Linux/Mac engine gauntlet, cluster testing

[jhellis3]

One can compile the library into cutechess by adding -config static to the qmake command.

[matthewlai]

Pepito's code looks very old (GCC 2.x era), and didn't look like it will compile without many changes.

Pepito 1.59 compiled fine using GCC 4.8.3. In a linux system, it just needed to add -lm to the linking command in the makefile: $(CC) -lm $(CFLAGS) -o pepitox $(OBJECTS)

Ah It's probably LLVM-specific then. Do you get any warnings with "-Wall"?

Just tried to add -Wall, oh yes, plenty of warnings - the engine runs stable though, played quite some test games against Phalanx with no issues.

Ah ok. Since you have it built already, would you mind running a few games at very short time control (say "level 0 0:10 0.1")?

I wouldn't mind fixing it for LLVM (if it's not TOO much work), but it would suck if I spend a lot of time to fix it only to find out that it can't play short games (which is mostly what I am interested in).

I am interested in how Phalanx does as well. It's in my gauntlet already.

Thanks

Started a quick test match at 0:10 + 0.1, all looking almost ok so far, no time forfeits, but ... game 96 shows an issue:

Warning: pepito forfeits by invalid result claim: 1/2-1/2 {Drawn game}
Finished game 96 (pepito vs r70): 0-1 {Black wins by adjudication: Invalid result claim}
Score of r70 vs pepito: 32 - 47 - 20 [0.424] 99

Thanks for running that!

Is it obvious why the draw was claimed? Maybe it treats more material combinations are insufficient for mate?

[matthewlai]

One can compile the library into cutechess by adding -config static to the qmake command.

Thanks. I'll keep that in mind if I decide to try cutechess.

[dusando]

Pepito's code looks very old (GCC 2.x era), and didn't look like it will compile without many changes.

Pepito 1.59 compiled fine using GCC 4.8.3. In a linux system, it just needed to add -lm to the linking command in the makefile: $(CC) -lm $(CFLAGS) -o pepitox $(OBJECTS)

Ah It's probably LLVM-specific then. Do you get any warnings with "-Wall"?

Just tried to add -Wall, oh yes, plenty of warnings - the engine runs stable though, played quite some test games against Phalanx with no issues.

Ah ok. Since you have it built already, would you mind running a few games at very short time control (say "level 0 0:10 0.1")?

I wouldn't mind fixing it for LLVM (if it's not TOO much work), but it would suck if I spend a lot of time to fix it only to find out that it can't play short games (which is mostly what I am interested in).

I am interested in how Phalanx does as well. It's in my gauntlet already.

Thanks

Started a quick test match at 0:10 + 0.1, all looking almost ok so far, no time forfeits, but ... game 96 shows an issue:

Warning: pepito forfeits by invalid result claim: 1/2-1/2 {Drawn game}
Finished game 96 (pepito vs r70): 0-1 {Black wins by adjudication: Invalid result claim}
Score of r70 vs pepito: 32 - 47 - 20 [0.424] 99

Other than this, there are all kinds of correct draws: 50 move rule, repetition, insufficient material. This is the game

[Event "?"]
[Site "?"]
[Date "2014.08.13"]
[Round "96"]
[White "pepito"]
[Black "r70"]
[Result "0-1"]

1. d4 Nf6 2. c4 e6 3. g3 d5 4. Bg2 dxc4 5. Nf3 c5 6. O-O Nc6 7. Qa4 Bd7 8. Qxc4
b5 9. Qd3 Rc8 10. dxc5 Bxc5 11. Nc3 Nb4 12. Qd1 Bc6 13. Qxd8+ Rxd8 14. Bg5 O-O
15. Rac1 a5 16. Nxb5 Bxb5 17. Rxc5 Bxe2 18. Ra1 Rd5 19. Rxd5 Nfxd5 20. Nd4 Ba6
21. Bd2 Rc8 22. b3 Nc3 23. a3 Nd3 24. Rf1 Rc5 25. f4 Nd5 26. Bf3 Kh8 27. Be2 Ne7
28. b4 axb4 29. axb4 Rd5 30. b5 Bb7 31. Bf3 Rd7 32. Bxb7 Rxb7 33. Ra1 h6 34. Ra3
Nb2 35. Bb4 Nd5 36. Bc5 Nc4 37. Ra7 Rxa7 38. Bxa7 g5 39. fxg5 hxg5 40. h4 Kg7
41. Bc5 gxh4 42. gxh4 Kg6 43. Ne2 Kh5 44. Bf2 Kg4 45. Kg2 e5 46. Kg1 f6 47. Kg2
f5 48. Kf1 f4 49. Be1 e4 50. Kg1 e3 51. Nd4 Nd2 52. Nc2 Nf3+ 53. Kf1 Ne5 54. Nd4
f3 55. h5 Nf7 56. Ba5 Nd6 57. h6 Nc4 58. Be1 Nf6 59. Kg1 Nh7 60. Ne6 Kf5
61. Nd4+ Ke4 62. Ne6 Ke5 63. Nc5 Kd4 64. Ne6+ Kd3 65. Nf4+ Kc2 66. Nd5 Kd1
67. Bh4 e2 68. Nc3+ Kd2 69. Ne4+ Kd3 70. Nc5+ Ke3 71. Nb3 Nd6 72. b6 Nb7
73. Bf2+ Kd3 74. Be1 Ke3 75. Bd2+ Ke4 76. Kf2 Nf6 77. Nc1 Nc5 78. Nxe2 fxe2
79. Kxe2 Nh7 80. Be3 Nb7 81. Bc1 Nc5 82. Bd2 Kd5 83. Be3 Nd7 84. Kd3 Ne5+
85. Kc3 Kc6 86. Kd2 Ng4 87. Bd4 Nxh6 88. Kc2 Nf5 89. Bf2 Nf8 90. Kd3 Nd7 91. Ke4
Nd6+ 92. Kf3 Kd5 93. Kf4 Nb7 94. Kf5 Ne5 95. Be3 Nd6+ 96. Kg5 Ke4 97. Bc5 Nb7
98. Bf2 Nd3 99. Bg1 Na5 100. Bh2 Nb2 101. Kf6 Kd5 102. Kf5 Nb7 103. Bg1 Nc4
104. Kf4 Ncd6 105. Be3 Nf7 106. Bf2 Ne5 107. Kg5 Nd3 108. Bg3 Ke4 109. Kf6 Ndc5
110. Kg6 Na5 111. Bf2 Ncb7 112. Kf6 Kd5 113. Kf5 Nd6+ 114. Kf4 Nc6 115. Be3 Nd8
116. Kg5 Ke4 117. Bg1 Nc4 118. Kf6 Ne5 119. Ke7 Nb7 120. Bf2 Kd5 121. Ke8 Ke6
122. Bd4 Nc4 123. Bg1 Nd2 124. Be3 Ne4 125. Kf8 Nf6 126. Kg7 Kf5 127. Bd4 Nd5
128. Bf2 Ke4 129. Kf7 Ke5 130. Bg3+ Kf5 131. Bc7 Nf6 132. Bf4 Nd8+ 133. Ke7 Nc6+
134. Kd6 Nb8 135. Be3 Ke4 136. Ke6 Nd5 137. Bc5 Nxb6 0-1

[dusando]

Pepito's code looks very old (GCC 2.x era), and didn't look like it will compile without many changes.

Pepito 1.59 compiled fine using GCC 4.8.3. In a linux system, it just needed to add -lm to the linking command in the makefile: $(CC) -lm $(CFLAGS) -o pepitox $(OBJECTS)

Ah It's probably LLVM-specific then. Do you get any warnings with "-Wall"?

Just tried to add -Wall, oh yes, plenty of warnings - the engine runs stable though, played quite some test games against Phalanx with no issues.

Ah ok. Since you have it built already, would you mind running a few games at very short time control (say "level 0 0:10 0.1")?

I wouldn't mind fixing it for LLVM (if it's not TOO much work), but it would suck if I spend a lot of time to fix it only to find out that it can't play short games (which is mostly what I am interested in).

I am interested in how Phalanx does as well. It's in my gauntlet already.

Thanks

Started a quick test match at 0:10 + 0.1, all looking almost ok so far, no time forfeits, but ... game 96 shows an issue:

Warning: pepito forfeits by invalid result claim: 1/2-1/2 {Drawn game}
Finished game 96 (pepito vs r70): 0-1 {Black wins by adjudication: Invalid result claim}
Score of r70 vs pepito: 32 - 47 - 20 [0.424] 99

Thanks for running that!

Is it obvious why the draw was claimed? Maybe it treats more material combinations are insufficient for mate?

Yes, quite possibly Pepito may consider more material combinations as drawn, just check the game posted. As of now this happened 3 times out of 395 games played.

[matthewlai]

Yes, quite possibly Pepito may consider more material combinations as drawn, just check the game posted. As of now this happened 3 times out of 395 games played.

Another possible explanation for that game is that it's trying to claim 50 moves rule, but there is a bug somewhere.

The last capture/pawn move was
87. ... Nxh6

So after
137. Bc5

That's exactly 50 full moves.

Maybe it made the move, then claimed 50-moves, but by the time xboard processes that, Phalanx already did Nxb4, presumably to avoid 50 moves.

The protocol has seen been changed to recommend "offer draw" before making the last move, then make it, and if the position is drawn after the move, xboard will grant the draw without asking the other engine.

[Max]

RedQueen didn't have any build instructions or Makefile, and just compiling all the source files didn't seem to work. I didn't spend too much time on this.

Try gcc (version 4.7.3/4.8.3/4.9.1 from macports) and redqueen 1.1.97 compiles from the command line

g++ -O3 *.cpp -o redqueen

-Max

[Max]

Current versions of Rodent do not compile/run on Linux.

Try removing the line #include „book_internal.c“ in rodent.c and compile without static linking in the Makefile. Rodent 1.4 compiles with gcc and runs fine on Mac.

-Max

[matthewlai]

Played 1000 games last night (5 duplicates).

"0 0:10 0" between Crafty and Phalanx.

Code: Select all

version 0057, Copyright (C) 1997-2010 Remi Coulom.
compiled Aug 13 2014 00:56:05.
This program comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under the terms and conditions of the GNU General Public License.
See http://www.gnu.org/copyleft/gpl.html for details.
ResultSet>off
1000 game(s) loaded, 0 game(s) with unknown result ignored.
00:00:00,00
00:00:00,00              
Rank Name                 Elo    +    - games score oppo. draws 
   1 Crafty-24.0          149   13   12  1000   84%  -149    9% 
   2 Phalanx XXIII Beta  -149   12   13  1000   16%   149    9% 

I guess that means my alge<=>SAN translation is OK now! (there were a few bugs with disambiguation).

On to testing with more engines!

[bob]

Note you can also cause Crafty to produce "long algebraic" (i.e. e7e5 or Ng1f3). Crafty will also accept such moves so you can test this in both directions to be sure you are producing and parsing both correctly.

the command is "output long"