2 e6 24. Nf4 Kd7 25. Ba5 Kc6 26.
Re1 Bd6 27. Nxe6 Ra8 28. Bc3 Ra3 29. Nd8+ Kb5 30. Rd1 Kc4 31. Rd4+ Kc5 32.
Ne6+ Bxe6 33. Rxd6 Kxd6 34. Bb4+ Ke5 35. Bxa3 Bd5 36. Bf8 g5 37. Be7 h5 38.
Bb4 Kf4 39. Be7 f5 40. Bf6 h4 41. Be7 h3 42. Bc5 g4 43. Bd6+ Ke4 44. Bb8 f4
45. Bc7 Bf7 46. Bd8 g3 47. fxg3 fxg3 48. hxg3 Kf3 49. g4 Kxg4 50. Kh2 Be6
51. Bf6 Kf3 52. Bh4 Ke2 53. Be7 Kf1 54. Bc5 Bg4 55. Bd4 Bf5[/pgn]
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: We are reaching the Max Odds that a top engine can give the top 3 GM's

Post by Chessqueen »

lkaufman wrote: Tue Jun 01, 2021 3:24 am
Chessqueen wrote: Tue Jun 01, 2021 3:03 am
I do NOT know exactly what is the minimum rating for an engine to beat either Stockfish13 NNUE or Komodo Dragon2 MCTS with a Knight odds, but it could be in the range of 2300 CCRL to 2400, anyway I tried Dragon 4.6 rated 2417 and it beat Stockfish easily probably other people will test engine lower than 2417 with Komodo Dragon 2 MCTS comtempt =150
[pgn]
[Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.05.31"]
[Round "?"]
[White "Stockfish_21052415_x64_bmi2"]
[Black "Dragon 4.6"]
[Result "1-0"]
[BlackElo "2417"]
[Time "19:21:59"]
[WhiteElo "3500"]
[TimeControl "900+1"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[Termination "normal"]
[PlyCount "83"]
[WhiteType "program"]
[BlackType "program"]

1. e4 e5 2. d4 Qh4 3. Nc3 Bb4 4. dxe5 Qxe4+ 5. Qe2 Bxc3+ 6. bxc3 Qxe2+ 7.
Bxe2 Nc6 8. Bb2 Nxe5 9. c4 f6 10. O-O-O Ne7 11. Rhe1 O-O 12. f4 N5c6 13.
Bf3 d6 14. g4 Ng6 15. c5 dxc5 16. Bxc6 bxc6 17. f5 Nf4 18. Re4 Nd5 19. c4
Nb6 20. Ba3 Ba6 21. Bxc5 Rfe8 22. Rde1 Rxe4 23. Rxe4 Bxc4 24. a3 Bf1 25. h4
Nd7 26. Bb4 c5 27. Ba5 Nb6 28. Re7 Nc4 29. Bxc7 Nxa3 30. g5 Nb5 31. Ba5 Bh3
32. gxf6 gxf6 33. Kd1 Bxf5 34. Rb7 a6 35. Rb6 Rd8+ 36. Ke2 Rd6 37. Rb8+ Kf7
38. h5 c4 39. Kf2 Rc6 40. Rb7+ Ke6 41. Rb8 c3 42. Bb6 {White resigns} *[/pgn]
When you post an engine game, you should always include the time limit at least, that's vital info. I'm guessing 15' + 10", but just a guess. I suppose it is single thread, but this is also important to state.
Since you have so many of the near or under 2000 ccrl engines and post games with them, perhaps you can tell what engine, if any, you are able to score about evenly against yourself in blitz, say 3' + 2"? I think you estimated your rating to be in the 1900 to 2000 range, so I'm guess it would be an engine a few hundred elo below that, but "a few" is not a number. Anyone else with a known blitz rating (server or FIDE) who can name an engine which they are evenly matched with, no matter how weak, please post here (and specify hardware and threads). Or it can be a Skill level on a strong engine.
I am tire today I just came from work and after that got an hour training with my Chess trainer online, I will try another day against an engine close to 1800 since CCRL rating is equal to 150 to 200 stronger than human. I strongly recommend you to try against Dragon 4.6 rated around 2417 by CCR, or Zahak rated 2415. Or Komodo Dragon 2 MCTS contempt = 150 against these two with a Knight odds. :roll:

256‑257 Dragon 4.6 2417 +8 −8 48.5% +11.0 24.6% 7652
47.4%
256‑257 Snitch 1.6.2 32-bit 2417 +10 −10 46.5% +27.8 25.0% 3997
56.6%
258 Zahak 3.0 64-bit 2415 +26 −26 66.6% −126.1 21.3% 559
54.7%