Chess variant tournament: Capablanca and similar

[enhorning]

The Bihasa - Bihasa game was also over surprisingly quick - most games between programs near the top last way longer than this:

[Event "Computer Chess Game"]
[Site "ENHORNING"]
[Date "2013.03.30"]
[Round "12"]
[White "Bihasa v3.4.3"]
[Black "Bihasa v3.2"]
[Result "0-1"]
[TimeControl "40/1200"]
[Variant "capablanca"]
[FEN "rnabqkbcnr/pppppppppp/10/10/10/10/PPPPPPPPPP/RNABQKBCNR w KQkq - 0 1"]
[SetUp "1"]

{--------------
r n a b q k b c n r
p p p p p p p p p p
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
P P P P P P P P P P
R N A B Q K B C N R
white to play
--------------}
1. Nh3 {+0.16/18} Nh6 {-0.12/18 1:00} 2. d4 {+0.12/16 31} d5 {-0.08/16 32}
3. f4 {+0.08/16 31} f5 {-0.08/17 42} 4. j4 {+0.16/15 55} Nc6 {-0.04/15 31}
5. Nd2 {+0.12/16 37} Ng4 {-0.20/16 28} 6. c3 {+0.16/16 28} h6 {-0.16/16 28}
7. Nf3 {+0.16/15 28} e6 {-0.16/15 28} 8. Nf2 {+0.16/16 26} Nxf2
{-0.20/15 42} 9. Bxf2 {+0.20/13 2.8} j5 {-0.12/15 33} 10. Cg3 {+0.12/15 46}
Ad6 {-0.12/15 44} 11. O-O {+0.20/15 30} Af7 {+0.46/18 37} 12. Ch3
{-0.24/16 40} Ai4 {+0.52/16 40} 13. e3 {-0.16/16 40} i5 {+1.02/16 26} 14.
jxi5 {-0.68/16 38} j4 {+1.36/16 30} 15. Cg1 {-1.24/15 36} j3 {+1.64/15 30}
16. i3 {-1.56/14 26} Qg6 {+2.26/15 26} 17. Qf1 {-0.96/12 33} hxi5
{+2.36/15 32} 18. Kj1 {-1.72/13 31} Ch5 {+4.14/15 31} 19. Ch3 {-3.28/13 30}
Qg4 {+24.18/17 25} 20. Bh4 {-299.82/13 28} Bxh4 {+299.83/14 23} 21. Nxh4
{-299.84/14 5} Axh2+ {+299.85/17 25} 22. Rxh2 {-299.86/16 4} Cxi3+
{+299.87/18 25} 23. Kj2 {-13.92/4 0.1} Ci4+ {+299.89/17 25} 24. Kj1
{-13.92/4 0.1} Cxh2+ {+299.91/17 25} 25. Ki1 {-13.92/3 0.1} Cxf1+
{+299.93/16 25} 26. Kj2 {-299.94/6 0.1} Qxh3 {+299.95/10 0.1} 27. Ng6+
{-299.96/4 0.1} Ke8 {+299.97/4 0.1} 28. gxh3 {-299.98/4 0.1} Ch1#
{+299.99/4 0.1}
{Xboard adjudication: Checkmate} 0-1

[enhorning]

... and current standings - getting close to halfway now, should get there at some point, and then start the rematches:

Code: Select all

Cross table, sorted by score percentage, Buchholz, SB

                              B B T N N S J P S B T S F M C H A
 1. Bihasa v3.2               # 1   1 1 1 1 1   1 1 1 1 1 1 1 1  100%  14.0 ( 84.0,  84.0)
 2. Bihasa v3.4.3             0 # 1 1 1 1 1 1 1 1 1 1 1     1 1   93%  13.0 ( 98.5,  84.5)
 3. TJchess10x8 1.1-x64         0 # 1 1 1 1 1 1   1 1 1 1 1 1 1   93%  13.0 ( 85.0,  72.0)
 4. NebiyuChess_1.43          0 0 0 # 1 1   1 1 1 1 1 1 0     1   69%   9.0 ( 97.0,  53.5)
 5. NebiyuChess_1.44          0 0 0 0 # 1 1 1 = 1   1 1 1 = 1 1   67%  10.0 (103.0,  49.0)
 6. Spartacus 1.02            0 0 0 0 0 # 1   1 1 1 1   1 = 1 1   61%   8.5 ( 98.0,  37.5)
 7. Joker80.np                0 0 0   0 0 # 1 1 1 1   = = 1 1     54%   7.0 ( 96.5,  34.0)
 8. Pair-o-Max 4.8S           0 0 0 0 0   0 #   0 1 1 1 1 1 1 1   50%   7.0 ( 95.5,  23.5)
 9. Smirf BC-173g-X             0 0 0 = 0 0   # 0 1 1 = 1 1 1 1   50%   7.0 ( 90.0,  26.3)
10. BigLion80 2.23x WB        0 0   0 0 0 0 1 1 # 0 1 = 1 = 1     43%   6.0 ( 98.0,  27.8)
11. TSCPtest                  0 0 0 0   0 0 0 0 1 # 0   1 1 1 1   36%   5.0 ( 98.5,  15.5)
12. Sjaak 524                 0 0 0 0 0 0   0 0 0 1 # 1 1 =   1   32%   4.5 (104.5,  15.5)
13. Fairy-Max 4.8S            0 0 0 0 0   = 0 = =   0 # 0 1 1 1   32%   4.5 (100.0,  16.0)
14. Max-Plus 4.8S             0   0 1 0 0 = 0 0 0 0 0 1 # 0 0 1   23%   3.5 (101.5,  18.0)
15. ChessV_WinBoard           0   0   = = 0 0 0 = 0 = 0 1 # 0     23%   3.0 ( 92.0,  18.0)
16. Heretic 0.3               0 0 0   0 0 0 0 0 0 0   0 1 1 # 0   14%   2.0 (102.5,   6.5)
17. ArcBishop80 1.01 WB       0 0 0 0 0 0   0 0   0 0 0 0   1 #    8%   1.0 (101.0,   2.0)

What surprises me is Max-Plus's weak performance so far.

[enhorning]

[Event "Computer Chess Game"]
[Site "ENHORNING"]
[Date "2013.03.31"]
[Round "2"]
[White "ArcBishop80"]
[Black "ChessV_WinBoard"]
[Result "0-1"]
[TimeControl "40/1200"]
[Variant "capablanca"]
[FEN "rnabqkbcnr/pppppppppp/10/10/10/10/PPPPPPPPPP/RNABQKBCNR w KQkq - 0 1"]
[SetUp "1"]

1. d3 {-0.04/7} Nh6 {-0.06/11 24} 2. e4 {-0.09/7 28} d5 {+0.24/11 29} 3.
exd5 {+0.18/7 28} Axi2 {+0.59/12 28} 4. Ci3 {+0.04/7 28} Ah4 {+0.59/11 28}
5. Cxi7 {+0.14/6 28} Ri8 {+0.55/10 28} 6. Qe4 {-0.73/6 28} e5 {+1.25/10 25}
7. Ch5 {-0.43/6 28} Ai2 {+1.58/10 24} 8. Qb4+ {-0.60/6 28} Be7
{+1.16/10 28} 9. Qxb7 {-0.68/6 16} Nd7 {+1.05/10 28} 10. Cxh6 {+0.00/6 23}
Axj1 {+1.05/9 28} 11. Bg4 {+0.81/7 28} Nc5 {-0.43/9 29} 12. Qxc7
{+1.81/6 28} Rxi1 {-1.59/8 29} 13. Bxj1 {+2.99/6 28} Rxj1 {-2.50/9 26} 14.
Cf5 {+3.00/6 28} Nd7 {-1.82/8 30} 15. Na3 {+2.87/5 28} Ci6 {-1.54/8 22} 16.
Ae2 {+2.71/6 28} Ci2 {-1.78/8 27}
{Xboard: Forfeit due to invalid move: f1a1 (f1a1) res=23} 0-1

... seems ArcBishop was trying to do a long castling, but didn't send the move in a format that Winboard recognized.

[enhorning]

Standings halfway through the Capablanca setup:

Code: Select all

Cross table, sorted by score percentage, Buchholz, SB

                              B B T N N S J S P B T S F C M A H
 1. Bihasa v3.2               # 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1  100%  16.0 (120.0, 120.0)
 2. Bihasa v3.4.3             0 # 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1   94%  15.0 (121.0, 105.0)
 3. TJchess10x8 1.1-x64       0 0 # 1 1 1 1 1 1 1 1 1 1 1 1 1 1   87%  14.0 (122.0,  91.0)
 4. NebiyuChess_1.43          0 0 0 # 1 1 0 1 1 1 1 1 1 1 0 1 1   69%  11.0 (125.0,  66.5)
 5. NebiyuChess_1.44          0 0 0 0 # 1 1 = 1 1 1 1 1 = 1 1 1   69%  11.0 (125.0,  63.0)
 6. Spartacus 1.02            0 0 0 0 0 # 1 1 1 1 1 1 1 = 1 1 1   66%  10.5 (125.5,  56.5)
 7. Joker80.np                0 0 0 1 0 0 # 1 1 1 1 1 = 1 = 1 1   62%  10.0 (126.0,  55.5)
 8. Smirf BC-173g-X           0 0 0 0 = 0 0 # 1 0 1 1 = 1 1 1 1   50%   8.0 (128.0,  37.8)
 9. Pair-o-Max 4.8S           0 0 0 0 0 0 0 0 # 0 1 1 1 1 1 1 1   44%   7.0 (129.0,  27.5)
10. BigLion80 2.23x WB        0 0 0 0 0 0 0 1 1 # 0 1 = = 1 0 1   37%   6.0 (130.0,  30.3)
11. TSCPtest                  0 0 0 0 0 0 0 0 0 1 # 0 1 1 1 1 1   37%   6.0 (130.0,  22.0)
12. Sjaak 524                 0 0 0 0 0 0 0 0 0 0 1 # 1 = 1 1 1   34%   5.5 (130.5,  20.0)
13. Fairy-Max 4.8S            0 0 0 0 0 0 = = 0 = 0 0 # 1 0 1 1   28%   4.5 (131.5,  20.0)
14. ChessV_WinBoard           0 0 0 0 = = 0 0 0 = 0 = 0 # 1 1 0   25%   4.0 (132.0,  22.0)
15. Max-Plus 4.8S             0 0 0 1 0 0 = 0 0 0 0 0 1 0 # 1 0   22%   3.5 (132.5,  22.5)
16. ArcBishop80 1.01 WB       0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 # 1   12%   2.0 (134.0,   8.0)
17. Heretic 0.3               0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 #   12%   2.0 (134.0,   7.5)

So, Bihasa hasn't dropped points to any other programs, and TJChess has only dropped points to Bihasa. Max-Plus's weak performance so far remains surprising to me.

Code: Select all

ResultSet-EloRating>ratings
Rank Name                  Elo    +    - games score oppo. draws
   1 Bihasa v3.2           505  255  255    16  100%   -32    0%
   2 Bihasa v3.4.3         410  217  217    16   94%   -26    0%
   3 TJchess10x8 1.1-x64   324  193  193    16   88%   -20    0%
   4 NebiyuChess_1.43      151  168  168    16   69%    -9    0%
   5 NebiyuChess_1.44      124  157  157    16   69%    -8   13%
   6 Spartacus 1.02        104  160  160    16   66%    -6    6%
   7 Joker80.np             68  156  156    16   63%    -4   13%
   8 Smirf BC-173g-X       -12  151  151    16   50%     1   13%
   9 Pair-o-Max 4.8S       -63  156  156    16   44%     4    0%
  10 TSCPtest             -110  157  157    16   38%     7    0%
  11 BigLion80 2.23x WB   -113  155  155    16   38%     7   13%
  12 Sjaak 524            -139  156  156    16   34%     9    6%
  13 Fairy-Max 4.8S       -167  152  152    16   28%    10   19%
  14 ChessV_WinBoard      -181  153  153    16   25%    11   25%
  15 Max-Plus 4.8S        -235  169  169    16   22%    15    6%
  16 Heretic 0.3          -331  188  188    16   13%    21    0%
  17 ArcBishop80 1.01 WB  -335  186  186    16   13%    21    0%
ResultSet-EloRating>los
                     Bi Bi TJ Ne Ne Sp Jo Sm Pa TS Bi Sj Fa Ch Ma He Ar
Bihasa v3.2             72 88 99 99 99 99 99 99 99 99 99 99 99 99 99 99
Bihasa v3.4.3        27    73 97 98 98 99 99 99 99 99 99 99 99 99 99 99
TJchess10x8 1.1-x64  11 26    91 94 96 98 99 99 99 99 99 99 99 99 99 99
NebiyuChess_1.43      0  2  8    59 66 76 92 97 98 98 99 99 99 99 99 99
NebiyuChess_1.44      0  1  5 40    57 69 90 95 98 98 99 99 99 99 99 99
Spartacus 1.02        0  1  3 33 42    62 85 93 97 97 98 99 99 99 99 99
Joker80.np            0  0  1 23 30 37    77 88 94 95 97 98 98 99 99 99
Smirf BC-173g-X       0  0  0  7  9 14 22    68 81 82 88 93 94 97 99 99
Pair-o-Max 4.8S       0  0  0  2  4  6 11 31    66 67 75 83 86 93 98 98
TSCPtest              0  0  0  1  1  2  5 18 33    51 60 70 74 86 96 96
BigLion80 2.23x WB    0  0  0  1  1  2  4 17 32 48    59 69 74 86 96 96
Sjaak 524             0  0  0  0  0  1  2 11 24 39 40    60 65 80 94 94
Fairy-Max 4.8S        0  0  0  0  0  0  1  6 16 29 30 39    55 72 91 92
ChessV_WinBoard       0  0  0  0  0  0  1  5 13 25 25 34 44    68 89 90
Max-Plus 4.8S         0  0  0  0  0  0  0  2  6 13 13 19 27 31    78 79
Heretic 0.3           0  0  0  0  0  0  0  0  1  3  3  5  8 10 21    51
ArcBishop80 1.01 WB   0  0  0  0  0  0  0  0  1  3  3  5  7  9 20 48

PGN file for the games that far can be downloaded from https://www.dropbox.com/s/xicymykn49eff ... alfway.pgn - please let me know if this works or not, it's the first time I've tried it!

[enhorning]

Amusing - very late castling by Bihasa. If it had wanted to, it could have won this game vs. Spartacus without ever moving its King or A1 rook:

[enhorning]

Also I have a suggestion if you want something to challenge Bihasa. Nebiyu is multi-processor so you can let it use 2 processors for the later stages. I mean I can not be responsible for the lack of MP support for other variant engines

Not for the main tournament, but I am playing with the idea of between rounds of the main tournament, running a super-tournament with the top 3 or 4 programs (Bihasa, TJChess, Nebiyu, and maybe Spartacus as things are right now), with longer time control, bigger hashes... and MP for any program happening to support it, either 4 or 6 cores... Might give Nebiyu a considerably better chance - you'd have to advise me on the best settings for the .ini file though.

[hgm]

Stange that Max-Plus is doing so poorly. In my 40 moves/min tests it beats Pair-o-Max by 77%, and I have to handicap it by time odds of a factor 4 (1 min vs 15 sec / 40 moves) to make it an even contest.

Does it have many time losses? In a fair number of games (1-2%) Max-Plus falls into an 'near-infinite think', which can last from minutes to hours. The patch I made for accurately updating total material on promotion (a case where it often happened) did not cure that. Strange thing is that in many cases this happens when the mate is within the horizon, where normally the depth shoots up to 28 in no time. In such cases it does not seem to produce any thinking output at all, after its opponent produced a -79.98 or +79.97 score at 28 ply. Very fishy, and I can never reproduce it.

Engines that are crash prone might actually get weaker when you give them more thinking time. With a fixed crash rate per time unit, given enough time they would crash in every game, while in fast games the can finish most of the games without crashing.

As to ArcBishop:

Indeed, it seems to send the castling in UCI FRC format. I never saw this problem before; there must be other games where ArcBishop has castled??? I am also not sure how WinBoard would treat KxR move; internally it uses this representation for FRC castlings, so it cannot be excluded that older versions of WinBoard accidentally would have understood this.

[enhorning]

Does it have many time losses?

3 losses on time for it. But two of those were lost positions, the third a draw, so only a small impact on the score. Fairy-max also had a loss on time, and Heretic has had a few.

[Daniel Shawul]

Also I have a suggestion if you want something to challenge Bihasa. Nebiyu is multi-processor so you can let it use 2 processors for the later stages. I mean I can not be responsible for the lack of MP support for other variant engines

Not for the main tournament, but I am playing with the idea of between rounds of the main tournament, running a super-tournament with the top 3 or 4 programs (Bihasa, TJChess, Nebiyu, and maybe Spartacus as things are right now), with longer time control, bigger hashes... and MP for any program happening to support it, either 4 or 6 cores... Might give Nebiyu a considerably better chance - you'd have to advise me on the best settings for the .ini file though.

Ok thanks! I will work on it till then and hopefully the additional processors close the gap with Bihasa. It would be interesting to see if deeper search can compensate for absence of evaluation. Right now on single core nebiyu gets about 4x more nps than bihasa but still looses very badly so it is not a sure thing it will catch up. It may be 200-300 elo below bihasa, but is definitely worth a shot.
Daniel

[enhorning]

Bihasa got its first loss... at a very quick look through the game, I am uncertain exactly where it went wrong, but its King got chased around a lot by the pair of Nebiyu's Knights in midgame.

[Event "Computer Chess Game"]
[Site "ENHORNING"]
[Date "2013.03.31"]
[Round "19"]
[White "Bihasa v3.4.3"]
[Black "NebiyuChess_1.43"]
[Result "0-1"]
[TimeControl "40/1200"]
[Variant "capablanca"]
[FEN "rnabqkbcnr/pppppppppp/10/10/10/10/PPPPPPPPPP/RNABQKBCNR w KQkq - 0 1"]
[SetUp "1"]

1. Nh3 {+0.16/18} Nc6 {-0.10/18 28} 2. d4 {+0.56/19 40} Nxd4 {-0.05/19 22}
3. Axi7 {+1.36/19 50} Nh6 {-0.05/19 23} 4. Axh8 {+1.48/19 26} Rxh8
{-0.05/19 29} 5. i4 {+1.28/18 34} d6 {-0.15/18 29} 6. i5 {+1.20/16 34} Nj5
{-0.20/18 29} 7. Qd2 {+0.92/15 49} e5 {+0.15/17 27} 8. c3 {+0.88/15 31} Ne6
{+0.15/16 29} 9. e4 {+0.60/15 45} h5 {+0.10/16 29} 10. Na3 {+0.48/14 43}
Nh4 {+0.60/17 24} 11. Qi7 {+0.24/16 40} Rh7 {+0.60/19 29} 12. Qc1
{+0.20/15 38} Qd7 {+1.45/18 24} 13. g4 {-0.44/15 36} Nc5 {+1.75/19 23} 14.
i6 {-0.40/15 25} Nxi6 {+1.85/19 30} 15. Ni5 {-0.52/16 33} Rh6 {+1.85/18 25}
16. Nxj7 {-0.84/14 32} Rf6 {+2.00/19 30} 17. j4 {-0.64/14 30} Rf4
{+2.60/17 30} 18. Ni5 {-1.32/15 26} Bh4 {+3.20/17 43} 19. j5 {-1.50/14 27}
Ng5 {+2.95/17 30} 20. j6 {-1.76/14 26} Bxi5 {+2.95/17 30} 21. h4
{-2.16/14 25} hxg4 {+3.45/17 24} 22. j7 {-1.76/11 23} g3 {+4.15/17 24} 23.
Qxf4 {-3.04/12 23} Qh3+ {+4.60/17 30} 24. Ke1 {-1.48/13 23} exf4
{+4.45/17 31} 25. hxi5 {-2.48/13 23} Nd3+ {+4.85/16 25} 26. Kd2
{-1.76/15 23} Nxe4+ {+3.15/18 41} 27. Kc2 {-3.52/16 23} Ne1+ {+3.90/19 33}
28. Kc1 {-5.12/16 23} gxf2 {+3.25/19 24} 29. Bxf2 {+0.00/15 5} Nd3+
{+4.40/21 24} 30. Kc2 {-1.20/16 25} Qxh1 {+4.20/21 30} 31. Rxh1
{-1.06/15 9} Ndxf2 {+4.70/22 28} 32. j8=Q {-1.72/17 26} Nxh1 {+3.80/22 39}
33. Bi6 {-2.40/17 26} Af5 {+5.15/24 27} 34. Rxh1 {-3.00/14 26} Ng5+
{+7.40/20 7} 35. Kc1 {-5.32/17 26} Nxi6 {+7.80/21 39} 36. Qj3 {-6.00/17 27}
Re8 {+7.85/20 39} 37. b4 {-7.00/17 27} Re3 {+10.35/24 38} 38. Qj2
{-9.20/19 27} Ad3+ {+11.40/24 32} 39. Kd1 {-9.72/18 27} Ng5 {+11.60/22 46}
40. Rh8 {-11.40/16 34} f6 {+12.30/22 40} 41. Qd2 {-13.20/20 1:01} Ae4
{+13.55/22 29} 42. Kc1 {-13.56/19 58} Nf3 {+14.95/22 29} 43. Qc2
{-16.28/18 55} Rxc3 {+16.80/26 26} 44. Qxc3 {-16.68/17 7} Axc3
{+16.80/24 7} 45. b5 {-17.08/17 52} Ne5 {+17.65/22 30} 46. b6
{-19.00/16 49} cxb6 {+17.95/22 30} 47. i6 {-18.76/14 27} f3 {+26.95/24 30}
48. i7 {-21.56/13 27} f2 {+299.84/25 25} 49. Rxg8+ {-299.86/14 27} Kf7
{+299.86/27 30} 50. Rf8+ {-299.88/14 27} Ke6 {+299.88/26 30} 51. Re8+
{-299.90/14 27} Kd5 {+299.90/26 30} 52. Rxe5+ {-299.92/14 25} dxe5
{+299.92/25 30} 53. i8=Q {-299.94/6 0.1} f1=C+ {+299.94/25 30} 54. Kc2
{-299.96/3 0.1} Ad4+ {+299.96/25 30} 55. Kd3 {-299.98/3 0.1} Cf3#
{+299.98/25 23}
{Xboard adjudication: Checkmate} 0-1

And current standings:

Code: Select all

Cross table, sorted by score percentage, Buchholz, SB

                              Bi Bi TJ Ne Ne Sp Jo Sm Pa TS Bi Sj Fa Ma Ch Ar He
 1. Bihasa v3.2               ## 1  1  11 1  11 11 1  11 11 11 1  11 1  1  1  1   100%  23.0 (255.5, 255.5)
 2. Bihasa v3.4.3             0  ## 11 1  10 11 1  11 1  11 1  11 1  1  1  11 1    91%  21.0 (261.0, 222.0)
 3. TJchess10x8 1.1-x64       0  00 ## 1  11 1  11 1  1  1  1  11 1  11 11 1  11   87%  20.0 (249.5, 184.5)
 4. NebiyuChess_1.44          00 0  0  ## 0  1  11 =  11 1  11 1  11 11 =1 1  11   75%  18.0 (255.0, 142.8)
 5. NebiyuChess_1.43          0  01 00 1  ## 1  0  11 1  1  1  11 1  01 1  11 1    73%  16.0 (253.0, 149.0)
 6. Spartacus 1.02            00 00 0  0  0  ## 11 1  1  1  11 1  1  11 =1 1  11   67%  15.5 (261.5, 117.0)
 7. Joker80.np                00 0  00 00 1  00 ## 1= 11 11 1  1  =1 =  1  1  1    56%  13.5 (300.5, 113.0)
 8. Smirf BC-173g-X           0  00 0  =  00 0  0= ## 1  1  01 11 =  11 11 1  11   56%  13.5 (263.5,  90.8)
 9. Pair-o-Max 4.8S           00 0  0  00 0  0  00 0  ## 1= 01 1  1= 1  1  1  11   43%  10.0 (268.5,  58.0)
10. TSCPtest                  00 00 0  0  0  0  00 0  0= ## 1= 0  1  10 11 1  11   37%   9.0 (278.0,  47.8)
11. BigLion80 2.23x WB        00 0  0  00 0  00 0  10 10 0= ## 1  =1 1  =  0  1    33%   7.5 (286.5,  57.0)
12. Sjaak 524                 0  00 00 0  00 0  0  00 0  1  0  ## 1  10 =1 1= 1    30%   7.0 (276.0,  37.0)
13. Fairy-Max 4.8S            00 0  0  00 0  0  =0 =  0= 0  =0 0  ## 0  11 1  11   30%   7.0 (270.5,  40.3)
14. Max-Plus 4.8S             0  0  00 00 10 00 =  00 0  01 0  01 1  ## 0  11 0    27%   6.5 (293.5,  51.8)
15. ChessV_WinBoard           0  0  00 =0 0  =0 0  00 0  00 =  =0 00 1  ## 11 0    21%   5.0 (286.0,  36.5)
16. ArcBishop80 1.01 WB       0  00 0  0  00 0  0  0  0  0  1  0= 0  00 00 ## 1=   14%   3.0 (253.0,  14.8)
17. Heretic 0.3               0  0  00 00 0  00 0  00 00 00 0  0  00 1  1  0= ##   10%   2.5 (291.5,  13.0)