Martin on the SF loss on time

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

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
User avatar
RJN
Posts: 303
Joined: Fri Jun 21, 2013 3:18 am
Location: Orion Spiral Arm

Martin on the SF loss on time

Post by RJN » Fri Oct 09, 2015 10:39 pm

i7-5930K @4.5GHz, H100i Hydro Cooler, 64GB DDR4 Corsair Dominator Platinum @3000MHz, ASUS X99 Deluxe mboard, 1TB EVO 850 SSD

User avatar
Eelco de Groot
Posts: 4268
Joined: Sun Mar 12, 2006 1:40 am
Location: Groningen

Re: Martin on the SF loss on time

Post by Eelco de Groot » Fri Oct 09, 2015 11:39 pm

TCEC archive link to the game: http://tcec.chessdom.com/archive.php?se=8&st=3&ga=6

Shortened pgn by Shredder, sorry the TCEC pgn from the above link was apparently too long for the forum software, so here you have the game, but not time use etc.

[pgn][Event "TCEC Season 8 - Stage 3"]
[Site "http://tcec.chessdom.com"]
[Date "2015.10.09"]
[Round "2"]
[White "Stockfish 071015L"]
[Black "Gull 3"]
[WhiteElo "3224"]
[BlackElo "3123"]
[ECO "E04"]
[Result "0-1"]

1. d4 d5 2. c4 e6 3. g3 Nf6 4. Bg2 dxc4 5. Nf3 a6 6. O-O
Nc6 7. e3 Bd7 8. Nc3 Rb8 9. Ne5 h5 10. Nxc4 b5 11. Ne5 Nxe5
12. dxe5 Ng4 13. f4 b4 14. Ne4 Bb5 15. Re1 Qxd1 16. Rxd1
Be7 17. Bd2 h4 18. Bf3 f5 19. exf6 Nxf6 20. Ng5 hxg3
21. hxg3 e5 22. fxe5 Nd7 23. Ne6 Nxe5 24. Nxc7+ Kf7
25. Bd5+ Kg6 26. Be4+ Kg5 27. Be1 Rhc8 28. Nxb5 Rxb5
29. Rac1 Rxc1 30. Rxc1 a5 31. Bd2 Bd6 32. b3 Kf6 33. Kg2
Ke6 34. Rc8 Ng4 35. Bf3 Ne5 36. e4 Rc5 37. Rxc5 Bxc5
38. Be2 Kf6 39. Bc1 Nc6 40. Bc4 Bd4 41. Bb5 Ne5 42. Kf1 Ke6
43. Ke2 Kd6 44. Be8 Ke7 45. Ba4 Kf6 46. Bd2 Ba7 47. Be8 Bc5
48. Bb5 Bd4 49. Ba6 Ke6 50. Bf4 Kd7 51. Bd3 Kc6 52. Bg5 Kc5
53. Ba6 Kc6 54. Be7 Bc5 55. Bh4 Bd6 56. Bg5 Nf7 57. Bf4 Ne5
58. Bc1 Nd7 59. Kf3 Ne5+ 60. Kg2 Bc7 61. Bf1 Bd6 62. Bf4
Ng6 63. Bd2 Ne5 64. Be2 Bc5 65. Ba6 Bd4 66. Kf1 Nf3 67. Bc1
Ne5 68. Ke2 Kd7 69. Bd3 Kc6 70. Bf4 Kb6 71. Bb1 Kc5 72. Bc1
Ba1 73. Ke3 Bd4+ 74. Kf4 Kd6 75. Kf5 g6+ 76. Kg5 Ke6
77. Bf4 Kf7 78. Bc2 Nf3+ 79. Kg4 Ne5+ 80. Kh4 Kg7 81. Kg5
Nf3+ 82. Kg4 Ne5+ 83. Kh4 Kf7 84. Bd1 Nd3 85. Be2 Nxf4
86. gxf4 Be3 87. Kg4 Bc5 88. Bc4+ Kf8 89. Kg5 Kg7 90. e5
Bb6 91. Kg4 Bc5 92. Kf3 Be7 93. Bd3 Bc5 94. Bc2 Kf7 95. Be4
Kg7 96. Bd5 Bb6 97. Kg3 Bd4 98. Kg4 Bc5 99. Be4 Be7
100. Bd3 Kf7 101. Bb1 Kg7 102. Kf3 Bc5 103. Kg3 Kf7
104. Be4 Be7 105. Kf3 Bd8 106. Kg4 Be7 107. Bd3 Kg7
108. Be2 Kg8 109. Kg3 Kf7 110. Kf3 Bd8 111. Ke4 Bh4
112. Bc4+ Ke7 113. Ba6 Bf2 114. Be2 Kf7 115. Bc4+ Ke7
116. Kf3 Bc5 117. Kg4 Kf8 118. Bd3 Kg7 119. Be2 Be7
120. Kf3 Bc5 121. Bf1 Be7 122. Bc4 Bc5 123. Bd3 Kf7
124. Bb5 Bd4 125. Bc4+ Kf8 126. Bd3 Kg7 127. Bf1 Kf7
128. Kg4 Bc5 129. Bg2 Be7 0-1[/pgn]

Fishcooking link (please post entire link in browser) https://groups.google.com/forum/?fromgr ... XwGwZt4KmQ
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan

User avatar
RJN
Posts: 303
Joined: Fri Jun 21, 2013 3:18 am
Location: Orion Spiral Arm

Re: Martin on the SF loss on time

Post by RJN » Sat Oct 10, 2015 10:25 pm

SF blows it again on time, with 1 ply to a TCEC win rule. Martin reports it went 88ms over, now the stuff has really hit the fan.....
i7-5930K @4.5GHz, H100i Hydro Cooler, 64GB DDR4 Corsair Dominator Platinum @3000MHz, ASUS X99 Deluxe mboard, 1TB EVO 850 SSD

User avatar
RJN
Posts: 303
Joined: Fri Jun 21, 2013 3:18 am
Location: Orion Spiral Arm

Re: Martin on the SF loss on time

Post by RJN » Sat Oct 10, 2015 11:20 pm

2nd video update from Martin:

https://vid.me/9XV0
i7-5930K @4.5GHz, H100i Hydro Cooler, 64GB DDR4 Corsair Dominator Platinum @3000MHz, ASUS X99 Deluxe mboard, 1TB EVO 850 SSD

S.Taylor
Posts: 8481
Joined: Thu Mar 09, 2006 2:25 am
Location: Jerusalem Israel

Re: Martin on the SF loss on time

Post by S.Taylor » Sun Oct 11, 2015 12:04 am

It's all a part of the fun anyway.
As Martin has often said, It's only for entertainment purposses. The results don't matter, as long as you enjoyed the games.

bnemias
Posts: 373
Joined: Thu Aug 14, 2008 1:21 am
Location: Albuquerque, NM

Re: Martin on the SF loss on time

Post by bnemias » Sun Oct 11, 2015 1:09 am

RJN wrote:SF blows it again on time, with 1 ply to a TCEC win rule. Martin reports it went 88ms over, now the stuff has really hit the fan.....
I haven't run flash in a while so I can't watch the videos. But my thought is it's possible that there's delay between the engine and the interface.

Probably adding an UCI option:
Move Overhead = 300

might fix it.

Dann Corbit
Posts: 11621
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: Martin on the SF loss on time

Post by Dann Corbit » Sun Oct 11, 2015 2:07 am

bnemias wrote:
RJN wrote:SF blows it again on time, with 1 ply to a TCEC win rule. Martin reports it went 88ms over, now the stuff has really hit the fan.....
I haven't run flash in a while so I can't watch the videos. But my thought is it's possible that there's delay between the engine and the interface.

Probably adding an UCI option:
Move Overhead = 300

might fix it.
The rules state:
Other Settings
Any configurable option not described above, are not adjusted in any way, except that "keep hash tables" or similar is usually enabled.

I agree that the change would probably fix it, but it is in violation of the rules.

thekingman
Posts: 35
Joined: Mon Mar 16, 2015 5:17 am

Re: Martin on the SF loss on time

Post by thekingman » Sun Oct 11, 2015 2:40 am

Dann Corbit wrote:
bnemias wrote:
RJN wrote:SF blows it again on time, with 1 ply to a TCEC win rule. Martin reports it went 88ms over, now the stuff has really hit the fan.....
I haven't run flash in a while so I can't watch the videos. But my thought is it's possible that there's delay between the engine and the interface.

Probably adding an UCI option:
Move Overhead = 300

might fix it.
The rules state:
Other Settings
Any configurable option not described above, are not adjusted in any way, except that "keep hash tables" or similar is usually enabled.

I agree that the change would probably fix it, but it is in violation of the rules.
I am not so certain. Here is the relevant section of the rules:
In the case of a serious, play-limiting bug (like crashing or interface communication problems) not discovered during the pre-Season testing, the engine can be updated once per Stage to fix this/these bug/bugs only.
I would argue that a losing half your games on time is play-limiting, and that under the rules, Stockfish is entitled to a single update to fix this.

carldaman
Posts: 2002
Joined: Sat Jun 02, 2012 12:13 am

Re: Martin on the SF loss on time

Post by carldaman » Sun Oct 11, 2015 4:06 am

S.Taylor wrote:It's all a part of the fun anyway.
As Martin has often said, It's only for entertainment purposses. The results don't matter, as long as you enjoyed the games.
(I detect a hint of sarcasm) ;)

Of course, such games cannot possibly be enjoyed, where competitors are being forfeited on technicalities. They need more flexibility and common sense in their rules (and rulings).

S.Taylor
Posts: 8481
Joined: Thu Mar 09, 2006 2:25 am
Location: Jerusalem Israel

Re: Martin on the SF loss on time

Post by S.Taylor » Sun Oct 11, 2015 4:19 am

carldaman wrote:
S.Taylor wrote:It's all a part of the fun anyway.
As Martin has often said, It's only for entertainment purposses. The results don't matter, as long as you enjoyed the games.
(I detect a hint of sarcasm) ;)

Of course, such games cannot possibly be enjoyed, where competitors are being forfeited on technicalities. They need more flexibility and common sense in their rules (and rulings).
_I_ meant it sarcasticly! yes!

But Martin didn't seem to care, a couple of years ago. He just kept himself "politically correct!"

Or maybe it really WAS pure entertainment in his mind, just to do it with no goals. If i ever spoke about results he made me feel guilty of being obsessed and don't just want to enjoy life!

:wink: :wink: yeah, you're right! :lol:

Post Reply