Galjoen version 0.29 available

[Werner Taelemans]

I've put a new version (v0.29) of my program on my website.

GUI: still no binaries for windows, as I ran into more problems than expected.
Probably next time.

Engine: it got a few bugfixes and it seems to run stable now. A win64 binary is included.
This version is a bit stronger than the previous version, about as strong as Jan Willem.

Here are some results:

Galjoen - Monarch: 51-114-35 (W-L-D)
Galjoen - Jan Willem: 84-87-29
Galjoen - Fairy Max: 102-59-39
Galjoen - Roce: 115-49-36

[Werner Taelemans]

Some changes are:

1) Gui: no change

2) Bugfixes: two bugfixes in the timemanager, one bugfix in uci-protocol handler concerning promotions.

3) Endgametable for KQKR. When looking at the games of v0.29 I saw that many points were lost in this endgame. Instead of adding extra knowledge, I decided to let the engine create it's own KQKR-egt at startup.

4) In bullet games it seems still a bit weaker than Jan Willem. In blitz probably a bit stronger: (26W-17L-7D)

[Ferdy]

Sample STS run.

Code: Select all

Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz
Physical Cores: 4, Logical Cores: 8
Engine: Galjoen_v0.30
Hash: 64, Threads: 1, time/pos: 0.199s
Number of positions in STS1-STS15_LAN_v2.epd: 1500
Max score = 1500 x 10 = 15000
Test duration: 00h:07m:22s
Expected time to finish: 00h:05m:43s
STS rating: 1591

  STS ID   STS1   STS2   STS3   STS4   STS5   STS6   STS7   STS8   STS9  STS10  STS11  STS12  STS13  STS14  STS15    ALL
  NumPos    100    100    100    100    100    100    100    100    100    100    100    100    100    100    100   1500
 BestCnt     44     20     33     29     62     45     15     20     34     65     18     28     38     30     13    494
   Score    492    281    401    358    671    557    249    304    396    709    304    352    457    395    252   6178
Score(%)   49.2   28.1   40.1   35.8   67.1   55.7   24.9   30.4   39.6   70.9   30.4   35.2   45.7   39.5   25.2   41.2
  Rating   1948   1008   1543   1351   2745   2237    866   1111   1520   2914   1111   1324   1792   1516    879   1591

Legend:
STS 01: Undermining
STS 02: Open Files and Diagonals
STS 03: Knight Outposts
STS 04: Square Vacancy
STS 05: Bishop vs Knight
STS 06: Re-Capturing
STS 07: Offer of Simplification
STS 08: Advancement of f/g/h Pawns
STS 09: Advancement of a/b/c Pawns
STS 10: Simplification
STS 11: Activity of the King
STS 12: Center Control
STS 13: Pawn Play in the Center
STS 14: Queens and Rooks to the 7th rank
STS 15: Avoid Pointless Exchange

The engine will output some log files. It would be better to create an option to enable/disable logging.

Played a sample game at 300s + 15s inc.
[pgn]
[Event "Computer Chess Game"]
[Site "INTEL-PC"]
[Date "2015.09.16"]
[Round "-"]
[White "F"]
[Black "Galjoen_v0.30"]
[Result "1-0"]
[TimeControl "300+15"]
[Annotator "1... -0.63"]

1. d4 e6 {-0.63/10 7} 2. Nf3 Nf6 {-0.19/12 7} 3. h3 Be7 {+0.07/11 7} 4.
Nbd2 d5 {+0.21/12 8} 5. g3 O-O {+0.36/12 8} 6. Bg2 Nc6 {+0.29/12 8} 7. O-O
a5 {+0.32/11 8} 8. Re1 a4 {+0.24/10 8} 9. c3 b6 {+0.26/10 8} 10. Qc2 Bd7
{+0.19/11 9} 11. e4 Ne8 {+0.04/10 9} 12. Nf1 dxe4 {+0.34/11 9} 13. Qxe4 Nd6
{+0.31/12 9} 14. Qd3 Ra7 {+0.16/10 9} 15. Bf4 f6 {+0.06/9 9} 16. g4 a3
{+0.59/9 9} 17. b3 f5 {+0.47/11 10} 18. Be5 fxg4 {+0.88/10 10} 19. hxg4 Ra5
{+0.79/11 10} 20. Rad1 Be8 {+0.72/10 10} 21. Ng3 Bg6 {+0.88/9 10} 22. Qe3
Qe8 {+0.74/9 10} 23. Bxd6 Bxd6 {+0.33/11 10} 24. Qxe6+ Qxe6 {+0.38/11 10}
25. Rxe6 Bf7 {+0.05/11 10} 26. Re2 Bd5 {+0.00/11 11} 27. Nf5 h5
{-0.43/12 11} 28. Nxd6 cxd6 {-0.58/12 11} 29. Nh2 Bxg2 {-0.62/12 11} 30.
Kxg2 Na7 {-0.79/13 11} 31. c4 hxg4 {-0.96/12 11} 32. Nxg4 Rg5 {-1.08/15 11}
33. Kg3 Rgf5 {-1.03/15 11} 34. Rd3 Rf4 {-1.21/13 11} 35. Re6 Nc8
{-1.68/13 11} 36. b4 R4f5 {-1.82/14 12} 37. Rxa3 Rh5 {-1.99/13 12} 38. Rf3
Rxf3+ {-2.38/13 12} 39. Kxf3 Kf7 {-2.14/13 12} 40. Re3 Ne7 {-2.08/13 12}
41. Ke2 Nc6 {-2.11/13 12} 42. b5 Nxd4+ {-1.25/15 12} 43. Kd3 Nf5
{-1.36/14 12} 44. Rf3 Kg8 {-1.31/13 12} 45. Ne3 Ne7 {-1.39/15 12} 46. a4 d5
{-1.74/14 12} 47. cxd5 Nxd5 {-2.31/14 12} 48. Nxd5 Rxd5+ {-2.32/17 12} 49.
Kc4 Rc5+ {-2.52/18 12} 50. Kb4 Rd5 {-2.62/17 12} 51. Rc3 Rd7 {-2.94/15 13}
52. Rc6 Rb7 {-3.59/17 13} 53. a5 bxa5+ {-4.19/19 13} 54. Kxa5 Ra7+
{-4.28/18 13} 55. Ra6 Rd7 {-4.54/17 13} 56. b6 Kh7 {-5.20/18 13} 57. Ra7
Rd5+ {-5.31/17 13} 58. Kb4 Rf5 {-4.22/16 13} 59. Ra2 Rf8 {-4.95/15 13} 60.
Kb5 g5 {-5.82/15 13} 61. b7 Kg6 {-6.70/16 13} 62. Ra8 Rxf2 {-5.20/14 13}
63. b8=Q Rf5+ {-18.73/11 13} 64. Kc4 Kh5 {-199.48/12 13} 65. Qh2+ Kg4
{-199.54/13 13} 66. Qg2+ Kf4 {-300.00/10 0.5} 67. Ra3 g4 {-300.00/9 0.4}
68. Rg3 Ke5 {-300.00/12 12} 69. Qd5+ Kf6 {-300.00/13 14} 70. Qd4+ Re5
{-300.00/11 1.2} 71. Re3 Kf7 {-300.00/6 0.2} 72. Qxe5 Kg6 {-300.00/6 0.2}
73. Qf4 Kh5 {-300.00/4 0.1} 74. Re6 Kh4 {-300.00/2 0.1} 75. Rh6#
{Xboard adjudication: Checkmate} 1-0[/pgn]

[Werner Taelemans]

Sample STS run.

Thanks! I downloaded your STS program. I will certainly use it when I'm working on my eval, during the next months.

The engine will output some log files. It would be better to create an option to enable/disable logging.

Yes, indeed. I've put a new v0.30 (same versionnumber) on my website with logfiles disabled now. In the next version I will add an extra option to do this.

[Ferdy]

I re-run sts, the one I posted before has wrong result due to buggy epd parser. The following is fine now.

Code: Select all

STS Rating v13.0
Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz
Number of cores: 8

Engine: Galjoen_v0.30
Hash: 128, Threads: 1, time/pos: 0.202s

Number of positions in STS1-STS15_LAN_v3.epd: 1500
Max score = 1500 x 10 = 15000
Test duration: 00h:07m:53s
Expected time to finish: 00h:05m:48s
STS rating: 1914

  STS ID   STS1   STS2   STS3   STS4   STS5   STS6   STS7   STS8   STS9  STS10  STS11  STS12  STS13  STS14  STS15    ALL
  NumPos    100    100    100    100    100    100    100    100    100    100    100    100    100    100    100   1500
 BestCnt     44     24     34     29     69     45     20     21     36     67     18     34     45     33     13    532
   Score    527    378    449    398    754    696    345    361    452    750    331    481    569    489    288   7268
Score(%)   52.7   37.8   44.9   39.8   75.4   69.6   34.5   36.1   45.2   75.0   33.1   48.1   56.9   48.9   28.8   48.5
  Rating   2104   1440   1756   1529   3114   2856   1293   1364   1770   3096   1231   1899   2291   1934   1039   1914

:: STS ID and Titles ::
STS 01: Undermining
STS 02: Open Files and Diagonals
STS 03: Knight Outposts
STS 04: Square Vacancy
STS 05: Bishop vs Knight
STS 06: Re-Capturing
STS 07: Offer of Simplification
STS 08: Advancement of f/g/h Pawns
STS 09: Advancement of a/b/c Pawns
STS 10: Simplification
STS 11: Activity of the King
STS 12: Center Control
STS 13: Pawn Play in the Center
STS 14: Queens and Rooks to the 7th rank
STS 15: Avoid Pointless Exchange

:: Top 5 STS with high result ::
1. STS 05, 75.4%, "Bishop vs Knight"
2. STS 10, 75.0%, "Simplification"
3. STS 06, 69.6%, "Re-Capturing"
4. STS 13, 56.9%, "Pawn Play in the Center"
5. STS 01, 52.7%, "Undermining"

:: Top 5 STS with low result ::
1. STS 15, 28.8%, "Avoid Pointless Exchange"
2. STS 11, 33.1%, "Activity of the King"
3. STS 07, 34.5%, "Offer of Simplification"
4. STS 08, 36.1%, "Advancement of f/g/h Pawns"
5. STS 02, 37.8%, "Open Files and Diagonals"

[Werner Taelemans]

These numbers look more realistic indeed. I was already worrying.