Dragon v4.7

[MichelG]

It has been a while, but Dragon version 4.7.0 is now available:

• New look, new icons
• Removed all references to paid version/registration. It is all free now.
• Solved crash if you used 32 or more processor cores (engine is unchanged otherwise)
• You can now download the 6 piece endings from view->endgame databases

The old site will be down per 1/7/2026 and now redirects to:
https://www.mdgsoftware.nl/draughts/index.html

[Krzysztof Grzelak]

Thank you for the new version of the program. Unfortunately, there are errors in the program.

1.The program is recording the moves incorrectly during the game.
I noticed that when I make a move with white, the program does not save it. And I will save it and the program will make a move with black.

2.Unable to generate the databases.
A black window appears and nothing happens.

Krzysztof

[pontel]

It has been a while, but Dragon version 4.7.0 is now available:

• New look, new icons
• Removed all references to paid version/registration. It is all free now.
• Solved crash if you used 32 or more processor cores (engine is unchanged otherwise)
• You can now download the 6 piece endings from view->endgame databases

The old site will be down per 1/7/2026 and now redirects to:
https://www.mdgsoftware.nl/draughts/index.html

Hello Michel,

The DXP mode isn't working, and the 1-ply balanced and 2-ply balanced options are unavailable.

Sincerely.

[Krzysztof Grzelak]

Hello Michel,

I have a request. Can you make a new version of the program free of errors? Unfortunately, the program contains serious errors.

Krzysztof

[pontel]

Hello Michel,

I have a request. Can you make a new version of the program free of errors? Unfortunately, the program contains serious errors.

Krzysztof

Hello Krzysztof Grzelak!

I think I managed to work around the problems temporarily until the version is updated, at least the DXP errors I found.

The workaround is to copy and overwrite the files from the current version into the previous version, except for the DragonDraughts interface executable. This way, you can use the v3.7.0 engine with the interface from the previous v3.6.2 version.

Best regards!

[Krzysztof Grzelak]

It doesn't quite work that well. I copied the database and unfortunately the program does not see this database (version 4.6.2). I also noticed there is a problem (in my opinion) with the use of multithreading.

[pontel]

It doesn't quite work that well. I copied the database and unfortunately the program does not see this database (version 4.6.2). I also noticed there is a problem (in my opinion) with the use of multithreading.

Hello Krzysztof Grzelak,

In my tests here, the 6-piece endgame database downloaded directly through Dragon seems to be working normally.

See the game below: on move 61, Dragon had an already clearly winning 6-against-2 position, but chose to simplify by sacrificing two pieces, reaching a winning 4-against-2 position already contained in the endgame database, even though the win then takes more than 20 moves.

Note: I do not consider this an error. I believe this would probably not happen if an endgame database with more pieces were available.

As for the problems with multithreading, I have seen similar issues in several versions when using AMD processors. My machine here uses an Intel processor, although an old one, and so far I have not had any problems.

Best regards...

__

[White "Moby Dam (May 15 2026)"]
[Black "Dragon Pro 4.7.0"]
[Event ""]
[Site ""]
[Round "42"]
[Date "2026.5.30"]
[Result "0-2"]
[GameType "20"]
[Setup "1"]
[FEN "W:W46,47,48,49,50,41,42,43,44,45,36,37,38,39,40,31,33,34,35,27:B23,16,17,19,20,11,12,13,14,15,06,07,08,09,10,01,02,03,04,05."]

1. 34-30 20-24 2. 37-32 12-18 3. 33-28 14-20 4. 39-33 07-12 5. 27-22 18x27
6. 31x22 01-07 7. 44-39 10-14 8. 49-44 16-21 9. 41-37 12-18 10. 37-31 18x27
11. 31x22 08-12 12. 22-18 13x22 13. 32-27 22x31 14. 36x27x16 23x32 15.
38x27 09-13 16. 30-25 12-18 17. 46-41 04-09 18. 41-37 02-08 19. 42-38 08-12
20. 37-31 03-08 21. 47-41 17-21 22. 31-26 21x32 23. 38x27 11-17 24. 27-21
18-23 25. 48-42 13-18 26. 33-28 23x32 27. 40-34 18-23 28. 34-29 23x34 29.
39x30 17-22 30. 42-37 32-38 31. 43x32 22-27 32. 32-28 27-31 33. 21-17 31x42
34. 41-37 42x31 35. 26x37 12x21 36. 16x27 09-13 37. 44-39 06-11 38. 45-40
13-18 39. 40-34 18-23 40. 28-22 07-12 41. 37-31 12-17 42. 31-26 17x28 43.
39-33 28x39 44. 34x43 11-16 45. 27-22 24-29 46. 22-17 20-24 47. 43-39 29-34
48. 39-33 16-21 49. 30x39 21x12 50. 50-45 12-17 51. 45-40 08-13 52. 40-34
13-18 53. 34-30 23-29 54. 33-28 05-10 55. 26-21 17x26 56. 28-23 19x28 57.
30x19 14x23 58. 39-34 29x40 59. 35x44 23-29 60. 44-40 29-34 61. 40x29 18-23
62. 29x18 26-31 63. 18-12 31-37 64. 12-07 37-41 65. 07-01 41-47 66. 01-07
10-14 67. 07-34 47-24 68. 34-43 28-33 69. 43-32 24-19 70. 32-27 33-39 71.
25-20 14x25 72. 27-21 39-44 73. 21-16 44-50 74. 16-02 19-14 75. 02-24 50-28
76. 24-29 25-30 77. 29-18 28-06 78. 18-29 30-35 79. 29-42 35-40 80. 42-38
40-45 81. 38-24 45-50 82. 24-42 06-39 83. 42-31 15-20 84. 31-27 20-25 85.
27-16 25-30 86. 16-27 14-37 87. 27-18 30-35 88. 18-12 50-45 89. 12-03 35-40
90. 03-08 40-44 91. 08-02 44-50 92. 02-16 37-23 93. 16-38 23-29 94. 38x24
45-29 95. 24x33x44 50x39 0-2

[Krzysztof Grzelak]

Hello Krzysztof Grzelak,

In my tests here, the 6-piece endgame database downloaded directly through Dragon seems to be working normally.

See the game below: on move 61, Dragon had an already clearly winning 6-against-2 position, but chose to simplify by sacrificing two pieces, reaching a winning 4-against-2 position already contained in the endgame database, even though the win then takes more than 20 moves.

Note: I do not consider this an error. I believe this would probably not happen if an endgame database with more pieces were available.

As for the problems with multithreading, I have seen similar issues in several versions when using AMD processors. My machine here uses an Intel processor, although an old one, and so far I have not had any problems.

Best regards...

I added the old ending databases from version 4.6.2 - and unfortunately they don't work properly. As for multithreading, it works, but I don't see Dragon increasing searches on 32 threads.

[Krzysztof Grzelak]

Hello Krzysztof Grzelak,

In my tests here, the 6-piece endgame database downloaded directly through Dragon seems to be working normally.

See the game below: on move 61, Dragon had an already clearly winning 6-against-2 position, but chose to simplify by sacrificing two pieces, reaching a winning 4-against-2 position already contained in the endgame database, even though the win then takes more than 20 moves.

Note: I do not consider this an error. I believe this would probably not happen if an endgame database with more pieces were available.

As for the problems with multithreading, I have seen similar issues in several versions when using AMD processors. My machine here uses an Intel processor, although an old one, and so far I have not had any problems.

Best regards...

Generate an ending database larger than 6 pieces.