API tools faq
paste
yurikvelo

TCEC Season 7 - Stage 1b | Round 11

yurikvelo
Oct 22nd, 2014
536
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 6.24 KB | None | 0 0
raw download clone embed print report
  1. http://tcec.chessdom.com/archive.php?se=7&st=1b&ga=73
  2. 13. .. b4 (H4 @ 4200 MN, D=28)
  3.  
  4. Analysis by Stockfish 181014 @ 13 300 MN, D=36
  5. 1. = (0.00): 13...exd4 14.Qxd4 c5 15.Qxd6 b4 16.Nb1 Rc8
  6. 2. = (0.00): 13...Na6 14.h4 b4 15.Nb1 h5 16.Nh3 hxg4 17.Ng5 Nc7
  7. 3. = (0.00): 13...b4 14.Nb1 Na6 15.Bxa6 Qxa6 16.dxe5 dxe5 17.Qxb4 Rab8
  8. 4. +/= (0.26): 13...Nbd7 14.g5 Ne8 15.d5 cxd5 16.exd5 Bf5 17.Bxb5 Rb8
  9.  
  10. ======================
  11. 13. .. b4 was OK
  12. ======================
  13. http://tcec.chessdom.com/archive.php?se=7&st=1b&ga=73
  14. 14. .. c5 (H4 @ 4270 MN, D=28)
  15. After 14 ..c5 Komodo's eval raise from 0.20 @ 14.6BN to 0.42 @ 1.4 BN
  16.  
  17. Analysis by Stockfish 181014 @ 8500 MN, D=35
  18. 1. = (0.18): 14...Na6 15.Bxa6 Qxa6 16.dxe5 dxe5 17.Qxb4 Rad8
  19. 2. +/= (0.26): 14...c5 15.d5 Bc8 16.Ne2 Ba6 17.Ng3 Bxf1
  20.  
  21. Analysis by Stockfish 181014 @ 34500 MN, D=36/54
  22. 1. +/= (0.37): 14...Na6 15.h4 h5 16.Nh3 Rab8 17.Bxa6 Qxa6
  23. 2. +/= (0.39): 14...Nbd7 15.Qxb4 Qxb4 16.axb4 h6 17.b5 exd4
  24. 3. +/= (0.55): 14...c5 15.d5 Bc8 16.Ne2 Nbd7 17.Ng3 Rb8
  25.  
  26. Analysis by Houdini 4 Pro x64 @ 5000 MN, D=26
  27. 1. = (0.04): 14...c5 15.d5 Bd7 16.h4 h5 17.g5 Ne8 18.f4 Bg4
  28. 2. = (0.18): 14...Na6 15.Bxa6 Qxa6 16.axb4 Qb6 17.h4 h5 18.gxh5 Nxh5
  29.  
  30. Analysis by Houdini 4 Pro x64 @ 5500 MN, D=26/90
  31. 1. = (0.18): 14...Na6 15.Bxa6 Qxa6 16.axb4 Qb6 17.h4 h5
  32. 2. +/= (0.36): 14...Nbd7 15.Qxb4 Qxb4 16.axb4 h6 17.dxe5 Nxe5
  33. 3. +/= (0.37): 14...c5 15.d5 Bd7 16.Ne2 h5 17.Ng3 hxg4
  34.  
  35. Analysis by Houdini 4 Pro x64 @ 80 000 MN, D=31
  36. 1. = (0.20): 14...Na6 15.Bxa6 Qxa6 16.axb4 Qb6
  37. 2. +/= (0.36): 14...c5 15.d5 Bd7 16.Ne2 Na6
  38. ======================
  39. 14. .. c5 was +0.14 over ..Na6 on 5B depth, but become -0.19 on 5.5B and finally -0.16 on 80B
  40.  
  41. Houdini spent 2:13 to reach 4.3BN, if he spent 2:53+ to get 5.5BN it would stick on ..c5
  42. Houdini realized losing on 15 and spent 18:27 to go 31BN deep. It was too late, he stabilized 0.20 loss confirmed by K8, SF and H4 itself
  43. ======================
  44.  
  45. ============================
  46. 16. .. Ba6 (H4 @ 4228 MN, D=27)
  47. After 16. .. Ba6 Komodo's eval raise from 0.41 @ 2.78BN to 0.48 @ 3.48 BN
  48.  
  49. Analysis by Houdini 4 Pro x64 @ 4228 MN, D=25
  50. 1. = (0.00): 16...h5 17.Ng3 hxg4 18.fxg4 Bxg4 19.Be2 Bh3
  51. 2. +/= (0.33): 16...Ba6 17.Ng3 Bxf1 18.Rdxf1 Nbd7 19.h4 Qb6
  52.  
  53. Analysis by Houdini 4 Pro x64 @ 7100 MN, D=26/53
  54. 1. +/= (0.30): 16...h5 17.Ng3 hxg4 18.fxg4 Nxg4
  55. 2. +/= (0.35): 16...Ba6 17.Ng3 Bxf1 18.Rdxf1 Nbd7
  56.  
  57. Analysis by Houdini 4 Pro x64 @ 20 500 MN, D=27/81
  58. 1. +/= (0.30): 16...Ba6 17.Ng3 Bxf1 18.Rdxf1 Nbd7
  59. 2. +/= (0.49): 16...h5 17.Ng3 hxg4 18.fxg4 Bxg4
  60. 3. +/= (0.55): 16...Qb6 17.Ng3 h6 18.h4 Ng8
  61.  
  62. Analysis by Stockfish 181014 @ 7 000MN, D=32/48 (stable on 1 to 7 BN)
  63. 1. +/= (0.44): 16...Ba6 17.Ng3 Bxf1 18.Rdxf1 Nbd7
  64. 2. +/= (0.49): 16...Nbd7 17.Ng3 Rg8 18.h4 Rb8
  65. ....
  66. 6. +/= (0.67): 16...h5 17.Ng3 hxg4 18.h4 gxf3
  67. ============================
  68. 15. .. Bc8 was best at any depth
  69. 16. .. Ba6 was OK
  70. 17 & 18 are forced moves
  71. ============================
  72. 19. .. Rab8 (H4 @ 2153 MN, D=26)
  73. After 19. .. Rab8 Komodo's eval raise from 0.44 @ 1.88BN to 0.64 @ 9.93 BN
  74. After 19. .. Rab8 Houdini's eval raise from 0.35 @ 2.72BN to 0.40 @ 2.15 BN
  75.  
  76. 19. .. Rab8 +Qb6 (or ..Qb6 + Rab8) is "the only" (forced) 19th move
  77.  
  78. 20-23 is forced line (at least for black, with Komodo's white choices)
  79. ============================
  80. 24. .. Nc7
  81. After 24. .. Nc7 Komodo's eval raise from 0.69 @ 2.8BN to 0.92 @ 7.6 BN
  82. After 24. .. Nc7 Houdini's eval raise from 0.38 @ 2.9BN to 0.45 @ 2.8 BN
  83. Komodo suggested 24. .. a5
  84.  
  85. Analysis by Houdini 4 Pro x64 @ 10 500 MN, D=28
  86. 1. +/= (0.33): 24...Nc7 25.Nb3 Nb5 26.Ne2 b3
  87. 2. +/= (0.39): 24...a5 25.f4 exf4 26.Rxf4 Ne5
  88.  
  89. Stockfish 181014 start preferring ..a5 after depth >0.78 BN and after >1.6 BN it become +0.11 over
  90. 1. +/= (0.60): 24...a5 25.Rh2 a4 26.Rfh1 Rg8 27.Nc4 Ra8
  91. 2. +/= (0.61): 24...Nc7 25.Nc4 Nb5 26.Ne2 b3 27.c3 Nc5
  92.  
  93. At D > 2.35BN ..Nc7 again better (+0.69 vs +0.76)
  94. At >3 BN (D=30) it is +0.72 on ..c5 (since Stockfish discovered 25. Rh4 reply found by Komodo) vs 0.75 for ..Nc7
  95. ============================
  96. at 24 .. point eval was already ~0.7 both by Komodo and Stockfish. Since 14.. c5 to 24 there were no better moves or variations.
  97. So the eval of 14 .. c5 should be ~0.7 if D=50 was possible.
  98. 25-26 are only moves, 27 confirmed by SF at D=31/50 (1BN) is 0.84 (2nd PV is 0.85)
  99. So the eval of 14 .. c5 should be ~0.84 if D=57 was possible.
  100. 28, 29 are only moves.
  101. After deep search >D31 on 30. .. Rxc6 vs a5, Stockfish even lowered eval to +0.65 and Komodo to +0.67
  102. After forced 31, 32 moves and clear 33. ..Ne6 even at low D, evaluation further dropped to 0.63
  103. ============================
  104. First alternative move was 35.. Ned4 vs Nc5
  105. 35. ..Ned4 (H4 @ 1850 MN, D=33)
  106. After 35. ..Ned4 Komodo's eval raise from 0.81 @ 1.25BN to 0.92 @ 2.89 BN
  107. Komodo preferred 35... Nc5 36. Rgg1 a6 37. Ke3
  108. Extra deep analyze confirm both Houdini and Stockfish consider 35... Nc5 0.35 & 0.55 better than 35. ..Ned4
  109.  
  110. Analysis by Stockfish 181014 @ D=41/78 @ 47 800 MN
  111. 1. +/= (0.53): 35...Nc5 36.Rgg1 a6 37.Rgf1 Nc7
  112. 2. +/- (0.93): 35...Ned4 36.Nxd4 Nxd4 37.Rf2 Rc7
  113. 3. +/- (1.08): 35...Nbd4 36.Nxd4 Nxd4 37.Rf2 Rc7
  114.  
  115. Analysis by Houdini 4 Pro x64 @ D=34/99, 79 500 MN
  116. 1. +/= (0.48): 35...Nc5 36.Rgg1 a6 37.Ke3 Nc7
  117. 2. +/- (0.83): 35...Ned4 36.Nxd4 Nxd4 37.Rf2 a6
  118. 3. +/- (0.83): 35...Nbd4 36.Nxd4 Nxd4 37.Rf2 a6
  119.  
  120. Conlusion (IMHO): Houdini lost game on Move 14, due to time management. 40 extra seconds could make a draw.
  121. Raise of eval between 14 and 24 was only due to horizon effect, not bad moves by Houdini.
  122. More raise after black's 16, 19 and 24 move than others because at this plyes more leafleting (plenty of choices for black), while other inter-nodes where mostly forced. After 16, 19, 24 horizon was coming much faster.
  123.  
  124. Possibility to turn into Draw-PV was on 35 Move, Komodo & Stockfish found it even at low depth, but Houdini again spent much less time than Komodo and made move +0.93
  125.  
  126. P.S. 35. ..Nc5 at extra D=44/78, 133.3 BN by Stockfish 181014 still high (0.79 @ 133 BN vs 0.53 @ 48 BN)
  127. 1. +/- (0.79): 35...Nc5 36.Rgg1 a6 37.Ke3 Nc7 38.Rgd1 Ke8
  128. 2. +/- (1.06): 35...Ned4 36.Nxd4 Nxd4 37.Rf2 Rc7 38.Ra6 Ke7
  129. 3. +/- (1.06): 35...Nbd4 36.Nxd4 Nxd4 37.Rf2 Rc7 38.Ra6 Ke7
  130.  
  131. So I guess 35. ..Nc5 wouldn't have saved Houdini really.
Advertisement
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!