SPCC: Testruns of Xiphos 0.5.3 and Lc0 JH.T6.532 finished

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
pohl4711
Posts: 1042
Joined: Sat Sep 03, 2011 5:25 am
Location: Berlin, Germany
Contact:

SPCC: Testruns of Xiphos 0.5.3 and Lc0 JH.T6.532 finished

Post by pohl4711 » Mon May 06, 2019 7:00 am

Testruns of Xhiphos 0.5.3 and Lc0 Net JH.T6.532 finished

https://www.sp-cc.de


(Perhaps you have to clear your browsercache or reload the website)

sovaz1997
Posts: 218
Joined: Sun Nov 13, 2016 9:37 am

Re: SPCC: Testruns of Xiphos 0.5.3 and Lc0 JH.T6.532 finished

Post by sovaz1997 » Mon May 06, 2019 8:47 am

Yes, it is better not to use the left networks in the TCEC SiFi. Who knows what the result will be in the very LTC. There are good main networks, they may turn out to be even slightly weaker, but risk will not use them.
Zevra chess engine
Binary, source and description here: https://gitlab.com/sovaz1997/zevra2/tags

jdart
Posts: 3804
Joined: Fri Mar 10, 2006 4:23 am
Location: http://www.arasanchess.org

Re: SPCC: Testruns of Xiphos 0.5.3 and Lc0 JH.T6.532 finished

Post by jdart » Mon May 06, 2019 12:42 pm

I was looking at the following game and was examining this position:



where Fire played the winning fork Ne8+. I was puzzled why my engines were showing a draw score. But if you look back in the game, I believe the last irreversible move was 36. .. h5, so the 50-move counter has actually been passed at this point. I'm not sure then why this was a win rather than being adjudicated as a draw.


RubiChess
Posts: 100
Joined: Fri Mar 30, 2018 5:20 am

Re: SPCC: Testruns of Xiphos 0.5.3 and Lc0 JH.T6.532 finished

Post by RubiChess » Mon May 06, 2019 2:47 pm

Seems that 39. ... g6 is the last irreversible move. So it is "on the edge" of the 50 moves rule.
Can I get the complete pgn from this game so I can test myself? Seems that I cannot just copy it from the post as the starting fen isn't there.

./Andreas

pohl4711
Posts: 1042
Joined: Sat Sep 03, 2011 5:25 am
Location: Berlin, Germany
Contact:

Re: SPCC: Testruns of Xiphos 0.5.3 and Lc0 JH.T6.532 finished

Post by pohl4711 » Mon May 06, 2019 3:00 pm

The latest pawnmove was 39...g6. That is a small, known bug in the LittleBlitzerGUI. The 50-move draw is given at ply 101, not 100. So, this can lead to win/loss instead of a draw. But it happens rarely. Some years ago, I checked my former Lightspeed-ratinglist for games with wrong results, based on that bug - the Elos of the engines were distorted only 1-2 points. So in practical testing this is meaningless.

jdart
Posts: 3804
Joined: Fri Mar 10, 2006 4:23 am
Location: http://www.arasanchess.org

Re: SPCC: Testruns of Xiphos 0.5.3 and Lc0 JH.T6.532 finished

Post by jdart » Mon May 06, 2019 3:58 pm

RubiChess wrote:
Mon May 06, 2019 2:47 pm
Seems that 39. ... g6 is the last irreversible move. So it is "on the edge" of the 50 moves rule.
Can I get the complete pgn from this game so I can test myself? Seems that I cannot just copy it from the post as the starting fen isn't there.

./Andreas
This is the PGN:

[Event "?"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "Fire 7.1 popc"]
[Black "Xiphos 0.5.3 bmi2"]
[Result "1-0"]
[SetUp "1"]
[FEN "r1bqkbnr/pp3ppp/2np4/1N2p3/2P1P3/8/PP3PPP/RNBQKB1R b KQkq - 0 1"]
[PlyCount "185"]

1... Be7 2. N5c3 Nf6 3. Be2 Nd4 4. Be3 Nxe2 5. Qxe2 O-O 6. O-O h6 7. f3 Be6 8.
Rd1 a6 9. a4 Nh5 10. Nd5 Bxd5 11. Rxd5 Nf4 12. Bxf4 exf4 13. a5 Rc8 14. Nc3 Qc7
15. Ra4 Bf6 16. Rb4 Bd4+ 17. Kh1 Rfe8 18. Qd2 Be3 19. Qd1 Re5 20. Ra4 Bc5 21.
h3 Qe7 22. Qb3 Qh4 23. Ra1 Rc7 24. Qc2 Rce7 25. Rxe5 dxe5 26. Rd1 Re6 27. Qd2
Kh7 28. Nd5 Rg6 29. b4 Bd4 30. Qe1 Rg3 31. Qf1 Rg6 32. c5 Rc6 33. Nb6 Qe7 34.
Nc4 Rc7 35. Nb6 Rc6 36. Qc4 h5 37. Qe2 Qe6 38. Nc4 Rc7 39. Rb1 g6 40. Nd6 Re7
41. Qd2 Rc7 42. Kh2 Rd7 43. Qe2 Be3 44. Qc2 Bd4 45. Rd1 Kh6 46. Kh1 Kg5 47. Rc1
Rc7 48. Kh2 Be3 49. Rd1 Bd4 50. Qb1 Rd7 51. Rf1 Kh6 52. Rc1 Rc7 53. Rf1 Rd7 54.
Rd1 Kg5 55. Qc2 Kh6 56. Re1 Kg7 57. Rb1 Kh7 58. Rc1 Be3 59. Rd1 Bd4 60. Rb1 Kh6
61. Rc1 Kg5 62. Nc4 Kh6 63. Nd6 Kg5 64. Kh1 Rc7 65. Rd1 Rd7 66. Rd2 Kh6 67. Qb1
Kg7 68. Qd3 Kh7 69. Qc2 Kh6 70. Qd3 Kh7 71. Qa3 Rc7 72. Rd1 Kh6 73. Rd2 Kh7 74.
Rd1 Kh6 75. Qa4 Kg5 76. Rb1 Kh6 77. Kh2 Kg5 78. Rc1 Be3 79. Rd1 Bd4 80. Rd3 Kh6
81. Qd1 Re7 82. Qc2 Rd7 83. Rd1 Kh7 84. Rf1 Kh6 85. Rh1 Kg7 86. Rc1 Qe7 87. Qa4
Rd8 88. Kh1 Qc7 89. Ne8+ Kf8 90. Nxc7 h4 91. c6 Be3 92. Rb1 Rb8 93. cxb7 Rxb7
1-0

Post Reply