On-line engine blitz tourney October

[flok]

It has just happened again, so check those logs.

It seems to handle 1+1 and 3+1 OK, it played several of those, but as soon as you ask it to play 5+1 it crashes later in the game.

Code: Select all

1667031485.005 Engine->Adapter: # dts: 4493017 depth 10 nodes 33982 lower_bound: 16812
1667031485.565 Engine->Adapter: # dts: 5052218 depth 11 nodes 38015 lower_bound: 6572
1667031486.347 Engine->Adapter:
1667031486.348 Engine->Adapter: ***ERROR*** A stack overflow in task pthread has been detected.
1667031486.348 Engine->Adapter:
1667031486.348 Engine->Adapter:
1667031486.357 Engine->Adapter: Backtrace:0x400819c2:0x3ffec7100x40084829:0x3ffec730 0x4008737e:0x3ffec750 0x40085f02:0x3ffec7c0 0x40084928:0x3ffec7f0 0x400848da:0x3ffd36c0 0x3ffd34bd:0x00000000  |<-CORRUPTED

After that it reboots. Added a workaround for the semingly hang after that.

Maybe I should add a qs depth limit?

Strange thing is: I keep track of the total depth (max_depth + qs_depth) and every time that gets a higher value, I check the stack-size. This time it complained about stack-size way before it ran out (it had still over 6kB left).

Don't think I can fix it today, so I consider running it on a normal system.

@joost are you willing to create a DogPC account for that?

[flok]

It seems to handle 1+1 and 3+1 OK, it played several of those, but as soon as you ask it to play 5+1 it crashes later in the game.

Might be a stack problem that manifests with higher thinking times due to more depth being reached. Going via some high stack water mark runtime checks isn't what I'm doing in my microcontroller engine. Instead, I use GCC's -fstack-usage feature for individual functions, then manually put together the call stack and calculate what maximum depth I can afford with the amount of stack that I configured. That's 23 for the main search and 10 for QS on top of that.

Right. I use freertos "around" my program, very convenient. I try to statically allocate everything possible. But as I also use libchess libchess (https://github.com/Mk-Chan/libchess), some dynamics are involved (e.g. it uses pthreads). I could patch of course but (initially) one of the main challenges was to use everything as much as possible "as is".

[Joost Buijs]

This time there were a lot of participants, like it used to be in the past.

Besides a small issue with the time-seal of Winter the tournament finished without problems.

Nightmare was very lucky this time. Considering the fact that the engine hasn't been changed for at least 1.5 years I didn't expect it to win. Goldbar suddenly played a bad move in an equal position, and against WASP the engine came out of book in a more or less won position.

There were 4 players with 6.5 points, BelFish, CSTal, Goldbar and Marvin.
The Baron and Waduuttie both got 6 points.

The new players, Cheers and KhepriChess didn't do bad either with 4 and 3,5 points.

Code: Select all

:Tourney Players: Round 9 of 9 
:
:     Name              Rating Score Perfrm Upset  Results 
:     ----------------- ------ ----- ------ ------ ------- 
:  1 +NightmareX        [2914]  7.5  [2976] [  43] +21w +06w +12b +10w =02b +04w =03b +08w =09b 
:  2 +BelFish           [3000]  6.5  [2909] [   0] +16w +11w =03b =04b =01w =05b +06w +10w =08b 
:  3 +ChessSystemTalX   [2901]  6.5  [2878] [  55] +22b +07w =02w =08b =04w +21b =01w =05b +10w 
:  4 +Goldbar           [2869]  6.5  [2776] [  81] +17w +19w +14b =02w =03b -01b =05w +16b +11w 
:  5 +Marvin            [2763]  6.5  [2836] [ 274] +27w +20w =11b +07w =10b =02w =04b =03w +15b 
:  6 +TheBaron          [2566]  6.0  [2618] [ 138] +28w -01b =08w +14w =07b +24w -02b +22w +20b 
:  7 +WaDuuttie         [2556]  6.0  [2594] [  52] +25w -03b +09w -05b =06w =12b +24b +21w +16w 
:  8 -WaspX             [2843]  5.5  [2719] [ 107] +23w =12w =06b =03w +24b +09w =10b -01b =02w
:  9 +atomNightmare     [2603]  5.5  [2627] [ 169] +24w =15w -07b +26w =11b -08b +14w +12b =01w 
: 10 +LeelaRB           [2832]  5.0  [2705] [   5] +13w +26w +15b -01b =05w +23b =08w -02b -03b 
: 11 +atomGoldbar       [2575]  5.0  [2584] [ 136] +18w -02b =05w +27b =09w +13b =12w =15w -04b 
: 12 +Winter            [2500]  5.0  [2529] [ 301] +31w =08b -01w =15w +17b =07w =11b -09w +23b 
: 13 +Spartacus         [2190]  5.0  [2397] [ 812] -10b +30w +19b -21b +20w -11w -22b +25w +26w 
: 14 -CTdccc            [2180]  5.0  [2503] [ 772] =15b +16w -04w -06b +26b +27w -09b =23b +22w 
: 15 +Amoeba            [2631]  4.5  [2492] [   0] =14w =09b -10w =12b =16w +20b +21w =11b -05w 
: 16 +Myrddin           [2321]  4.5  [2462] [ 216] -02b -14b +20w +28w =15b +22b +23w -04w -07b
: 17 +Prophet           [2238]  4.5  [2224] [ 131] -04b +29w +31b -23w -12w =19b -20w +32b +27b 
: 18 +DogPC             [2128]  4.5  [2193] [ 372] -11b -24b -21w -20b +25w +26w =28b +31w +29b 
: 19 -Cheers            [2500]  4.0  [2191] [   0] +29w -04b -13w +25b -22w =17w -27b +24w =21b 
: 20 +rpiDumb           [2382]  4.0  [2255] [   0] +32w -05b -16b +18w -13b -15w +17b +27w -06w 
: 21 +rpiArminius       [2278]  4.0  [2412] [ 111] -01b +28w +18b +13w =23b -03w -15b -07b =19w 
: 22 +atomFloyd         [2251]  4.0  [2284] [ 249] -03w +25b +28b -24w +19b -16w +13w -06b -14b 
: 23 +Barbarossa        [2211]  4.0  [2276] [  60] -08b +31w +29b +17b =21w -10w -16b =14w -12w 
: 24 +Joker             [2151]  4.0  [2280] [ 100] -09b +18w +30b +22b -08w -06b -07w -19b +32w
: 25 +KingSlayer        [2068]  4.0  [2044] [   0] -07b -22w +32b -19w -18b +29w +30w -13b +31b 
: 26 +KhepriChess       [2500]  3.5  [2096] [   0] +30w -10b =27w -09b -14w -18b +32b +28w -13b 
: 27 +feligres          [2180]  3.5  [2201] [ 480] -05b +32w =26b -11w +30b -14b +19w -20b -17w 
: 28 +Isa               [2085]  3.5  [2081] [  21] -06b -21b -22w -16b +29w +31b =18w -26b +30w 
: 29 +microMax          [1902]  3.0  [1922] [   0] -19b -17b -23w +32w -28b -25b +31w +30b -18w 
: 30 +micahGB           [1791]  2.0  [1839] [   7] -26b -13b -24w +31b -27w +32w -25b -29w -28b 
: 31 +atomSargon        [1798]  0.5  [1711] [   0] -12b -23b -17w -30w =32b -28w -29b -18b -25w 
: 32 -Skipper           [1682]  0.5  [1756] [  58] -20b -27b -25w -29b =31w -30b -26w -17w -24b
:
:     Average Rating    2387.2

[flok]

The new players, Cheers and KhepriChess didn't do bad either with 4 and 3,5 points.

Dog(PC) is also brand new!
With 4.5 points not to shaby either imho.

[Joost Buijs]

The new players, Cheers and KhepriChess didn't do bad either with 4 and 3,5 points.

Dog(PC) is also brand new!
With 4.5 points not to shaby either imho.

Of course you are right.

[JVMerlino]

Indeed, very tough competition (as usual). Myrddin's highlight was a hard-fought draw as Black against the much higher-rated Amoeba, which pleased me greatly. No idea about Amoeba's hardware, but Myrddin was running with 8 cores.

[pgn][Event "ICS rated blitz match"]
[Site "nightmare-chess.nl"]
[Date "2022.10.29"]
[Round "-"]
[White "Amoeba"]
[Black "Myrddin"]
[Result "1/2-1/2"]
[WhiteElo "2625"]
[BlackElo "2321"]
[TimeControl "300+1"]
[Annotator "8... +0.22"]

1. d4 Nf6 2. Bg5 Ne4 3. Bh4 g5 4. f3 gxh4 5. fxe4 c5 6. e3 Bh6 7. Qd3 Nc6
8. dxc5 Qa5+ {+0.22/20 8} 9. Qc3 Bxe3 {+0.08/22 12} 10. Qxa5 Nxa5
{+0.30/23 20} 11. Nc3 Bxc5 {+0.30/22 12} 12. O-O-O Nc6 {+0.46/20 8} 13. Nf3
h3 {+0.47/20 8} 14. gxh3 d6 {+0.56/22 22} 15. Nd5 Rb8 {+0.50/22 15} 16. c3
e6 {+0.69/20 7} 17. b4 exd5 {+0.78/22 17} 18. bxc5 dxe4 {+0.72/24 17} 19.
Ng5 d5 {+0.58/25 44} 20. Rxd5 f5 {+0.63/23 17} 21. h4 h6 {+1.10/21 6} 22.
Nh3 Be6 {+1.09/23 18} 23. Rd6 Bxa2 {+0.92/19 11} 24. Nf4 Rd8 {+0.87/18 9}
25. Rf6 Kd7 {+0.71/17 5} 26. Kb2 Ke7 {+0.56/19 7} 27. Rxf5 Rd2+
{+0.50/20 10} 28. Kc1 e3 {+0.66/21 15} 29. Be2 Kd8 {+1.10/16 5} 30. Rd1
Rxd1+ {+0.80/20 12} 31. Kxd1 Bb3+ {+0.76/21 10} 32. Kc1 a5 {+0.80/22 6} 33.
Kb2 a4 {+0.74/21 4} 34. c4 Rg8 {+0.83/19 9} 35. Kc3 Rg1 {+0.60/18 8} 36.
Bd3 Ne7 {+0.48/18 9} 37. Re5 Rc1+ {+0.45/19 10} 38. Kb2 Re1 {+0.08/19 8}
39. Re6 h5 {+0.00/19 7} 40. h3 Kd7 {+0.00/21 10} 41. Re4 Nf5 {+0.00/17 4}
42. Re5 Nxh4 {+0.00/18 10} 43. Rxh5 Nf3 {-0.14/18 7} 44. Ng2 Rg1
{-0.36/18 21} 45. Nxe3 Nd2 {-0.48/18 11} 46. Be2 Rg3 {+0.00/15 3} 47. Nf5
Rg2 {-0.37/16 5} 48. Bg4 Nxc4+ {-0.41/16 6} 49. Kc3 Ke8 {-0.39/17 5} 50.
Kb4 Rb2 {-0.25/16 8} 51. Nd4 Nd2 {-0.37/17 2.7} 52. Nxb3 axb3 {-0.59/19 8}
53. Re5+ Kf7 {-0.53/21 2.5} 54. Re3 Kf6 {-0.32/20 5} 55. Bd1 Rb1
{-0.43/20 2.8} 56. Bxb3 Kg6 {-0.64/22 12} 57. h4 Kh5 {-0.80/17 2.2} 58. Rd3
Ne4 {-1.18/20 4} 59. Rd4 Nf6 {-1.21/19 5} 60. Ka4 Ra1+ {-0.74/18 2.1} 61.
Kb5 Rb1 {-0.71/22 5} 62. Kb4 Rf1 {-1.30/15 2.5} 63. Bf7+ Kh6 {-1.51/16 3}
64. Kb5 Rb1+ {-1.15/18 2.0} 65. Ka4 Re1 {-0.95/18 2.0} 66. Kb5 Rb1+
{+0.00/29 1.9} 67. Ka4 Re1 {+0.00/28 1.9} 68. Ka5 Ne4 {-0.76/16 1.9} 69.
Kb5 Rb1+ {-0.67/16 1.9} 70. Rb4 Rxb4+ {+0.00/28 1.9} 71. Kxb4 Nxc5
{+0.00/50 4} 72. Kxc5 Kh7 {+0.00/1} 73. Be8 Kh8 {+0.00/1} 74. Kb6 Kh7
{+0.00/1} 75. Kxb7 Kh8 {+0.00/1} 76. h5 Kh7 {+0.00/1} 77. Kb6 Kh8 {+0.00/1}
78. Kc7 Kh7 {+0.00/1} 79. Kb6 Kh8 {+0.00/1} 80. Kc7 Kh7 {+0.00/1} 81. Kd6
Kh8 {+0.00/1} 82. Bg6 Kg8 {+0.00/1} 83. h6 Kh8 {+0.00/1} 84. Kc5 Kg8
{+0.00/1} 85. Kb4 Kh8 {+0.00/1} 86. Bd3 Kg8 {+0.00/1} 87. h7+ Kh8 {+0.00/1}
88. Kc5 Kg7 {+0.00/1} 89. Bc2 Kh8 {+0.00/1} 90. Bg6 Kg7 {+0.00/1} 91. Kd6
Kh8 {+0.00/1} 92. Be4 Kg7 {+0.00/1} 93. Kc6 Kh8 {+0.00/1} 94. Kb6 Kg7
{+0.00/1} 95. Bg6 Kh8 {+0.00/1} 96. Kc6 Kg7 {+0.00/1} 97. Kc5 Kh8 {+0.00/1}
98. Kd4 Kg7 {+0.00/1} 99. Ke5 Kh8 {+0.00/1} 100. Ke4 Kg7 {+0.00/1} 101. Kf5
Kh8 {+0.00/1} 102. Ke4 Kg7 {+0.00/1} 103. Kf5 Kh8 {+0.00/1} 104. Ke4 Kg7
{+0.00/1}
{Game drawn by repetition} 1/2-1/2
[/pgn]

[Graham Banks]

Why not run this weekly?

[Modern Times]

Why not run this weekly?

I would say that is asking too much of Lisebeth the tournament director. People do have a life outside of chess and want to do other things on Saturday evenings.

[smatovic]

Why not run this weekly?

I would say that is asking too much of Lisebeth the tournament director. People do have a life outside of chess and want to do other things on Saturday evenings.

Do they?

--
Srdja