MCGE against the Magellan

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

Post by Yarc »

spacious_mind wrote:Playing 40/2hrs is a tough slog you are doing well there.

To not lose patience nowadays I typically stick to 60/30 (30 seconds per move) and I have to play two games simultaneously to keep my mind on the games. :P

Best regards
Thanks spacious_mind, but I don't mind doing this long time control once in a while. I now have the final game of the match to write up, so

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

Post by Yarc »

Finally, I have the last game of my 10 game match:

This one was the longest of the match at 88 moves, with all of the games played through to the end rather than me adjudicating.

I really thought this game was going to be a draw, but one of the players made a blunder in the end game stage. I expect you can guess which one.


[Event "40m/2h MCGE v Magellan Champion Book"]
[Date "2018.01.09"]
[Round "10"]
[White "MCGE"]
[Black "Magellan"]
[Result "1-0"]
[ECO "D02"]

1.d4 d5 2.Nf3 Nf6 3.g3 e6 4.Bg2 c6 5.O-O Bd6 6.Nc3 Nbd7 7.Bg5 Qb6 8.b3 O-O 9.Qd3 e5 10.Na4 Qa5 11.dxe5 Nxe5 12.Nxe5 Bxe5 13.Rad1 b5 14.Bd2 Qa6 15.Nc3 Rd8 16.a4 bxa4 17.Nxa4 Re8 18.Bc3 Bxc3 19.Nxc3 Qb6 20.Ra1 Nd7 21.e3 Ne5 22.Qd4 Bf5 23.Rfc1 Bg4 24.Ra2 Nf3+ 25.Bxf3 Bxf3 26.Qd3 Bh5 27.Ra6 Qb7 28.Rca1 Bg6 29.Qd2 Bh5 30.Nd1 Bg6 31.Qc3 Rec8 32.Nb2 Bh5 33.Nd3 Be2 34.Qd2 Bxd3 35.Qxd3 Re8 36.Qc3 Rec8 37.Qc5 Rc7 38.Kg2 h6 39.R1a2 Kh8 40.R2a3 Kh7 41.R3a5 Kg8 42.Ra1 Kh8 43.c4 dxc4 44.bxc4 Qb8 45.Kf1 Qb2 46.Qd6 Qb7 47.Kg2 Rcc8 48.Qc5 Rc7 49.h3 Qb3 50.Kg1 Qb2 51.Qd6 Rac8 52.Kg2 Qb7 53.Qc5 Ra8 54.R1a2 Qb1 55.Qe5 Rd7 56.Rb2 Qg6 57.c5 Rc8 58.Rba2 Rcc7 59.R2a4 Qd3 60.Rd4 Rxd4 61.exd4 Qxa6 62.Qxc7 Qd3 63.Qxc6 Qxd4 64.Qc8+ Kh7 65.Qf5+ Kg8 66.c6 g6 67.Qc2 Qd8 68.c7 Qa8+ 69.Kg1 Qc8 70.Qc5 a5 71.Qd6 Qxh3 72.Qd8+ Kg7 73.c8=Q Qxc8 74.Qxc8 a4 75.Qa6 h5 76.Qxa4 g5 77.Qc6 f6 78.Qe6 Kg6 79.Qg8+ Kf5 80.Qe8 h4 81.f3 g4 82.Qe4+ Kg5 83.gxh4+ Kh6 84.Qxg4 Kh7 85.h5 f5 86.Qg6+ Kh8 87.h6 f4 88.Qg7# 1-0

Final position:
[fen]7k/6Q1/7P/8/5p2/5P2/8/6K1 w - - 1 88[/fen]

It looked a very evenly matched game until black's 66th move when it blundered!

In this position white has just advanced its Pawn to c6:
[fen]6k1/p4pp1/2P4p/5Q2/3q4/6PP/5PK1/8 w - - 0 66[/fen]

A pawn on this square should ring alarm bells especially with a queen keeping a watchful eye over it! The Magellan here seemed more concerned with White's Queen and moved 66..... g6?? Better would have been to attack the advancing pawn with 66.... Qc4. This blunder really lost it the game when it could have possibly held for a draw.

So the final score is 7.5 - 2.5 to the MCGE, or put another way
MCGE +7 =1 -2

The Magellan has a tournament rating of 2228 at

https://www.schach-computer.info/wiki/i ... o-Liste_En

Using this rating I calculate gives the MCGE a ELO of around 2418 in this match. This seems to tally reasonably well with what I have read. I originally used Selective Search rating of 2206, but was not sure at what time controls were used to obtain this rating. Either way, I would say the MCGE has a rating of 2400 +/- a small margin.

The Magellan, despite its much slower hardware and making a few blunders, still managed to gain 2.5 points from the match, so in my opinion did pretty well.

In this match the MCGE did slightly better with the black pieces, winning 4 games and drawing once. With White it managed to win 3 games and lost twice.

I wonder if the pending engine update for the MCGE will have much impact on its playing strength?

Now I must do some work Regards
Ray
User avatar
spacious_mind
Senior Member
Posts: 4018
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Yes, its situations like this where the old programs have problems. Often it is not because they cannot find the best move, but they just don't have the speed to search deep enough in a given time setting to spot the risk properly.

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, its situations like this where the old programs have problems. Often it is not because they cannot find the best move, but they just don't have the speed to search deep enough in a given time setting to spot the risk properly.

Regards
I think so. The Magellan engine on MCGE's hardware would be interesting to see.

Regards
Ray
Post Reply