Author: Uri Blass
Date: 22:13:17 10/11/04
Go up one level in this thread
On October 11, 2004 at 19:03:05, Albert Silver wrote:
>On October 11, 2004 at 15:49:07, Enrique Irazoqui wrote:
>
>>I have Chrilly's permission to post these remarks he emailed me about Hydra:
>>
>>"Winning against GMs has to a certain extent already lost its charm. We are
>>already in the position were we can only lose. I personally think that this
>>expectation overestimates the capabilities of the program. But it is a matter of
>>fact that this version played 8 games against top GMs, won 6 and made 2 draws.
>>Gives an Elo-rating of >2900."
>>
>>"Leonxto (Leontxo García, well known chess journalist from Spain) wrote in El
>>Pais: "y ademas muestra uns unaudita "comprension" de la estrategia" (and it
>>also shows an unheard of "understanding" of the strategy). Actually it has no
>>idea at all about this. It has some simple rules like "It is beneficial to
>>attack the area around the opponent king, to attack opponent pieces, to control
>>the center. It knows a little bit what abut bad/good bishops, when is the knight
>>stronger than the bishop". The big surprise for me was, that even this simple
>>rules were sufficient to know that the exchange of the light-squared bishop of
>>Ponomariov in the first match was not the best idea. The evaluation went up a
>>1/4 of a pawn. Generally I saw no GM move which had some flavour of geniality (I
>>am speaking only of the Hydra-games). The moves which really surprised me (e.g.
>>Qe2 after Nc4 in the last match) were done by Hydra."
>
>Thanks for sharing. As to the remarks on its knowledge, I must say I'm not
>terribly surprised. Similar remarks have been noted on Fritz when it was a far
>more 'ignorant' program. The depth of the search allows it to see the
>consequences of its moves sufficiently to compensate for a lack of knowledge,
>and my oh my is Hydra going deeply. Even Ed Schroeder has noted this on Rebel,
>and the latest versions have the knowledge parameter at a mere 100 when the
>maximum goes all the way to 500.
Note that in the case of Rebel based on my understanding reducing the knowledge
parameter only means that Rebel does not use the knowledge in the qsearch.
Rebel does full evaluation and not lazy evaluation at every earlier ply so
it is misleading to claim that Rebel does not know because of smaller knowledge
and if Rebel can find something with knowledge of 500 then it can also find it
with knowledge of 100 and the only demage is going to be that it may find it
some plies later.
Note also that seeing deeper means better positional moves and not only tactics
and IMO if you give computers to play random moves that are not tactical
mistakes they will probably not play better than 1800.
Based on my experience even weak players can play games with no tactical
mistakes if the opponent does not cause them problems.
Here is an example of a game that I played on the chessbase server when even
1438 player at 3 minutes per game failed to do tactical mistakes for more than
50 moves when I did not play moves that caused him problems.
Interesting to note that with 9.1 seconds to finish the game against 18 seconds
of the opponent I blundered by allowing my opponent to force mate in 4 and my
opponent found the first 3 moves but missed the last mating move.
The fact that I knew about the premove and my opponent did not know about it
helped me to win and you can see that I played for a win and refused to repeat
moves even when I had inferior position and I knew that I was not using the best
mouse so moving the mouse to some piece that is in the other side of the board
may take expensive seconds.
Here is the game analyzed by yace
[Event "Rated game, 3m + 0s"]
[Site "Main Playing Hall"]
[Date "2004.10.06"]
[Round "?"]
[White "Uri_Blass"]
[Black "Lagdamys"]
[Result "1-0"]
[ECO "B01"]
[WhiteElo "1909"]
[BlackElo "1438"]
[Annotator ",agur"]
[PlyCount "167"]
[EventDate "2004.10.06"]
[TimeControl "180"]
1. e4 {[%emt 0:00:01]} d5 {[%emt 0:00:01]} 2. exd5 {[%emt 0:00:01]} Qxd5 {
[%emt 0:00:00]} 3. Nc3 {[%emt 0:00:01]} Qa5 {[%emt 0:00:00]} 4. Nf3 {
[%emt 0:00:01]} Nf6 {[%emt 0:00:00]} 5. Be2 {[%emt 0:00:03]} Bg4 {
last book move [%emt 0:00:01]} 6. O-O {[%emt 0:00:01]} Bxf3 {[%emt 0:00:00]} 7.
Bxf3 {[%emt 0:00:01]} c6 {[%emt 0:00:00]} 8. d3 {[%emt 0:00:02]} Nbd7 {
[%emt 0:00:00]} 9. Bd2 {[%emt 0:00:01]} e6 {[%emt 0:00:01]} 10. a3 {
[%emt 0:00:08]} Qc7 {[%emt 0:00:02]} 11. Re1 {[%emt 0:00:02]} Bd6 {
[%emt 0:00:01]} 12. h3 {[%emt 0:00:03]} O-O {[%emt 0:00:00]} 13. Ne4 {
[%emt 0:00:01]} Nxe4 {[%emt 0:00:01]} 14. Bxe4 {[%emt 0:00:03]} Nf6 {
[%emt 0:00:00]} 15. Bf3 {[%emt 0:00:02]} Rad8 {[%emt 0:00:01]} 16. Bg5 {
[%emt 0:00:03]} Be7 {[%emt 0:00:03]} 17. Qd2 {[%emt 0:00:03]} h6 {
[%emt 0:00:01]} 18. Bf4 {[%emt 0:00:10]} Bd6 {[%emt 0:00:04]} 19. Bxd6 {
[%emt 0:00:16]} Qxd6 {[%emt 0:00:01]} 20. Re2 {[%emt 0:00:08]} Nd5 {
[%emt 0:00:04]} 21. Bxd5 {[%emt 0:00:06]} cxd5 {[%emt 0:00:05]} 22. Rae1 {
[%emt 0:00:03]} Rc8 {[%emt 0:00:03]} 23. c3 {[%emt 0:00:02]} a6 {[%emt 0:00:08]
} 24. Qe3 {[%emt 0:00:03]} b5 {[%emt 0:00:04]} 25. Qe5 {[%emt 0:00:02]} Qxe5 {
[%emt 0:00:05]} 26. Rxe5 {[%emt 0:00:00]} a5 {[%emt 0:00:04]} 27. d4 {
[%emt 0:00:06]} b4 {[%emt 0:00:03]} 28. axb4 {[%emt 0:00:03]} axb4 {
[%emt 0:00:01]} 29. cxb4 {[%emt 0:00:02]} Rb8 {[%emt 0:00:03]} 30. R5e2 {
[%emt 0:00:08]} Rxb4 {[%emt 0:00:01]} 31. Rd2 {[%emt 0:00:01]} Rfb8 {
[%emt 0:00:02]} 32. Rb1 {[%emt 0:00:04]} Rb3 {[%emt 0:00:09]} 33. f3 {
[%emt 0:00:04]} R3b4 {[%emt 0:00:06]} 34. Kf2 {[%emt 0:00:01]} Kf8 {
[%emt 0:00:04]} 35. Ke3 {[%emt 0:00:05]} Ke7 {[%emt 0:00:00]} 36. Kd3 {
[%emt 0:00:03]} Rb3+ {[%emt 0:00:02]} 37. Kc2 {[%emt 0:00:04]} R3b4 {
[%emt 0:00:12]} 38. b3 {[%emt 0:00:01]} Kd6 {[%emt 0:00:04]} 39. Rd3 {
[%emt 0:00:02]} R8b5 {[%emt 0:00:04]} 40. Rb2 {[%emt 0:00:05]} e5 {
[%emt 0:00:03]} 41. dxe5+ {[%emt 0:00:01]} Kxe5 {[%emt 0:00:02]} 42. Re3+ {
[%emt 0:00:02]} Kd6 {[%emt 0:00:02]} 43. Rd3 {[%emt 0:00:04]} Rc5+ {
[%emt 0:00:01]} 44. Kb1 {[%emt 0:00:03]} d4 {[%emt 0:00:04]} 45. Ka2 {
[%emt 0:00:03]} Kd5 {[%emt 0:00:03]} 46. Rbd2 {[%emt 0:00:02]} Ra5+ {
[%emt 0:00:02]} 47. Kb2 {[%emt 0:00:01]} Rab5 {[%emt 0:00:03]} 48. g3 {
[%emt 0:00:01]} f5 {[%emt 0:00:02]} 49. f4 {[%emt 0:00:01]} g6 {[%emt 0:00:01]}
50. Rd1 {[%emt 0:00:02]} Ke4 {[%emt 0:00:02]} 51. R1d2 {[%emt 0:00:01]} Kd5 {
[%emt 0:00:01]} 52. Rf3 {[%emt 0:00:02]} Ke4 {[%emt 0:00:03]} 53. Rdd3 {
[%emt 0:00:03]} ({-1.27 Yace 0.99.87:} 53. Rfd3 h5 54. Kc2 Rc5+ 55. Kb2 Ra5 56.
Kb1 Ra7 57. h4 Rc7 58. Kb2 Rc8 59. Ka3 g5 {[%eval -30,12]}) 53... h5 {
[%emt 0:00:04]} 54. h4 {[%emt 0:00:01]} ({-2.37 Yace 0.99.87:} 54. Kc1 Ra5 55.
Kd1 Rb6 56. Kc2 Rc6+ 57. Kb1 Rca6 58. b4 Ra4 59. Kc2 Rb6 60. Kd1 Ra1+ 61. Ke2
Ra2+ 62. Rd2 Rxd2+ 63. Ke1 {[%eval -127,12]}) 54... Kd5 {[%emt 0:00:02]} ({
-0.05 Yace 0.99.87:} 54... Ra4 55. Kc2 Ra2+ 56. Kd1 Rb7 57. b4 Rxb4 58. Rb3
Rxb3 59. Rxb3 Rg2 60. Ra3 d3 {[%eval -237,13]}) 55. Kc2 {[%emt 0:00:01]} Ke4 {
[%emt 0:00:01]} 56. Kd2 {[%emt 0:00:01]} ({-2.24 Yace 0.99.87:} 56. Rf1 Kd5 57.
Re1 Kd6 58. Ra1 Kc5 59. Re1 Rb8 60. Re6 R8b6 61. Re7 Rxb3 62. Rxb3 Re6 63. Rxe6
Kd5 {[%eval -4,13]}) 56... Kd5 {[%emt 0:00:01]} ({-0.04 Yace 0.99.87:} 56...
Ra5 57. Kc2 Rb7 58. Kb2 Rb8 59. Rxd4+ Kxd4 60. b4 {[%eval -224,14]}) 57. Ke2 {
[%emt 0:00:00]} ({-1.16 Yace 0.99.87:} 57. Rfe3 Kd6 58. Re8 Kd7 59. Re2 Ra5 60.
Ke1 Ra6 61. Kf2 Re6 62. Rxe6 Kc7 {[%eval -4,12]}) 57... Ke4 {[%emt 0:00:01]}
58. Kf2 {[%emt 0:00:00]} Kd5 {[%emt 0:00:01]} ({-0.03 Yace 0.99.87:} 58... Rb6
59. Rd2 Rxb3 60. Re2+ Kd5 61. Rxb3 Rxb3 62. Re5+ Kd6 63. Ra5 Rc3 64. Ra8 Kd7
65. Ra6 Rc6 66. Ra8 {[%eval -116,13]}) 59. Kf1 {[%emt 0:00:01]} ({
-1.15 Yace 0.99.87:} 59. Rfe3 Rc5 60. Rd2 Rc6 61. Re5+ Kd6 62. Re8 Kd7 63. Ra8
Re6 64. Ra7+ Kd6 65. Rd3 Kd5 66. Rd7+ Ke4 {[%eval -3,12]}) 59... Ke4 {
[%emt 0:00:01]} 60. Ke1 {[%emt 0:00:00]} ({-2.14 Yace 0.99.87:} 60. Kf2 Rc5 61.
Rd2 Rc3 62. Re2+ Kd5 63. Rb2 d3 64. Re3 Rbxb3 65. Rd2 Rc2 66. Rexd3+ Kc6 {
[%eval -115,13]}) 60... Kd5 {[%emt 0:00:00]} ({-0.04 Yace 0.99.87:} 60... Ra5
61. Kf2 Ra2+ 62. Kg1 Rb7 63. b4 Rxb4 64. Rb3 Rxb3 65. Rxb3 d3 66. Rb4+ Kf3 67.
Kh1 d2 68. Kh2 d1=Q+ 69. Kh3 {[%eval -214,14]}) 61. Kd2 {[%emt 0:00:00]} ({
-1.12 Yace 0.99.87:} 61. Rfe3 Kd6 62. Ke2 Ra5 63. Rd2 Ra1 64. Red3 Kd5 65. Re3
{[%eval -4,12]}) 61... Ke4 {[%emt 0:00:00]} 62. Kc1 {[%emt 0:00:01]} ({
-1.84 Yace 0.99.87:} 62. Ke2 Ra5 63. Rd2 Ra3 64. Rb2 Ra1 65. Rd3 Rh1 66. Rdd2
Rb1 67. Rxb1 Kd5 {[%eval -112,13]}) 62... Kd5 {[%emt 0:00:00]} ({
-0.08 Yace 0.99.87:} 62... Ra5 63. Kd1 Rb7 64. Kd2 Rba7 65. b4 Ra2+ 66. Kd1
Ra1+ 67. Kc2 R7a2+ 68. Kb3 Ra3+ 69. Kc2 R1a2+ 70. Kb1 Rxd3 71. Rxd3 Kxd3 72.
Kxa2 g5 73. fxg5 Kc3 74. g6 f4 75. gxf4 d3 76. g7 d2 77. g8=Q d1=Q {
[%eval -184,13]}) 63. Kb2 {[%emt 0:00:00]} ({-1.58 Yace 0.99.87:} 63. Rfe3 Rb8
64. Kd2 Ra8 65. Re7 Ra2+ 66. Kc1 Ra3 67. Kc2 Kd6 68. Rg7 Raxb3 69. Rxb3 Kc5 {
[%eval -8,12]}) 63... Rc5 {[%emt 0:00:01]} ({-0.07 Yace 0.99.87:} 63... Ke4 64.
Kc2 Rc5+ 65. Kd2 Ra5 66. Ke1 Rb7 67. Ke2 Ra2+ 68. Kd1 Rba7 69. b4 Ra1+ 70. Kc2
{[%eval -158,14]}) 64. Ka3 {[%emt 0:00:00]} Rb6 {[%emt 0:00:02]} 65. b4 {
[%emt 0:00:01]} ({-1.56 Yace 0.99.87:} 65. Rfe3 Ra5+ 66. Kb2 Ra4 67. Re5+ Kd6
68. Kc1 Rab4 69. Kd1 Ra6 70. Ke2 Ra2+ 71. Kf3 Rb2 72. Ree3 {[%eval -4,11]})
65... Rc4 {[%emt 0:00:02]} 66. Rb3 {[%emt 0:00:04]} Ra6+ {[%emt 0:00:01]} 67.
Kb2 {[%emt 0:00:01]} Rac6 {[%emt 0:00:03]} 68. b5 {[%emt 0:00:01]} Rc2+ {
[%emt 0:00:01]} 69. Ka3 {[%emt 0:00:01]} Rb6 {[%emt 0:00:02]} 70. Kb4 {
[%emt 0:00:01]} Rc4+ {[%emt 0:00:02]} 71. Ka5 {[%emt 0:00:02]} Rb8 {
[%emt 0:00:03]} 72. b6 {[%emt 0:00:00]} ({-#4 Yace 0.99.87:} 72. Ra3 Kc5 73.
Ka6 Rb6+ 74. Ka7 Rxb5 75. Rf1 Rc3 76. Rfa1 Kc4 77. Ka6 Rxa3+ 78. Rxa3 Rb4 79.
Ra4 {[%eval -155,12]}) 72... Ra8+ {[%emt 0:00:01]} 73. Kb5 {[%emt 0:00:01]}
Rc5+ {[%emt 0:00:01]} 74. Kb4 {[%emt 0:00:01]} Raa5 {[%emt 0:00:02]} 75. b7 {
[%emt 0:00:01]} Rab5+ {[%emt 0:00:01]} ({4.22 Yace 0.99.87:} 75... Rcb5# {
[%eval -32766,7]}) 76. Ka3 {[%emt 0:00:01]} Rxb3+ {[%emt 0:00:02]} ({
11.11 Yace 0.99.87:} 76... Ke4 77. Rxb5 Rxb5 78. Rb3 Rxb7 79. Rxb7 Kf3 80. Rb3+
Ke2 81. Rb6 d3 82. Re6+ Kf3 83. Rxg6 d2 84. Rd6 Kxg3 {[%eval 382,11]}) 77. Rxb3
{[%emt 0:00:00]} Ra5+ {[%emt 0:00:03]} 78. Kb4 {[%emt 0:00:01]} Kc6 {
[%emt 0:00:04]} 79. b8=Q {[%emt 0:00:02]} Rb5+ {[%emt 0:00:00]} ({
#5 Yace 0.99.87:} 79... Ra1 80. Kc4 Rc1+ 81. Kxd4 Rd1+ 82. Ke3 Re1+ 83. Kd2 Ra1
84. Qb5+ Kc7 85. Qe5+ Kc8 86. Qxa1 {[%eval 1545,10]}) 80. Qxb5+ {[%emt 0:00:02]
} Kd6 {[%emt 0:00:01]} 81. Qc5+ {[%emt 0:00:00]} Ke6 {[%emt 0:00:01]} 82. Qxd4
{[%emt 0:00:00]} Ke7 {[%emt 0:00:01]} 83. Rc3 {[%emt 0:00:00]} Ke6 {
[%emt 0:00:01]} 84. Rc5 {Zeit (Lag: Av=0.27s, max=0.5s) [%emt 0:00:00]} 1-0
Uri
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.