French for engines?

It suddenly occured to me that engine book authors assume that your engine is going to slug it out in tactical skirmishes in the Sicilian. Here's what happened to my poor engine playing versus the French. How can I edit this line so that its more favourable to the engine?


[Event "Rated game, 5m + 0s"]
[Site "Engine Room"]
[Date "2007.04.02"]
[Round "?"]
[White "z00t, Engine"]
[Black "Wolfman, Rybka 2.2 mp 32-bit"]
[Result "0-1"]
[ECO "C14"]
[WhiteElo "2479"]
[BlackElo "2603"]
[PlyCount "92"]
[EventDate "2007.04.02"]

1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. Bg5 Be7 5. e5 Nfd7 6. Bxe7 Qxe7 7. f4 O-O 8.
Nf3 c5 9. Qd2 Nc6 10. dxc5 Nxc5 11. O-O-O a6 12. a3 b5 13. Be2 b4 14. axb4 Nxb4
15. Kb1 Qc7 16. h4 Bd7 17. Nd4 Rab8 18. Rh3 Qa5 19. Re3 Rb6 20. h5 Rfb8 21. Nb3
Nxb3 22. cxb3 h6 23. Rf3 Rc8 24. Rg3 Bb5 25. Bf1 Rbc6 26. f5 exf5 27. e6 Rxe6
28. Rf3 Re5 29. Rf4 f6 30. g3 Rd8 31. Qf2 Kh8 32. Rfd4 Nc6 33. R4d2 Bxf1 34.
Qxf1 d4 35. Na4 Red5 36. Qf4 Qb5 37. Qc7 R5d7 38. Qf4 Nb4 39. Qf1 d3 40. Qf3
Nc2 41. Qf4 Rc8 42. Ka2 Rd4 43. Rxc2 dxc2 44. Qxd4 Rb8 45. Nc5 cxd1=Q 46. Qxd1
Qxc5 0-1

Originally posted by z00t
It suddenly occured to me that engine book authors assume that your engine is going to slug it out in tactical skirmishes in the Sicilian. Here's what happened to my poor engine playing versus the French. How can I edit this line so that its more favourable to the engine?


[Event "Rated game, 5m + 0s"]
[Site "Engine Room"]
[Date "2007.04.02"]
[Round "?"] ...[text shortened]... 2. Ka2 Rd4 43. Rxc2 dxc2 44. Qxd4 Rb8 45. Nc5 cxd1=Q 46. Qxd1
Qxc5 0-1

Why don't you just stop playing useless engine matches?

Originally posted by cmsMaster
Why don't you just stop playing useless engine matches?

They are not useless, the Sicilian games are of a high quality and can show you what can happen in standard theory. As I only play a few human games, I watch the engines responses to standard openings.

The top engines are usually the ones where the strong human has edited the opening book to play to that engines style.

On the human side I plan to become a rook before Christmas. What's your status on the server?

Originally posted by z00t
They are not useless, the Sicilian games are of a high quality and can show you what can happen in standard theory. As I only play a few human games, I watch the engines responses to standard openings.

The top engines are usually the ones where the strong human has edited the opening book to play to that engines style.

On the human side I plan to become a rook before Christmas. What's your status on the server?

Bishop - which means the same rank as you have right now. I plan to become a top GM by Christmas.

Originally posted by z00t
It suddenly occured to me that engine book authors assume that your engine is going to slug it out in tactical skirmishes in the Sicilian. Here's what happened to my poor engine playing versus the French. How can I edit this line so that its more favourable to the engine?


[Event "Rated game, 5m + 0s"]
[Site "Engine Room"]
[Date "2007.04.02"]
[Round "?"] ...[text shortened]... 2. Ka2 Rd4 43. Rxc2 dxc2 44. Qxd4 Rb8 45. Nc5 cxd1=Q 46. Qxd1
Qxc5 0-1

I've drawn a lot vs. the French in there. It's funny because I've seen my book go up to 50 moves deep in the Sicilian and 5 in the French.

Imagine an engine vs human GM event for $1m and the human found such holes in the engines book? This is like Kasparov's famous 1. e3 see the article

http://www.chessbase.com/support/support.asp?pid=20

Originally posted by Falco Lombardi
I've drawn a lot vs. the French in there. It's funny because I've seen my book go up to 50 moves deep in the Sicilian and 5 in the French.

50 move book = waste of time.

I would have thought the exchange variation would be best for engines. It avoids most of positionally nuanced play that is present in a lot of other french variations.

No, the exchange variation scores pretty badly for engines too.

Originally posted by exigentsky
No, the exchange variation scores pretty badly for engines too.

I really can't see how a perfectly symmetrical position with initiative can be bad for white ... drawish perhaps, but not bad

.. of course this comes from someone who last had experience with the french roughly a decade ago so perhaps it does play out badly in practise 😉

[Event "Rated game, 5m + 0s"]
[Site "Engine Room"]
[Date "2007.04.03"]
[Round "?"]
[White "Falco_Lombardi, Rybka 2.3.1 32-bit"]
[Black "Sternenozean, Rybka 2.2 mp 32-bit"]
[Result "1-0"]
[ECO "C18"]
[WhiteElo "2591"]
[BlackElo "2635"]
[Annotator "0.64;0.38"]
[PlyCount "117"]
[EventDate "2007.04.03"]
[TimeControl "300"]

{Rybka 2.2 mp 32-bit: 15.7 ply; 187kN/s AMD Athlon(tm) 64 X2 Dual Core
Processor 5200+ 2613MHz, (2 threads)} 1. e4 {0} e6 {0.00/0 0} 2. d4 {0} d5 {
0.00/0 0} 3. Nc3 {0} Bb4 {0.00/0 0} 4. e5 {0} c5 {0.00/0 0} 5. a3 {0} Bxc3+ {
0.00/0 0} 6. bxc3 {0} Ne7 {0.00/0 0} 7. Qg4 {0} O-O {0.00/0 0} 8. Bd3 {0} Nbc6
{0.00/0 0} 9. Qh5 {0} Ng6 {0.00/0 0} 10. Nf3 {0} Qc7 {0.00/0 0} 11. Be3 {0}
Nce7 {0.00/0 0} 12. h4 {0} Bd7 {0.00/0 0} 13. Qg4 {0} f5 {0.00/0 0} 14. Qh3 {0}
cxd4 {0.38/12 10} 15. cxd4 {0} Qc3+ {0.54/13 8} 16. Ke2 {0.64/11 2} h5 {
0.58/14 5} 17. Bd2 {0.80/13 10} Qc6 {(Qc7) 0.57/14 20} 18. Rhc1 {
(Ta1-b1) 0.65/11 19} f4 {0.60/13 10} 19. Kf1 {0.85/13 9} Nf5 {0.68/13 6} 20.
Kg1 {0.90/12 3} Qa4 {(b5) 0.75/12 6} 21. Rab1 {1.17/12 10} Rac8 {(b6) 0.92/12 1
} 22. Rxb7 {1.44/10 3} Rb8 {1.15/14 10} 23. Rcb1 {1.59/13 1} Rxb7 {1.25/12 3}
24. Rxb7 {1.86/13 5} Bc8 {1.38/14 5} 25. Rb3 {2.00/13 2} Qc6 {(Ba6) 1.77/14 15}
26. Ng5 {(Ld2-b4) 2.36/12 8} Nge7 {(Qd7) 1.78/13 5} 27. Bb4 {2.45/12 5} Re8 {
(Qe8) 1.96/13 0} 28. Qf3 {(Lb4-c5) 2.61/12 5} g6 {1.95/14 4} 29. Qxf4 {
2.63/11 1} Ba6 {(Qd7) 2.06/13 4} 30. Rc3 {2.85/13 7} Bc4 {2.11/15 0} 31. Ne4 {
2.91/13 7} dxe4 {2.17/15 0} 32. Rxc4 {3.05/13 4} Qd5 {2.39/15 8} 33. Qxe4 {
3.12/15 5} a5 {2.49/14 0} 34. Bc3 {3.30/14 14} Qxe4 {2.61/15 19} 35. Bxe4 {
3.38/18 18} Nxh4 {2.90/18 0} 36. Bxa5 {3.68/15 2} Nhf5 {3.12/14 3} 37. c3 {
3.84/14 3} Nd5 {(Rb8) 3.09/15 2} 38. a4 {3.90/14 6} h4 {3.19/15 4} 39. Bb4 {
4.27/15 3} Ra8 {(Nfe7) 3.19/15 0} 40. a5 {4.46/12 2} Ra6 {3.34/14 1} 41. Bxd5 {
(Tc4-c8+) 4.90/15 3} exd5 {3.52/14 1} 42. Rc7 {5.13/16 20} g5 {(Ng7) 3.98/19 0}
43. Rd7 {(Tc7-b7) 5.13/13 3} g4 {3.55/16 5} 44. Rxd5 {(Td7-b7) 5.46/15 4} Kg7 {
(Kf7) 3.63/14 3} 45. Rb5 {(Td5-d8) 5.37/12 3} Ra8 {3.95/15 7} 46. Rb6 {
5.96/15 14} Nh6 {(Kh7) 4.13/16 9} 47. a6 {(Tb6-f6) 6.26/8 1} h3 {
(Nf5) 4.13/13 3} 48. f3 {(Kg1-h2) 6.52/9 2} gxf3 {4.11/11 3} 49. gxh3 {
7.01/12 0} Nf7 {4.13/14 5} 50. c4 {(Kg1-f2) 5.77/12 11} Ng5 {3.44/13 2} 51. Kf2
{6.94/11 0} Kh7 {(Nxh3+) 4.13/13 5} 52. Be1 {(h3-h4) 6.39/10 2} Rf8 {4.13/12 4}
53. Ke3 {(a6-a7) 6.77/13 3} f2 {(Nxh3) 4.13/14 2} 54. Bxf2 {7.14/9 1} Rf3+ {
(Nxh3) 4.13/14 1} 55. Ke2 {4.63/8 0} Rxh3 {6.09/17 16} 56. Be3 {8.45/15 0} Rh2+
{8.26/17 4} 57. Kd3 {9.71/15 7} Rg2 {(Nh3) 8.78/18 5} 58. d5 {9.21/10 2} Nf3 {
9.36/16 0} 59. e6 {
Sternenozean,Rybka 2.2 mp 32-bit gibt auf (Lag: Av=0.29s, max=1.0s) 11.39/10 2
} 1-0

What white should have done is go for 12. Bd3 rather than the 12. a3 which is an error. There are GM games such as Short-Korchnoi so that is probably an oversight from the book author.

1.e4 e6
White resigns

No need to prolong your suffering. 🙂