OliThink 5.4.0 has been published with an big leap in strength for only 3 lines of code

[OliverBr]

This is announcement of the next version: OliThink 5.4.3
(http://brausch.org/home/chess)

Here is an example game:

[pgn][Event "Computer Chess Game"]
[Site "Olivers-MacBook-2.local"]
[Date "2020.06.21"]
[Round "3"]
[White "OliThink 5.4.3"]
[Black "Fruit 2.1"]
[Result "1-0"]
[TimeControl "40/60"]
[Annotator "2. +0.05 1... -0.22"]

1. Nf3 Nc6 2. d4 d5 3. e3 Nf6 4. Nc3 e6 5. Ne5 Nxe5 6. dxe5 Nd7 7. f4 Bb4
8. Bd2 O-O 9. Qh5 Nc5 10. a3 Bxc3 11. Bxc3 g6 12. Qh6 Ne4 13. h4 Nxc3 14.
h5 Qe7 15. bxc3 Rb8 16. Bd3 Qd7 17. g4 Re8 18. f5 Qe7 19. fxg6 fxg6 20.
hxg6 b6 21. Kd2 Ba6 22. Raf1 Qg7 23. Qxg7+ Kxg7 24. Rxh7+ Kg8 25. Bxa6 b5
26. Rxc7 Rf8 27. Rxf8+ Rxf8 28. Bxb5 a5 29. c4 dxc4 30. Bxc4 Re8 31. Rh7
Rd8+ 32. Ke2 Re8 33. a4 Kf8 34. Rh8+ Ke7 35. g7 Rg8 36. Rxg8 Kf7 37. Ra8
Kxg7 38. Ra7+ Kf8 39. Bxe6 Ke8 40. g5 Kd8 41. g6 Ke8 42. g7 Kd8 43. g8=Q#
{Xboard adjudication: Checkmate} 1-0
[/pgn]

[OliverBr]

In Version 5.4.4, "Bus Error 10" won't happen anymore when ply >= 64. It sometimes wins against Glaurung:

[pgn][Event "Computer Chess Game"]
[Site "Olivers-MacBook-2.local"]
[Date "2020.06.21"]
[Round "4"]
[White "Glaurung 2.1"]
[Black "OliThink 5.4.4"]
[Result "0-1"]
[TimeControl "40/120"]
[Annotator "10. -0.45 9... -0.16"]

1. c4 e5 2. g3 Nf6 3. Bg2 Bc5 4. Nc3 Nc6 5. a3 a6 6. e3 O-O 7. Nge2 d6 8.
h3 Ba7 9. d3 h5 {-0.16/12 1.8} 10. O-O {-0.45/15 6} Bd7 {-0.16/12 4} 11. f4
{-0.25/15 6} Re8 {-0.06/11 2.4} 12. fxe5 {-0.25/15 5} dxe5 {-0.28/12 3} 13.
Nd5 {+0.01/14 9} Nxd5 {-0.18/14 3} 14. cxd5 {+0.33/16 4} Ne7 {+0.05/14 4}
15. Nc3 {+0.60/16 5} g6 16. Qf3 {+0.39/16 4} f5 17. d6 {-0.15/15 10} Nc6
{+0.62/14 2.6} 18. dxc7 {+0.13/15 4} Qxc7 19. Nd5 {-0.25/15 8} Qd6 20. g4
{-0.39/14 7} hxg4 {+0.60/14 0.1} 21. hxg4 {-0.39/14 2.7} f4 {+0.60/12 0.2}
22. g5 {-0.49/12 2.2} Kg7 {+0.81/13 2.7} 23. Qe2 {-0.56/13 4} Rh8 24. Nf6
{-1.21/12 3} Na5 {+1.47/13 5} 25. Bd2 {-1.78/12 2.2} Nb3 {+2.19/13 3} 26.
Rae1 {-2.23/12 2.3} Nxd2 {+2.34/14 2.1} 27. Qxd2 {-2.49/12 0.9} fxe3
{+2.22/14 5} 28. Rxe3 {-2.58/13 2.3} Bf5 {+2.55/13 2.0} 29. Rf2
{-2.54/13 4} Qd4 {+2.63/14 1.8} 30. Ref3 {-2.86/14 3} Rad8 31. Kf1
{-2.70/13 2.9} Rh4 {+3.08/13 4} 32. Qc2 {-2.78/13 1.9} Bxd3+ {+3.26/15 2.8}
33. Qxd3 {-3.45/16 2.8} Qxd3+ {+3.57/16 3} 34. Rxd3 {-3.43/13 0.2} Rxd3
{+3.53/15 4} 35. Ne8+ {-3.45/16 3} Kg8 {+3.53/15 2.1} 36. Nf6+ {-3.80/16 5}
Kh8 {+3.66/17 4} 37. Rf3 {-3.92/16 1.9} Rd1+ {+5.02/18 9} 38. Ke2
{-3.11/6 0.1} Rg1 {+5.16/16 10} 39. Bf1 {-4.72/15 2.6} Rh2+ {+4.99/16 6}
40. Kd3 {-4.76/16 1.8} Rxg5 {+4.73/15 5} 41. Ne4 {-4.92/16 7} Rf5
{+5.11/17 0.2} 42. Be2 {-5.31/18 4} Rxf3+ 43. Bxf3 {-5.31/15 0.3} Rxb2
{+5.17/17 2.2} 44. Ng5 {-5.35/18 6} b5 45. Bd5 {-5.54/18 5} Bc5
{+5.40/16 2.9} 46. Nf3 {-5.86/17 5} e4+ {+5.54/16 2.7} 47. Kxe4
{-5.54/18 5} Re2+ 48. Kd3 {-5.72/17 3} Re3+ {+5.71/15 3} 49. Kc2
{-5.92/18 4} Rxa3 {+5.72/15 2.1} 50. Ng5 {-6.03/17 6} Bd4 {+5.78/15 4} 51.
Ne6 {-5.70/17 2.9} Bf6 {+6.08/16 4} 52. Nf4 {-6.21/15 4} g5 {+6.25/16 4}
53. Nd3 {-6.66/17 6} g4 54. Kd2 {-7.09/16 5} b4 {+7.72/15 4} 55. Ke2
{-7.33/15 3} b3 {+8.15/16 4} 56. Kf2 {-8.11/15 3} b2 {+8.88/16 2.9} 57.
Nxb2 {-11.74/18 2.4} g3+ {+9.81/15 1.9} 58. Kf1 {-81.47/19 2.1} Bxb2
{+10.40/14 4} 59. Be6 {-84.64/19 2.4} Rd3 {+11.66/14 4} 60. Kg2
{-90.59/19 2.3} Be5 {+12.69/14 1.9} 61. Kf1 {-99.16/19 2.2} Rd1+
{+15.60/13 4} 62. Ke2 {-1000.14/17 0.8} g2 {+15.60/13 2.6} 63. Kxd1
{-1000.13/15 0.2} g1=Q+ {+15.60/12 4} 64. Ke2 {-1000.12/15 0.6} Qh2+
{+15.60/11 4} 65. Kd3 {-1000.11/14 0.2} Qh7+ {+15.70/11 3} 66. Ke2
{-1000.10/13 0.2} Qe4+ {+1000.08/11 5} 67. Kd2 {-1000.09/11 0.2} Bd6
{+1000.07/11 5} 68. Kc3 {-1000.07/11 0.2} Bb4+ {+1000.06/11 4} 69. Kb3
{-1000.05/11 0.2} Qd4 {+1000.05/10 0.2} 70. Ka2 {-1000.04/6 0.1} Qd2+
{+1000.04/10 0.1} 71. Kb3 {-1000.03/6 0.1} Qc3+ {+1000.03/11 0.5} 72. Ka2
{-1000.02/6 0.1} Qc2+ {+1000.02/11 0.3} 73. Ka1 {-1000.01/6 0.1} Bc3#
{+1000.01/1 0.1}
{Xboard adjudication: Checkmate} 0-1
[/pgn]

Still there are a couple of bugs, e.g. in another game vs Glaurung OliThink played, which has been lost after 14 ...Kg8???

[d]r4rk1/pp1nb1p1/1qp1pn2/4p2p/4P3/P1N3PP/1PP1QPB1/R1BR2K1 b - - 1 14

The funny part is, that it won't play the move when analyzing the position. This is the eval:

Code: Select all

./olithink544 
r4rk1/pp1nb1p1/1qp1pn2/4p2p/4P3/P1N3PP/1PP1QPB1/R1BR2K1 b - - 1 14
 1    54      0        94  a7a5 
 2    14      0       441  a7a5 g1h1 
 3    26      0      2464  a7a5 g1h1 a5a4 
 4    15      0      4723  a7a5 g1h1 a5a4 h3h4 
 5    23      0     10604  a7a5 g1h1 g8h8 h3h4 a5a4 
 6   -17      2     65556  a7a5 a1b1 e7c5 c1e3 b6c7 h3h4 
 7   -13      4    123348  a7a5 a1b1 e7c5 c1e3 c5e3 f2e3 b6c7 
 8   -22     10    298025  a7a5 a1b1 e7c5 c1e3 c5d4 g1h1 a5a4 h3h4 
 9   -17     25    778187  a7a5 g1h1 a5a4 e2c4 b6f2 d1d7 f2e1 h1h2 f6d7 c4e6 f8f7 
10   -39    117   4144161  g8f7 a1b1 e7c5 c3a4 b6b5 e2b5 c6b5 a4c5 d7c5 f2f3 
11   -30    172   6376449  g8f7 b2b4 a7a5 c1e3 b6a6 e2a6 a8a6 g2f1 b7b5 a1b1 a6a8 
12   -58    803  31390383  b6a6 e2a6 b7a6 c1e3 f8b8 b2b4 c6c5 c3e2 c5b4 a3b4 e7b4 a1a6 
13   -64   1227  48910980  b6a6 e2a6 b7a6 c1e3 a6a5 f2f4 a8b8 b2b3 c6c5 g2f3 g7g6 f3e2 g8h8 
14   -62   2689 111038881  e7c5 c3a4 b6b5 e2b5 c6b5 a4c5 d7c5 f2f3 c5a4 g2f1 a7a6 c2c4 h5h4 g3h4 
15   -57   5065 212926260  b6c7 e2c4 g8f7 c1e3 f8c8 f2f4 b7b5 c4a2 e7c5 e3c5 d7c5 d1f1 f7g8 b2b4 c7d6 

[OliverBr]

I couldn't not reproduce the bug with "14. Kg8", but have the guess that a bad timed interruption of ponder could be the reason. I did some changes in 5.4.5.

[OliverBr]

5.4.6 is (finally) published.
Some protocol features have been adapted and we got rid off an asymmetrical evaluation.

[Dann Corbit]

It would be nice if you put your home page link in your sig.
Then old guys like me could remember where to look.

[Dann Corbit]

Suggestion:
compress the games before sending

[Dann Corbit]

Suggestion:
set the priority of scorpio to high.
It will give you 5% more games

[Dann Corbit]

sorry, last two suggestions were posted to the wrong thread

[OliverBr]

It would be nice if you put your home page link in your sig.
Then old guys like me could remember where to look.

Can you please navigate me to the section where I can write my signature?
PS: 5.4.8 is out

[Dann Corbit]

It would be nice if you put your home page link in your sig.
Then old guys like me could remember where to look.

Can you please navigate me to the section where I can write my signature?
PS: 5.4.8 is out

Step 1:

Step 2: