Andscacs - New version 0.921 with source

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

User avatar
cdani
Posts: 2204
Joined: Sat Jan 18, 2014 10:24 am
Location: Andorra

Re: Andscacs - New version 0.921 with source

Post by cdani »

elcabesa wrote:
cdani wrote: About the license, Elcabesa just told me that the WinGroupAffinity part of the code taken from Stockfish is incompatible with public domain license.
I'm not a GPL expert, so I really don't know all the implication of licenses of pieces of code that we can find in internet.
Sorry yes, I mean you suggested it :-)
tttony
Posts: 268
Joined: Sun Apr 24, 2011 12:33 am

Re: Andscacs - New version 0.921 with source

Post by tttony »

cdani wrote:
tttony wrote:Cool!

The source has no license so I guess it's under public domain?
About the license, Elcabesa just told me that the WinGroupAffinity part of the code taken from Stockfish is incompatible with public domain license. Maybe the best is to remove it from Andscacs. In fact nobody is using it that I know. What do you think about? Thanks.
Since is a strong program I think you should choose a license: https://choosealicense.com/

Anyway a code without license has copyright https://blog.codinghorror.com/pick-a-li ... y-license/
User avatar
cdani
Posts: 2204
Joined: Sat Jan 18, 2014 10:24 am
Location: Andorra

Re: Andscacs - New version 0.921 with source

Post by cdani »

pferd wrote:Thanks for this nice gift. :D

I tried to make it compile under clang under Linux and ran into a couple of these errors:

Code: Select all

analisis.cpp:491:3: error: cannot jump from this goto statement to its label
                goto fianalisi;
                ^
analisis.cpp:505:6: note: jump bypasses variable initialization
        int mirarfinsprofunditat = MaxProfunditat;
            ^
analisis.cpp:811:7: warning: unused variable 'alphaori' [-Wunused-variable]
                int alphaori = alpha;
                    ^
analisis.cpp:1505:5: warning: unused label 'jugadapotserdolenta' [-Wunused-label]
                                jugadapotserdolenta:
After some quick fixing it compiled but hit this error. (https://stackoverflow.com/a/32351440/6763189)

Code: Select all

gdb ./andscacs                                                      :(
GNU gdb (GDB) 8.0.1
...
bestmove f8e7 ponder e4d2

Program received signal SIGSEGV, Segmentation fault.
eval_KPK &#40;ss=0x705370 <ssbase>, ie=<optimized out>, color=<optimized out>) at finals.cpp&#58;485
485             if (!win&#41; &#123;
&#40;gdb&#41; 
The version compiled with gcc works without flaws.
I used gcc also to compile it for linux.
I will clean those warnings, thanks.
About the error at finals.cpp, at first sight I don't see what can cause the error, specially taking into account that works well with other compiles. I will think about it. Thanks.
carldaman
Posts: 2283
Joined: Sat Jun 02, 2012 2:13 am

Re: Andscacs - New version 0.921 with source

Post by carldaman »

Thanks a lot, Daniel. Good luck the rest of the way. :D

CL
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

MacOS exe

Post by MikeB »

cdani wrote:Hello. So here it is the version for the second stage of TCEC. Is really a minor version, with luck +5 elo.

Changes:
* Improved various pawn structure evaluation parameters.
* Move search improvements.
* Improvement of various imbalances.
* Improvements in some endgames.

http://www.andscacs.com/downloads/andscacs0921.zip

As promised the source code:
www.andscacs.com/downloads/andscacs921src.zip

Is a Visual Studio 2013 solution. Also it contains a makefile for Linux.
Michael will try to make a Mac version, thanks to him!
Daniel ,

What a great treat you gave to the CC community today. I love your engine a very unique style and very strong.

This run was just for fun and it was very quick tc (24 sec base)
K11.2.2 has really picked up its game with these micro time controls but I believe this may be a slightly lucky run. I checked the pgn and there was no losses on time by engine. Almost 500s point higher than Fruit 2.2.1 in this run.

Code: Select all

300 of 756 games completed...
time control&#58; 24+.40 seconds
Date&#58; 11/04/17 &#58; 19&#58;51&#58;38
300 game&#40;s&#41; loaded

Rank Name                  Rating   &#916;     +    -     #     &#931;    &#931;%     W    L    D   W%    =%   OppR 
---------------------------------------------------------------------------------------------------------
   1 Komodo 11.2.2 64-bit   3496   0.0   72   72   151  143.0  94.7  136    1   14  90.1   9.3  2968 
   2 Andscacs 0.921         3226 270.1   54   54   149   98.0  65.8   85   38   26  57.0  17.4  3059 
   3 Crafty-25.2            2941 285.7   53   53   150   46.5  31.0   36   93   21  24.0  14.0  3155 
   4 Fruit 2.2.1            2737 204.2   65   65   150   12.5   8.3    4  129   17   2.7  11.3  3221 
---------------------------------------------------------------------------------------------------------
  &#931; = total score, 1 point for win, 1/2 point for draw
300 game&#40;s&#41; loaded
[pgn][Event "Computer chess game"]
[Site "Mac-Pro"]
[Date "Sat Nov 4 19:35:37 2017"]
[Round "7"]
[White "Komodo 11.2.2 64-bit"]
[Black "Andscacs 0.921"]
[Result "0-1"]
[TimeControl "2400+40"]
[FEN "r1bqkb1r/pp1ppppp/2n2n2/2p5/5P2/5NP1/PPPPP2P/RNBQKB1R w KQkq - 0 1"]
[SetUp "1"]

1. c4 g6 2. Nc3 Bg7 3. e4 d6 4. d3 Bg4 5. Bg2 O-O 6. Be3 Nd4 7. O-O a6 8.
Bxd4 cxd4 9. Ne2 Nd7 10. Qd2 e5 11. h3 Bxf3 12. Rxf3 b5 13. Rc1 Qb6 14.
Rff1 Rfc8 15. cxb5 Rxc1 16. Rxc1 axb5 17. b4 Bh6 18. Rc2 Qa7 19. Nc1 Nf8
20. Bf3 exf4 21. Ne2 Nd7 22. Kg2 Ne5 23. gxf4 d5 24. Rc5 Nxf3 25. Kxf3
dxe4+ 26. dxe4 Bf8 27. Rd5 Qa3+ 28. Kg2 Bxb4 29. Qxd4 Qxa2 30. Qd1 Be7 31.
Qd2 b4 32. Qxa2 Rxa2 33. Kf2 Bh4+ 34. Ke3 Ra3+ 35. Kd2 Rxh3 36. Rb5 Rh1 37.
Ke3 Rb1 38. Nd4 Be7 39. f5 Rc1 40. e5 Bc5 41. f6 h5 42. Kd2 Rc3 43. Rb8+
Kh7 44. e6 Bxd4 45. e7 Re3 46. Rxb4 Bxf6 47. Kxe3 Bxe7 48. Rb7 Bc5+ 49. Kf4
Kg7 50. Kg5 Be3+ 51. Kh4 Kf6 52. Rb3 Bf2+ 53. Kh3 Kg7 54. Kg2 Bh4 55. Kf3
g5 56. Rb8 Kg6 57. Rg8+ Kf5 58. Rh8 g4+ 59. Kg2 Kg6 60. Rg8+ Kh7 61. Rb8
Kg7 62. Rb5 Kg6 63. Rb8 Bf6 64. Kg3 Bg7 65. Re8 Kg5 66. Kf2 h4 67. Rg8 Kg6
68. Ke3 h3 69. Rd8 h2 70. Rd1 Bd4+ 71. Ke2 Bg1 72. Rd6+ Kg5 73. Rd5+ f5 74.
Rd8 h1=Q 75. Rg8+ Kf4 76. Rc8 Bc5 77. Rd8 Qf3+ 78. Ke1 Bb4+ 79. Rd2 Qg2 80.
Kd1 Qxd2#
{White Checkmated} 0-1[/pgn]

nice solid game...

https://www.dropbox.com/s/wbxwoa3g3idnl ... S.zip?dl=1
User avatar
cdani
Posts: 2204
Joined: Sat Jan 18, 2014 10:24 am
Location: Andorra

Re: MacOS exe

Post by cdani »

Thanks also to you for the tournament and the Mac compile! I added it to the zip file.
Damir
Posts: 2801
Joined: Mon Feb 11, 2008 3:53 pm
Location: Denmark
Full name: Damir Desevac

Re: Andscacs - New version 0.921 with source

Post by Damir »

cdani wrote:
Damir wrote:Where do I have to change the things you just mentioned, so I could get Andscacs parameters ? :)
In the es.cpp file.
The problem is that if I add a few parameters it will not make much sense. And adding a lot of them is just a lot of work :-) I'm sorry.
Ouch looks like I will have to go into Andscacs source and make these changes. The es.cpp file you just mentioned is there. :? :?
Too bad I am complete patzer regarding this. :(
thunderstruck
Posts: 50
Joined: Tue Jul 19, 2016 10:19 am

Re: Andscacs - New version 0.921 with source

Post by thunderstruck »

Hi Daniel,

Version 0.921 doesn't work with Cutechess GUI in Windows and Linux.
Doesn't start(segfault).
Even when I start it from a teminal it crashes with the UCI command that sends Cutechess GUI ; go wtime 300000 btime 300000 movestogo 40
0.92 did work in Cutechess GUI in Windows and Linux.

Kind regards,

Antonio
Volker Pittlik
Posts: 619
Joined: Wed Mar 08, 2006 9:10 pm
Location: Murten / Morat, Switzerland
Full name: Volker Pittlik

Re: Andscacs - New version 0.921 with source

Post by Volker Pittlik »

thunderstruck wrote:Hi Daniel,

Version 0.921 doesn't work with Cutechess GUI in Windows and Linux.
Doesn't start(segfault).
Even when I start it from a teminal it crashes with the UCI command that sends Cutechess GUI ; go wtime 300000 btime 300000 movestogo 40
0.92 did work in Cutechess GUI in Windows and Linux.

Kind regards,

Antonio
I can confirm that. However, version 0.921 does work in xboard.

Code: Select all

New game &#40;0&#41;&#58; Andscacs 0.92-Andscacs 0.921 &#40;w&#41;
21634 >first &#58; computer
21634 >first &#58; name Andscacs 0.921
21634 >second&#58; computer
21634 >second&#58; name Andscacs 0.92
21635 >first &#58; time 1000
21635 >first &#58; otim 1000
book hit = &#40;NULL&#41;
21635 >first &#58; go
nps&#58; w=-1, b=-1
21965 <first &#58; move d2d4
21965 >second&#58; time 1000
21966 >second&#58; otim 977
book hit = &#40;NULL&#41;
21966 >second&#58; usermove 21966 >second&#58; d2d4
21966 >second&#58; go
22196 <second&#58; move d7d5
22196 >first &#58; time 977
22196 >first &#58; otim 986
book hit = &#40;NULL&#41;
22196 >first &#58; usermove 22196 >first &#58; d7d5
22530 <first &#58; move c1f4
22530 >second&#58; time 986
22530 >second&#58; otim 953
Volker
User avatar
cdani
Posts: 2204
Joined: Sat Jan 18, 2014 10:24 am
Location: Andorra

Re: Andscacs - New version 0.921 with source

Post by cdani »

thunderstruck wrote:Hi Daniel,

Version 0.921 doesn't work with Cutechess GUI in Windows and Linux.
Doesn't start(segfault).
Even when I start it from a teminal it crashes with the UCI command that sends Cutechess GUI ; go wtime 300000 btime 300000 movestogo 40
0.92 did work in Cutechess GUI in Windows and Linux.

Kind regards,

Antonio
Solved! There was a bug with the movestogo keyword. Thanks for reporting!
If anyone will play a tournament with moves per time, should update the version.
http://www.andscacs.com/downloads/andscacs0921.zip

I also updated the source file.