Why is Houdini 4 so clueless?

ernst · Post by **ernst** » Mon Mar 03, 2014 11:05 pm

Time and again I see houdini being clueless against stockfish of the danger that is upon it until it is too late. To be honest it frustates me that a €60 engine gets outplayed so often by a free engine.

An example.

Code: Select all

&#91;Event "Rated game, 5 min"&#93;
&#91;Site "Engine Room"&#93;
&#91;Date "2014.03.03"&#93;
&#91;Round "?"&#93;
&#91;White "Stockfish 160214 64"&#93;
&#91;Black "Houdini 4 Pro x64"&#93;
&#91;Result "1-0"&#93;
&#91;ECO "B92"&#93;
&#91;WhiteElo "2560"&#93;
&#91;BlackElo "2582"&#93;
&#91;Annotator "0.12;0.00"&#93;
&#91;PlyCount "152"&#93;
&#91;EventDate "2014.03.03"&#93;
&#91;TimeControl "300"&#93;

1. e4 &#123;B 0&#125; c5 &#123;B 0&#125; 2. Nf3 &#123;B 0&#125; d6 &#123;B 0&#125; 3. d4 &#123;B 0&#125; cxd4 &#123;B 0&#125; 4. Nxd4 &#123;
B 0&#125; Nf6 &#123;B 0&#125; 5. Nc3 &#123;B 0&#125; a6 &#123;B 0&#125; 6. Be3 &#123;B 0&#125; e5 &#123;B 0&#125; 7. Nb3 &#123;B 0&#125; Be6 &#123;
B 0&#125; 8. Be2 &#123;B 0&#125; Be7 &#123;B 0&#125; 9. O-O &#123;B 0&#125; O-O &#123;B 0&#125; 10. Qd3 &#123;B 0&#125; Nc6 &#123;B 0&#125; 11.
a3 &#123;0.12/23 7&#125; d5 &#123;B 0 &#40;Tc8&#41;&#125; 12. exd5 &#123;0.38/25 5&#125; Nxd5 &#123;B 0&#125; 13. Nxd5 &#123;0.37/
27 9&#125; Qxd5 &#123;B 0&#125; 14. Rad1 &#123;0.30/28 5&#125; Qxd3 &#123;B 0&#125; 15. Bxd3 &#123;0.32/28 5&#125; Rac8 &#123;B
0 &#40;f5&#41;&#125; 16. Nc5 &#123;0.27/25 7&#125; Bxc5 &#123;B 0&#125; 17. Bxc5 &#123;0.33/28 5&#125; Rfd8 &#123;B 0 &#40;Tfe8&#41;&#125;
18. Bb6 &#123;0.26/26 5&#125; Rd7 &#123;B 0&#125; 19. Rfe1 &#123;0.30/28 5&#125; f6 &#123;B 0 &#40;g6&#41;&#125; 20. Bc4 &#123;0.41/
27 12&#125; Re7 &#123;B 0&#125; 21. Bf1 &#123;0.36/27 6&#125; Kf7 &#123;B 0&#125; 22. c4 &#123;0.27/27 6&#125; Nb8 &#123;B 0 
&#40;Td7&#41;&#125; 23. b3 &#123;0.37/25 7&#125; Nd7 &#123;B 0 &#40;Td7&#41;&#125; 24. Be3 &#123;0.37/27 7&#125; Nc5 &#123;B 0&#125; 25.
Bxc5 &#123;0.19/29 8&#125; Rxc5 &#123;0.00/21 3&#125; 26. b4 &#123;0.19/31 4 &#40;f3&#41;&#125; Rcc7 &#123;0.00/25 7&#125; 27.
c5 &#123;0.18/32 0&#125; Red7 &#123;0.00/24 4&#125; 28. f3 &#123;0.19/31 2&#125; g6 &#123;0.00/25 5 &#40;Lb3&#41;&#125; 29. h4
&#123;0.13/29 7 &#40;g3&#41;&#125; Bb3 &#123;0.00/25 5&#125; 30. Rxd7+ &#123;0.13/30 0&#125; Rxd7 &#123;0.02/24 1&#125; 31. Re3
&#123;0.13/31 3&#125; Be6 &#123;-0.03/27 2&#125; 32. Rd3 &#123;0.15/30 2&#125; Rd4 &#123;0.00/27 4&#125; 33. Rxd4 &#123;0.
15/31 1&#125; exd4 &#123;0.00/27 0&#125; 34. Kf2 &#123;0.19/32 5&#125; Bd5 &#123;0.00/29 0&#125; 35. Bd3 &#123;0.15/33
4&#125; Ke6 &#123;0.00/28 0 &#40;f5&#41;&#125; 36. g3 &#123;0.00/33 5&#125; f5 &#123;0.00/28 0 &#40;Lc6&#41;&#125; 37. a4 &#123;0.25/
34 4 &#40;Ke2&#41;&#125; Bc6 &#123;0.00/28 4 &#40;Kd7&#41;&#125; 38. a5 &#123;0.25/37 5&#125; Kd5 &#123;0.00/28 0&#125; 39. Ke2 &#123;
0.25/37 4&#125; Ba4 &#123;0.00/27 5 &#40;Ke6&#41;&#125; 40. Kd2 &#123;0.57/35 5 &#40;f4&#41;&#125; Bd7 &#123;0.00/30 4&#125; 41.
f4 &#123;0.65/38 0&#125; h5 &#123;0.00/31 3 &#40;h6&#41;&#125; 42. Bf1 &#123;1.64/36 5&#125; Bc6 &#123;0.00/33 0&#125; 43. Kc2
&#123;1.80/39 4 &#40;Bd3&#41;&#125; Ba4+ &#123;0.84/26 18&#125; 44. Kc1 &#123;1.80/47 0 &#40;Kb2&#41;&#125; Bc6 &#123;0.00/23 4 
&#40;d3&#41;&#125; 45. Kd2 &#123;2.95/39 4 &#40;Kb2&#41;&#125; d3 &#123;0.84/22 3&#125; 46. Bxd3 &#123;1.80/42 0&#125; Kd4 &#123;0.86/
26 4&#125; 47. Be2 &#123;1.80/46 0&#125; Be4 &#123;0.90/26 6 &#40;Ld5&#41;&#125; 48. b5 &#123;1.81/48 6&#125; axb5 &#123;0.90/
27 0&#125; 49. Bxb5 &#123;1.81/50 3&#125; Kxc5 &#123;0.88/26 0&#125; 50. Be8 &#123;1.81/51 3&#125; Bf3 &#123;0.89/26 0&#125;
51. Bxg6 &#123;1.81/50 6&#125; Bg4 &#123;0.89/29 0&#125; 52. Bf7 &#123;1.81/55 4&#125; Kb4 &#123;0.94/28 0 &#40;Kb5&#41;&#125;
53. Bd5 &#123;1.81/59 4&#125; Kxa5 &#123;0.92/29 0&#125; 54. Bxb7 &#123;1.81/59 3&#125; Kb5 &#123;0.89/30 1&#125; 55.
Bc8 &#123;1.81/63 2 &#40;Kd3&#41;&#125; Kb4 &#123;0.90/30 4 &#40;Kc6&#41;&#125; 56. Be6 &#123;1.81/61 4 &#40;Kd3&#41;&#125; Kc5 &#123;0.
88/31 3&#125; 57. Kd3 &#123;1.81/50 0 &#40;Kc3&#41;&#125; Kd6 &#123;0.91/30 4&#125; 58. Bg8 &#123;1.81/50 0 &#40;Bc4&#41;&#125;
Kc5 &#123;0.92/30 4 &#40;Ld1&#41;&#125; 59. Ke3 &#123;1.81/60 3 &#40;Ba2&#41;&#125; Bd1 &#123;0.89/30 3 &#40;Kd6&#41;&#125; 60. Bf7 &#123;
1.81/47 3 &#40;Be6&#41;&#125; Kd6 &#123;0.92/29 3&#125; 61. Kf2 &#123;1.81/59 0 &#40;Bg6&#41;&#125; Kc5 &#123;0.90/30 4 &#40;Ke7&#41;
&#125; 62. Be6 &#123;1.81/61 3&#125; Bg4 &#123;0.89/31 0&#125; 63. Ke3 &#123;1.81/61 4&#125; Kd6 &#123;0.89/32 0&#125; 64.
Ba2 &#123;1.81/64 2 &#40;Bc8&#41;&#125; Kc5 &#123;0.89/31 3 &#40;Ke7&#41;&#125; 65. Bf7 &#123;1.81/56 3 &#40;Bb3&#41;&#125; Kd6 &#123;0.
91/31 4 &#40;Ld1&#41;&#125; 66. Kd4 &#123;1.82/60 7&#125; Bf3 &#123;0.91/30 0 &#40;Le2&#41;&#125; 67. Bb3 &#123;1.82/54 5 
&#40;Bc4&#41;&#125; Be2 &#123;1.05/29 3&#125; 68. Bg8 &#123;1.82/55 4 &#40;Bc2&#41;&#125; Bf1 &#123;1.86/28 3 &#40;Lg4&#41;&#125; 69. Bh7
&#123;1.82/46 2 &#40;Bd5&#41;&#125; Bh3 &#123;1.87/27 3 &#40;Ke6&#41;&#125; 70. Bg6 &#123;1.82/55 2&#125; Bg4 &#123;1.86/28 0&#125; 71.
Kc4 &#123;1.82/62 13&#125; Ke6 &#123;2.24/29 0 &#40;Le2+)&#125; 72. Be8 &#123;1.82/60 2&#125; Be2+ &#123;3.50/29 19 
&#40;Lf3&#41;&#125; 73. Kc5 &#123;1.82/61 2&#125; Ke7 &#123;3.30/28 2 &#40;Lg4&#41;&#125; 74. Ba4 &#123;1.82/56 2&#125; Bf3 &#123;3.21/
26 1 &#40;Ke6&#41;&#125; 75. Kd4 &#123;1.82/56 1 &#40;Bb3&#41;&#125; Kd6 &#123;3.04/25 3 &#40;Ke6&#41;&#125; 76. Bb5 &#123;1.82/57 1
&#40;Bc2&#41;&#125; Bd1 &#123;3.08/28 7 Houdini 4 Pro x64 resigns&#125; 1-0

mwyoung · Post by **mwyoung** » Mon Mar 03, 2014 11:24 pm

ernst wrote:Time and again I see houdini being clueless against stockfish of the danger that is upon it until it is too late. To be honest it frustates me that a €60 engine gets outplayed so often by a free engine.

An example.

Code: Select all

&#91;Event "Rated game, 5 min"&#93;
&#91;Site "Engine Room"&#93;
&#91;Date "2014.03.03"&#93;
&#91;Round "?"&#93;
&#91;White "Stockfish 160214 64"&#93;
&#91;Black "Houdini 4 Pro x64"&#93;
&#91;Result "1-0"&#93;
&#91;ECO "B92"&#93;
&#91;WhiteElo "2560"&#93;
&#91;BlackElo "2582"&#93;
&#91;Annotator "0.12;0.00"&#93;
&#91;PlyCount "152"&#93;
&#91;EventDate "2014.03.03"&#93;
&#91;TimeControl "300"&#93;

1. e4 &#123;B 0&#125; c5 &#123;B 0&#125; 2. Nf3 &#123;B 0&#125; d6 &#123;B 0&#125; 3. d4 &#123;B 0&#125; cxd4 &#123;B 0&#125; 4. Nxd4 &#123;
B 0&#125; Nf6 &#123;B 0&#125; 5. Nc3 &#123;B 0&#125; a6 &#123;B 0&#125; 6. Be3 &#123;B 0&#125; e5 &#123;B 0&#125; 7. Nb3 &#123;B 0&#125; Be6 &#123;
B 0&#125; 8. Be2 &#123;B 0&#125; Be7 &#123;B 0&#125; 9. O-O &#123;B 0&#125; O-O &#123;B 0&#125; 10. Qd3 &#123;B 0&#125; Nc6 &#123;B 0&#125; 11.
a3 &#123;0.12/23 7&#125; d5 &#123;B 0 &#40;Tc8&#41;&#125; 12. exd5 &#123;0.38/25 5&#125; Nxd5 &#123;B 0&#125; 13. Nxd5 &#123;0.37/
27 9&#125; Qxd5 &#123;B 0&#125; 14. Rad1 &#123;0.30/28 5&#125; Qxd3 &#123;B 0&#125; 15. Bxd3 &#123;0.32/28 5&#125; Rac8 &#123;B
0 &#40;f5&#41;&#125; 16. Nc5 &#123;0.27/25 7&#125; Bxc5 &#123;B 0&#125; 17. Bxc5 &#123;0.33/28 5&#125; Rfd8 &#123;B 0 &#40;Tfe8&#41;&#125;
18. Bb6 &#123;0.26/26 5&#125; Rd7 &#123;B 0&#125; 19. Rfe1 &#123;0.30/28 5&#125; f6 &#123;B 0 &#40;g6&#41;&#125; 20. Bc4 &#123;0.41/
27 12&#125; Re7 &#123;B 0&#125; 21. Bf1 &#123;0.36/27 6&#125; Kf7 &#123;B 0&#125; 22. c4 &#123;0.27/27 6&#125; Nb8 &#123;B 0 
&#40;Td7&#41;&#125; 23. b3 &#123;0.37/25 7&#125; Nd7 &#123;B 0 &#40;Td7&#41;&#125; 24. Be3 &#123;0.37/27 7&#125; Nc5 &#123;B 0&#125; 25.
Bxc5 &#123;0.19/29 8&#125; Rxc5 &#123;0.00/21 3&#125; 26. b4 &#123;0.19/31 4 &#40;f3&#41;&#125; Rcc7 &#123;0.00/25 7&#125; 27.
c5 &#123;0.18/32 0&#125; Red7 &#123;0.00/24 4&#125; 28. f3 &#123;0.19/31 2&#125; g6 &#123;0.00/25 5 &#40;Lb3&#41;&#125; 29. h4
&#123;0.13/29 7 &#40;g3&#41;&#125; Bb3 &#123;0.00/25 5&#125; 30. Rxd7+ &#123;0.13/30 0&#125; Rxd7 &#123;0.02/24 1&#125; 31. Re3
&#123;0.13/31 3&#125; Be6 &#123;-0.03/27 2&#125; 32. Rd3 &#123;0.15/30 2&#125; Rd4 &#123;0.00/27 4&#125; 33. Rxd4 &#123;0.
15/31 1&#125; exd4 &#123;0.00/27 0&#125; 34. Kf2 &#123;0.19/32 5&#125; Bd5 &#123;0.00/29 0&#125; 35. Bd3 &#123;0.15/33
4&#125; Ke6 &#123;0.00/28 0 &#40;f5&#41;&#125; 36. g3 &#123;0.00/33 5&#125; f5 &#123;0.00/28 0 &#40;Lc6&#41;&#125; 37. a4 &#123;0.25/
34 4 &#40;Ke2&#41;&#125; Bc6 &#123;0.00/28 4 &#40;Kd7&#41;&#125; 38. a5 &#123;0.25/37 5&#125; Kd5 &#123;0.00/28 0&#125; 39. Ke2 &#123;
0.25/37 4&#125; Ba4 &#123;0.00/27 5 &#40;Ke6&#41;&#125; 40. Kd2 &#123;0.57/35 5 &#40;f4&#41;&#125; Bd7 &#123;0.00/30 4&#125; 41.
f4 &#123;0.65/38 0&#125; h5 &#123;0.00/31 3 &#40;h6&#41;&#125; 42. Bf1 &#123;1.64/36 5&#125; Bc6 &#123;0.00/33 0&#125; 43. Kc2
&#123;1.80/39 4 &#40;Bd3&#41;&#125; Ba4+ &#123;0.84/26 18&#125; 44. Kc1 &#123;1.80/47 0 &#40;Kb2&#41;&#125; Bc6 &#123;0.00/23 4 
&#40;d3&#41;&#125; 45. Kd2 &#123;2.95/39 4 &#40;Kb2&#41;&#125; d3 &#123;0.84/22 3&#125; 46. Bxd3 &#123;1.80/42 0&#125; Kd4 &#123;0.86/
26 4&#125; 47. Be2 &#123;1.80/46 0&#125; Be4 &#123;0.90/26 6 &#40;Ld5&#41;&#125; 48. b5 &#123;1.81/48 6&#125; axb5 &#123;0.90/
27 0&#125; 49. Bxb5 &#123;1.81/50 3&#125; Kxc5 &#123;0.88/26 0&#125; 50. Be8 &#123;1.81/51 3&#125; Bf3 &#123;0.89/26 0&#125;
51. Bxg6 &#123;1.81/50 6&#125; Bg4 &#123;0.89/29 0&#125; 52. Bf7 &#123;1.81/55 4&#125; Kb4 &#123;0.94/28 0 &#40;Kb5&#41;&#125;
53. Bd5 &#123;1.81/59 4&#125; Kxa5 &#123;0.92/29 0&#125; 54. Bxb7 &#123;1.81/59 3&#125; Kb5 &#123;0.89/30 1&#125; 55.
Bc8 &#123;1.81/63 2 &#40;Kd3&#41;&#125; Kb4 &#123;0.90/30 4 &#40;Kc6&#41;&#125; 56. Be6 &#123;1.81/61 4 &#40;Kd3&#41;&#125; Kc5 &#123;0.
88/31 3&#125; 57. Kd3 &#123;1.81/50 0 &#40;Kc3&#41;&#125; Kd6 &#123;0.91/30 4&#125; 58. Bg8 &#123;1.81/50 0 &#40;Bc4&#41;&#125;
Kc5 &#123;0.92/30 4 &#40;Ld1&#41;&#125; 59. Ke3 &#123;1.81/60 3 &#40;Ba2&#41;&#125; Bd1 &#123;0.89/30 3 &#40;Kd6&#41;&#125; 60. Bf7 &#123;
1.81/47 3 &#40;Be6&#41;&#125; Kd6 &#123;0.92/29 3&#125; 61. Kf2 &#123;1.81/59 0 &#40;Bg6&#41;&#125; Kc5 &#123;0.90/30 4 &#40;Ke7&#41;
&#125; 62. Be6 &#123;1.81/61 3&#125; Bg4 &#123;0.89/31 0&#125; 63. Ke3 &#123;1.81/61 4&#125; Kd6 &#123;0.89/32 0&#125; 64.
Ba2 &#123;1.81/64 2 &#40;Bc8&#41;&#125; Kc5 &#123;0.89/31 3 &#40;Ke7&#41;&#125; 65. Bf7 &#123;1.81/56 3 &#40;Bb3&#41;&#125; Kd6 &#123;0.
91/31 4 &#40;Ld1&#41;&#125; 66. Kd4 &#123;1.82/60 7&#125; Bf3 &#123;0.91/30 0 &#40;Le2&#41;&#125; 67. Bb3 &#123;1.82/54 5 
&#40;Bc4&#41;&#125; Be2 &#123;1.05/29 3&#125; 68. Bg8 &#123;1.82/55 4 &#40;Bc2&#41;&#125; Bf1 &#123;1.86/28 3 &#40;Lg4&#41;&#125; 69. Bh7
&#123;1.82/46 2 &#40;Bd5&#41;&#125; Bh3 &#123;1.87/27 3 &#40;Ke6&#41;&#125; 70. Bg6 &#123;1.82/55 2&#125; Bg4 &#123;1.86/28 0&#125; 71.
Kc4 &#123;1.82/62 13&#125; Ke6 &#123;2.24/29 0 &#40;Le2+)&#125; 72. Be8 &#123;1.82/60 2&#125; Be2+ &#123;3.50/29 19 
&#40;Lf3&#41;&#125; 73. Kc5 &#123;1.82/61 2&#125; Ke7 &#123;3.30/28 2 &#40;Lg4&#41;&#125; 74. Ba4 &#123;1.82/56 2&#125; Bf3 &#123;3.21/
26 1 &#40;Ke6&#41;&#125; 75. Kd4 &#123;1.82/56 1 &#40;Bb3&#41;&#125; Kd6 &#123;3.04/25 3 &#40;Ke6&#41;&#125; 76. Bb5 &#123;1.82/57 1
&#40;Bc2&#41;&#125; Bd1 &#123;3.08/28 7 Houdini 4 Pro x64 resigns&#125; 1-0

I have seen the same reaction many times on my You tube channel when some are watching the Stockfish vs Houdini 4 match.

Houdini 4's eval is showing no problems at all, and it gets blind sided by Stockfish. The H4 fans go nuts ever time, thinking something is wrong with the H4 setup.

Stockfish just sees things that Houdini 4 does not, and sometimes sees them for many moves, before Houdini 4's eval catches the tactic Stockfish saw much earlier.

Clearly Stockfish's search depth is more then eye candy.

acase · Post by **acase** » Mon Mar 03, 2014 11:34 pm

This happened all of the time when Houdini was on top, and it was the other programs that were "blind".

mwyoung · Post by **mwyoung** » Mon Mar 03, 2014 11:34 pm

[pgn]
[Event "Rated game, 5 min"]
[Site "Engine Room"]
[Date "2014.03.03"]
[Round "?"]
[White "Stockfish 160214 64"]
[Black "Houdini 4 Pro x64"]
[Result "1-0"]
[ECO "B92"]
[WhiteElo "2560"]
[BlackElo "2582"]
[Annotator "0.12;0.00"]
[PlyCount "152"]
[EventDate "2014.03.03"]
[TimeControl "300"]

1. e4 {B 0} c5 {B 0} 2. Nf3 {B 0} d6 {B 0} 3. d4 {B 0} cxd4 {B 0} 4. Nxd4 {
B 0} Nf6 {B 0} 5. Nc3 {B 0} a6 {B 0} 6. Be3 {B 0} e5 {B 0} 7. Nb3 {B 0} Be6 {
B 0} 8. Be2 {B 0} Be7 {B 0} 9. O-O {B 0} O-O {B 0} 10. Qd3 {B 0} Nc6 {B 0} 11.
a3 {0.12/23 7} d5 {B 0 (Tc8)} 12. exd5 {0.38/25 5} Nxd5 {B 0} 13. Nxd5 {0.37/
27 9} Qxd5 {B 0} 14. Rad1 {0.30/28 5} Qxd3 {B 0} 15. Bxd3 {0.32/28 5} Rac8 {B
0 (f5)} 16. Nc5 {0.27/25 7} Bxc5 {B 0} 17. Bxc5 {0.33/28 5} Rfd8 {B 0 (Tfe8)}
18. Bb6 {0.26/26 5} Rd7 {B 0} 19. Rfe1 {0.30/28 5} f6 {B 0 (g6)} 20. Bc4 {0.41/
27 12} Re7 {B 0} 21. Bf1 {0.36/27 6} Kf7 {B 0} 22. c4 {0.27/27 6} Nb8 {B 0
(Td7)} 23. b3 {0.37/25 7} Nd7 {B 0 (Td7)} 24. Be3 {0.37/27 7} Nc5 {B 0} 25.
Bxc5 {0.19/29 8} Rxc5 {0.00/21 3} 26. b4 {0.19/31 4 (f3)} Rcc7 {0.00/25 7} 27.
c5 {0.18/32 0} Red7 {0.00/24 4} 28. f3 {0.19/31 2} g6 {0.00/25 5 (Lb3)} 29. h4
{0.13/29 7 (g3)} Bb3 {0.00/25 5} 30. Rxd7+ {0.13/30 0} Rxd7 {0.02/24 1} 31. Re3
{0.13/31 3} Be6 {-0.03/27 2} 32. Rd3 {0.15/30 2} Rd4 {0.00/27 4} 33. Rxd4 {0.
15/31 1} exd4 {0.00/27 0} 34. Kf2 {0.19/32 5} Bd5 {0.00/29 0} 35. Bd3 {0.15/33
4} Ke6 {0.00/28 0 (f5)} 36. g3 {0.00/33 5} f5 {0.00/28 0 (Lc6)} 37. a4 {0.25/
34 4 (Ke2)} Bc6 {0.00/28 4 (Kd7)} 38. a5 {0.25/37 5} Kd5 {0.00/28 0} 39. Ke2 {
0.25/37 4} Ba4 {0.00/27 5 (Ke6)} 40. Kd2 {0.57/35 5 (f4)} Bd7 {0.00/30 4} 41.
f4 {0.65/38 0} h5 {0.00/31 3 (h6)} 42. Bf1 {1.64/36 5} Bc6 {0.00/33 0} 43. Kc2
{1.80/39 4 (Bd3)} Ba4+ {0.84/26 18} 44. Kc1 {1.80/47 0 (Kb2)} Bc6 {0.00/23 4
(d3)} 45. Kd2 {2.95/39 4 (Kb2)} d3 {0.84/22 3} 46. Bxd3 {1.80/42 0} Kd4 {0.86/
26 4} 47. Be2 {1.80/46 0} Be4 {0.90/26 6 (Ld5)} 48. b5 {1.81/48 6} axb5 {0.90/
27 0} 49. Bxb5 {1.81/50 3} Kxc5 {0.88/26 0} 50. Be8 {1.81/51 3} Bf3 {0.89/26 0}
51. Bxg6 {1.81/50 6} Bg4 {0.89/29 0} 52. Bf7 {1.81/55 4} Kb4 {0.94/28 0 (Kb5)}
53. Bd5 {1.81/59 4} Kxa5 {0.92/29 0} 54. Bxb7 {1.81/59 3} Kb5 {0.89/30 1} 55.
Bc8 {1.81/63 2 (Kd3)} Kb4 {0.90/30 4 (Kc6)} 56. Be6 {1.81/61 4 (Kd3)} Kc5 {0.
88/31 3} 57. Kd3 {1.81/50 0 (Kc3)} Kd6 {0.91/30 4} 58. Bg8 {1.81/50 0 (Bc4)}
Kc5 {0.92/30 4 (Ld1)} 59. Ke3 {1.81/60 3 (Ba2)} Bd1 {0.89/30 3 (Kd6)} 60. Bf7 {
1.81/47 3 (Be6)} Kd6 {0.92/29 3} 61. Kf2 {1.81/59 0 (Bg6)} Kc5 {0.90/30 4 (Ke7)
} 62. Be6 {1.81/61 3} Bg4 {0.89/31 0} 63. Ke3 {1.81/61 4} Kd6 {0.89/32 0} 64.
Ba2 {1.81/64 2 (Bc8)} Kc5 {0.89/31 3 (Ke7)} 65. Bf7 {1.81/56 3 (Bb3)} Kd6 {0.
91/31 4 (Ld1)} 66. Kd4 {1.82/60 7} Bf3 {0.91/30 0 (Le2)} 67. Bb3 {1.82/54 5
(Bc4)} Be2 {1.05/29 3} 68. Bg8 {1.82/55 4 (Bc2)} Bf1 {1.86/28 3 (Lg4)} 69. Bh7
{1.82/46 2 (Bd5)} Bh3 {1.87/27 3 (Ke6)} 70. Bg6 {1.82/55 2} Bg4 {1.86/28 0} 71.
Kc4 {1.82/62 13} Ke6 {2.24/29 0 (Le2+)} 72. Be8 {1.82/60 2} Be2+ {3.50/29 19
(Lf3)} 73. Kc5 {1.82/61 2} Ke7 {3.30/28 2 (Lg4)} 74. Ba4 {1.82/56 2} Bf3 {3.21/
26 1 (Ke6)} 75. Kd4 {1.82/56 1 (Bb3)} Kd6 {3.04/25 3 (Ke6)} 76. Bb5 {1.82/57 1
(Bc2)} Bd1 {3.08/28 7 Houdini 4 Pro x64 resigns} 1-0

[/pgn]

carldaman · Post by **carldaman** » Mon Mar 03, 2014 11:49 pm

This is a clear sign that H4 is being outsearched by SF. I've been noticing the same thing consistently in my tests for the last 2 months.

The shoe is on the other foot now, as they say

CL

mwyoung · Post by **mwyoung** » Tue Mar 04, 2014 12:08 am

acase wrote:This happened all of the time when Houdini was on top, and it was the other programs that were "blind".

Part of the problem and disbelief is what he posted. Houdini was #1 for a long time, and cost a lot of money. Some can not wrap their heads around a free program. With a reputation of less then stellar play, and a even worst reputation for a accurate evaluation. Doing this to Houdini.

If stockfish was called something else, and cost $200. It would not be so shocking to some.

I understand that Stockfish is the real deal, because I follow engine development.

But most do not.

mwyoung · Post by **mwyoung** » Tue Mar 04, 2014 2:28 am

carldaman wrote:This is a clear sign that H4 is being outsearched by SF. I've been noticing the same thing consistently in my tests for the last 2 months.

The shoe is on the other foot now, as they say

CL

I have observed the same thing Carl. I finally had to put this on my You Tube Channel. To explain what was happening in my broadcast match on You Tube to the H4 fans. For most who do not follow engine development they thought something was wrong with Houdini 4.

But there is no problem with Houdini 4, it is what is right about Stockfish in 2014.

Mark Young You Tube Channel:

Why does Stockfish beat Houdini 4?

My observation studying the games are this.

1. Houdini 4 is blind to many King attacks launched by Stockfish. This is seen in the evaluation scores by both programs.

2. Stockfish badly out searches Houdini 4 in the endgames, and is blind to losing positions in some endgames. This is also shown by the programs evaluation scores.

It comes down to search depth, and king safety.

Or in other words, Stockfish can see deeper hidden tactics that Houdini 4 misses. When this happens in a game, Stockfish can win quickly and suddenly.

To the dismay of H4 fans.

shrapnel · Post by **shrapnel** » Tue Mar 04, 2014 8:41 am

Everyone havin' fun bashing H4 ?

H4 wouldn't make these mistakes given fast, modern hardware and the use of Syzygy EGTB and more time.
OP disparages H4 based on the basis of a 5 minute game !?

Dude, you're just checking out Engine Time Management at that TC, NOTHING else !

notyetagm · Post by **notyetagm** » Tue Mar 04, 2014 9:23 am

mwyoung wrote:
acase wrote:This happened all of the time when Houdini was on top, and it was the other programs that were "blind".
Part of the problem and disbelief is what he posted. Houdini was #1 for a long time, and cost a lot of money. Some can not wrap their heads around a free program. With a reputation of less then stellar play, and a even worst reputation for a accurate evaluation. Doing this to Houdini.

If stockfish was called something else, and cost $200. It would not be so shocking to some.

I understand that Stockfish is the real deal, because I follow engine development.

But most do not.

That was before. This is now.

This is not your father's Stockfish.

Sedat Canbaz · Post by **Sedat Canbaz** » Tue Mar 04, 2014 9:29 am

Just my 2 cents over this issue,

Any engine performance is highly depending on:
1)Hardware speed
2)Time control
3)Opening book usage
4)MP or single core
5)Test conditions (I suggest the latest engines versions to be tested and as MP of course !)
6)Live broadcasting tools should not be allowed on same Tournament PC
...

Based on SCCT results,
The latest Stockfish versions are winner in almost all my tournaments, probably the strongest engine in the world!
http://abrok.eu/stockfish/

For example,
On LS rating (ultra fast time control

) Houdini 4's s performance is best
On FCP rating (40/5) Houdini 4's s performance is best
On CCRL rating (40/40) Houdini 4's s performance is best
On SSDF rating (40/120) Komodo's performance is best
On CEGT rating (40/120) Komodo's performance is best
On SCCT rating (overall) Stockfish's performance is best
*It seems, Stockfish likes a lot my test conditions

Hope helps...

Best,
Sedat

Why is Houdini 4 so clueless?

Why is Houdini 4 so clueless?

Re: Why is Houdini 4 so clueless?

Re: Why is Houdini 4 so clueless?

Re: Why is Houdini 4 so clueless?

Re: Why is Houdini 4 so clueless?

Re: Why is Houdini 4 so clueless?

Re: Why is Houdini 4 so clueless?

Re: Why is Houdini 4 so clueless?

Re: Why is Houdini 4 so clueless?

Re: Why is Houdini 4 so clueless?