Glaurung 2.0.1 Windows build available

[Thomas Cutter]

Hello Tord

FYI, I compiled it myself (MS Visual Studio 2005) and couldn't run it under the Fritz 10 GUI, but it worked under Fritz 9 GUI.

Greetings Thomas

[Werner]

Hi Thomas,
I think you missed the solution:


see here http://www.talkchess.com/forum/viewtopi ... 27&t=17979

You have to delete the init string before compiling!

[LJC]

Hmmm - I can't get this compile working on my Athlons.

When I try to create a new uci engine, Fritz 10 crashes.

same thing happened to me 2 :.(

[Graham Banks]

Hmmm - I can't get this compile working on my Athlons.

When I try to create a new uci engine, Fritz 10 crashes.

same thing happened to me 2 :.(

It's fixed now.

[LJC]

Hmmm - I can't get this compile working on my Athlons.

When I try to create a new uci engine, Fritz 10 crashes.

same thing happened to me 2 :.(

It's fixed now.

I meant the same thing happens to me when I try to load it in DF 10 GUI, "engine can not load"

[Jim Ablett]

Hi Lonnie,

I meant the same thing happens to me when I try to load it in DF 10 GUI, "engine can not load"

I'm to blame for the confusion. I put out so many different
builds trying to solve the Chessbase crash bug, now one one knows
which is which.

If you download the last/latest builds on my homepage you should
find it will load ok now in Fritz/Chessbase.

http://homepages.tesco.net/henry.ablett/jims.html

regards,
Jim.

[Eelco de Groot]

This was a nice win from Glaurung Mjolnir against HIARCS 11.2, not very spectacular maybe, but good endgame technique from Glaurung and a few surprising moves for a Blitz game. It was the only two game match that Glaurung managed to win but the overall results were still better than against the Belka monster.

Matchresults to follow later.

[Event "8 Minutes/Game + 2 Seconds/Move"]
[Site "Engine Match"]
[Date "2007.11.26"]
[Round "1"]
[White "HIARCS 11.2 SP Default"]
[Black "Glaurung 2.0.1 Mjolnir 001b"]
[Result "0-1"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Bc4
Nbd7 7. Bg5 e6 8. O-O Qb6 9. Bb3 Be7 10. Re1 O-O 11. a3 Nc5
12. Qf3 Bd7 13. Be3 Rfe8 14. Red1 Rac8 15. Rac1 a5 16. a4
Qa6 17. Kh1 Nxb3 18. cxb3 Red8 19. Kg1 b6 20. Qg3 Qb7
21. f3

{[d]2rr2k1/1q1bbppp/1p1ppn2/p7/P2NP3/1PN1BPQ1/1P4PP/2RR2K1 b - -

Only after 23 seconds does HIARCS judge that Black has at least equality here, the eval very slowly starts climbing over zero}

21... e5 22. Nf5 Bxf5 23. exf5 d5

{[d]2rr2k1/1q2bppp/1p3n2/p2ppP2/P7/1PN1BPQ1/1P4PP/2RR2K1 w - -
Glaurung has sacrificed the advantage of the Bishop pair but in exchange has gained control over the center. Or at least so it seems. Nevertheless, 23... d5 is not correct I think because White could have at least equalized if it had dared to take the pawn on e5 with its Queen, in spite of the threatening forkmoves. But the position is difficult to judge, and Glaurung changes its mind a few times on the position. Not on Qxe5 as best move but it needs 36 seconds (50% CPU) to find it. It does not consider 24. Bg5 what Hiarcs played, only 24. Nb5 at first:

2rr2k1/1q2bppp/1p3n2/p2ppP2/P7/1PN1BPQ1/1P4PP/2RR2K1 w - -

Engine: Glaurung 2.0.1 default (64 MB)
by Tord Romstad

16.01 0:24 -0.84 24.Nb5 Qb8 25.Bh6 Nh5 26.Qg4 g6
27.Bd2 Nf6 28.Qg5 Re8 29.fxg6 hxg6
30.b4 axb4 31.Rxc8 Rxc8 32.Bxb4 Bxb4
33.Qxf6 (8.456.083) 351

16.08 0:36 +0.23 24.Qxe5 Bd6 25.Qd4 Bc5 26.Qd3 d4
27.Ne4 dxe3 28.Qxd8+ Rxd8 29.Rxd8+ Bf8
30.Nxf6+ gxf6 31.Rcc8 Kg7 32.Rxf8 Qd5
33.Rg8+ Kh6 34.Rgd8 Qxf5 (12.905.577) 352

17.01 1:02 +0.23 24.Qxe5 Bd6 25.Qd4 Bc5 26.Qd3 d4
27.Ne4 dxe3 28.Qxd8+ Rxd8 29.Rxd8+ Bf8
30.Nxf6+ gxf6 31.Rcc8 Kg7 32.Rxf8 Qd5
33.Rg8+ Kh6 34.Rgd8 Qxf5 (22.408.868) 356

18.01 2:04 +0.17 24.Qxe5 Bd6 25.Qd4 Bc5 26.Qd3 d4
27.Ne4 dxe3 28.Qxd8+ Rxd8 29.Rxd8+ Bf8
30.Nxf6+ gxf6 31.Rcc8 Kg7 32.Rxf8 Qd5
33.Rg8+ Kh6 34.Rgd8 Qxf5 35.Rc1 (44.990.858) 360

19.01 3:28 -0.72 24.Qxe5 Bd6 25.Qd4 Bc5 26.Qd3 d4
27.Ne4 dxe3 28.Qxd8+ Rxd8 29.Rxd8+ Bf8
30.Nxf6+ gxf6 31.Rcc8 Kg7 32.Rxf8 Qd7
33.Rg8+ Kh6 34.Rc1 e2 35.Re1 Qd4+
36.Kh1 Qxb2 (75.936.398) 364

20.01 21:04 +0.52 24.Qxe5 Bd6 25.Qd4 Bc5 26.Qd3 Qd7
27.Bd4 Qc7 28.Qd2 Qd7 29.Bxc5 bxc5
30.Nb5 Qxf5 31.Qxa5 Qf4 32.Na7 Qe3+
33.Kh1 Rb8 34.Rxc5 Qxb3 (458.867.509) 363

Engine: HIARCS 11.2 SP Default (64 MB)
by Mark John Uniacke

16/39 4:13 +0.51 24.Qxe5 Bd6 25.Qd4 Qe7 26.Qd2 Qe5
27.f4 Qxf5 28.Nxd5 Nxd5 29.Rxc8 Qxc8
30.Qxd5 Be7 31.Qf3 Rxd1+ 32.Qxd1 Qe6
33.Bd4 Qd6 34.f5 (30.624.763) 120

17/41 11:15 +0.62 24.Qxe5 Bc5 25.Bd4 Re8 26.Qf4 Qd7
27.Kh1 Bd6 28.Qh4 Qxf5 29.Bxf6 gxf6
30.Rd4 (81.657.400) 120

18/45 30:02 +0.71 24.Qxe5 Bd6 25.Qd4 Qe7 26.Qd2 Qe5
27.f4 Qxf5 28.Nxd5 Nxd5 29.Rxc8 Qxc8
30.Qxd5 Be7 31.Qf3 Rxd1+ 32.Qxd1 Qe6
33.Qd3 g6 34.Bd4 (208.095.480) 115 }

24. Bg5

{[d]2rr2k1/1q2bppp/1p3n2/p2ppPB1/P7/1PN2PQ1/1P4PP/2RR2K1 b - -

The threat passes. No Qxe5. Maybe this Bg5 could work if White presses for a King Attack but it leaves the center to Black. How do you counter a threat through the center? Should you try to hold the center or press for an attack on the wings?}

24... Nh5 25. Qg4 Bxg5
26. Qxg5 Nf6 27. Nb5 d4 28. Na3 Rxc1 29. Rxc1 h6 30. Qg3 e4
31. fxe4 Nxe4 32. Qe5 d3 33. Nc4 Qc6 34. Re1 d2 35. Rd1 Kh7
36. f6 g6 37. h3 Rd3 38. Qe7 Qd5 39. Kh2 Rxb3 40. Ne5 Nd6
41. Nxf7 Nxf7 42. Rxd2 Qe5+ 43. Qxe5 Nxe5 44. Re2 Nc4
45. Re8 Nd6 46. Rd8 Nf7 47. Rf8 Ng5 48. h4 Ne4 49. f7 Kg7

{[d]5R2/5Pk1/1p4pp/p7/P3n2P/1r6/1P4PK/8 w - -

The f7-pawn will not reach its destination, the King is stronger but Black loses his extra piece in the process. But because pawn b2 is also lost Black is still winning}

50. Re8 Kxf7 51. Rxe4 Rxb2 52. Rc4 Rb4 53. Rc7+ Ke6
54. Rc6+ Kf5 55. h5 gxh5 56. Rxh6 Kg5 57. Rd6 h4 58. Kh3
Kf5 59. Rh6 Ke5 60. Rc6 Kd5 61. Rf6 Kc5 62. Rf5+ Kc6
63. Rf6+ Kb7 64. g4 hxg3 65. Rf1 Rxa4 66. Kxg3 b5 67. Kf3
Rh4 68. Ke2 a4 69. Kd3 b4 70. Rf7+ Kc6 71. Rf6+ Kd5
72. Rf5+ Kd6 73. Rf6+ Ke5 74. Ra6 b3 75. Ra5+ Kd6 76. Kc3
Kc6 77. Kb2 Rh2+ 78. Kb1 Ra2 79. Rh5 Kb6 80. Rh8 Kb5
81. Rb8+ Kc4 82. Rc8+ Kb4 83. Rb8+ Ka3 84. Rd8 Rb2+ 85. Ka1
Rg2 86. Rd1 b2+

{[d]8/8/8/8/p7/k7/1p4r1/K2R4 w - -

HIARCS resigns}

0-1

[LJC]

Hi Lonnie,

I meant the same thing happens to me when I try to load it in DF 10 GUI, "engine can not load"

I'm to blame for the confusion. I put out so many different
builds trying to solve the Chessbase crash bug, now one one knows
which is which.

If you download the last/latest builds on my homepage you should
find it will load ok now in Fritz/Chessbase.

http://homepages.tesco.net/henry.ablett/jims.html

regards,
Jim.

works now .. thanks for the hard work in resolving this :.)

[Eelco de Groot]

Intermediate Result Belka 1.8.11 - Glaurung 2.0.1 Mjolnir 15-7

Second match 68 % - 32 %

First match 80 % - 20 %

Difference 12 % ≈ 84 elopoints in TPR[/color]

Too few games to be significant but I think it would depend more on results against other programs, with more threads and on different platforms.

Two matches played under the same conditions against Hiarcs 11.2 SP, 22 games again for Glaurung 2.0.1 Mjolnir instead of 20, I can think up a reason for that: it is only possible to compare the percentages directly with the Belka matches if the programs play the same number of games and another reason to play more games with the alternative could be because results against the alternative settings are a bit more valuable -because more rare, and you hope to prove the settings perform better, not worse, if they prove better you would like those percentages with different settings to be the more precise ones... Anyway 22 is not much different from 20, of course you can not arbitrarily lower the number of games in the default match.

The results of the matches against Hiarcs 11.2 Default (no learning for Hiarcs enabled) are then as follows;



Mjolnir match

First Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 2-0 (+2 =0 -0)


Second Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

A draw 1-1 (+1 =0 -1)


Third Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

A draw 1-1 (+0 =2 -0)


Fourth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 1½-½ (+1 =1 -0)


Fifth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 1½-½ (+1 =1 -0)


Sixth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 1½-½ (+1 =1 -0)


Seventh Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

A draw 1-1 (+1 =0 -1)


Eighth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

A draw 1-1 (+1 =0 -1)


Ninth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

A draw 1-1 (+0 =2 -0)

Tenth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

A draw 1-1 (+0 =2 -0)


Eleventh Run, Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 001b, Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game with 2 seconds Fischer increment. 64 Mb hash for both programs.

Glaurung Mjolnir wins 1½-½ (+1 =1 -0)


Intermediate Result Hiarcs 11.2 Default - Glaurung 2.0.1 Mjolnir 13-9


----------------------------------------------------------------------------------------------------------------

Default match

First Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

A draw 1-1 (+0 =2 -0)


Second Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 1½-½ (+1 =1 -0)



Third Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

A draw 1-1 (+0 =2 -0)


Fourth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 2-0 (+2 =0 -0)


Fifth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 1½-½ (+1 =1 -0)


Sixth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 2-0 (+2 =0 -0)


Seventh Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 1½-½ (+1 =1 -0)


Eighth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

Glaurung 2.0.1 default wins 1½-½ (+1 =1 -0)


Ninth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 1½-½ (+1 =1 -0)


Tenth Run, Hiarcs 11.2 Default - Glaurung 2.0.1 default,
Shredder GUI, HS Rybka.bkt tournament mode, 8 minutes per game
with 2 seconds Fischer increment. 64 Mb hash for both programs.

Hiarcs 11.2 wins 1½-½ (+1 =1 -0)



Intermediate Result: Hiarcs 11.2 Default - Glaurung 2.0.1 default 14-6

Glaurung 2.0.1 Mjolnir 9/22 = 41%

Glaurung 2.0.1 default 6/20 = 30%

Difference 11% ≈ 77 elopoints in TPR

[LJC]

M confused .. I haven't drank enough beer !! \O/ ..I.