English Closed System- which chess engine best for analysis

[LaurenceChen]

[d] r1bq1rk1/1pp1npb1/3p1npp/p3p3/2P5/P1NP2P1/1P1NPPBP/1RBQ1RK1 w - - 0 11
In this English closed position... the book move is b4. Are there any chess engines able to understand how to play maneuvering chess in closed position such as above?

[LaurenceChen]

This is the analysis by Toga II 1.4 beta 5c

New game - Toga II 1.4 beta5c
r1bq1rk1/1pp1npb1/3p1npp/p3p3/2P5/P1NP2P1/1P1NPPBP/1RBQ1RK1 w - - 0 1

Analysis by Toga II 1.4 beta5c:

1. = (0.10): 11.Nde4 Nxe4 12.Nxe4 Nf5 13.Kh1 Bd7 14.f4 Bc6 15.fxe5 Bxe5 16.Nc3 Bxg2+ 17.Kxg2 h5 18.e4 Nd4 19.Be3 Re8 20.Qa4 c6 21.Kh1
2. = (0.11): 11.Qb3 Nc6 12.Nf3 Bd7 13.Bd2 Qb8 14.Be3

(, 08.05.2009)

[Terry McCracken]

[d] r1bq1rk1/1pp1npb1/3p1npp/p3p3/2P5/P1NP2P1/1P1NPPBP/1RBQ1RK1 w - - 0 11
In this English closed position... the book move is b4. Are there any chess engines able to understand how to play maneuvering chess in closed position such as above?

How about Rybka 3?

[AdminX]

Yes, Rybka 3 will see it. But ...

Analysis by Rybka 3:

1.Nf3 c6 2.Qb3
= (0.06) Depth: 6 00:00:00 7kN

1.Nde4 Nxe4 2.Nxe4[] c6 3.Bd2 Qc7 4.Nc3 Nf5 5.e3 Be6 6.Ne4 Rab8 7.Qa4
= (0.14) Depth: 6 00:00:00 10kN
= (0.06) Depth: 14 00:00:04 760kN

1.b3 Nf5 2.e3 c6 3.Nde4 Be6 4.a4 Ne8 5.Ba3 Ne7 6.Nd2 f5 7.Qc2 Nf6
= (0.10) Depth: 14 00:00:08 1310kN
= (0.08) Depth: 17 00:00:27 4146kN

1.b4 axb4 2.axb4[] c6 3.b5 Be6 4.Qb3 Qd7 5.bxc6 bxc6 6.Nde4 Nxe4 7.Nxe4 Nf5 8.e3 Qc7 9.Qb6 Qd7 10.Bd2
= (0.12) Depth: 17 00:01:14 11056kN
= (0.12) Depth: 17 00:01:14 11056kN

So will Zappa Mexico

Analysis by Zappa Mexico II:

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nexd5 4.Nxd5 Nxd5
= (0.00) Depth: 7/24 00:00:01 83kN

1.Nb3 Nf5 2.Bd2 a4 3.Na1 c6 4.Nxa4 d5
= (0.07) Depth: 7/28 00:00:01 226kN
= (-0.10) Depth: 8/28 00:00:01 298kN

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nfxd5 4.Nxd5 Nxd5
= (-0.06) Depth: 8/30 00:00:01 375kN
= (-0.14) Depth: 10/35 00:00:02 1546kN

1.Nb3 c6 2.Bd2 a4 3.Nc1 d5 4.cxd5 cxd5
= (-0.13) Depth: 10/35 00:00:03 2336kN

1.Qb3 Nf5 2.Qa2 c6 3.b4 axb4 4.Rxb4 Qe7 5.Rb1 Rd8 6.Nde4 Nxe4 7.Bxe4
= (-0.07) Depth: 10/35 00:00:04 3920kN
= (-0.07) Depth: 15/49 00:00:59 111mN

1.b4 axb4 2.Rxb4 d5 3.Qc2 d4 4.Nce4 Nxe4 5.Nxe4 Ra7 6.Qb3 b6
= (-0.05) Depth: 15/49 00:01:06 125mN
= (-0.05) Depth: 15/49 00:01:06 125mN

Note: Thanks to kgburcham, I find that Zappa seems to spot tactics much faster than Rybka in alot of these test positions.

[Terry McCracken]

Yes, Rybka 3 will see it. But ...

Analysis by Rybka 3:

1.Nf3 c6 2.Qb3
= (0.06) Depth: 6 00:00:00 7kN

1.Nde4 Nxe4 2.Nxe4[] c6 3.Bd2 Qc7 4.Nc3 Nf5 5.e3 Be6 6.Ne4 Rab8 7.Qa4
= (0.14) Depth: 6 00:00:00 10kN
= (0.06) Depth: 14 00:00:04 760kN

1.b3 Nf5 2.e3 c6 3.Nde4 Be6 4.a4 Ne8 5.Ba3 Ne7 6.Nd2 f5 7.Qc2 Nf6
= (0.10) Depth: 14 00:00:08 1310kN
= (0.08) Depth: 17 00:00:27 4146kN

1.b4 axb4 2.axb4[] c6 3.b5 Be6 4.Qb3 Qd7 5.bxc6 bxc6 6.Nde4 Nxe4 7.Nxe4 Nf5 8.e3 Qc7 9.Qb6 Qd7 10.Bd2
= (0.12) Depth: 17 00:01:14 11056kN
= (0.12) Depth: 17 00:01:14 11056kN

So will Zappa Mexico

Analysis by Zappa Mexico II:

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nexd5 4.Nxd5 Nxd5
= (0.00) Depth: 7/24 00:00:01 83kN

1.Nb3 Nf5 2.Bd2 a4 3.Na1 c6 4.Nxa4 d5
= (0.07) Depth: 7/28 00:00:01 226kN
= (-0.10) Depth: 8/28 00:00:01 298kN

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nfxd5 4.Nxd5 Nxd5
= (-0.06) Depth: 8/30 00:00:01 375kN
= (-0.14) Depth: 10/35 00:00:02 1546kN

1.Nb3 c6 2.Bd2 a4 3.Nc1 d5 4.cxd5 cxd5
= (-0.13) Depth: 10/35 00:00:03 2336kN

1.Qb3 Nf5 2.Qa2 c6 3.b4 axb4 4.Rxb4 Qe7 5.Rb1 Rd8 6.Nde4 Nxe4 7.Bxe4
= (-0.07) Depth: 10/35 00:00:04 3920kN
= (-0.07) Depth: 15/49 00:00:59 111mN

1.b4 axb4 2.Rxb4 d5 3.Qc2 d4 4.Nce4 Nxe4 5.Nxe4 Ra7 6.Qb3 b6
= (-0.05) Depth: 15/49 00:01:06 125mN
= (-0.05) Depth: 15/49 00:01:06 125mN

Note: Thanks to kgburcham, I find that Zappa seems to spot tactics much faster than Rybka in alot of these test positions.

Thanks Ted, I was only guessing as I don't have the program.

This position is easy for humans but less so for computers. More work in this area is required rather than just adding it to their books.

[michiguel]

Yes, Rybka 3 will see it. But ...

Analysis by Rybka 3:

1.Nf3 c6 2.Qb3
= (0.06) Depth: 6 00:00:00 7kN

1.Nde4 Nxe4 2.Nxe4[] c6 3.Bd2 Qc7 4.Nc3 Nf5 5.e3 Be6 6.Ne4 Rab8 7.Qa4
= (0.14) Depth: 6 00:00:00 10kN
= (0.06) Depth: 14 00:00:04 760kN

1.b3 Nf5 2.e3 c6 3.Nde4 Be6 4.a4 Ne8 5.Ba3 Ne7 6.Nd2 f5 7.Qc2 Nf6
= (0.10) Depth: 14 00:00:08 1310kN
= (0.08) Depth: 17 00:00:27 4146kN

1.b4 axb4 2.axb4[] c6 3.b5 Be6 4.Qb3 Qd7 5.bxc6 bxc6 6.Nde4 Nxe4 7.Nxe4 Nf5 8.e3 Qc7 9.Qb6 Qd7 10.Bd2
= (0.12) Depth: 17 00:01:14 11056kN
= (0.12) Depth: 17 00:01:14 11056kN

So will Zappa Mexico

Analysis by Zappa Mexico II:

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nexd5 4.Nxd5 Nxd5
= (0.00) Depth: 7/24 00:00:01 83kN

1.Nb3 Nf5 2.Bd2 a4 3.Na1 c6 4.Nxa4 d5
= (0.07) Depth: 7/28 00:00:01 226kN
= (-0.10) Depth: 8/28 00:00:01 298kN

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nfxd5 4.Nxd5 Nxd5
= (-0.06) Depth: 8/30 00:00:01 375kN
= (-0.14) Depth: 10/35 00:00:02 1546kN

1.Nb3 c6 2.Bd2 a4 3.Nc1 d5 4.cxd5 cxd5
= (-0.13) Depth: 10/35 00:00:03 2336kN

1.Qb3 Nf5 2.Qa2 c6 3.b4 axb4 4.Rxb4 Qe7 5.Rb1 Rd8 6.Nde4 Nxe4 7.Bxe4
= (-0.07) Depth: 10/35 00:00:04 3920kN
= (-0.07) Depth: 15/49 00:00:59 111mN

1.b4 axb4 2.Rxb4 d5 3.Qc2 d4 4.Nce4 Nxe4 5.Nxe4 Ra7 6.Qb3 b6
= (-0.05) Depth: 15/49 00:01:06 125mN
= (-0.05) Depth: 15/49 00:01:06 125mN

Note: Thanks to kgburcham, I find that Zappa seems to spot tactics much faster than Rybka in alot of these test positions.

Thanks Ted, I was only guessing as I don't have the program.

This position is easy for humans but less so for computers. More work in this area is required rather than just adding it to their books.

Toga analysis is pathetic. Nde4??? oh gee.... Black plays Nh5 or Ne8 and white loses two tempi with a further f5. Zappa analysis is not good either at all. b4 axb4 Rxb4????.

Rybka analysis is more reasonable.

Miguel

[Terry McCracken]

Yes, Rybka 3 will see it. But ...

Analysis by Rybka 3:

1.Nf3 c6 2.Qb3
= (0.06) Depth: 6 00:00:00 7kN

1.Nde4 Nxe4 2.Nxe4[] c6 3.Bd2 Qc7 4.Nc3 Nf5 5.e3 Be6 6.Ne4 Rab8 7.Qa4
= (0.14) Depth: 6 00:00:00 10kN
= (0.06) Depth: 14 00:00:04 760kN

1.b3 Nf5 2.e3 c6 3.Nde4 Be6 4.a4 Ne8 5.Ba3 Ne7 6.Nd2 f5 7.Qc2 Nf6
= (0.10) Depth: 14 00:00:08 1310kN
= (0.08) Depth: 17 00:00:27 4146kN

1.b4 axb4 2.axb4[] c6 3.b5 Be6 4.Qb3 Qd7 5.bxc6 bxc6 6.Nde4 Nxe4 7.Nxe4 Nf5 8.e3 Qc7 9.Qb6 Qd7 10.Bd2
= (0.12) Depth: 17 00:01:14 11056kN
= (0.12) Depth: 17 00:01:14 11056kN

So will Zappa Mexico

Analysis by Zappa Mexico II:

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nexd5 4.Nxd5 Nxd5
= (0.00) Depth: 7/24 00:00:01 83kN

1.Nb3 Nf5 2.Bd2 a4 3.Na1 c6 4.Nxa4 d5
= (0.07) Depth: 7/28 00:00:01 226kN
= (-0.10) Depth: 8/28 00:00:01 298kN

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nfxd5 4.Nxd5 Nxd5
= (-0.06) Depth: 8/30 00:00:01 375kN
= (-0.14) Depth: 10/35 00:00:02 1546kN

1.Nb3 c6 2.Bd2 a4 3.Nc1 d5 4.cxd5 cxd5
= (-0.13) Depth: 10/35 00:00:03 2336kN

1.Qb3 Nf5 2.Qa2 c6 3.b4 axb4 4.Rxb4 Qe7 5.Rb1 Rd8 6.Nde4 Nxe4 7.Bxe4
= (-0.07) Depth: 10/35 00:00:04 3920kN
= (-0.07) Depth: 15/49 00:00:59 111mN

1.b4 axb4 2.Rxb4 d5 3.Qc2 d4 4.Nce4 Nxe4 5.Nxe4 Ra7 6.Qb3 b6
= (-0.05) Depth: 15/49 00:01:06 125mN
= (-0.05) Depth: 15/49 00:01:06 125mN

Note: Thanks to kgburcham, I find that Zappa seems to spot tactics much faster than Rybka in alot of these test positions.

Thanks Ted, I was only guessing as I don't have the program.

This position is easy for humans but less so for computers. More work in this area is required rather than just adding it to their books.

Toga analysis is pathetic. Nde4??? oh gee.... Black plays Nh5 or Ne8 and white loses two tempi with a further f5. Zappa analysis is not good either at all. b4 axb4 Rxb4????.

Rybka analysis is more reasonable.

Miguel

Whoa! I didn't even look at Zappa closely that is terrible! Whites plan is to expand his pawns on the Queen-Side. A lot of engines don't have a clue!

No question humans are better at chess but you have to match their tactics, combinations, which is very very hard to do.

[Dr.Wael Deeb]

Yes, Rybka 3 will see it. But ...

Analysis by Rybka 3:

1.Nf3 c6 2.Qb3
= (0.06) Depth: 6 00:00:00 7kN

1.Nde4 Nxe4 2.Nxe4[] c6 3.Bd2 Qc7 4.Nc3 Nf5 5.e3 Be6 6.Ne4 Rab8 7.Qa4
= (0.14) Depth: 6 00:00:00 10kN
= (0.06) Depth: 14 00:00:04 760kN

1.b3 Nf5 2.e3 c6 3.Nde4 Be6 4.a4 Ne8 5.Ba3 Ne7 6.Nd2 f5 7.Qc2 Nf6
= (0.10) Depth: 14 00:00:08 1310kN
= (0.08) Depth: 17 00:00:27 4146kN

1.b4 axb4 2.axb4[] c6 3.b5 Be6 4.Qb3 Qd7 5.bxc6 bxc6 6.Nde4 Nxe4 7.Nxe4 Nf5 8.e3 Qc7 9.Qb6 Qd7 10.Bd2
= (0.12) Depth: 17 00:01:14 11056kN
= (0.12) Depth: 17 00:01:14 11056kN

So will Zappa Mexico

Analysis by Zappa Mexico II:

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nexd5 4.Nxd5 Nxd5
= (0.00) Depth: 7/24 00:00:01 83kN

1.Nb3 Nf5 2.Bd2 a4 3.Na1 c6 4.Nxa4 d5
= (0.07) Depth: 7/28 00:00:01 226kN
= (-0.10) Depth: 8/28 00:00:01 298kN

1.b4 axb4 2.Rxb4 d5 3.cxd5 Nfxd5 4.Nxd5 Nxd5
= (-0.06) Depth: 8/30 00:00:01 375kN
= (-0.14) Depth: 10/35 00:00:02 1546kN

1.Nb3 c6 2.Bd2 a4 3.Nc1 d5 4.cxd5 cxd5
= (-0.13) Depth: 10/35 00:00:03 2336kN

1.Qb3 Nf5 2.Qa2 c6 3.b4 axb4 4.Rxb4 Qe7 5.Rb1 Rd8 6.Nde4 Nxe4 7.Bxe4
= (-0.07) Depth: 10/35 00:00:04 3920kN
= (-0.07) Depth: 15/49 00:00:59 111mN

1.b4 axb4 2.Rxb4 d5 3.Qc2 d4 4.Nce4 Nxe4 5.Nxe4 Ra7 6.Qb3 b6
= (-0.05) Depth: 15/49 00:01:06 125mN
= (-0.05) Depth: 15/49 00:01:06 125mN

Note: Thanks to kgburcham, I find that Zappa seems to spot tactics much faster than Rybka in alot of these test positions.

Thanks Ted, I was only guessing as I don't have the program.

This position is easy for humans but less so for computers. More work in this area is required rather than just adding it to their books.

Toga analysis is pathetic. Nde4??? oh gee.... Black plays Nh5 or Ne8 and white loses two tempi with a further f5. Zappa analysis is not good either at all. b4 axb4 Rxb4????.

Rybka analysis is more reasonable.

Miguel

Whoa! I didn't even look at Zappa closely that is terrible! Whites plan is to expand his pawns on the Queen-Side. A lot of engines don't have a clue!

No question humans are better at chess but you have to match their tactics, combinations, which is very very hard to do.

The first ever to agree with you on this issue Terry
Dr.D

[MattieShoes]

My extremely crap engine:
Nf3 Nd7 h3 f5 Ra1 f4 e3 fxe3 fxe3 Nc5 d4 e4

resulting in
[d] r1bq1rk1/1pp1n1b1/3p2pp/p1n5/2PPp3/P1N1PNPP/1P4B1/R1BQ1RK1 w - - 0 7

I can't even laugh at the pathetic analysis since I'm so terrible at chess.

[Dann Corbit]

The only engine on earth that is good at analyzing openings is Rybka.[1]

In order to trust any chess engine on an opening position, an hour is a good threshold, and if you intend to actually play the move in an important game, overnight is a much better bet.

Code: Select all

Analysis from C:\q\epd\sub\english.epd   
5/8/2009 9:10:10 PM Level: 1900 Seconds
Analyzing engine: Rybka 3

1) "English Closed Syst 
    Avoid move: 
    Best move (Rybka 3): b2-b4
    Not found in: 31:40
      2	00:00	         489	500.736	+0.15	Qd1b3
      2	00:00	         619	633.856	+0.24	Qd1c2
      3	00:00	         797	816.128	+0.24	Qd1c2
      4	00:00	       1.702	108.928	+0.10	Qd1c2
      4	00:00	       2.900	185.600	+0.15	Nd2e4
      5	00:00	       4.218	134.976	+0.12	Nd2e4 Nf6xe4
      6	00:00	       7.838	127.398	+0.14	Nd2e4 Nf6xe4 Nc3xe4 Ne7f5 Bc1d2
      7	00:00	      11.370	147.378	+0.13	Nd2e4 Nf6xe4 Nc3xe4 Ne7f5 Bc1d2 Qd8e7
      8	00:00	      19.546	141.951	+0.10	Nd2e4 Nf6xe4 Nc3xe4 Ne7f5 Bc1d2 Qd8e7 Rf1e1
      8	00:00	      26.838	146.181	+0.12	Qd1b3 Ne7f5 e2e3 Rf8e8 Nd2e4
      9	00:00	      35.977	156.767	+0.07	Qd1b3 Ne7f5 e2e3 Rf8e8 Rf1e1 c7c6
     10	00:00	     106.648	170.636	+0.13	Qd1b3 Ne7f5 e2e3 c7c6 Nd2e4 Qd8e7 Ne4xf6+
     11	00:01	     129.016	172.695	+0.07	Qd1b3 Ne7f5 e2e3 c7c6 Nd2e4 Qd8e7 Ne4xf6+ Bg7xf6 Nc3e4
     11	00:01	     177.634	171.439	+0.11	Nd2e4 Nf6xe4 Nc3xe4 Ne7f5 Bc1d2 Qd8e7 e2e3 Bc8d7 Qd1e2
     12	00:01	     286.789	177.552	+0.14	Nd2e4 Nf6xe4 Nc3xe4 c7c6 Bc1d2 Bc8e6 Qd1c2 Ne7f5 e2e3 b7b6
     13	00:02	     413.448	183.356	+0.05	Nd2e4 Nf6xe4 Nc3xe4 c7c6 Bc1d2 Bc8e6 Qd1c2 Ne7f5 e2e3 b7b6 Rb1e1
     13	00:03	     698.020	194.920	+0.07	b2b3 Ne7f5 e2e3 c7c6 Nd2e4 Bc8e6 Qd1e2 d6d5 Ne4c5 Qd8e7
     14	00:05	   1.025.856	188.595	+0.09	b2b3 Ne7f5 e2e3 c7c6 Nd2e4 Bc8e6 Qd1e2 Nf6xe4 d3xe4 Nf5e7
     15	00:08	   1.508.859	184.772	+0.09	b2b3 Ne7f5 e2e3 c7c6 Nd2e4 Bc8e6 Qd1e2 Nf6xe4 d3xe4 Nf5e7 a3a4 Qd8c7 Bc1a3 Rf8e8
     16	00:25	   4.492.869	182.040	+0.08	b2b3 c7c6 Nd2e4 Nf6h7 Qd1c2 Qd8c7 Nc3a4 Ne7f5 Bc1b2 Bc8e6 Qc2d2 Rf8e8 e2e3
     16	00:36	   6.558.057	183.102	+0.10	b2b4 a5xb4 a3xb4 c7c6 b4b5 Bc8e6 Qd1b3 Rf8e8 Nc3e4 Nf6xe4 Nd2xe4 Ne7f5 e2e3 d6d5 b5xc6 b7xc6 c4xd5 Be6xd5
     17	00:51	   8.646.198	174.464	+0.12	b2b4 a5xb4 a3xb4 c7c6 b4b5 Bc8e6 Qd1b3 Qd8d7 Nd2e4 Nf6xe4 Nc3xe4 Ne7f5 e2e3 Ra8a7 b5xc6 b7xc6 Bc1b2 Ra7b7 Qb3c2
     18	01:25	  14.707.188	176.453	+0.15	b2b4 a5xb4 a3xb4 c7c6 b4b5 Bc8e6 Qd1b3 d6d5 Bc1b2 Ne7f5 Qb3b4 Qd8d6 Qb4xd6 Nf5xd6 b5xc6 b7xc6 Nc3d1 e5e4 c4xd5 c6xd5 Bb2e5
     19	02:05	  21.275.633	173.893	+0.11	b2b4 a5xb4 a3xb4 c7c6 b4b5 Bc8e6 Qd1b3 d6d5 Bc1b2 Ne7f5 Qb3b4 Qd8d6 Qb4xd6 Nf5xd6 b5xc6 b7xc6 Nc3d1 e5e4 c4xd5 c6xd5 Bb2e5 e4xd3 e2xd3 Nf6e8 Be5xg7
     20	04:05	  43.284.609	180.429	+0.11	b2b4 a5xb4 a3xb4 c7c6 b4b5 Bc8e6 Qd1b3 d6d5 Bc1b2 Ne7f5 b5xc6 b7xc6 Rf1c1 Nf5d4 Qb3d1 Qd8a5 c4xd5 c6xd5 e2e3 Nd4c6 Nc3a4 Rf8c8
     21	09:28	  98.955.469	178.472	+0.11	b2b4 a5xb4 a3xb4 c7c6 b4b5 Bc8e6 Bc1b2 d6d5 b5xc6 b7xc6 c4xd5 c6xd5 Nc3b5 Nf6d7 e2e4 Nd7b6 e4xd5 Be6xd5 Bg2xd5 Ne7xd5 Nd2c4 Qd8g5 Nc4xb6 Nd5xb6 Qd1b3 Rf8d8 Nb5c7
     22	17:29	 186.171.484	181.672	+0.14	b2b4 a5xb4 a3xb4 c7c6 b4b5 Bc8e6 Bc1b2 d6d5 b5xc6 b7xc6 c4xd5 c6xd5 Nc3b5 Nf6e8 Nd2f3 d5d4 Nf3d2 Ra8a5 Nb5a3 Ne8d6 Na3c4 Ra5a6 Qd1c2 Ne7d5 Bb2a3 Nd5c3
     23	26:23	 281.594.897	182.160	+0.11	b2b4 a5xb4 a3xb4 c7c6 b4b5 Bc8e6 Bc1b2 d6d5 b5xc6 b7xc6 c4xd5 c6xd5 Nc3b5 Nf6e8 Nd2f3 d5d4 Qd1c1 Ra8c8 Qc1d2 Ne8c7 Bb2a3 Nc7d5 Rf1c1 Rc8xc1+ Qd2xc1 Qd8d7 Ba3d6
   5/8/2009 9:37:16 PM, Time for this analysis: 00:26:59, Rated time: 31:40

0 of 1 matching moves
5/8/2009 9:37:16 PM, Total time: 12:27:06 AM
Rated time: 31:40 = 1900 Seconds

[1] Except for Crafty 20.0 with Jeremiah Pennery's "Draw Detection" code compiled in for extremely closed positions {where it has no equal}.

Just for fun, try this position with your favorite chess engine:
[d]r1bqk1nr/pppp1ppp/2n5/b7/2BpP3/2P2N2/P4PPP/RNBQ1RK1 b kq - bm Nge7;

Chess engines are really awful in gambits. Rybka is an exception. Even the evaluation numbers seem reasonable.