Rodent 1.4

[PK]

...is available at http://www.pkoziol.cal24.pl/rodent/rodent_download.htm

Developing this version has been a rocky ride - I started with a serious rewrite, noticed that it doesn't gain anything despite positive tests of all the separate changes, rolled back and added only conservative modifications, leaving rewrite for later. So what's new within current Rodent?

- important bugfix of weak pawn eval
- hidden passers eval
- deleting buggy pruning of pawn captures
- tuned late move pruning
- internal iterative deepening at cut nodes
- a couple of minor optimizations (i.e. reuse null move eval in razoring)

As usual, I can co only Windows compiles, so I'd be grateful for a Linux compile.

[Graham Banks]

Thanks Pawel.

[phenri]

As usual, I can co only Windows compiles, so I'd be grateful for a Linux compile.

Hi,

How do you manage to compile with missing files? This is a great mystery to me.

Code: Select all

$ make
g++ -c -g rodent.c  -s -w -Wfatal-errors -pipe -s -DNDEBUG -Ofast -static -fno-rtti \
                       -finline-functions -fprefetch-loop-arrays -flto -fwhole-program
rodent.c:8:27: fatal error: book_internal.c: No such file or directory
compilation terminated.
make: *** [rodent.o] Error 1

[velmarin]

Just exclude the archive.

At least in Visual Studio.
Compiles and works well.

Pawel makes a nice job.

[Jamal Bubker]

Thanks Pawel !!

[Werner]

Hi Pawel,
I have some Problems here under Arena 3.0. The engine crashes (looses on time) and then does not Play anymore. All games after that are finished after the opening. Now trying what happens when I restart the engines after each game.

regards
Werner

[pgn][Event "Rodent2"]
[Site "I7"]
[Date "2014.05.02"]
[Round "2"]
[White "Scorpio 2.77 x64 1CPU"]
[Black "Rodent 1.4 x64"]
[Result "1-0"]
[BlackElo "2200"]
[ECO "B19"]
[Opening "Caro-Kann (Hauptvariante), Spassky, 10.Qxd3 Ngf6"]
[Time "09:26:24"]
[Variation "Classical, Spassky variation"]
[WhiteElo "2200"]
[TimeControl "40/480:40/480:40/480"]
[Termination "time forfeit"]
[PlyCount "179"]
[WhiteType "program"]
[BlackType "program"]

1. e4 c6 2. d4 d5 3. Nc3 dxe4 4. Nxe4 Bf5 5. Ng3 Bg6 6. h4 h6 7. Nf3 Nd7 8.
h5 Bh7 9. Bd3 {+0.10/21 34} Bxd3 {-0.04/21 15} 10. Qxd3 {+0.16/21 13} Ngf6
{0.00/21 15} 11. Qb3 {-0.02/19 13} Qb6 {+0.08/21 13} 12. Be3 {-0.02/19 13}
e6 {+0.16/20 12} 13. Nf1 {-0.03/19 13} Qxb3 {+0.24/21 13} 14. axb3
{+0.06/23 11} Be7 {+0.20/19 11} 15. Ke2 {+0.04/20 18} O-O {+0.24/21 12} 16.
N1d2 {+0.05/20 13} c5 {+0.28/20 13} 17. Nc4 {+0.01/19 11} Rfc8 {+0.28/20
15} 18. c3 {+0.04/19 11} Nd5 {+0.20/20 12} 19. Nce5 {+0.04/19 13} cxd4
{+0.36/21 13} 20. Bxd4 {-0.02/22 13} Nc5 {+0.24/21 15} 21. Nd2 {+0.05/21
12} a5 {+0.12/20 15} 22. g3 {+0.05/19 12} b6 {0.00/20 15} 23. f4 {+0.20/20
12} Re8 {+0.04/21 15} 24. Bxc5 {+0.20/21 14} Bxc5 {0.00/22 15} 25. Ne4
{+0.27/22 12} Bf8 {+0.04/22 15} 26. Kf3 {+0.23/19 12} f6 {0.00/22 12} 27.
Ng6 {-0.02/21 12} Rad8 {+0.20/23 13} 28. Rad1 {-0.02/22 15} f5 {+0.08/23
16} 29. Nd2 {-0.02/23 29} b5 {+0.08/24 18} 30. Ne5 {+0.06/22 12} Nf6
{+0.08/20 13} 31. Kg2 {-0.02/23 14} Bc5 {-0.04/22 14} 32. Ndf3 {+0.16/21
10} Rxd1 {-0.16/22 24} 33. Rxd1 {+0.20/20 3} Nxh5 {-0.04/23 14} 34. Nd7
{+0.39/23 13} Be7 {-0.04/23 13} 35. Nd4 {+0.26/22 16} g5 {-0.04/22 12} 36.
Nxe6 {+0.46/21 15} gxf4 {-0.04/22 15} 37. Nxf4 {+0.37/23 17} Nxf4+
{-0.04/23 13} 38. gxf4 {+0.37/22 4} Rd8 {-0.03/24 19} 39. Rd5 {+0.29/25 39}
b4 {-0.03/24 14} 40. Kf3 {+0.61/24 13} h5 {-0.24/25 19} 41. c4 {+0.54/24 9}
h4 {-0.40/25 18} 42. c5 {+0.93/24 13} Kf7 {-0.44/24 11} 43. c6 {+0.76/25
11} Ke6 {-0.40/23 17} 44. Re5+ {+0.78/25 8} Kd6 {-0.28/23 6} 45. Rxf5
{+0.70/25 9} Rh8 {-0.36/23 10} 46. Ne5 {+0.65/24 11} Rh7 {-0.24/25 11} 47.
Kg2 {+0.66/25 12} h3+ {-0.48/23 9} 48. Kh2 {+0.54/26 12} Kc7 {-0.44/25 18}
49. Nf3 {+1.01/24 10} Kxc6 {-0.44/25 11} 50. Rxa5 {+0.90/23 9} Bc5
{-0.40/25 12} 51. Ne5+ {+1.07/24 9} Kb6 {-0.36/25 10} 52. Ra1 {+1.16/22 10}
Kb5 {-0.36/25 11} 53. Rd1 {+0.98/24 13} Rh8 {-0.44/26 13} 54. Rf1 {+0.98/24
23} Rh5 {-0.40/25 12} 55. Nc4 {+0.98/26 11} Ba7 {-0.52/26 11} 56. Ra1
{+0.94/25 11} Bc5 {-0.44/28 11} 57. Rd1 {+1.09/26 9} Rh7 {-0.44/24 13} 58.
f5 {+1.35/25 10} Rh8 {-0.48/23 9} 59. f6 {+0.95/25 17} Ba7 {-1.12/24 11}
60. Ne5 {+1.47/27 9} Bb8 {-1.16/25 10} 61. Rd5+ {+1.40/31 10} Kb6 {-1.24/26
9} 62. Kh1 {+1.20/31 8} Rc8 {-0.92/27 11} 63. Nd3 {+1.45/26 16} Rf8
{-0.92/25 12} 64. Rf5 {+1.11/27 36} Kc6 {-1.08/26 10} 65. Nxb4+ {+1.20/27
10} Kd7 {-1.08/23 10} 66. Nd3 {+1.20/26 9} Bd6 {-1.08/25 12} 67. Rf3
{+1.12/26 18} h2 {-1.20/22 10} 68. b4 {+0.58/23 9} Rg8 {-1.04/24 11} 69.
Nc5+ {+1.02/22 10} Kc6 {-0.88/25 11} 70. Rf1 {+1.05/24 8} Rg4 {-0.80/25 18}
71. f7 {+0.62/25 10} Rxb4 {-0.84/25 12} 72. Nd3 {+0.51/26 31} Rb8 {-0.84/27
10} 73. Rf6 {+0.51/26 7} Kd7 {-0.96/25 17} 74. Nf4 {+0.51/27 7} Rc8
{-1.00/25 8} 75. b3 {+0.81/24 12} Ra8 {-0.88/28 14} 76. Rf5 {+0.51/25 5}
Rf8 {-0.88/29 8} 77. Nh3 {+0.15/22 5} Ke6 {-0.44/23 12} 78. Rf3 {+0.14/23
8} Be7 {-0.44/23 8} 79. Nf4+ {+0.16/24 8} Kxf7 {-0.44/24 12} 80. Kxh2
{+0.14/23 7} Bd6 {-0.40/24 11} 81. Kg2 {+0.14/24 11} Kf6 {-0.48/22 10} 82.
Ne2+ {+0.14/22 9} Ke5 {-0.28/21 13} 83. Rd3 {+0.14/21 8} Bb4 {-0.36/22 12}
84. Re3+ {+0.15/21 11} Kd5 {-0.12/23 16} 85. Rh3 {+0.16/20 11} Ke4
{-0.03/23 12} 86. Nc3+ {+0.03/20 25} Bxc3 {0.00/32 11} 87. Rxc3 {0.00/33
10} Kd4 {0.00/44 10} 88. Rh3 {0.00/34 11} Rc8 {0.00/52 10} 89. b4 {0.00/28
10} Rb8 {0.00/62 9} 90. Rh4+ {0.00/30 8} ... {Schwarz überschreitet die
Zeit} 1-0 [/pgn]

[PK]

@Werner, what exact settings are You using? I have never had such problems under Arena 2.1 and Rodent is capable of playing like 10s + 0.25s increment per game under LittleBlitzer, so I suspect Arena glitch.

[Werner]

Hi Pawel,
thanks for the answer. I am running tournaments with 40 moves in 8 minutes, ponder = off.
From the game it looks like that it happens in draw positions - where the engine wants to stop the game and give it a draw ?

regards

[Werner]

Hi,
now 50 games played under these conditions. There are 2 games lost in time, both in a clear draw position. And with Arena feature: start always engines new, all games where played without more problems.
regards
Werner

[PK]

@Werner, perhaps I begin to understand what's happening here. I just need one little hint. In the games that Rodent lost on time, did it display suspiciously big depths playing a couple of moves before losing on time?