The winboard.debug (edited to leave only the communication with Bihasa) is this:
Code: Select all
StartChildProcess (dir="../Bihasa") Bihasa
549537 >first : xboard
protover 2
549537 <first : # Bihasa 1.0
549537 <first : # Compiled: Sep 17 2011
549537 <first : # Can only play Capablanca and gothic chess variants
549537 <first :
549537 <first : # Reading initialization file...
549537 <first :
549537 <first : # Hash : 49 Mb
549537 <first : # Eval Cache : 16 Mb
549537 <first : # Resign : OFF
549537 <first : # Resign Score : -1200
549537 <first : # Resign Score Count : 5
549537 <first : # Contempt Value : 0
549537 <first : # Ponder : ON
549537 <first :
549537 <first : # Pawn_Value 100 100
549537 <first : # Knight_Value 315 315
549537 <first : # Bishop_Value 325 325
549537 <first : # Rook_Value 500 500
549537 <first : # ArchBishop_Value 980 980
549537 <first : # Chancellor_Value 1015 1015
549537 <first : # Queen_Value 1150 1150
549537 <first :
549537 <first : # Closing initialization file...
549537 <first :
549537 <first :
549537 <first :
549599 <first : feature ping=1 time=1 setboard=1 name=1 colors=1
549599 >first : accepted ping
549599 >first : accepted time
549599 >first : accepted setboard
549599 >first : accepted name
549599 >first : accepted colors
549599 <first : feature draw=0 sigint=0 sigterm=0 reuse=0 analyze=1 ics=1
549599 >first : accepted draw
549599 >first : accepted sigint
549599 >first : accepted sigterm
549599 >first : accepted reuse
549599 >first : accepted analyze
549599 >first : accepted ics
549599 <first : feature variants="capablanca, gothic"
549599 >first : accepted variants
549599 <first : feature myname="Bihasa 1.0"
549599 >first : accepted myname
549599 <first : feature done=1
549599 >first : accepted done
549615 >first : new
random
549615 >first : variant capablanca
549615 >first : ics -
549615 >first : level 40 1 0
549615 >first : post
549615 >first : hard
549615 >first : easy
549615 >first : ping 1
549615 >first : force
549615 >first : setboard rnbqckabnr/pppppppppp/10/10/10/10/PPPPPPPPPP/RNBQCKABNR w KQkq - 0 1
549677 <first : # received variant command capablanca
549740 <first : # moves per session = 40
549740 <first : # time per session = 1
549740 <first : # inc per move = 0
549740 <first : # Total Initial Time = 60 sec.
549740 <first : # Conventional Time Control.
549740 <first : # Moves per session = 40 moves.
549740 <first : # Minutes per game = 1.00 min.
549740 <first : # Increment = 0 ms.
549740 <first : pong 1
549802 <first : rnbqckabnr/pppppppppp/10/10/10/10/PPPPPPPPPP/RNBQCKABNR w KQkq -
New game (0): Bihasa 1.0-Fairy-Max 4.8R (w)
554560 >first : computer
554560 >first : name Fairy-Max 4.8R
554560 >second: computer
554560 >first : black
554560 >first : time 6000
554560 >first : otim 6000
554560 >first : white
554560 >first : go
554560 <first : # TimeLeft=60000.00 ms. MaxTime=60000.00 ms AllocatedTime=60000.00 ms.
554560 <first : # ply score time nodes Pv line
554560 <first : # -----------------------------------
554560 <first : 1 47 0 32 b1c3
554576 <first : 2 0 0 202 b1c3 b8c6
554576 <first : 3 30 1 309 b1c3
554576 <first : 3 47 1 454 b1c3 b8c6 i1h3
554576 <first : 4 0 1 1020 b1c3 b8c6 i1h3 i8h6 (1.0)
554576 <first : 5 24 1 1482 b1c3 b8c6 i1h3 i8h6 d2d4 (1.0)
554591 <first : 6 0 1 2815 b1c3 b8c6 i1h3 i8h6 d2d4 d7d5 (1.0)
554607 <first : 7 22 1 5682 b1c3 b8c6 i1h3 i8h6 d2d4 g7g6 c1e3 (1.0)
554607 <first : 8 12 3 14765 b1c3 b8c6 i1h3 i8h6 d2d4 g7g6 d4d5 c6d4 (1.9)
554669 <first : 9 16 10 47613 b1c3 b8c6 i1h3 i8h6 d2d4 g7g6 d4d5 c6d4 c1f4 d7d6 (3.5)
555028 <first : 10 0 45 224436 b1c3 b8c6 i1h3 d7d5 d2d4 g7g6 g2g3 h8d4 h1d5 i8h6 (4.2)
555574 <first : 11 8 101 511731 b1c3 i8h6 i1h3 d7d6 g2g3 b8c6 d2d3 c8g4 c1g5 h6i4 g5f4 (2.2)
556058 <first : 12 7 149 756853 b1c3 i8h6 i1h3 d7d6 g2g3 b8c6 d2d3 c8g4 c1g5 h6i4 g5f4 e7e5 j2j3 e5f4 h1c6 (1.5)
557087 <first : 13 7 252 1259429 b1c3 i8h6 i1h3 d7d6 g2g3 b8c6 d2d3 c8g4 c1g5 h6i4 g5f4 e7e5 j2j3 e5f4 h1c6 (1.7)
560504 <first : 14 0 594 2941092 b1c3 i8h6 i1h3 d7d6 g2g3 b8c6 d2d3 g7g6 c1g5 c8f5 h1d5 h6i4 g5f4 h8d4 (2.4)
565262 <first : 15 5 1070 5232875 b1c3 i8h6 i1h3 d7d6 g2g3 b8c6 d2d3 g7g6 h1d5 c6b4 d5f3 h8d4 c1g5 h6i4 g5f4 c7c6 (1.8)
574871 <first : 16 5 2031 9843909 b1c3 i8h6 i1h3 b8c6 g2g3 g7g6 d2d3 a7a6 j2j4 d7d6 c1f4 c8f5 h3g5 e7e5 f4d2 j7j5 (1.9)
593498 <first : 17 6 3893 18663774 b1c3 i8h6 i1h3 b8c6 g2g3 g7g6 d2d3 a7a6 c1f4 d7d6 a2a4 c8f5 h3g5 h8d4 e2e3 d4b6 h1d5 j7j5 (1.9)
614573 >first : result 0-1 {Black wins on time}
614573 >first : quit
This is the first time I tried Bihasa from a setup position; I don't know if it is related to that. The previous Bihasa tests I did were all gauntlets (from the standard opening position), and there Bihasa did always remain loaded as gauntlet engine, so it could be I missed the error in the first game.
Some other remarks w.r.t. protocol:
*) Please no space before 'gothic' in the variants feature. Current WB does not care, because it matches variant names very sloppily (using strstr(featuredVariants, chosenVariant), so it would even recognize 'pseudogothic' as if the engine played gothic, but formally the space would be part of the variant name.. So when future WB versions would do more strict matching, it would no longer recognize ' gothic' as gothic.
*) You nicely use # to comment out the engine debug output with # to disarm it, but this is only fully effective when you also put debug=1 in the features. (The idea was that when a GUI answers you rejected debug, you know it will not recognize the # as special, and you could refrain from sending such lines. Not terribly important, I admit, because most likely you would want to take the risk and send them anyway.)
*) When you echo the FEN, there is no #.
, the TC type was changed when receiving positions from setboard . I have not encountered this since I do my testing from pgn files. I will fix this including those additional comments.
