TogaII 1.4.1SE breaks 3000 elo barrier!

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

Moderator: Ras

ArmyBridge

Re: TogaII 1.4.1SE breaks 3000 elo barrier!

Post by ArmyBridge »

Ovyron wrote:
Tony Thomas wrote:So what's new?
I do not only care about rating, but about playing style. If this new Toga is going to be very active and aggressive, yet that strong, then it'll be good new news to me :)
I think this Toga is less agressive than other Toga version like Toga 3.2.1SE, but I've check only few games vs Rybka 2.3.2a, till now seems that all programs tend to be more solid.. dam Rybka :lol: .. well I´waiting for Junior 11. I hope that it keep its agressive style, I dont want to see like-rybka-style engines everywhere :shock:
Yarget

Re: TogaII 1.4.1SE breaks 3000 elo barrier!

Post by Yarget »

I´waiting for Junior 11. I hope that it keep its agressive style, I dont want to see like-rybka-style engines everywhere
True, Junior is in my opinion one of the most original engines out there. It would really be a pity if Junior (like Fritz) will become another strong "mainsteam engine".

Regards
Per
Henrik Dinesen
Posts: 877
Joined: Wed Mar 08, 2006 9:52 pm
Location: Denmark

Re: TogaII 1.4.1SE breaks 3000 elo barrier!

Post by Henrik Dinesen »

Yarget wrote:
I´waiting for Junior 11. I hope that it keep its agressive style, I dont want to see like-rybka-style engines everywhere
True, Junior is in my opinion one of the most original engines out there. It would really be a pity if Junior (like Fritz) will become another strong "mainsteam engine".

Regards
Per
As long as so much talk is about ratinglists, and less flattering remarks about engines that are not no. 1 -3 strengthwise, and therefore isn't suited to fight for the operators ego, ehh rating, on playchess or similar, it's no wonder if we begin to see more and more similar styles among many different engines...:(
Henrik
Edward German

Re: My first experimental games (PGN)

Post by Edward German »

Chris Formula wrote:1. Fabien Letouzey/Thomas Gaksch for the great Fruit/TogaII code
2. CCRL group (Shaun Brewer, Graham Banks, etc) for the test support
3. Daniel Charles for the 8cpu test support
4. Denis Mendoza for the compile support
My special thanks too!

My first experimental games.

Today I played about 24 games on playchess. First games with default settings Toga 1.4.1SE T2 on my Core 2 Duo 2,66 GHz lost. The new Toga played extrem aggressive with a sharp pruning. I my opinion that is not the best setting, sorry.

As last I played with this experimental (µ) Settings:

[OPTIONS]
Verification Reduction=4
Material=120
Piece Activity=95
Piece Square Activity=95
King Safety=105

T2µ = 2 Threads; µ = The new experimental Settings.

My last 4 games on playchess with my Nickname Solist, PGN:

[Event "Wertungspartie, 3m + 0s"]
[Site "Maschinenraum"]
[Date "2008.02.12"]
[Round "?"]
[White "Hedlund, Rybka 2.3.2a mp"]
[Black "Solist, Toga II 1.4.1SE T2µ"]
[Result "1-0"]
[ECO "B18"]
[WhiteElo "2723"]
[BlackElo "2508"]
[Annotator "0.42;0.23"]
[PlyCount "39"]
[EventDate "2008.02.12"]
[TimeControl "180"]

{Solista.ctg, oreDeviceObjects} 1. Nc3 {B/0 0} d5 {B/0 0} 2. e4 {B/0 0} c6 {
B/0 0} 3. d4 {B/0 0} dxe4 {B/0 0} 4. Nxe4 {B/0 0} Bf5 {B/0 0} 5. Ng3 {B/0 0}
Bg6 {B/0 0} 6. Nf3 {B/0 0} Nf6 {B/0 0} 7. h4 {0.42/13 9} h6 {(Sh5) B/0 0} 8.
Ne5 {B/0 0} Bh7 {B/0 0} 9. Bd3 {B/0 0} Bxd3 {B/0 0} 10. Qxd3 {B/0 0} e6 {B/0 0}
11. Bf4 {B/0 0} Nbd7 {0.23/14 6} 12. O-O-O {B/0 0} Bd6 {0.30/14 6} 13. Kb1 {
0.28/13 3} Nb6 {(Dc7) 0.28/15 5} 14. c4 {0.35/12 6} O-O {(Lc7) 0.22/15 0} 15.
Qe2 {0.42/13 13} Bc7 {0.20/14 6} 16. Bg5 {1.43/14 11} Na4 {1.50/14 10} 17. Qc2
{1.60/14 2} Qe7 {(Sxb2) 2.41/14 10} 18. Nh5 {2.34/13 4} Qb4 {(Tfd8) 3.84/15 11}
19. Bxf6 {2.97/13 1} gxf6 {3.94/15 4} 20. Nxf6+ {
Solist,Toga II 1.4.1SE T2µ gibt auf 3.36/15 0} 1-0

[Event "Wertungspartie, 4m + 0s"]
[Site "Maschinenraum"]
[Date "2008.02.12"]
[Round "?"]
[White "Solan, Rybka 2.3.2a mp"]
[Black "Solist, Toga II 1.4.1SE T2µ"]
[Result "1/2-1/2"]
[ECO "C42"]
[WhiteElo "2546"]
[BlackElo "2501"]
[Annotator "-0.03;0.34"]
[PlyCount "86"]
[EventDate "2008.02.12"]
[TimeControl "240"]

{Rybka 2.3.2a mp: 17.1 ply; 236kN/s Dual Core AMD Opteron(tm) Processor 185 ,
(2 threads), Solista.ctg, oreDeviceObjects} 1. e4 {B/0 0} e5 {B/0 0} 2. Nf3 {
B/0 0} Nf6 {B/0 0} 3. Nxe5 {B/0 0} d6 {B/0 0} 4. Nf3 {B/0 0} Nxe4 {B/0 0} 5. d4
{B/0 0} d5 {B/0 0} 6. Bd3 {B/0 0} Nc6 {B/0 0} 7. O-O {B/0 0} Be7 {B/0 0} 8. c4
{B/0 0} Nb4 {B/0 0} 9. Be2 {B/0 0} O-O {B/0 0} 10. Nc3 {B/0 0} Bf5 {B/0 0} 11.
a3 {B/0 0} Nxc3 {B/0 0} 12. bxc3 {B/0 0} Nc6 {B/0 0} 13. Re1 {B/0 0} Re8 {B/0 0
} 14. Bf4 {B/0 0} dxc4 {B/0 0} 15. Bxc4 {B/0 0} Bd6 {B/0 0} 16. Rxe8+ {B/0 0}
Qxe8 {0.34/14 2} 17. Ng5 {B/0 0} Bg6 {B/0 0} 18. Bxd6 {B/0 0} cxd6 {B/0 0} 19.
h4 {B/0 0} Qe7 {B/0 0} 20. Qg4 {B/0 0} h6 {B/0 0} 21. Nh3 {B/0 0} Qe4 {B/0 0}
22. Qg3 {B/0 0} Rc8 {B/0 0} 23. Qxd6 {-0.03/14 7} Qc2 {B/0 0} 24. Qg3 {
-0.12/14 2} Na5 {B/0 0} 25. Bd5 {-0.13/14 1} Qxc3 {B/0 0} 26. Qxc3 {-0.18/16 7}
Rxc3 {B/0 0} 27. Nf4 {-0.16/16 2} Bh7 {B/0 0} 28. Ne2 {-0.21/16 11} Rc2 {
-0.13/17 6} 29. Rc1 {-0.20/18 2} Rxc1+ {-0.08/19 11} 30. Nxc1 {-0.14/21 6} Kf8
{-0.12/20 4} 31. Kf1 {-0.17/20 3} b5 {-0.09/18 7} 32. Ke2 {-0.12/20 6} Bf5 {
(Ke7) -0.06/19 2} 33. Kd2 {-0.05/16 1} Be6 {(Ke7) -0.12/18 6} 34. Bxe6 {
-0.10/15 0} fxe6 {-0.11/18 2} 35. Ne2 {0.00/19 3} Ke7 {-0.08/19 2} 36. Kc3 {
0.00/20 5} Nc6 {-0.11/21 2} 37. Kd3 {0.00/21 2} a5 {(Kd6) -0.02/19 14} 38. Nc3
{0.00/19 4} Na7 {0.00/20 1} 39. h5 {0.00/20 5} Kd6 {0.00/22 2} 40. Ne4+ {
0.00/21 1} Kd5 {0.00/23 6} 41. Nc3+ {0.00/26 3} Kd6 {0.00/27 3} 42. Ne4+ {
0.00/23 2} Kd5 {0.00/28 7} 43. Nc3+ {
Solist,Toga II 1.4.1SE T2µ bietet Remis 0.00/28 6} Kd6 {0.00/28 5} 1/2-1/2

[Event "Wertungspartie, 5m + 0s"]
[Site "Maschinenraum"]
[Date "2008.02.12"]
[Round "?"]
[White "Solist, Toga II 1.4.1SE T2µ"]
[Black "Phil Harris, Rybka 2.3.2a mp"]
[Result "1/2-1/2"]
[ECO "B65"]
[WhiteElo "2503"]
[BlackElo "2742"]
[Annotator "0.25;0.28"]
[PlyCount "217"]
[EventDate "2008.02.12"]
[TimeControl "300"]

{Rybka 2.3.2a mp: 20.2 ply; 544kN/s Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.
40GHz 3690MHz, (4 threads), Solista.ctg, oreDeviceObjects} 1. e4 {B/0 0} c5 {
B/0 0} 2. Nf3 {B/0 0} d6 {B/0 0} 3. d4 {B/0 0} cxd4 {B/0 0} 4. Nxd4 {B/0 0} Nc6
{B/0 0} 5. Nc3 {B/0 0} Nf6 {B/0 0} 6. Bg5 {B/0 0} e6 {B/0 0} 7. Qd2 {B/0 0} Be7
{B/0 0} 8. O-O-O {B/0 0} Nxd4 {B/0 0} 9. Qxd4 {B/0 0} O-O {B/0 0} 10. f4 {B/0 0
} h6 {0.28/15 7} 11. Bh4 {B/0 0} a6 {0.26/14 2} 12. g4 {(Be2) B/0 0} Qa5 {
0.14/15 13} 13. Qg1 {(Qd3) B/0 0} d5 {0.19/14 9} 14. exd5 {(e5) 0.25/14 11}
exd5 {0.02/15 11} 15. h3 {(Qg2) 0.04/15 5} Bc5 {0.00/14 2} 16. Bf2 {
(Qg2) 0.08/16 6} Bxf2 {0.03/15 3} 17. Qxf2 {0.08/16 0} Ne4 {0.10/16 7} 18. Nxe4
{0.00/16 11} dxe4 {0.03/18 2} 19. Bc4 {(a3) -0.02/16 5} Be6 {0.12/15 5} 20.
Bxe6 {0.31/16 11} fxe6 {0.18/18 8} 21. Kb1 {(a3) 0.32/16 2} Qc7 {0.16/17 13}
22. Rhf1 {0.28/15 7} Rac8 {0.16/17 1} 23. Rde1 {0.23/15 12} e3 {0.01/17 23} 24.
Qxe3 {0.24/18 0} Qxc2+ {0.00/16 2} 25. Ka1 {0.20/17 0} Rc6 {0.00/17 4} 26. Rc1
{0.21/17 7} Qa4 {0.00/17 0} 27. Rcd1 {(Rxc6) 0.06/16 15} Qc4 {0.00/15 4} 28. h4
{(Rf3) 0.00/16 9} Rd6 {0.00/15 4} 29. Rc1 {0.17/16 8} Qd4 {-0.10/17 1} 30. Qf3
{(Qg3) 0.12/16 6} Rb6 {-0.09/16 3} 31. Qe2 {0.00/15 11} Rb4 {-0.02/18 6} 32.
Qxe6+ {(g5) 0.00/15 3} Kh7 {-0.08/14 1} 33. Rb1 {-0.10/16 8} Rf6 {-0.03/15 1}
34. Qe7 {(Qe8) 0.00/16 11} Qe4 {-0.13/17 4} 35. Qd7 {-0.07/16 3} Rd4 {
-0.08/17 10} 36. Qc7 {(Qc8) -0.18/16 0} Rxf4 {-0.33/14 3} 37. Rxf4 {
(g5) -0.17/15 3} Qxf4 {-0.28/16 1} 38. Qxb7 {-0.25/14 4} Qxg4 {-0.32/17 0} 39.
Qxa6 {-0.28/15 5} h5 {-0.34/17 3} 40. Qa5 {(Qb5) -0.30/13 9} Qxh4 {-0.31/13 1}
41. Qf5+ {-0.31/14 3} Kh6 {-0.22/18 6} 42. Rf1 {(Qe6+) -0.39/14 0} Qg4 {
-0.74/16 5} 43. Qb1 {-0.46/13 4} h4 {-0.78/16 2} 44. Rg1 {-0.41/14 5} Qh5 {
-0.74/17 1} 45. Qc1+ {(Qf1) -0.33/12 4} g5 {-0.82/15 1} 46. a3 {
(Qe1) -0.63/13 8} Rd3 {-0.95/15 2} 47. Qe1 {-0.57/13 4} Qf7 {-1.14/16 1} 48.
Qe5 {-0.71/13 4} Rd5 {-1.22/17 0} 49. Qh8+ {-0.72/13 3} Kg6 {-1.32/18 14} 50.
Qc3 {-0.89/15 0} Qd7 {-1.32/17 5} 51. Qc2+ {-0.83/14 0} Kh6 {-1.42/17 8} 52.
Rc1 {(Rf1) -0.88/15 0} h3 {-1.56/13 1} 53. Qe2 {(Qc3) -1.04/13 3} Rd2 {
-1.76/14 1} 54. Qe5 {(Qf3) -1.10/13 2} h2 {-1.45/15 2} 55. Qh8+ {-0.87/14 1}
Kg6 {-0.71/18 1} 56. Qg8+ {-0.88/14 2} Kh5 {-0.69/20 1} 57. Qh8+ {-0.13/13 2}
Kg4 {-0.64/20 1} 58. Rc4+ {0.00/13 1} Kf3 {-0.71/20 11} 59. Qf6+ {0.00/14 3}
Kg2 {-0.70/21 1} 60. Qxg5+ {0.00/16 0} Kh1 {-0.70/21 1} 61. Qf4 {0.00/16 2}
Rd1+ {-0.70/21 2} 62. Ka2 {0.00/16 1} Qd3 {-0.70/21 3} 63. Rc3 {-0.08/14 0}
Qd5+ {-0.70/19 0} 64. Rc4 {-0.06/15 3} Re1 {-0.70/19 7} 65. Kb3 {
(a4) -0.37/14 0} Qd3+ {-0.89/10 0} 66. Ka2 {-0.43/14 4} Qf1 {-1.00/18 4} 67.
Qc7 {-0.23/11 2} Qf5 {-1.16/18 1} 68. a4 {-0.52/12 1} Qb1+ {-1.09/17 1} 69. Kb3
{(Ka3) -0.55/13 1} Re3+ {-1.09/14 0} 70. Rc3 {-0.49/14 2} Qe4 {-1.18/17 0} 71.
Rxe3 {-0.46/13 2} Qxe3+ {-1.10/21 0} 72. Ka2 {-0.47/15 2} Qe6+ {-1.10/22 1} 73.
b3 {-0.47/15 1} Kg2 {-1.10/22 1} 74. Qc2+ {-0.49/15 1} Kg1 {-1.10/24 1} 75.
Qc5+ {-0.48/16 1} Kf1 {-1.10/24 0} 76. Qf8+ {-0.48/15 1} Kg2 {-1.16/24 2} 77.
Qg7+ {-0.49/17 0} Kf2 {-1.16/24 2} 78. Qf8+ {-0.49/13 0} Kg3 {-1.16/24 1} 79.
Qb8+ {-0.49/17 1} Kg2 {-1.16/25 1} 80. Qa8+ {-0.49/18 1} Kg1 {-1.16/26 1} 81.
Qa7+ {-0.49/18 0} Kf1 {-1.10/26 5} 82. Qh7 {-0.49/19 0} Qe2+ {-1.10/25 4} 83.
Ka3 {-0.49/12 1} Qe5 {-1.10/25 3} 84. Qh3+ {(Qb1+) -0.49/15 0} Kg1 {-1.10/21 2}
85. Qg4+ {0.00/13 1} Kf2 {-1.10/25 0} 86. Qh4+ {-0.49/15 1} Kf3 {-1.10/26 1}
87. Qh3+ {-0.49/15 2} Ke2 {-1.10/26 1} 88. Qg2+ {(Qg4+) -0.49/15 1} Ke3 {
-1.10/25 0} 89. Qh3+ {-0.49/13 1} Kd2 {-1.10/26 0} 90. Qg2+ {-0.49/13 1} Ke1 {
-1.10/27 3} 91. Qc6 {-0.49/14 0} Kf2 {-1.10/25 0} 92. Qc2+ {-0.49/12 1} Ke3 {
-1.10/25 2} 93. Qc1+ {-0.48/12 1} Ke2 {-1.10/26 2} 94. Qc2+ {-0.48/12 1} Kf3 {
-1.10/26 0} 95. Qc6+ {(Qd3+) -0.48/13 1} Kg3 {-1.10/19 0} 96. Qg6+ {-0.48/13 1}
Kf4 {-1.10/21 1} 97. Qh6+ {-0.47/12 1} Kg4 {-1.10/23 2} 98. Qg6+ {-0.36/12 1}
Kh3 {0.00/24 1} 99. Qh6+ {0.00/17 1} Kg2 {0.00/27 1} 100. Qc6+ {0.00/18 1} Kf1
{0.00/28 1} 101. Qf3+ {(Qc1+) 0.00/17 1} Kg1 {0.00/27 0} 102. Qg4+ {
(Qd1+) 0.00/22 2} Kh1 {0.00/25 0} 103. Kb4 {(Qd1+) 0.00/14 1} Qe1+ {-0.14/14 0}
104. Kc4 {0.00/16 1} Qf1+ {0.00/21 0} 105. Kc3 {0.00/17 1} Qf6+ {0.00/22 1}
106. Kd3 {0.00/16 1} Qd6+ {0.00/22 0} 107. Ke2 {(Kc3) 0.00/15 1} Qa6+ {
0.00/18 0} 108. Ke1 {(Ke3) 0.00/16 1} Qd3 {0.00/20 0} 109. Kf2 {
(Lag: Av=0.19s, max=0.7s) 0.00/17 1} 1/2-1/2

[Event "Wertungspartie, 4m + 0s"]
[Site "Maschinenraum"]
[Date "2008.02.12"]
[Round "?"]
[White "Solist, Toga II 1.4.1SE T2µ"]
[Black "Solan, Rybka 2.3.2a mp"]
[Result "1/2-1/2"]
[ECO "D15"]
[WhiteElo "2513"]
[BlackElo "2553"]
[Annotator "0.34;0.29"]
[PlyCount "107"]
[EventDate "2008.02.12"]
[TimeControl "240"]

{Rybka 2.3.2a mp: 17.0 ply; 233kN/s Dual Core AMD Opteron(tm) Processor 185 ,
(2 threads), Solista.ctg, oreDeviceObjects} 1. d4 {B/0 0} d5 {B/0 0} 2. c4 {
B/0 0} c6 {B/0 0} 3. Nf3 {B/0 0} Nf6 {B/0 0} 4. Nc3 {B/0 0} a6 {B/0 0} 5. e3 {
B/0 0} b5 {B/0 0} 6. b3 {B/0 0} Bg4 {B/0 0} 7. Be2 {B/0 0} e6 {B/0 0} 8. O-O {
B/0 0} Nbd7 {B/0 0} 9. Bb2 {B/0 0} Bd6 {B/0 0} 10. h3 {0.34/14 5} Bh5 {B/0 0}
11. Ne5 {B/0 0} Bxe2 {B/0 0} 12. Nxe2 {B/0 0} Qc7 {B/0 0} 13. cxd5 {B/0 0} cxd5
{B/0 0} 14. Rc1 {B/0 0} Qb8 {B/0 0} 15. Nxd7 {B/0 0} Nxd7 {B/0 0} 16. e4 {B/0 0
} dxe4 {B/0 0} 17. d5 {B/0 0} O-O {B/0 0} 18. dxe6 {B/0 0} Nc5 {B/0 0} 19. Nf4
{B/0 0} Ra7 {0.29/13 6} 20. Bd4 {0.31/14 9} Bxf4 {0.28/16 2} 21. Bxc5 {
0.30/16 7} Bxc1 {0.31/17 4} 22. Bxf8 {0.47/16 1} Qxf8 {0.35/17 11} 23. Qxc1 {
0.44/18 0} Re7 {0.36/16 3} 24. exf7+ {0.43/15 3} Rxf7 {0.36/17 10} 25. Qc3 {
(Qe3) 0.46/17 0} Qd6 {0.30/14 4} 26. Re1 {0.72/16 9} Rc7 {0.50/15 1} 27. Qe3 {
0.58/16 4} Re7 {0.47/16 3} 28. f3 {0.84/15 6} h6 {0.45/16 2} 29. fxe4 {
0.73/16 6} Qe5 {0.45/16 2} 30. Qd3 {(Qd2) 0.78/16 4} Re6 {0.44/15 9} 31. a4 {
0.68/16 0} Rd6 {0.44/14 2} 32. Qc2 {(Qe3) 0.65/16 5} Kh7 {0.39/13 3} 33. axb5 {
(Qe2) 0.76/16 9} axb5 {0.40/13 3} 34. b4 {0.73/17 3} Rd4 {0.12/15 8} 35. Qc3 {
(Qc5) 0.40/17 7} Kg8 {0.12/14 8} 36. Qa1 {0.40/16 0} Qd6 {0.17/15 6} 37. e5 {
0.73/16 7} Qxb4 {0.15/17 1} 38. e6 {0.83/17 6} Qc5 {0.17/18 2} 39. Kh1 {
0.48/16 5} Rd8 {0.07/17 2} 40. e7 {0.34/17 9} Re8 {0.05/18 8} 41. Qa2+ {
0.22/18 0} Qc4 {0.05/18 1} 42. Qa6 {0.16/15 2} Kh7 {0.02/18 2} 43. Qa5 {
0.05/16 4} b4 {0.02/18 2} 44. Qf5+ {0.05/17 2} Kg8 {0.02/18 2} 45. Qd7 {
(Qe5) 0.06/16 1} Qf7 {0.00/15 4} 46. Kh2 {0.04/16 2} b3 {0.00/16 3} 47. Re2 {
0.05/16 3} Kh7 {0.00/16 0} 48. Qd6 {(Qd3+) 0.03/16 4} g6 {0.00/15 5} 49. Kg3 {
(Qe5) 0.00/15 6} h5 {0.00/14 6} 50. Kh2 {(Qe5) 0.00/17 0} Kg7 {0.00/15 2} 51.
Qd4+ {(Kg1) 0.00/16 2} Qf6 {0.00/13 1} 52. Re5 {(Qb4) 0.00/17 4} Kg8 {0.00/16 3
} 53. Qd5+ {(Qc4+) 0.00/17 0} Kh7 {
Solist,Toga II 1.4.1SE T2µ bietet Remis 0.00/16 3} 54. Qd4 {
(Lag: Av=0.22s, max=0.5s) 0.00/17 0} 1/2-1/2

All in all, I hold with this "µ-Settings" as last 3 draws against Rybka MP. In my opinion, here (Toga 1.4.1SE) is "Verification Reduction=4" much better than "Verification Reduction=5". I am sure!

My Best!

PS: Is it possible to compile my Setting as "µ"-Setting by Eduard Nemeth?
Edward German

Re: BTW...

Post by Edward German »

My strongest Setting for Toga 1.4.1SE to analyses under Shredder Classic are that:

[ENGINE]
Name=Toga II 1.4.1SE T2e2
Author=Thomas Gaksch and Fabien Letouzey
Filename=C:\Programme\ShredderChess\Shredder 10 UCI\Engines\TogaIIv1.4.1SE\toga141se-2cpu.exe
[OPTIONS]
Verification Reduction=4
Toga Lazy Eval Margin=225
Toga King Safety Margin=1300


Best!
User avatar
Ovyron
Posts: 4562
Joined: Tue Jul 03, 2007 4:30 am

Re: TogaII 1.4.1SE breaks 3000 elo barrier!

Post by Ovyron »

ArmyBridge wrote:I think this Toga is less agressive than other Toga version like Toga 3.2.1SE, but I've check only few games vs Rybka 2.3.2a
You don't need a lot of games to check the engine's style. When I watched Toga 3.2.1SE with Eelco's settings, I watched fireworks on the board, game after game, so it never let me down.

Now I've tested 1.4.1SE and I can say that it seems too materialistic. It contrasts with how 3.2.1SE sacrificed material to keep the initiative while 1.4.1SE doesn't risk it and goes for a quieter game.

I can still say that 1.4.1SE is the strongest Toga, score-wise, but performance isn't enough to me.

It reminds me of the story with Rybka Winfinder and normal Rybka, in where Winfinder plays a more interesting chess, while Rybka has been playing duller and duller with each new version. Perhaps there's no way to make Toga style's better without hurting its strength and I understand that, so I'm already happy with my Toga 3.2.1SE :) (I'm going to test 1.4.1SE with Eelco and the new Eduard's settings, to see if there's still hope ;) .)
Jouni
Posts: 3730
Joined: Wed Mar 08, 2006 8:15 pm
Full name: Jouni Uski

Re: TogaII 1.4.1SE breaks 3000 elo barrier!

Post by Jouni »

Yes latest version of Toga is tactically weaker than 3.1.2 or 1.4 as You can
see fast by running testsuite like Arasan. It's speed is around Rybka level now. But seems to play about 1.4 level in single version (blitz):

Code: Select all


                     
1   Toga II 1.4 beta5c       ½½½½½10½½½½01½½1½½111½½011½0½½00½½0½½11½  21.5/40
2   Toga II 1.4.1SE          ½½½½½01½½½½10½½0½½000½½100½1½½11½½1½½00½  18.5/40

Against Fritz10 both versions scored same results.
Jouni
gerold
Posts: 10121
Joined: Thu Mar 09, 2006 12:57 am
Location: van buren,missouri

Re: TogaII 1.4.1SE breaks 3000 elo barrier!

Post by gerold »

Jouni wrote:Yes latest version of Toga is tactically weaker than 3.1.2 or 1.4 as You can
see fast by running testsuite like Arasan. It's speed is around Rybka level now. But seems to play about 1.4 level in single version (blitz):

Code: Select all


                     
1   Toga II 1.4 beta5c       ½½½½½10½½½½01½½1½½111½½011½0½½00½½0½½11½  21.5/40
2   Toga II 1.4.1SE          ½½½½½01½½½½10½½0½½000½½100½1½½11½½1½½00½  18.5/40

Against Fritz10 both versions scored same results.
Jouni
Thanks for the post Jouni.

I have it the other way around.

TogaII 1.4.1 SE 2cpu. 50.5/97 20-16-61
TogaII 1.4 5c beta 2cpu 46.5/97 16-20-61

Lots of draw games.:) I think these two are the
same program. I can't tell the differ.

Good luck,

Gerold.
Chris Formula

Re: BTW...

Post by Chris Formula »

Thanks for sharing your settings, Eduard. I wish this setting would really improve the play strength after running statistical number of games/matches.
User avatar
Denis P. Mendoza
Posts: 415
Joined: Fri Dec 15, 2006 9:46 pm
Location: Philippines

Re: BTW...

Post by Denis P. Mendoza »

Edward German wrote:My strongest Setting for Toga 1.4.1SE to analyses under Shredder Classic are that:

[ENGINE]
Name=Toga II 1.4.1SE T2e2
Author=Thomas Gaksch and Fabien Letouzey
Filename=C:\Programme\ShredderChess\Shredder 10 UCI\Engines\TogaIIv1.4.1SE\toga141se-2cpu.exe
[OPTIONS]
Verification Reduction=4
Toga Lazy Eval Margin=225
Toga King Safety Margin=1300


Best!
Hello Eduard,

The last time we met here was also related to your strong Toga settings. I still remember that good old TLEM=225. I never forgot that trademark setting of yours:D. Good to hear some good reports about the new Togas.

Thomas' and Chris' versions are almost equal. Not much difference in the codes except for insertion of Chris' 3.1.2SE scheme in the Toga MP version (which worked too). The aggressiveness of TogaII 3.1.2SE wasn't seen in the new 1.4.1SE release. My hunch is that, it could be caused by the inclusion of Search Razoring technique, an idea from Tord Romstad's Glaurung. I may be wrong. But it's one of the best additions from the first previous MP version TogaII 1.4 beta 4a, that is why TogaII 1.4 beta5c was released.Thomas has not made a UCI-option yet, maybe because it is still a beta release, so you can't enable or disable it.

Since Toga already made a big leap from Toga single cpu to MP, reaching/surpassing ~3000 elo, it may just need some tuning for tweakers like you. Adding some important UCI-options might create a more customized strong personality for these new Togas. And your feedbacks will be greatly appreciated by Thomas, while working on his engine.

I haven't touched the codes since 1.4.1SE release, but I already managed to add 2 UCI-options while writing this post today. Give me another day or two. I'm still recuperating from a flu, but I'll send you a build with more editable UCI-options. I'm sure you'll like it.

Thanks.

Denis