ChessGenius Exclusive v Star diamond 10 Game Match

This forum is for general discussions and questions, including Collectors Corner and anything to do with Computer chess.

Moderators: Harvey Williamson, Steve B, Watchman

Forum rules
This textbox is used to restore diagrams posted with the fen tag before the upgrade.
Post Reply
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

ChessGenius Exclusive v Star diamond 10 Game Match

Post by Yarc »

Following on from my MCGE v Magellan match, I have decided to carry out another unfair match. This time pitting my MCGE against the Star diamond(SD). It's unfair because of the processing power difference, but interesting nonetheless because these older machines still manage a surprise or two. The games also highlight possible weaknesses in the new Exclusive machine.

The strongest machines in my small collection are the Millennium ChessGenius machines, the Magellan and the Star diamond. I'm not interested in watching a match between the same program, and the other dedicated machines in my collection are much weaker so this might not be so interesting. Maybe the Saitek Cosmos(2020ELO) might manage a draw against the MCGE, but the Excalibur Grandmaster at around 1800ELO is probably stretching the possibility. Maybe a handicapped match is more in order for these machines.

As with the match against the Magellan, this is ten games at 40m/2h using the Champion book. I have not tried the London book with the Exclusive, but suspect against other machines, this would yield more points. This seems to be the case with the MCG Pro, and since they are using the same engine it seems a fair assumption. Maybe with the Exclusive's extra processing grunt, this difference is smaller?

Anyway, here is the first game with them playing the Vienna opening. The first thing I notice with the SD is because of it's reasonably large book of 128,000 moves the openings are followed much further than with the Magellan which I believe has a repertoire in the order of 50,000.

[Event "40m/2h Champion Book"]
[Round "1"]
[White "MCGE"]
[Black "SD"]
[Result "1-0"]
[ECO "C28"]

1.e4 e5 2.Bc4 Nf6 3.d3 Nc6 4.Nc3 Bb4 5.Nf3 d6 6.O-O Bxc3 7.bxc3 Na5 8.Bb3 Nxb3 9.axb3 O-O 10.Bg5 h6 11.Bh4 Qe7 12.Nd2 Bd7 13.Qf3 Rfe8 14.Bxf6 Qxf6 15.Qxf6 gxf6 16.Nc4 Kg7 17.f4 Bg4 18.f5 Be2 19.Rfe1 Bh5 20.h3 d5 21.exd5 b5 22.Ne3 c6 23.dxc6 a5 24.c7 Rec8 25.Nd5 Ra7 26.g4 Raxc7 27.Nxc7 Rxc7 28.Rxa5 Rxc3 29.Re2 Bg6 30.fxg6 fxg6 31.Rxb5 h5 32.g5 fxg5 33.Rbxe5 1-0

Final position:
[fen]8/6k1/6p1/4R1pp/8/1PrP3P/2P1R3/6K1 w - - 0 33[/fen]

A hopeless situation for black and it resigned. I've said it before but I wish my other machines offered resignation. Anyway, a short game and a good victory for the MCGE which pounced on the SD's inferior moves.

The first curious move was by the SD in this position:
[fen]r3r1k1/pppb1p2/3p1p1p/4p3/2N1P3/1PPP4/2P2PPP/R4RK1 w - - 1 16[/fen]
16. ... Kg7?! a bit early to bring the King into play.

Another curious move by the SD was 17. ... Bg4?
[fen]r3r3/ppp2pk1/3p1p1p/4p3/2N1PPb1/1PPP4/2P3PP/R4RK1 w - - 1 18[/fen]
In response the MCGE advance it's pawn 18. f5 trapping Black's bishop on the right-hand side of the board. Things are starting to go wrong for the SD now!

At black's 22nd move we have another oddity:
[fen]r3r3/p1p2pk1/5p1p/1p1PpP1b/8/1PPPN2P/2P3P1/R3R1K1 w - - 1 22[/fen]
Here, the SD played 22. ... c6?! The MCGE lapped up this offering. Had the SD given up at this point? Well, it was all over after a few more moves! Perhaps this match was too unfair?

1-0 to the MCGE Regards
Ray
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

On to game number 2:

In this game they played the Sicilian, Dragon, Yugoslav, Rauser variation and the Stardiamond showed that it wasn't going to be a complete push-over. That said, the result was a draw.

[Event "40m/2h Champion Book"]
[Round "2"]
[White "SD"]
[Black "MCGE"]
[Result "1/2-1/2"]
[ECO "B76"]

1.e4 c5 2.Nf3 d6 3.d4 Nf6 4.Nc3 cxd4 5.Nxd4 g6 6.Be3 Bg7 7.f3 Nc6 8.Qd2 O-O 9.O-O-O d5 10.exd5 Nxd5 11.Nxc6 bxc6 12.Bd4 Bxd4 13.Qxd4 Qb6 14.Nxd5 cxd5 15.Qxd5 Be6 16.Qg5 Rfc8 17.Bd3 Bxa2 18.Qxe7 Rab8 19.Qa3 Qe3+ 20.Rd2 Be6 21.Be4 Qxa3 22.bxa3 Ba2 23.Rdd1 Rb3 24.Rhe1 Rxa3 25.Bb7 Rc7 26.Kb2 Ra5 27.Re8+ Kg7 28.Ba8 Be6 29.Be4 Rh5 30.h3 Rb5+ 31.Kc1 Rg5 32.g4 Rgc5 33.Rdd8 Bb3 34.Rd2 Ba4 35.Red8 Bc6 36.Bxc6 R5xc6 37.h4 Rc4 38.Ra8 h5 39.gxh5 gxh5 40.Rh2 Rf4 41.Rh3 Rd4 42.Rh2 Rdc4 43.Rg2+ Kh7 44.Rh2 Rf4 45.Rh3 Rd4 46.Rh2 Kg7 47.Rh3 f6 48.Rg3+ Kh7 49.Rh3 Rdc4 50.Rh2 Rf4 51.Rh3 Ra4 52.Kb2 Rb4+ 53.Kc1 Rd4 54.Rh1 Ra4 55.Kd2 Ra2 56.Rc1 Rc4 57.Kd3 Rxh4 58.c4 Ra3+ 59.Rc3 Rf4 60.Rxa3 Rxf3+ 61.Kd4 Rxa3 62.c5 Ra1 63.c6 Rc1 64.Rxa7+ Kg6 65.Kd5 h4 66.Ra8 h3 67.Rh8 Rc3 68.Kd6 Rd3+ 69.Kc5 Rc3+ 70.Kb6 Rb3+ 71.Ka6 Rc3 72.Kb5 Kg7 73.Rh4 Kg6 74.Kb6 Kg5 75.Rh8 Rb3+ 76.Ka6 Rc3 77.Kb7 Rb3+ 78.Kc8 Ra3 79.Kd7 Rd3+ 80.Kc8 Ra3 81.Rh7 Kg6 1/2-1/2

Final position:
[fen]2K5/7R/2P2pk1/8/8/r6p/8/8 w - - 30 82[/fen]
I adjourned the game at this point. They were both seeing it as a draw, and I didn't want to wait for the inevitable!

No obvious blunders in this game and it showed that it is worth carrying on with the match.

1.5 to 0.5 to the MCGE Regards
Ray
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

Game 3:

In this game they started with the Queen's Pawn opening.

[Event "40m/2h Champion Book"]
[Round "3"]
[White "MCGE"]
[Black "SD"]
[Result "1-0"]
[ECO "D02"]

1.d4 d5 2.Nf3 Nf6 3.g3 g6 4.Bg2 Nc6 5.O-O Bg7 6.Nc3 Ne4 7.Nxe4 dxe4 8.Ng5 Qxd4 9.Bxe4 Qb6 10.c3 h6 11.Nf3 Bh3 12.Re1 f5 13.Bd3 O-O-O 14.Qb3 Qxb3 15.axb3 Bg4 16.Bb5 Bxf3 17.exf3 Ne5 18.Kg2 Rd5 19.Bc4 Rdd8 20.Rxa7 Kb8 21.Be3 Nc6 22.Ra4 b6 23.Bb5 Ne5 24.Bf4 Kb7 25.Bxe5 Bxe5 26.Rxe5 Rh7 27.Rd4 Rxd4 28.cxd4 c6 29.Bc4 Kc7 30.b4 b5 31.Bb3 Kd7 32.f4 h5 33.d5 Rg7 34.dxc6+ Kxc6 35.Kh3 Kc7 36.Kh4 e6 37.Bxe6 Kb6 38.Re2 Kc6 39.Kg5 h4 40.Kf6 Ra7 41.Kxg6 hxg3 42.fxg3 1-0

Final Position:
[fen]8/r7/2k1B1K1/1p3p2/1P3P2/6P1/1P2R2P/8 w - - 0 42[/fen]

In this position the SD resigned.

It started to go wrong for the SD at move 15:
[fen]2kr3r/ppp1p1b1/2n3pp/5p2/8/1PPB1NPb/1P2PP1P/R1B1R1K1 w - - 0 15[/fen]
White had just taken back the Queen with 15 axb3 and black replied
15. .... Bg4? This would lead to the eventual loss of the Pawn on a7 with the MCGE in a strong position.

A relatively short game in which there were no obvious blunders but the SD was outplayed.

2.5 - .5 to the MCGE Regards
Ray
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

Game 4.

In this game they played the King's Indian, Four Pawns Attack. The SD manages another draw against the mighty MCGE.

[Event "40m/2h Champion Book"]
[Round "4"]
[White "SD"]
[Black "MCGE"]
[Result "1/2-1/2"]
[ECO "E76"]

1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f4 c5 6.d5 O-O 7.Nf3 a6 8.Bd3 b5 9.e5 dxe5 10.fxe5 Ng4 11.Qe2 Qc7 12.Bf4 Nd7 13.d6 exd6 14.exd6 Qa5 15.O-O bxc4 16.Bxc4 Nb6 17.Bd5 Nxd5 18.Nxd5 Qb5 19.Qxb5 axb5 20.h3 Nf6 21.Ne7+ Kh8 22.Nxc8 Rfxc8 23.Ne5 Ra7 24.Rfd1 Nd7 25.Nxf7+ Kg8 26.Nh6+ Kh8 27.Rd2 Rf8 28.Be3 Re8 29.Bg5 Bd4+ 30.Kh1 Kg7 31.Rf1 Bf6 32.Bxf6+ Nxf6 33.Ng4 Ne4 34.Rd3 Rd7 35.Kg1 Rxd6 36.Rxd6 Nxd6 37.Rd1 Nf7 38.Kf2 Re4 39.Rd5 Rb4 40.b3 c4 41.Rd7 cxb3 42.axb3 Kf8 43.Rd3 Ng5 44.Ne5 Ne4+ 45.Ke2 Nc5 46.Rd8+ Kg7 47.Nd3 Nxd3 48.Rxd3 Kf6 49.Re3 Rd4 50.Rc3 Ke6 51.Rc7 Rb4 52.Rxh7 Rxb3 53.Rb7 g5 54.Rb8 Ke5 55.Re8+ Kf4 56.Rf8+ Kg3 57.Rg8 Kxg2 58.h4 Kg3 59.hxg5 Kg4 60.g6 Kf5 61.Kd2 b4 62.Kc2 Rc3+ 63.Kd2 Rg3 64.g7 Kg6 65.Rb8 Kxg7 66.Rxb4 Kf6 67.Re4 Kf5 1/2-1/2

Final position:
[fen]8/8/8/5k2/4R3/6r1/3K4/8 w - - 3 68[/fen]
I declared this a draw. Both machines were consistently showing a score of 0.00 so they knew it as well.

This was a surprising game because I really thought the SD was going to push for a win.

At move 18 White really looked to be in a better position. Black's pieces seemed to be loitering around the edges!
[fen]r1b2rk1/5pbp/p2P2p1/q1pN4/5Bn1/5N2/PP2Q1PP/R4RK1 w - - 0 18[/fen]

Alas, the SD was unable to calculate its way through and a draw was the result. I have seen this in other games where the SD has a good advantage but is unable to push for the win.

3 - 1 to the MCGE Regards
Ray
User avatar
spacious_mind
Senior Member
Posts: 3999
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Hi Ray,

Since with MCGE you are supposed to be able to adjust the MHz in 50 increments, you might want to if you are still in the mood after your match is complete play MCGE 50 MHz against Star Diamond and see how the results then compare.

regards
Nick
User avatar
spacious_mind
Senior Member
Posts: 3999
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Actually the bar for MCGE to beat had already been set by Lars Sandin (SSDF):

Millenium ChessGenius ARM 48 MHz - Novag Star Sapphire, ended 13-7
Millenium ChessGenius Pro ARM 140 MHz-Novag Star Sapphire, ended 12½-7½

Same time level as what you are playing.

Regards
Nick
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

Game 5.

The MCGE decides it doesn't like just drawing and gains another win, but not without making some questionable moves.

They opened with the Queen's Gambit Accepted, Janowsky-Larsen variation.

[Event "40m/2h Champion Book"]
[Round "5"]
[White "MCGE"]
[Black "SD"]
[Result "1-0"]
[ECO "D25"]

1.d4 d5 2.c4 dxc4 3.Nf3 Nf6 4.e3 Bg4 5.Bxc4 e6 6.h3 Bh5 7.Qb3 Bxf3 8.gxf3 Nbd7 9.Qxb7 c5 10.dxc5 Bxc5 11.Rg1 Rb8 12.Qc6 Rb6 13.Qa4 O-O 14.Qc2 Ne5 15.Ke2 Bb4 16.b3 Qa8 17.e4 Rd8 18.Be3 Rbd6 19.Qb2 Nxf3 20.Kxf3 Qxe4+ 21.Ke2 Nh5 22.Rg4 Qh1 23.Nc3 Qxh3 24.Rag1 f5 25.Rxg7+ Nxg7 26.Nd5 Qh5+ 27.Kf1 Qd1+ 28.Kg2 Be1 29.Nf4 R8d7 30.Kh2 Qf3 31.Rxe1 Qe4 32.Rg1 Rf7 33.Qc3 h6 34.f3 Qxf3 35.Qe5 Kh7 36.Qxd6 Qxe3 37.Rg2 Qe4 38.Re2 Rd7 39.Qxd7 Qxf4+ 40.Kg1 Qg3+ 41.Kf1 Qf4+ 42.Ke1 Qc1+ 43.Qd1 Qf4 44.Qd2 Qh4+ 45.Kd1 Qe7 46.Qe3 Qd6+ 47.Kc2 a5 48.Bxe6 a4 49.Qa7 axb3+ 50.axb3 Qc6+ 51.Bc4 h5 52.Qb8 Kg6 53.Qg3+ 1-0

Final position:
[fen]8/6n1/2q3k1/5p1p/2B5/1P4Q1/2K1R3/8 w - - 3 53[/fen]
The SD resigned, which is just as well!

At move 14 White (MCGE) has just moved its Queen from a4:
[fen]3q1rk1/p2n1ppp/1r2pn2/2b5/2B5/4PP1P/PPQ2P2/RNB1K1R1 w Q - 7 14[/fen]
Not the best position to be in with underdeveloped Knight and Bishop. The King exposed, and the King-side pawns in disarray! But it is still early in the game so anything could happen.

Then, at move 15 the MCGE makes another questionable move:
[fen]3q1rk1/p4ppp/1r2pn2/2b1n3/2B5/4PP1P/PPQ2P2/RNB1K1R1 w Q - 8 15[/fen]
Here it moves 15.Ke2? To protect the f pawn, but why not develop a Knight to protect it!? The MCGE likes to live dangerously!

At move 19 it's looking very dicey for the MCGE:
[fen]q2r2k1/p4ppp/3rpn2/4n3/1bB1P3/1P2BP1P/PQ2KP2/RN4R1 w - - 4 19[/fen]
The SD looks strong with its pieces poised for attack. On the other hand, the MCGE is looking weak, submissive. How did it win this game?

Then, at move 22:
[fen]3r2k1/p4ppp/3rp3/7n/1bB1q1R1/1P2B2P/PQ2KP2/RN6 w - - 3 22[/fen]
The SD decides to move 22. .... Qh1? Was it thinking it could then move to e1? White simply replied moving its Knight to c3! The Black Queen is now under attack, Black's Bishop on b4 has its path blocked and finally White is developing an important piece!

After Black takes the h3 pawn and White moves its Rook to g1, the MCGE is now looking in a much stronger position. This is a very interesting game:
[fen]3r2k1/p4ppp/3rp3/7n/1bB3R1/1PN1B2q/PQ2KP2/6R1 w - - 1 24[/fen]

And then at move 28 we have a blunder by the SD, what a shame!
[fen]3r2k1/p5np/3rp3/3N1p2/1bB5/1P2B3/PQ3PK1/3q2R1 w - - 5 28[/fen]
Here the SD moved 28. .... Be1?? It really needed to worry about White's Queen on b2 and the yet to be exposed Rook on g1. Black's King is now looking very vulnerable and it was only time before the MCGE forced the SD into resignation!

Not a perfect game for the MCGE although in the end it pounced on the SD's inferior moves and managed to turn the game around.

4 - 1 to the MCGE Regards
Ray
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

spacious_mind wrote:Hi Ray,

Since with MCGE you are supposed to be able to adjust the MHz in 50 increments, you might want to if you are still in the mood after your match is complete play MCGE 50 MHz against Star Diamond and see how the results then compare.

regards
Hi spacious_mind,

This would be interesting. My sense is that the MCGE would not win, but then I would also probably need to use the London Book option. In my experience with the MCGP, this played better against the SD with the London book.

Regards
Ray
User avatar
spacious_mind
Senior Member
Posts: 3999
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Yarc wrote:
spacious_mind wrote:Hi Ray,

Since with MCGE you are supposed to be able to adjust the MHz in 50 increments, you might want to if you are still in the mood after your match is complete play MCGE 50 MHz against Star Diamond and see how the results then compare.

regards
Hi spacious_mind,

This would be interesting. My sense is that the MCGE would not win, but then I would also probably need to use the London Book option. In my experience with the MCGP, this played better against the SD with the London book.

Regards
Ray
Yes the curiosity behind this is that MCG 48 MHz beat Star Sapphire at the same time settings that you are playing quite comfortably and SS and SD are pretty much identical other than one is travel and the other desktop.

So theoretically speaking MCGE should at 50 MHz have similar results. At 150 MHz MCGE should have similar results as MCGP (140 MHz).

Yep I am playing in my Division 2 the new book and I am not really that impressed with it either for computer vs computer matches. I also tend to think that the original London book was probably better configured to the London style of play and possibly better in computer vs computer matches.

Regards
Nick
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

spacious_mind wrote: Yes the curiosity behind this is that MCG 48 MHz beat Star Sapphire at the same time settings that you are playing quite comfortably and SS and SD are pretty much identical other than one is travel and the other desktop.

So theoretically speaking MCGE should at 50 MHz have similar results. At 150 MHz MCGE should have similar results as MCGP (140 MHz).

Yep I am playing in my Division 2 the new book and I am not really that impressed with it either for computer vs computer matches. I also tend to think that the original London book was probably better configured to the London style of play and possibly better in computer vs computer matches.

Regards
We also have to consider that my 10 game matches are not enough to get a very accurate result. I think because of the vast number of book openings in the Champion Book (300,000) and the 128,000 in the SD, there is going to be a large random variation in possible results depending on what particular variations are followed and how each program is able to handle them. What I have considered, but not yet undertaken, is a match with no opening books. This might then reduce the possible variation in results and give a better idea of performance between the machines.

More is not always better.

Regards
Ray
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

Game 6.

It's beginning to look very one-sided, which is to be expected. Here we have another win for the MCGE. Neither machine made obvious blunders and I feel that the SD's inability to calculate as far ahead as the MCGE is probably the main cause for it to have lost.

Opening: Queen's Gambit Declined, Semi-Slav, Bogolyubov variation.

[Event "40m/2h Champion Book"]
[Round "6"]
[White "SD"]
[Black "MCGE"]
[Result "0-1"]
[ECO "D46"]

1.d4 Nf6 2.c4 c6 3.Nc3 d5 4.e3 e6 5.Nf3 Nbd7 6.Bd3 Be7 7.O-O dxc4 8.Bxc4 b5 9.Bd3 b4 10.Ne4 Bb7 11.Nxf6+ Nxf6 12.Ne5 O-O 13.Qf3 Qc7 14.Ng4 Nxg4 15.Qxg4 c5 16.dxc5 Qxc5 17.Qh3 h6 18.Rd1 Rfd8 19.Qg4 Qg5 20.Qxg5 Bxg5 21.Rd2 Rac8 22.h3 Rd5 23.f4 Bh4 24.g3 Bxg3 25.Kf1 Rh5 26.Rd1 Rxh3 27.e4 Bxf4 28.Bxf4 Rf3+ 29.Ke2 Rxf4 30.Ke3 Rg4 31.a3 bxa3 32.Rxa3 Rd8 33.e5 Rd5 34.Rxa7 Rxe5+ 35.Kd2 Rg2+ 36.Kc3 Rc5+ 37.Kb4 Rc8 38.b3 Rb8 39.Kc5 Bd5 40.Rda1 Rc8+ 41.Kd4 Kf8 42.Ke3 Rc3 43.Kd4 Rxb3 44.Rc1 Rg4+ 45.Ke5 Bb7 46.Be2 Re3+ 47.Kd6 Rd4+ 48.Kc7 Rxe2 49.Rxb7 g5 50.Kc6 Kg7 51.Re7 h5 52.Rf1 Rc2+ 53.Kb5 Rf4 54.Re1 Kf6 55.Ra7 h4 56.Re3 g4 0-1

Final position when SD resigned:
[fen]8/R4p2/4pk2/1K6/5rpp/4R3/2r5/8 w - - 0 57[/fen]

The result so far is showing how good the MCGE is, but it's also showing that it's not infallible. There is fight left yet in the SD, watch this space!

5 - 1 to the MCGE Regards
Ray
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

Game 7.

The next game was the second longest in the match at 83 moves and ended up a draw. This is the 3rd draw so far and I think showing just how good the SD is even against such strong opposition.

They played the Ruy Lopez, Closed variation.

[Event "40m/2h Champion Book"]
[Round "7"]
[White "MCGE"]
[Black "SD"]
[Result "1/2-1/2"]
[ECO "C84"]

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.d3 b5 6.Bb3 Be7 7.O-O Bb7 8.Nc3 d6 9.Nd5 Na5 10.Nxe7 Qxe7 11.Bg5 Nxb3 12.axb3 h6 13.Bh4 g5 14.Bg3 Nh5 15.Nd2 Nxg3 16.fxg3 a5 17.Qf3 O-O 18.Qf6 Qxf6 19.Rxf6 Kg7 20.Raf1 a4 21.b4 a3 22.bxa3 Rxa3 23.Nb1 Ra2 24.R6f2 Rb2 25.c3 Rxf2 26.Kxf2 f5 27.Ke3 fxe4 28.Rxf8 Kxf8 29.dxe4 Kf7 30.g4 Bc8 31.h3 Be6 32.Nd2 Bc4 33.Nxc4 bxc4 34.g3 Ke6 35.Kd2 c6 36.Ke2 d5 37.Ke3 Kd6 38.Kf3 d4 39.Ke2 d3+ 40.Kd2 c5 41.bxc5+ Kxc5 42.Kd1 Kd6 43.Ke1 Kd7 44.Kd1 Kc6 45.Kd2 Kc7 46.Ke3 Kd8 47.Kd2 Ke7 48.Ke3 Kd6 49.Kd2 Kc6 50.Kd1 Kc5 51.Kc1 Kb5 52.Kb1 Kc6 53.Kc1 Kc7 54.Kd2 Kd8 55.Ke3 Kd7 56.Kd2 Ke6 57.Ke3 Ke7 58.Kd2 Kf7 59.Ke3 Kg7 60.Kd2 Kf6 61.Ke3 Kg6 62.Kd2 h5 63.gxh5+ Kxh5 64.Ke3 Kg6 65.Kd2 Kf7 66.Ke3 Kf6 67.Kd2 Ke6 68.Ke3 Kf7 69.Kd2 Ke7 70.Ke1 Kf8 71.Kd2 Ke8 72.Kd1 Kd7 73.Kd2 Kd6 74.Ke1 Ke7 75.Kd1 Kd8 76.Kd2 Kd7 77.Ke1 Ke6 78.Kd1 g4 79.h4 Kf6 80.Ke1 Kf7 81.Kd1 Kg7 82.Ke1 Kf8 83.Kd2 1/2-1/2

Final position when I declared it a draw:
[fen]5k2/8/8/4p3/2p1P1pP/2Pp2P1/3K4/8 w - - 8 83[/fen]

This is the first game so far in both matches I have played where the MCGE blundered! However, this blunder was seen by Stockfish calculating umpteen moves ahead so I think a little unfair to judge the MCGE on this!

Here is the position:
[fen]8/2p2k2/3p3p/1p2p1p1/1Pb1P1P1/2P1K2P/3N2P1/8 w - - 3 33[/fen]
At move 33 the MCGE as white played 33. Nxc4 and black replied by taking the Knight. Stockfish shows a very large winning score for Black, but neither of these machines can see that!

A few moves later at move 42 we have this position with Black to move:
[fen]8/8/7p/2k1p1p1/2p1P1P1/2Pp2PP/8/3K4 w - - 1 42[/fen]

If the SD were able to see far enough ahead it would start moving its King around b5, a4, a3 etc. White would lose the Pawn on c3 and ultimately Black would get its c pawn to the queening square. But the SD cannot see this and and the game ends up with much toing and froing until I call it a draw.

An almost win for the SD!

5.5 - 1.5 to the MCGE Regards
Ray
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

Game 8.

Not content with 3 draws, the machines produce another one! Watch out MCGE, the SD is upsetting your rating!

In this game they played the Queen's Gambit Declined, Semi-Slav.

[Event "40m/2h Champion Book"]
[Round "8"]
[White "SD"]
[Black "MCGE"]
[Result "1/2-1/2"]
[ECO "D45"]

1.d4 d5 2.c4 c6 3.Nc3 Nf6 4.e3 e6 5.Nf3 Bd6 6.Bd3 dxc4 7.Bxc4 b5 8.Bd3 Nbd7 9.Ne4 Nxe4 10.Bxe4 Bb7 11.Qc2 Qb6 12.Bxh7 c5 13.Be4 cxd4 14.Bxb7 Qxb7 15.exd4 Rc8 16.Qe2 Qd5 17.O-O a6 18.Bd2 O-O 19.Bg5 Nb6 20.Rfe1 Rc7 21.b3 Rc3 22.Qd2 Rfc8 23.Rab1 Qf5 24.Rbc1 Bb4 25.Qf4 Rxf3 26.gxf3 Qxf4 27.Bxf4 Bxe1 28.Rxe1 Rc2 29.a3 Rc3 30.Rb1 Nd5 31.Bd6 Rxf3 32.Kg2 Rd3 33.Bc5 Nf4+ 34.Kf1 Rh3 35.Kg1 Ne2+ 36.Kf1 Nc3 37.Kg2 Rxh2+ 38.Kxh2 Nxb1 39.Bb4 a5 40.Bxa5 Nxa3 41.Bb4 Nc2 42.Bc5 b4 43.Kg3 Na1 44.Bxb4 Nxb3 45.Bc3 Kf8 46.Kf4 Ke7 47.Ke5 Nc1 48.d5 Nd3+ 49.Ke4 Nxf2+ 50.Kf3 Nd3 51.dxe6 f6 52.Ke4 Nc5+ 53.Kf5 Nb7 54.Bd2 Nd6+ 55.Kg6 Kxe6 56.Kxg7 f5 57.Bf4 Nc8 58.Kg6 Ne7+ 59.Kh5 Kd5 60.Kh4 Ke4 61.Bd6 Nc6 62.Kg3 Nd4 63.Bb8 Ne2+ 64.Kg2 Nc3 65.Kf2 Nd5 66.Bd6 Nf6 67.Bc7 Nd7 68.Bd6 Nb6 69.Kg3 Nc4 70.Bc7 Ne3 71.Kf2 Nd1+ 72.Kg3 Nb2 73.Kf2 Nd3+ 74.Kg3 Nc5 75.Bf4 Ne6 76.Bb8 Nd4 77.Kf2 Nc6 78.Bd6 Na7 79.Ke2 Nc8 1/2-1/2

Final position which I declared as drawn:
[fen]2n5/8/3B4/5p2/4k3/8/4K3/8 w - - 46 80[/fen]

Just a long hard battle, the SD fighting off the MCGE, and the latter not really shining.

Current score 6 - 2 to the MCGE Regards
Ray
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

Game 9.

The MCGE feeling a little disappointed with its performance decides it's time to win.

Here, the machines played the Queen's Indian, Nimzowitsch variation:

[Event "40m/2h Champion Book"]
[Round "9"]
[White "MCGE"]
[Black "SD"]
[Result "1-0"]
[ECO "E15"]

1.d4 Nf6 2.c4 e6 3.Nf3 b6 4.g3 Ba6 5.b3 Bb4+ 6.Bd2 Be7 7.Bg2 c6 8.Bc3 O-O 9.O-O d5 10.Ne5 Nfd7 11.Nxd7 Nxd7 12.Nd2 Qc7 13.Rc1 Ba3 14.Rc2 Rae8 15.e3 Nf6 16.e4 Rd8 17.e5 Nd7 18.Nb1 Be7 19.cxd5 cxd5 20.Bb4 Qxc2 21.Qxc2 Bxb4 22.Rc1 Rc8 23.Qb2 Rxc1+ 24.Qxc1 Rc8 25.Qd1 f6 26.f4 fxe5 27.fxe5 h6 28.Bh3 Nf8 29.Bg4 Be7 30.h4 Bb4 31.Kf2 Rc6 32.Ke3 Rc7 33.Be2 Bxe2 34.Kxe2 Kf7 35.Kd3 Kg8 36.a3 Be7 37.Nc3 Bxa3 38.Nb5 Rc1 39.Qd2 Ra1 40.Nxa7 Rg1 41.Qf2 Rd1+ 42.Ke2 Ra1 43.Kf3 Rc1 44.Qd2 Rf1+ 45.Kg2 Rb1 46.Qd3 Rc1 47.Qb5 Rc2+ 48.Kh3 g5 49.hxg5 hxg5 50.Nc6 Rc1 51.Qxb6 Re1 52.Qd8 Rc1 53.Qxg5+ Kf7 54.Nd8+ Ke8 55.b4 Rh1+ 56.Kg2 Rb1 57.Nxe6 Rb2+ 58.Kh3 Nxe6 59.Qg6+ Kd8 60.Qxe6 Bxb4 61.Qxd5+ 1-0

Final position when the SD resigned:
[fen]3k4/8/8/3QP3/1b1P4/6PK/1r6/8 w - - 0 61[/fen]

It's very interesting watching these games and quite pleasurable when I am able to pick up on little things that these machines miss. For example, in this position at move 15
[fen]4rrk1/p1qn1ppp/bpp1p3/3p4/2PP4/bPB3P1/P1RNPPBP/3Q1RK1 w - - 6 15[/fen]
White(MCGE) played 15. e3. This move is okay but if only it had seen that by moving 15. b4 it would have trapped Black's Bishop on a3!

The MCGE just outplayed the SD in this game fair and square.

Score so far 7 - 2 to the MCGE Regards
Ray
User avatar
Yarc
Senior Member
Posts: 1051
Joined: Sat Apr 11, 2009 8:13 am
Location: United Kingdom

Post by Yarc »

Game 10.

Here is the final game of the match, and the longest at 85 moves. The SD manages another draw making it 5 in total!

In this game they opened with the Ruy Lopez, Cozio Defence deferred.

[Event "40m/2h Champion Book"]
[Round "10"]
[White "SD"]
[Black "MCGE"]
[Result "1/2-1/2"]
[ECO "C70"]

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nge7 5.c3 g6 6.d4 exd4 7.cxd4 Bg7 8.Nc3 d5 9.exd5 Nxd5 10.Qb3 Nxc3 11.Bxc6+ bxc6 12.Qxc3 O-O 13.O-O Qd6 14.Re1 Be6 15.Ne5 Rfd8 16.Nxc6 Bxd4 17.Nxd4 Qxd4 18.Qxc7 Rac8 19.Qf4 Rc2 20.Qxd4 Rxd4 21.b3 Rd5 22.Bh6 Rb2 23.h4 Rh5 24.Bg5 h6 25.Bf6 Rd2 26.Red1 Rhd5 27.Rxd2 Rxd2 28.Kf1 Bg4 29.a3 Rd6 30.Be5 Rd3 31.b4 Rb3 32.f3 Bf5 33.Ke2 Bd7 34.Kd2 f5 35.Bc3 Ba4 36.Kd3 g5 37.h5 f4 38.Kd2 Kf7 39.Be5 Be8 40.Bd4 Bb5 41.Bh8 Bd7 42.Ke1 Re3+ 43.Kf2 Kg8 44.Bf6 Be8 45.Bb2 Rd3 46.Ke1 Rb3 47.Bd4 Bxh5 48.Kd2 Be8 49.Bf6 g4 50.Bc3 gxf3 51.gxf3 Kh7 52.a4 h5 53.Kc2 Bf7 54.Rg1 Kh6 55.Bg7+ Kh7 56.Be5 Be6 57.Rg7+ Kh6 58.Ra7 Kg5 59.Rxa6 Bc4 60.b5 Rxf3 61.Ra5 Bd5 62.b6 Be4+ 63.Kd2 Rb3 64.Bc7+ Kg4 65.Re5 Bc6 66.Re6 Bxa4 67.Rf6 f3 68.Rf4+ Kh3 69.Rxa4 f2 70.Ra1 Kg2 71.Kc2 Rb5 72.Kd3 f1=Q+ 73.Rxf1 Kxf1 74.Kc4 Rb2 75.Kd5 h4 76.Kc6 h3 77.b7 Kg2 78.b8=R Rc2+ 79.Kd7 Rd2+ 80.Ke6 Re2+ 81.Kf5 Rf2+ 82.Kg4 h2 83.Bxh2 Kxh2 84.Rd8 Kg2 85.Re8 Rf7 1/2-1/2

Final position when I declared it a draw:
[fen]4R3/5r2/8/8/6K1/8/6k1/8 w - - 4 86[/fen]

This is another game where the SD playing White had the upper hand, and for much of the game. The MCGE seemed to be struggling to fend off the SD's moves. I can only think that this opening was not to the MCGE's liking.

Despite the SD having a lower ELO rating than the Magellan, it has managed to score the same amount of points against the MCGE with 7.5 to 2.5. This is very good result for the SD even though it didn't win a single game. Of course, we could say that the MCGE did not lose a single game in this match.

The MCGE's score of 7.5 out of 10 gives it a ELO rating difference of about 191. I always used the chess ratings from Eric Hallsworth's excellent Selective Search magazine, however in my post for the MCGE v Magellan I quoted from Wiki-Elo which gives the ratings at tournament controls. I think Eric's results from Selective Search are an average over many games at different controls and may also include games against humans. Take your pick from these results:

Using Selective Search ratings:
MCGE v SD - MCGE Rating = 2173 + 191 = 2364 ELO
MCGE v Magellan - MCGE Rating = 2206 + 191 = 2397 ELO
Using Wiki-ELO
MCGE v SD - MCGE Rating = 2153 + 191 = 2344 ELO
MCGE v Magellan - MCGE Rating = 2228 + 191 = 2419 ELO

Average rating across both matches using Selective Search values: 2380
Average rating across both matches using Wiki-Elo values: 2382

Interesting! Using either set of figures gives near enough the same result.

I am tempted to go with Selective search ratings as this gives a ELO difference of only 33 points between the Magellan and SD, which given the current result seems better.

More games are required to establish an accurate rating, and of course the pending update to the MCGE will boost its strength, so a rematch will be required.

At the end of these matches I am happy with all three machines.

Regards
Ray
Post Reply