Reply
 
LinkBack Thread Tools Display Modes
  #1   Report Post  
Old September 9th 10, 01:12 AM posted to rec.games.chess.misc,rec.games.chess.computer
external usenet poster
 
First recorded activity by ChessBanter: Mar 2009
Posts: 3,536
Default Examples of much weaker programs beating stronger ones?

I don't think they exist. That's because computers use the same
search engines at the brute force level, and the only intelligence is
the evaluation function. So unless a stronger computer walks into a
mating net that lies beyond it's search ply, and those must be rare (I
would imagine), I don't see how a much weaker chess program (say 300
Elo points weaker) can beat a stronger one.

Any examples?

RL

Here's what I found from the net...I'd like to see how Super Const.
beat Mephisto, probably though the three points are mostly draws.

http://ssdf.bosjo.net/rlwww101.txt

217 Mephisto Lyon 68000 12 MHz 2004 21 -21 1121
57% 1951
294 Novag Super Constellation 6502 4 MHz 1629 18 -18 1583
34% 1746

217 Mephisto Lyon 68000 12 MHz, 2004 match against Super Const.
17-3
  #2   Report Post  
Old September 9th 10, 07:41 AM posted to rec.games.chess.misc,rec.games.chess.computer
external usenet poster
 
First recorded activity by ChessBanter: Dec 2009
Posts: 1,146
Default Examples of much weaker programs beating stronger ones?

On Sep 8, 8:12*pm, raylopez99 wrote:

I don't think they exist. *That's because computers use the same
search engines at the brute force level, and the only intelligence is
the evaluation function. *So unless a stronger computer walks into a
mating net that lies beyond it's search ply, and those must be rare (I
would imagine), I don't see how a much weaker chess program (say 300
Elo points weaker) can beat a stronger one.



You appear to have forgotten about the fact that chess programs
often 'override' their calculation routines and eval functions while
in
the opening book. So then, theoretically at least, it is possible
for
a program with a great eval function and fine calculation speed to
blindly stumble into a lost position right in the opening! Of
course,
cases of this happening in real life are likely to be few and far
between since commercial programmers may do extensive testing.
But lest we forget, the current top program once saw its creator
answer a question regarding the program's endgame knowledge by
saying he thought it wasn't important for results. Now, imagine a
programmer whose attitude is that 'any decent opening book will
do.' Such a program could be subject to occasional (if infrequent)
upsets by much weaker programs before the brute-force engine
even begins to crank out torque.

Now, I expect such upsets to be quite rare since so many people
are obsessed with the chess openings and they are so very heavily
studied. But 300 points is not a huge gap. I recently was sent a
game in which a gap twice this large was surmounted between
human players for a draw, but just one simple improvement in the
endgame would have made this upset draw into an upset win!


  #3   Report Post  
Old September 9th 10, 08:58 AM posted to rec.games.chess.misc,rec.games.chess.computer
external usenet poster
 
First recorded activity by ChessBanter: Mar 2009
Posts: 3,536
Default Examples of much weaker programs beating stronger ones?

On Sep 9, 9:41*am, The Master wrote:

* Now, I expect such upsets to be quite rare since so many people
are obsessed with the chess openings and they are so very heavily
studied. *But 300 points is not a huge gap. * I recently was sent a
game in which a gap twice this large was surmounted between
human players for a draw, but just one simple improvement in the
endgame would have made this upset draw into an upset win!


We're talking computers Minor, not humans. Can you find me a game
with a 600 Elo difference that was won by the weaker machine?

RL
  #4   Report Post  
Old September 9th 10, 02:39 PM posted to rec.games.chess.misc,rec.games.chess.computer
Banned
 
First recorded activity by ChessBanter: Dec 2008
Posts: 2,402
Default Examples of much weaker programs beating stronger ones?


We're talking computers Minor, not humans. *Can you find me a game
with a 600 Elo difference that was won by the weaker machine?


GetClub once beat Jester 2 years back when Jester was arround 20 times
stronger than GetClub.

Today, even after 20 fold improvements GetClub finds it difficult to
beat Jester.

So Luck may favour a weak engine.

Bye
Sanny

If you have not played Chess at GetClub

You have never played any nice game in your life.

So, Play Chess at: http://www.GetClub.com/Chess.html


  #5   Report Post  
Old September 9th 10, 07:01 PM posted to rec.games.chess.computer
external usenet poster
 
First recorded activity by ChessBanter: May 2006
Posts: 179
Default Examples of much weaker programs beating stronger ones?

On Wed, 8 Sep 2010 17:12:15 -0700 (PDT), raylopez99
wrote:

I don't think they exist. That's because computers use the same
search engines at the brute force level, and the only intelligence is
the evaluation function.


I realize you're sort of joking here, but this statement is untrue.
Evaluation and search together provide "intelligence", and a program
without search (eg CPP1) is not going to be very intelligent. Plus,
searches are different in every program. Most programs use some sort
of forward pruning, so most modern programs are not pure "brute force"
programs. A stronger program may prune a good move that a weaker
program doesn't, and that could be a possible downfall for the
stronger program.

So unless a stronger computer walks into a
mating net that lies beyond it's search ply, and those must be rare (I
would imagine), I don't see how a much weaker chess program (say 300
Elo points weaker) can beat a stronger one.

Any examples?


Here are three examples of my weak program beating programs 300 ELO
above it, and one example of a draw against a program over 1000 ELO
above it (a lucky draw by threefold repetition against the strong
program Thinker). The win against Freyr was on time in a drawn
position, so I'm not sure if that counts using your criteria.

[Event "GoodEnginesAgain"]
[Site "TONY-SUA7L25HR1"]
[Date "2005.06.04"]
[Round "1"]
[White "TonysChess 0.01a"]
[Black "TSCP 1.81"]
[Result "1-0"]
[ECO "A70"]
[PlyCount "118"]
[EventDate "2005.??.??"]

1. Nf3 Nf6 2. d4 e6 3. c4 c5 4. d5 exd5 5. cxd5 d6 6. Nc3 g6 7. e4 Bg4
8. Bb5+
Nbd7 9. O-O Bg7 10. Bf4 Qe7 11. h3 Bxf3 12. Qxf3 h5 13. Bg5 O-O-O 14.
Bxd7+
Rxd7 15. Rae1 Rhd8 16. e5 dxe5 17. Ne4 Rxd5 18. Nxf6 Rd3 19. Re3 Rxe3
20. fxe3
Qe6 21. Ne4 f6 22. Bxf6 Bxf6 23. Qxf6 Qxf6 24. Rxf6 Rd1+ 25. Kh2 Ra1
26. Nc3 b5
27. Rc6+ Kd8 28. Rxc5 b4 29. Nd5 Rxa2 30. Nxb4 Rxb2 31. Nc6+ Ke8 32.
Rxe5+ Kd7
33. Nxa7 Rb4 34. e4 h4 35. Re7+ Kxe7 36. Nc6+ Kd6 37. Nxb4 Ke5 38. g3
g5 39.
Nd3+ Kxe4 40. Ne1 Kf5 41. gxh4 gxh4 42. Nf3 Ke4 43. Nxh4 Kf4 44. Ng2+
Kf3 45.
Kg1 Kg3 46. h4 Kg4 47. Kh2 Kf5 48. Kh3 Ke5 49. Kg4 Kf6 50. h5 Kf7 51.
Kf5 Kg7
52. Kg5 Kh7 53. h6 Kg8 54. Kg6 Kh8 55. Nf4 Kg8 56. Ng2 Kh8 57. Nh4 Kg8
58. Nf5
Kh8 59. Nd4 Kg8 1-0

[Event "GoodEnginesAgain"]
[Site "TONY-SUA7L25HR1"]
[Date "2005.06.04"]
[Round "1"]
[White "TonysChess 0.01a"]
[Black "SmallPotato 0.6.1"]
[Result "1-0"]
[ECO "D14"]
[PlyCount "90"]
[EventDate "2005.??.??"]

1. c4 c6 2. Nc3 d5 3. cxd5 cxd5 4. d4 Nf6 5. Nf3 Nc6 6. e3 a6 7. Bd3
Bg4 8. O-O
e5 9. Be2 Bxf3 10. Bxf3 e4 11. Be2 Bd6 12. h3 O-O 13. f3 b5 14. fxe4
dxe4 15.
Qc2 Nb4 16. Qb1 Re8 17. a4 bxa4 18. Nxa4 Nbd5 19. Nc5 Bxc5 20. dxc5
Qc7 21. Qc2
a5 22. Bd2 Reb8 23. Kf2 Qe5 24. Rxa5 Rxa5 25. Bxa5 Nxe3 26. Kxe3 Rxb2
27. Qc4
Qg3+ 28. Kd4 Qxg2 29. Qa6 Rb8 30. h4 Qg3 31. Qd6 Qxd6+ 32. cxd6 Kf8
33. Bc4 Ke8
34. Re1 Kf8 35. Ra1 h6 36. Bc7 Rc8 37. Bb5 Kg8 38. Bc6 h5 39. Rc1 Rf8
40. d7
Nxd7 41. Bxd7 f5 42. Ke5 g6 43. Bd6 Rf7 44. Be6 Kg7 45. Bxf7 e3 1-0

[Event "GoodEnginesAgain"]
[Site "TONY-SUA7L25HR1"]
[Date "2005.06.04"]
[Round "1"]
[White "TonysChess 0.01a"]
[Black "Freyr 1.0.67"]
[Result "1-0"]
[ECO "D38"]
[PlyCount "191"]
[EventDate "2005.??.??"]

1. d4 Nf6 2. c4 e6 3. Nc3 Bb4 4. Nf3 d5 5. Qa4+ Nc6 6. Ne5 Bd7 7. Nxd7
Qxd7 8.
e3 e5 9. dxe5 Bxc3+ 10. bxc3 Nxe5 11. Qxd7+ Nexd7 12. Rb1 O-O-O 13.
Ba3 Ne4 14.
Bb2 dxc4 15. Bxc4 Nd2 16. Kxd2 Ne5+ 17. Kc2 Nxc4 18. Bc1 Rhe8 19. Rb4
Re4 20.
Kb3 Nd2+ 21. Kb2 Re6 22. Rf4 f6 23. Ra4 Kb8 24. Rd4 Rxd4 25. cxd4 Rb6+
26. Kc2
Ne4 27. f3 Rc6+ 28. Kb3 Nf2 29. Rg1 f5 30. Bd2 g6 31. a3 g5 32. Rf1
Nd3 33. e4
f4 34. g3 fxg3 35. hxg3 Rb6+ 36. Kc3 Nb2 37. Bxg5 Na4+ 38. Kc4 Nb2+
39. Kc3
Na4+ 40. Kd3 Rb3+ 41. Kc4 Rb2 42. e5 h5 43. e6 b5+ 44. Kd5 Nb6+ 45.
Ke4 Re2+
46. Kf5 Nc4 47. Rc1 Nd6+ 48. Kf6 Rf2 49. f4 Rd2 50. Kg6 Rxd4 51. Kxh5
Rd3 52.
a4 Rxg3 53. Bd8 Ne8 54. axb5 Ng7+ 55. Kh6 Nxe6 56. Bg5 Nd4 57. Rb1 Rg2
58. Bf6
Nf5+ 59. Kh5 Ne3 60. Re1 Nc4 61. Bd4 Rd2 62. Re8+ Kb7 63. Be5 Na3 64.
f5 Nxb5
65. f6 Rf2 66. Kg6 Rg2+ 67. Kh5 Rf2 68. Kg6 Rf1 69. f7 Kc6 70. f8=Q
Rxf8 71.
Rxf8 a5 72. Ra8 Kb6 73. Bf6 Nd6 74. Bd8 Nb5 75. Kg7 Nd4 76. Kf7 Nc6
77. Bf6 Kc5
78. Ke8 Kb6 79. Kf8 Nb4 80. Kg7 Kb5 81. Bd8 Nd5 82. Rb8+ Kc6 83. Bg5
a4 84. Rb2
Kc5 85. Ra2 Kb5 86. Rd2 c6 87. Rb2+ Nb4 88. Bd2 c5 89. Rb1 a3 90. Kg6
Kc4 91.
Kf5 Kd3 92. Bxb4 cxb4 93. Rxb4 Kc2 94. Rc4+ Kb1 95. Ke6 Ka1 96. Kd7
1-0

[Event "GoodEnginesAgain"]
[Site "TONY-SUA7L25HR1"]
[Date "2005.06.04"]
[Round "1"]
[White "TonysChess 0.01a"]
[Black "Thinker 4.7a"]
[Result "1/2-1/2"]
[ECO "D02"]
[PlyCount "113"]
[EventDate "2005.??.??"]

1. Nf3 d5 2. d4 c6 3. Nc3 Nf6 4. e3 e6 5. Bd3 Nbd7 6. Ng5 h6 7. Nf3
Bd6 8. O-O
O-O 9. e4 dxe4 10. Nxe4 Nxe4 11. Bxe4 e5 12. dxe5 Nxe5 13. Nxe5 Bxe5
14. Qxd8
Rxd8 15. Re1 Re8 16. g3 Bg4 17. c3 Bc7 18. Bd2 Rad8 19. Bc2 Bb6 20.
Rxe8+ Rxe8
21. Re1 Rxe1+ 22. Bxe1 Bf3 23. a3 Kf8 24. b3 Ke7 25. c4 Bd4 26. a4 Ke6
27. h3
f5 28. Kf1 a6 29. Bd2 h5 30. a5 g6 31. h4 c5 32. Bc1 Bc3 33. Be3 Bb4
34. Kg1
Kd6 35. Bh6 Bxa5 36. Bf8+ Kc6 37. Be7 Bb4 38. Kf1 Bc3 39. Bf8 Bb4 40.
Be7 Bc3
41. Bf8 Bd4 42. Bh6 b6 43. Ke1 Kd7 44. Bf8 Bc3+ 45. Kf1 Ke6 46. Bh6
Kd6 47.
Bf4+ Kd7 48. Bh6 Kd6 49. Bf4+ Kd7 50. Bh6 Kc8 51. Bf8 Kc7 52. Be7 Kd7
53. Bf8
Ke8 54. Bd6 Kd8 55. Bf8 Kc7 56. Be7 Kd7 57. Bf8 1/2-1/2



Tony


RL

Here's what I found from the net...I'd like to see how Super Const.
beat Mephisto, probably though the three points are mostly draws.

http://ssdf.bosjo.net/rlwww101.txt

217 Mephisto Lyon 68000 12 MHz 2004 21 -21 1121
57% 1951
294 Novag Super Constellation 6502 4 MHz 1629 18 -18 1583
34% 1746

217 Mephisto Lyon 68000 12 MHz, 2004 match against Super Const.
17-3




  #6   Report Post  
Old September 9th 10, 08:25 PM posted to rec.games.chess.computer
external usenet poster
 
First recorded activity by ChessBanter: Jan 2006
Posts: 470
Default Examples of much weaker programs beating stronger ones?

On Sep 9, 11:01*am, Tony M wrote:
On Wed, 8 Sep 2010 17:12:15 -0700 (PDT), raylopez99

wrote:
I don't think they exist. *That's because computers use the same
search engines at the brute force level, and the only intelligence is
the evaluation function. *


I realize you're sort of joking here, but this statement is untrue.
Evaluation and search together provide "intelligence", and a program
without search (eg CPP1) is not going to be very intelligent. *Plus,
searches are different in every program. *Most programs use some sort
of forward pruning, so most modern programs are not pure "brute force"
programs. *A stronger program may prune a good move that a weaker
program doesn't, and that could be a possible downfall for the
stronger program.

So unless a stronger computer walks into a
mating net that lies beyond it's search ply, and those must be rare (I
would imagine), I don't see how a much weaker chess program (say 300
Elo points weaker) can beat a stronger one.


It is indeed very rare for a weaker program to beat a considerably
stronger one, but it does happen. My program, Myrddin, has never done
it (unless you allow for the stronger engine crashing, which has
happened twice). However, it does have perfectly legitimate DRAWS
against engines that are MUCH stronger.

Here are two. The first is against Komodo, probably rated 1000 points
stronger. The other is against Hermann, probably about 600 points
stronger.

[Event "BASEMENT_QUAD_2009-11-7 (1)"]
[Site "BASEMENT_QUAD"]
[Date "2010.08.03"]
[Round "14.1"]
[White "Komodo 1.2"]
[Black "Myrrdin 0.83k 64bits"]
[Result "1/2-1/2"]
[TimeControl "3600"]
[Annotator "29. +0.44 25... -1.51"]
[Number "262"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 O-
O 8.
c3 d5 9. exd5 Nxd5 10. Nxe5 Nxe5 11. Rxe5 c6 12. d4 Bd6 13. Re1 Qh4
14. g3
Qh3 15. Be3 Bg4 16. Qd3 Rae8 17. Nd2 Re6 18. a4 bxa4 19. Rxa4 f5 20.
Qf1
Qh5 21. f4 Rfe8 22. Rxa6 Rxe3 23. Rxe3 Rxe3 24. Rxc6 Qe8 25. Rxd6 Re1
{-1.51/16} 26. Bxd5+ Kf8 {-1.23/18} 27. Qxe1 Qxe1+ {-1.11/18} 28. Nf1
Be2
{-1.34/17 1:16} 29. Bg2 {+0.44/26 1:21} Bxf1 {-1.15/17 25} 30. Bxf1
{+0.19/26 1:19} h5 {-1.15/14 1:59} 31. Rd8+ {+0.60/20 1:17} Kf7
{-1.03/17 58} 32. Rh8 {+0.38/23 1:15} Kf6 {-1.18/15 2:12} 33. d5
{+0.06/21 1:14} Qe3+ {-0.82/15 53} 34. Kg2 {+0.00/25 1:12} Qd2+
{-1.00/16 23} 35. Kh3 {+0.00/28 1:10} Qxd5 {-0.84/14 2:48} 36. Re8
{+0.00/26 1:08} g5 {-0.62/15 55} 37. fxg5+ {+0.00/23 1:06} Kxg5
{-0.76/15 1:10} 38. c4 {+0.00/25 1:05} Qd2 {-0.49/16 4:42} 39. Rg8+
{+0.00/21 1:03} Kf6 {-0.16/14 1:47} 40. Bg2 {+0.00/23 1:02} Qxb2
{-0.04/13 1:26} 41. Rf8+ {-0.09/22 1:00} Kg5 {+0.00/18 4} 42. Rg8+
{-0.09/25 58} Kf6 {+0.00/19 45} 43. Rf8+ {+0.00/31 57} Kg5 {+0.00/15
39}
44. Rg8+ {+0.00/37 55} Kf6 {+0.00/21 56}
{Draw by Repetition} 1/2-1/2

[Event "BASEMENT_QUAD_2009-11-7 (1)"]
[Site "BASEMENT_QUAD"]
[Date "2010.08.04"]
[Round "15.1"]
[White "Myrrdin 0.83k 64bits"]
[Black "Hermann 2.536"]
[Result "1/2-1/2"]
[TimeControl "3600"]
[Annotator "18. +0.21 17... -0.17"]
[Number "282"]

1. d4 d5 2. c4 e6 3. Nc3 c6 4. Nf3 dxc4 5. e3 b5 6. a4 Bb4 7. Bd2 a5
8.
axb5 Bxc3 9. Bxc3 cxb5 10. b3 Bb7 11. bxc4 b4 12. Bb2 Nf6 13. Bd3 Nbd7
14.
O-O O-O 15. Qc2 Qc7 16. e4 e5 17. Rfe1 exd4 {-0.17/14 1:30} 18. Bxd4
{+0.21/12} Nh5 19. c5 {+0.04/11 5:02} Nf4 20. Bb5 {+0.27/13 54} Ba6
{+0.05/14 1:24} 21. Bxd7 {+0.15/13 2:55} Qxd7 {-0.02/13 1:21} 22. Be5
{-0.02/15 22} Nd3 23. Red1 {+0.03/15} Qb5 24. Bd6 {-0.04/14} Rfe8 25.
Nd4
{+0.22/15} Qc4 26. Qxc4 {+0.00/15 1:18} Bxc4 {+0.13/17 3:53} 27. c6
{-0.21/15 5} Rxe4 28. Nf5 {-0.20/14 1:37} Ree8 {+0.00/15 1:04} 29. c7
{-0.56/13 1:18} b3 {+0.42/15 1:03} 30. Ne7+ {-0.56/14 1:01} Kh8
{+0.45/16 1:01} 31. Ra3 {-0.60/14 2:13} a4 {+0.76/15 59} 32. f4
{-0.58/14 1:32} b2 {+0.78/16 58} 33. Rc3 {+0.03/14 33} Nc1 {+0.88/17
56}
34. Rxc4 {-0.78/14 40} b1=Q {+0.21/15 55} 35. Kh1 {-0.78/11 34} a3
{+0.90/16 54} 36. Rcxc1 {-0.26/14 30} Qb7 {+0.89/16 52} 37. Bxa3
{-0.53/14 3:21} h6 38. Bd6 {-0.55/14} f6 39. h3 {-0.53/10 3:01} Kh7
40. Rd2
{+0.00/17 2:42} Rxe7 {+0.04/17 47} 41. Bxe7 {+1.21/13 16} Rc8
{+0.00/15 46}
42. Bd6 {+1.13/14 1:10} Qa6 {-0.04/15 45} 43. Kh2 {+1.35/13 15} Qb5
{-0.24/15 44} 44. Rcc2 {+1.19/13 4} Qb3 45. Rb2 {+1.12/12 2.4} Qc3 46.
g3
{+1.13/15 24} h5 {-0.21/17 41} 47. Rg2 {+1.21/13} h4 48. gxh4
{+1.63/14 59}
Qd3 {-0.22/18 39} 49. Rbd2 {+1.51/14 48} Qc4 {-0.22/15 38} 50. Rg3
{+1.33/12 36} Qe6 {-0.22/14 37} 51. Ra3 {+1.41/14 1:45} Qe1 {-0.08/17
36}
52. Rg2 {+1.33/14 42} Qd1 {+0.00/15 35} 53. Ra6 {+1.30/14 31} Re8
{+0.00/13 34} 54. h5 {+1.02/12 16} Qxh5 {+0.00/14 33} 55. Rc6
{+0.81/12 35}
Qf5 {+0.27/13 32} 56. Rc5 {+0.66/12 1:24} Qd7 {+0.00/15 31} 57. Rd5
{+0.81/12 28} Qc6 {+0.27/13 31} 58. Rd3 {+0.73/12 41} Rg8 {+0.07/14
58} 59.
Rgd2 {+0.68/12 25} Ra8 {+0.14/14 28} 60. Rg3 {+0.79/12 1:09} Qd7
{+0.03/15 28} 61. Rc3 {+0.97/12 35} Rc8 {+0.07/13 27} 62. Rcc2
{+0.79/13 36} Qa4 {+0.06/14 26} 63. Rc5 {+0.72/13} Qb3 64. Rh5+
{+0.87/13 14} Kg6 {+0.00/18 25} 65. Rhd5 {+0.79/14 1:14} Qc4 66. f5+
{+1.07/12} Kh5 67. R5d4 {+1.03/14 22} Qc6 {+0.00/18 23} 68. Bg3
{+0.83/14 21} Rxc7 {+0.00/16 23} 69. Bxc7 {+0.80/15} Qxc7+ 70. Kg2
{+0.80/16 19} Qc6+ 71. Rd5 {+0.80/16 12} Qb7 {+0.00/21 21} 72. Rd3
{+0.83/14 17} Qc6 {+0.00/20 20} 73. Kf2 {+0.83/14 11} Qb6+ {+0.00/17
20}
74. Ke2 {+0.82/15 25} Qb1 {+0.00/17 19} 75. R5d4 {+0.63/14 11} Qc2+
{+0.00/13 19} 76. Kf1 {+0.23/13 24} Qc1+ {+0.00/16 18} 77. Kg2
{+0.17/14 1:13} Qc5 {+0.00/17 18} 78. Rd8 {+0.18/13 11} Qc2+ {+0.00/17
18}
79. Kf1 {+0.00/13 12} Qc1+ {+0.00/18 17} 80. Rd1 {+0.00/14 9} Qf4+
{+0.00/17 17} 81. Kg2 {+0.00/15} Qe4+ 82. Kg3 {+0.00/15} Qe3+ 83. Kg2
{+0.00/16} Qe4+ 84. Kh2 {+0.00/14 19} Qf4+ {+0.00/20 15} 85. Kh1
{+0.00/15 8} Qf3+ {+0.00/16 15} 86. Kh2 {+0.00/17} Qf2+ 87. Kh1
{+0.00/16}
Qf3+ 88. Kh2 {+0.00/20} Qf2+ 89. Kh1 {+0.00/5 58} Qf3+
{Draw by repetition} 1/2-1/2

Myrddin has also drawn DanaSah and Amyan, and both are about as strong
as Hermann.

jm
  #7   Report Post  
Old September 9th 10, 09:55 PM posted to rec.games.chess.misc,rec.games.chess.computer
external usenet poster
 
First recorded activity by ChessBanter: Dec 2009
Posts: 1,146
Default Examples of much weaker programs beating stronger ones?

On Sep 9, 3:58*am, raylopez99 wrote:

On Sep 9, 9:41*am, The Master wrote:

* Now, I expect such upsets to be quite rare since so many people
are obsessed with the chess openings and they are so very heavily
studied. *But 300 points is not a huge gap. * I recently was sent a
game in which a gap twice this large was surmounted between
human players for a draw, but just one simple improvement in the
endgame would have made this upset draw into an upset win!


We're talking computers Minor, not humans. *Can you find me a game
with a 600 Elo difference that was won by the weaker machine?



That depends... on how much you are willing to pay.

BTW, that was very sneaky of you, substituting '600' where you
originally had '300'! Sneaky tactical tricks like that may work on
the weak-minded (i.e. TK, MH, and their ilk) but your Jedi mind
tricks won't work on me. (Oh, and you are supposed to use The
Force for *good*, not evil. Duh.)
  #8   Report Post  
Old September 9th 10, 10:10 PM posted to rec.games.chess.misc,rec.games.chess.computer
external usenet poster
 
First recorded activity by ChessBanter: Dec 2009
Posts: 1,146
Default Examples of much weaker programs beating stronger ones?

On Sep 9, 9:39*am, Sanny wrote:

We're talking computers Minor, not humans. *Can you find me a game
with a 600 Elo difference that was won by the weaker machine?


GetClub once beat Jester 2 years back when Jester was arround 20 times
stronger than GetClub.

Today, even after 20 fold improvements GetClub finds it difficult to
beat Jester.

So Luck may favour a weak engine.



In this case, it may be that the operator had something to do
with the outcome. I think what Phil-Ray is looking for is a game
in which a significantly weaker program won on its own, sans any
hanky-panky or interference or incompetence by the computer
operator.

One way to find such games is to use the google search engine.
Advanced techniques like typing two words into a search engine
are far beyond the abilities of Phil-Ray, because after typing them
in correctly you must press the 'enter' key and then examine the
results. Note that this process involves more than one easy step,
thus exceeding his limited abilities.
  #9   Report Post  
Old September 9th 10, 10:31 PM posted to rec.games.chess.misc,rec.games.chess.computer
external usenet poster
 
First recorded activity by ChessBanter: Mar 2009
Posts: 3,536
Default Examples of much weaker programs beating stronger ones?

On Sep 10, 12:10*am, The Master wrote:

* One way to find such games is to use the google search engine.
Advanced techniques like typing two words into a search engine
are far beyond the abilities of Phil-Ray, because after typing them
in correctly you must press the 'enter' key and then examine the
results. *Note that this process involves more than one easy step,
thus exceeding his limited abilities. *


Why don't you Google it and post then? I'll shut up if you do.

Cordially,

Phil (but you can call me Ray)
  #10   Report Post  
Old September 9th 10, 10:33 PM posted to rec.games.chess.misc,rec.games.chess.computer
external usenet poster
 
First recorded activity by ChessBanter: Mar 2009
Posts: 3,536
Default Examples of much weaker programs beating stronger ones?

On Sep 9, 4:39*pm, Sanny wrote:

GetClub once beat Jester 2 years back when Jester was arround 20 times
stronger than GetClub.

Today, even after 20 fold improvements GetClub finds it difficult to
beat Jester.

So Luck may favour a weak engine.

Bye
Sanny


Please post a game where your weak "GetClub" beats your strong
"Jester".

That would be interesting. Maybe if more people visit your site (if
you make it interesting, like showing a weak engine beating a strong)
you might get more popular. Less people visiting, less popular. So
try something unique like what I suggest.

RL
Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Can an open source chess program become number one? Mapleleaf rec.games.chess.computer (Computer Chess) 20 June 6th 08 03:21 AM
Computer vs. human matches and Chess256 M Winther rec.games.chess.misc (Chess General) 159 September 14th 07 11:21 AM
Computer vs. human matches and Chess256 Guy Macon rec.games.chess.computer (Computer Chess) 155 September 8th 07 03:20 PM
An old article about early chess computers! Quite amusing rant. [email protected] rec.games.chess.computer (Computer Chess) 0 January 21st 07 06:32 AM


All times are GMT +1. The time now is 01:45 AM.

Powered by vBulletin® Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Copyright 2004-2019 ChessBanter.
The comments are property of their posters.
 

About Us

"It's about Chess"

 

Copyright © 2017