List of bugfree, opensource Linux and MacOSX engines

[Roland Chastain]

This is really great how quick you fix things.

Thank you. It was a very simple modification.

Code: Select all

  /*
  signal(SIGTERM, SIG_IGN);
  signal(SIGINT, SIG_IGN);
  */

But there is still a small issue:

Hippocampe did not win one single game against Quokka 2.1, but drew 61 times. All draws were 3-fold repetition.
It looks to me as hippocampe still has a small bug here in the detection of 3-fold repetition.

I will take another look into the code and see if I can do something.

[hgm]

Fortunately my engine has no weaknesses whatsoever.
On the other hand, it doesn't have any strengths either.

With regard to chess theory, it knows nothing, except for piece values and a general idea of where to put them. It's as dumb as a doorknob

That does count as a weakness. Or rather, a large number of weaknesses. If you lose to TSCP because it doesn't recognize the danger of TSCP's passers, then passers are your weakness. If you allow the opponent to surround your King unhindred as long as you don't see the mate, then King Safety is a weakness.

The nice thing is that when you cure a weakness, the engine automatically starts using it as a strength against other opponents that have that weakness.

[hgm]

@Oliver

Did you already have time to try KingSlayer? Latest source is at http://hgm.nubati.net/cgi-bin/gitweb.cg ... ;a=summary . It comes with a Linux Makefile.

And N.E.G. now should not play any illegal moves anymore. (It now resigns when it doesn't know what to play.) Latest source was posted here at http://talkchess.com/forum3/viewtopic.p ... 7&start=33 .

[mvanthoor]

Fortunately my engine has no weaknesses whatsoever.
On the other hand, it doesn't have any strengths either.

With regard to chess theory, it knows nothing, except for piece values and a general idea of where to put them. It's as dumb as a doorknob

That does count as a weakness. Or rather, a large number of weaknesses.

I know. My post was meant as a joke

The nice thing is that when you cure a weakness, the engine automatically starts using it as a strength against other opponents that have that weakness.

And that's the reason why self-play testing always overestimates your rating gain

[OliverBr]

@Oliver

Did you already have time to try KingSlayer? Latest source is at http://hgm.nubati.net/cgi-bin/gitweb.cg ... ;a=summary . It comes with a Linux Makefile.

First I have tried KingSlayer.. Very interesting engine!
unfortunately still there seems to be a promotion bug. (e2e1r is not a white promotion, is it?)

Code: Select all

Finished game 31 (KingSlayer 0.00 vs OliThink 3.0.7): 0-1 {White makes an illegal move: e2e1r}
Score of KingSlayer 0.00 vs OliThink 3.0.7: 5 - 2 - 0  [0.714] 7
Started game 55 of 2000 (KingSlayer 0.00 vs OliThink 3.0.7)
Finished game 24 (OliThink 3.0.7 vs KingSlayer 0.00): 1-0 {Black makes an illegal move: a7a8q}
Score of KingSlayer 0.00 vs OliThink 3.0.7: 5 - 3 - 0  [0.625] 8

The first game, it looks as KingSlayer wants to promote for the opponent, funny underpromotion anyway
[pgn]
[Event "?"]
[Site "?"]
[Date "2020.11.14"]
[Round "16"]
[White "KingSlayer 0.00"]
[Black "OliThink 3.0.7"]
[Result "0-1"]

1. Nc3 c5 2. e4 Nc6 3. Nge2 Qb6 4. Nf4 e6 5. Bc4 Nf6 6. O-O Bd6 7. d3 Bxf4
8. Bxf4 e5 9. Be3 d6 10. Kh1 Be6 11. b3 Qa5 12. Nd5 Bxd5 13. Bxd5 Nxd5 14. exd5
Nd4 15. Bd2 Qb5 16. g4 Rc8 17. c3 Qxd3 18. cxd4 cxd4 19. f4 Rc2 20. Bb4 Qe4+
21. Qf3 Qxf3+ 22. Rxf3 h5 23. Rh3 Kd7 24. fxe5 dxe5 25. Rxh5 Rxh5 26. gxh5 d3
27. Rd1 e4 28. a4 Rf2 29. d6 a5 30. Bxa5 Kxd6 31. Bb4+ Kd5 32. Bf8 g5 33. h6 Rf3
34. Be7 g4 35. h4 Rf5 36. Bg5 Rxg5 37. hxg5 e3 38. Rxd3+ Ke4 39. Rd7 e2 40. Re7+
Kd3 41. h7 Kd2 42. h8=Q e1=R+ 0-1
[/pgn]

And the second game:
[pgn]
[Event "?"]
[Site "?"]
[Date "2020.11.14"]
[Round "12"]
[White "OliThink 3.0.7"]
[Black "KingSlayer 0.00"]
[Result "1-0"]

1. d4 d5 2. Bf4 c6 3. e3 Qb6 4. b3 Nf6 5. Bd3 Bg4 6. f3 Be6 7. Nc3 Nh5 8. Nge2
Nd7 9. Na4 Qa5+ 10. Qd2 Qxd2+ 11. Kxd2 b5 12. Nc5 Nxc5 13. dxc5 Nxf4 14. exf4 g6
15. c3 Bg7 16. Rhe1 Bf6 17. g4 Bh4 18. Ng3 a5 19. a4 b4 20. cxb4 Bf6 21. bxa5
Bxa1 22. Rxa1 Rxa5 23. b4 Ra8 24. a5 h5 25. gxh5 gxh5 26. f5 Bd7 27. h4 e6
28. Re1 Ke7 29. fxe6 fxe6 30. f4 Raf8 31. Ke3 Rb8 32. Rb1 Rhg8 33. Nxh5 Be8
34. Be2 d4+ 35. Kxd4 Rd8+ 36. Ke3 Rd5 37. Bf3 Rf5 38. Rf1 Bxh5 39. Bxc6 e5
40. Ke4 Rxf4+ 41. Rxf4 exf4 42. a6 Rf8 43. a7 f3 44. Ke3 f2 45. a8=Q 1-0

[/pgn]

[hgm]

Umm, it seems like KingSlayer here just repeats the previous move, which is indicative for that it could not find a move itself (so that the old move is left in the gameMove buffer). It cannot be a coincidence that this happens both times after a promotion. I will check it out. Tomorrow.

[OliverBr]

@Oliver
And N.E.G. now should not play any illegal moves anymore. (It now resigns when it doesn't know what to play.) Latest source was posted here at http://talkchess.com/forum3/viewtopic.p ... 7&start=33 .

What an engine! It's amazing and perfectly stable. It will be added to the list.
There hasn't been not one resign in 2000 games.
ELO is probably about 600:

Code: Select all

   # PLAYER                  :  RATING  ERROR  POINTS  PLAYED   (%)     W    D     L  D(%)  CFS(%)
   1 Hippocampe 0.4.2.0.1    :     406     21  1820.0    2000  91.0  1701  238    61  11.9     100
   2 N.E.G. 1.3b             :       0   ----   180.0    2000   9.0    61  238  1701  11.9     ---

White advantage = 10.70 +/- 11.32
Draw rate (equal opponents) = 27.00 % +/- 1.81

May I proudly present N.E.G's first great win, being down two pawns:

[pgn]
[Event "?"]
[Site "?"]
[Date "2020.11.14"]
[Round "42"]
[White "N.E.G. 1.3b"]
[Black "Hippocampe 0.4.2.0.1"]
[Result "1-0"]

1. c4 e5 2. g3 d6 3. Nf3 Be7 4. Qa4+ Kf8 5. Nc3 Be6 6. e4 h5 7. Bd3 c6 8. b4 a6
9. h4 Qc7 10. a3 f6 11. Bb2 Qb6 12. Qa5 Qd8 13. Qxd8+ Bxd8 14. a4 Bg4 15. Nh2
Bc8 16. f3 c5 17. bxc5 dxc5 18. Nd5 Nh6 19. Bc3 g6 20. f4 exf4 21. gxf4 Nd7
22. Nf3 Kf7 23. e5 fxe5 24. Nxe5+ Nxe5 25. Be4 Bf5 26. Bxf5 gxf5 27. Bxe5 Re8
28. d3 Ng4 29. Rb1 Nxe5 30. fxe5 Ke6 31. Rxb7 Ba5+ 32. Kf1 Kxe5 33. Re7+ Kd4
34. Kg1 Rg8+ 35. Kf1 Kxd3 36. Re3+ Kd4 37. Kf2 Bd8 38. Rd1+ Kxc4 39. Rc3# 1-0
[/pgn]

PS: download repository with link and Makefile would be very nice.

Let's see if N.E.G can handle neuroGrape...
EDIT: neuroGrape has a timeout issue. This is quite annoying and makes is almost impossible to test.

[OliverBr]

For everybody: An update is coming soon. We will have some additions, but also 2 engines that will be removed:

-Claudia 0.5.1
-Fridolin 3.10

Unfortunately, those engines crashed during further tests :/

[OliverBr]

This one hurts: Wyldchess 1.5 crashed after having played over 100.000s games altogether. (Same happened to Murka3 64 once, btw).
So it's a very very rare bug. I hope the author Michael B. can fix this.

Code: Select all

Terminating process of engine WyldChess(34266)
Finished game 18982 (WyldChess vs OliThink 5.9.1): 0-1 {White's connection stalls}

[pgn]
[Event "?"]
[Site "?"]
[Date "2020.11.15"]
[Round "678"]
[White "WyldChess"]
[Black "OliThink 5.9.1"]
[Result "0-1"]

1. Nf3 d6 2. a3 c6 3. c4 Nf6 4. Nc3 e5 5. e4 c5 6. Be2 Nc6 7. d3 Be7 8. O-O Nd4
9. b4 O-O 10. bxc5 Nxf3+ 11. Bxf3 dxc5 12. Rb1 b6 13. Re1 Bd7 14. Nd5 Nxd5
15. exd5 Bd6 16. Qd2 f6 17. Bb2 Kh8 18. Qe2 b5 19. cxb5 Qa5 20. Be4 Rae8 21. a4
f5 22. Bf3 Qxa4 23. Bc3 Rb8 24. Ra1 Qxb5 25. Rxa7 Ra8 26. Rea1 Rxa7 27. Rxa7 Rd8
28. Qd1 Qb8 29. Qa1 Bb5 30. Be2 Be8 31. f4 c4 32. Kh1 Bc5 33. Ra2 Bd4 34. Bxd4
exd4 35. dxc4 Qxf4 36. Bd3 g6 37. Qf1 Qg5 38. Ra7 Qe3 39. Rb7 h5 40. Ra7 Qd2
41. Re7 Kg8 42. c5 Bf7 43. d6 Qc3 44. Rc7 Be6 45. Bb5 Ra8 46. Re7 Ra1 47. Rxe6
d3 48. Bxd3 Rxf1+ 49. Bxf1 Qc1 50. Kg1 Qxc5+ 51. Kh1 Qc1 52. Kg1 Kf7 53. d7 Qd2
54. Re1 Qxd7 55. Kh1 Qd5 56. h3 Kf6 57. Rc1 Qd2 58. Ra1 Kg5 59. Rb1 Kh4 60. Ba6
g5 61. Bb5 f4 62. Rg1 Kg3 63. Bc6 g4 64. Rf1 Qe2 65. Kg1 Qe6 66. Bf3 Qb6+
67. Kh1 Qa6 68. Kg1 gxh3 0-1
[/pgn]

see also https://github.com/Mk-Chan/WyldChess/issues/46

[Guenther]

Congratulations, it looks as you actually have fixed the castling issue. The linux executable you provide in your link is running now against OliPow 2.2.3 and there are no issues until now.

Good! Thank you for the confirmation.

3) Sometimes, I am having problems interrupting cutechess-cli when hippocampe is playing. It looks as it may having problems being interrupted (with "quit" or SIGNAL). It is quite important that engines interrupt their search when die GUI/manager wants them, too.

The "quit" command was already supported (if I am not mistaken) but, indeed, the signals were ignored. It should be fixed now. Please try this new version: hippocampe.zip

PS: Hippocampe's playing strength is really something. It is really the first and only stable engine in this "sector".

Yes.

For your other suggestions: yes, I will create a GitHub or GitLab repository. For the version number, let's say that previous versions were 0.4.2.0.1 beta. If the latest code is satisfying, it will be the official 0.4.2.0.1. :wink:

Would you mind to add my WIN compile of Hippocampe in your current or future download place too?
I don't think it would make sense to make it available at my download site, when you did the code changes
and already have other OS compiles. They shouldn't be splattered over various download sites.

It's here (fresh compiled this morning)
https://rwbc-chess.de/Downloads/Engines ... e_04201.7z