asmFish online again

[CositasBuenas]

https://github.com/lantonov/asmFish

[Damir]

It has not been updated for over a month.. which means it is not online..
Someone put the old files there that's all. The engine is no longer beeing developed...

[CositasBuenas]

Not true. I see recent compiles dating 2016/07/12.

[Guenther]

It has not been updated for over a month.. which means it is not online..
Someone put the old files there that's all. The engine is no longer beeing developed...

That's wrong. All important SF patches since 05/11 are comitted to new asmFish and pedantFish compilations (date today).
There are 2 contributors. One of them is Lyudmil Antonov, who posts here.

[Damir]

Last time I looked there were not any new patches.. It must have been updated just recently. I have known that asmFish was online for quite some time, but there weren't any patches available..

[Damir]

It also says that the last update from Mohammed Li was:

The latest version of asmFish by Mohammed Li (04.11.2016) on the top of the page. Mohammed is not working on the engine himself, that some other people have decided to pick up on his work, that is something else, but he himself is not involved anymore...

[mbabigian]

I noticed that if I give asmFish the following command line:

bench 1024 4 18 default depth

the program does not terminate automatically when finished. This prevents running tests using scripts, BuildTester (software.farseer.org), etc. Is there a way to convince asmFish to terminate after the bench command has completed?

Thanks.

[Jesse Gersenson]

Is there a way to convince asmFish to terminate after the bench command has completed?

One hack would be to use the linux command 'timeout', which kills a process if it runs longer than X seconds.

timeout 8 ./command

[Nordlandia]

[pgn][Event "60+30"]
[Site "DESKTOP-DD1B2QU"]
[Date "2016.12.11"]
[Round "1"]
[White "Houdini 5"]
[Black "SF"]
[Result "1/2-1/2"]
[BlackElo "3200"]
[ECO "A36"]
[Opening "English"]
[Time "08:31:16"]
[Variation "Symmetrical, Botvinnik System, 5...d6, 8.d3 a6"]
[WhiteElo "3200"]
[TimeControl "3600+30"]
[Termination "normal"]
[PlyCount "57"]
[WhiteType "program"]
[BlackType "program"]

1. c4 g6 2. Nc3 c5 3. g3 Bg7 4. Bg2 Nc6 5. e4 d6 6. d3 Nf6 7. Nge2 O-O 8.
O-O a6 9. h3 Rb8 10. f4 Ne8 11. Be3 Nd4 12. Qd2 b5 13. Rae1 b4 14. Nd1 b3
15. Bxd4 bxa2 16. Bxg7 a1=Q 17. Bc3 Qa4 18. f5 {-0.23/26 322} f6 19. Qh6
{-0.15/30 298} Rf7 20. Ne3 {-0.15/30 0} Nc7 21. fxg6 {-0.15/31 0} hxg6 22.
e5 {-0.15/31 0} Rh7 23. Qxg6+ {-0.15/32 251} Rg7 24. Qh6 {-0.15/33 66} dxe5
25. Nd5 {-0.15/33 0} Rh7 26. Qg6+ {-0.15/33 0} Rg7 27. Qh6 {-0.15/36 55}
Rh7 28. Qg6+ {-0.15/36 126} Rg7 29. Qh6 {-0.15/37 338 3-fold repetition}
1/2-1/2[/pgn]

Lichess Analysis: https://nb.lichess.org/AKnRFkWv

[pgn][Event "60+30"]
[Site "DESKTOP-DD1B2QU"]
[Date "2016.12.11"]
[Round "2"]
[White "Asmfish 07.12.16"]
[Black "Houdini 5 Pro"]
[Result "1-0"]
[BlackElo "3200"]
[ECO "A36"]
[Opening "English"]
[Time "09:25:59"]
[Variation "Symmetrical, Botvinnik System, 5...d6, 8.d3 a6"]
[WhiteElo "3200"]
[TimeControl "3600+30"]
[Termination "normal"]
[PlyCount "121"]
[WhiteType "program"]
[BlackType "program"]

1. c4 g6 2. Nc3 c5 3. g3 Bg7 4. Bg2 Nc6 5. e4 d6 6. d3 Nf6 7. Nge2 O-O 8.
O-O a6 9. h3 Rb8 10. f4 Ne8 11. Be3 Nd4 12. Qd2 b5 13. Rae1 b4 14. Nd1 b3
15. Bxd4 bxa2 16. Bxg7 a1=Q 17. Bc3 Qa4 18. f5 f6 {-0.15/27 325} 19. Qh6
Rf7 {-0.15/29 178} 20. Ne3 Nc7 {-0.15/27 26} 21. e5 dxe5 {-0.06/27 176} 22.
fxg6 Rg7 {-0.15/29 206} 23. Bxe5 Rxg6 {-0.15/28 35} 24. Bxc7 Qxc7 {-0.06/31
268} 25. Qh4 Rxb2 {-0.15/30 208} 26. Nd5 Qd6 {-0.15/30 0} 27. Nef4 Rxg2+
{-0.15/32 0} 28. Nxg2 Rg7 {-0.13/30 0} 29. Ngf4 Qac6 {-0.13/33 161} 30. Re2
Rf7 {-0.13/33 241} 31. Rfe1 Qcd7 {-0.13/34 0} 32. g4 e5 {-0.65/28 308} 33.
Nh5 Qd8 {-0.75/24 26} 34. Rf1 f5 {-1.23/29 571} 35. Qe1 Qe8 {-0.91/27 90}
36. Rb2 Qed8 {-1.01/28 0} 37. Rff2 Bb7 {-0.51/30 584} 38. Rb6 Bc6 {-1.39/27
334} 39. Ng3 Qh4 {-0.92/27 159} 40. Nxf5 Rxf5 {-0.84/29 0} 41. gxf5 Kf7
{-0.84/33 0} 42. Qb1 e4 {-0.75/25 79} 43. Rxc6 Qxc6 {-1.02/30 126} 44. dxe4
Qxh3 {-1.54/27 159} 45. Rg2 Qh4 {-1.86/28 90} 46. e5 Qd4+ {-2.07/28 237}
47. Kh1 Qxe5 {-2.54/28 199} 48. f6 Qh5+ {-2.92/24 97} 49. Rh2 Qf3+
{-3.07/21 7} 50. Kg1 Kf8 {-3.12/25 0} 51. Rg2 Qe8 {-3.42/26 62} 52. f7
Qexf7 {-3.56/25 70} 53. Qb8+ Qe8 {-3.57/22 0} 54. Rg8+ Kxg8 {-3.94/25 47}
55. Qxe8+ Kg7 {-4.28/24 25} 56. Qe7+ Kg6 {-5.10/22 42} 57. Qxc5 Qg3+
{-6.08/24 36} 58. Kf1 Qe5 {-6.50/21 38} 59. Qg1+ Kf5 {-90.00/29 83} 60.
Qf2+ Ke6 {-149.99/39 37} 61. Qb6 {Black resigns} 1-0
[/pgn]

Lichess Analysis: https://nb.lichess.org/gGWgNNRC#0

[lantonov]

I noticed that if I give asmFish the following command line:

bench 1024 4 18 default depth

the program does not terminate automatically when finished. This prevents running tests using scripts, BuildTester (software.farseer.org), etc. Is there a way to convince asmFish to terminate after the bench command has completed?

Thanks.

This problem is fixed in today's (2016-12-17) version of asmFish