My crushing defeat of Excalibur GM

[Reinfeld]

Observations, no particular order:

1) Yes, I am a golden god.

2) I'd love to hear Queegmeister's opinion of this game, since he's been messing around with Igor, which is supposedly the same under the hood.

3) Usually when I manage to beat a machine, it's an eke-it-out affair where I play like a frightened rat, plug up everything in a sort of closed-position ziggurat and inch toward an endgame. This is one of those rare games where I felt like a berserker.

4) The machine is gigantic. It gives you a tournament feel, but given the tolerance of the average spouse, it's hard to find a place to put it. I rarely play it for that reason, though I love the size and the pieces.

5) It shuts itself off after a certain amount of down time (a few other tabletops do this), which makes it a nice option for correspondence-style play and cat tours. You can make a few moves, let it hum and snore, then think for a day, press it back to life and take your next shot. That was my chief advantage, obviously. This was not a speed-chess game.

6) The glue on the faux-felt piece bottoms is hilariously feeble. I have a stack of piece bottoms - they look like lost poker chips.

7) The claim of 100 levels is misleading - typical Excalibur tackiness. The distinctions are pretty small. The reality is basic: the usual standard time controls, plus the usual handicaps.

I have the standard 12 mhz model. I set Excalibur at 3 minutes/move. The game took a couple of days. I played with several manuals at hand (no computers), which ceased to matter after move 7.

I was proudest of moves 12, 14 and 17, which I found after long thought. Excalibur blundered badly at move 15 - for once, I saw a very long sequence, and didn't make too many errors. This is a Petroff, an opening I don't like. Only four diagrams, I promise:

[White "Reinfeld"]
[Black "Excalibur GM"]
[ECO "C43"]
[Result "1-0"]

1.e4 e5 2.Nf3 Nf6 3.d4 Nxe4 4.Bd3 d5 5.Nxe5 Bd6 6.Nd2 Bxe5 7.dxe5
Nc6 (Excalibur leaves the books).

[fen]r1bqk2r/ppp2ppp/2n5/3pP3/4n3/3B4/PPPN1PPP/R1BQK2R w KQkq - 1 8[/fen]

From here, I was on my own, thinking only of staying even.

8.O-O Nxd2 9.Bxd2 O-O 10.Re1 Re8 11.f4 Qh4

[fen]r1b1r1k1/ppp2ppp/2n5/3pP3/5P1q/3B4/PPPB2PP/R2QR1K1 w - - 1 12[/fen]

I looked at this for a long time and tried to be a chess player. I found no obvious tactical sequence. I thought of Silman and tried to run through the imbalances. I've got two bishops, he doesn't. I'm ahead in development. He's got one good minor piece - the N on c6. Suppose I take away good squares for that piece?

12.c3 Qg4 (invites an exchange) 13.Qc2 (refuses, forcing him back) Qh4

[fen]r1b1r1k1/ppp2ppp/2n5/3pP3/5P1q/2PB4/PPQB2PP/R3R1K1 w - - 3 14[/fen]

I wanted to move the rook up to e3, but then I stopped, thinking his Q could grab the seemingly undefended pawn on f4. Then I realized he couldn't.

14.Re3 Be6 15.Rg3

Here's the losing moment, at Black's 15th:

[fen]r3r1k1/ppp2ppp/2n1b3/3pP3/5P1q/2PB2R1/PPQB2PP/R5K1 w - - 6 15[/fen]

Inexplicably, Ex GM plays 15...Na5? After that, I could see a long way out. No sacs or fireworks at the front end, just obvious forcing moves....

16.f5 Bd7 17.Bg5! Qa4 18.b3 Qc6 19.Bf6
Qb6+ (Ex is dead lost at this point) 20.Kh1 g6 21.fxg6 hxg6 22.Bxg6 Qxf6 23.exf6 Nxb3 24.axb3
Kf8 25.Bh7 Bg4 26.Rxg4 Re1+ 27.Rxe1 Re8 28.Rg8# 1-0

- R.

[LWSteve]

Your game took 2 days. (aprox.)

48 x 60 = 2880 minutes

2880 / 28 moves = 103 minutes per move for you

The Grandmaster = 3 minutes per move.

And you used opening manuals and J. Silman chess books...

Are you Captain Kangaroo or sumpin like dat?

hahahahahahahahahahaha!!!!!

LW

[spacious_mind]

Hi Reinfeld,

Don't let Stevie phase you . Nice game.

I am the opposite to you perhaps, as you write that you mostly win when you try to keep the game closed and equal to the endgame.

Most people seem to use this tactic against their chess computers.

But I cannot bring myself to play like that (follower of Marshall's school of chess). Actually, it is in these situations that I mostly mess up (lack of self control and patience) and lose.

In my opinion your premise of following Silman's imbalance theories are sound tactics against dedicated's. This is an approach that works well. The old dedicated computers seem to get uncomfortable with pins and piece pressures and space disadvantages and invariably this upsets their play plans and they loose tempo by trying to make adjustments (ie backward moves or unnecessary precautionary defensive moves or silly exchanges of their good pieces). This gives you the opportunities to win spectacularly

Best regards,

Nick

[Queegmeister]

Well, it looks like you were planning well and took advantage of the computers' "artificial" moves like Qh4. I like your c3 and you spotted the tactical Rook lift to e3 which was indeed winning. My chess teacher said to stay behind your pawns and use them as barricades and then offensive weapons once you can safely advance them. This is one way to win.

Elementary Watson, elementary. The pawn Phalanx.

Nice game !!

The GM is nice to play it on g/75 where your time is limited but not TOO limited. And I wish all my chess computers were as large like the GM. At this time limit - I can usually play another computer on g/75 as well so I make it a simul and it is more fun.

If you want to see the game I posted vs the GM - it was a Danish Gambit and about as bloody as it gets. I enjoy crushing a computer in an open position which is not USUALLY possible with anything above strict Swedish 2250 Elo. They just see too much. And we get tired, they do not.

As far as the GM goes , I think it is maybe 1900 elo and I have 2 so I will probably play Dr. and put a 24 MHZ crystal in it making it maybe 2000 elo. Which will send the townsfolk, pitchforks in hand,( and Steve B) to my house to burn it down.

I will tell you a secret - the GM and Igor can't understand the King's Indian attack so if you play it right - you can have an easy victory. Also , play the Nimzo-Indian as Black and take the Knight on c3 then blockade and the Exacalibur's will be easy to trap - the Bishop's can be easily neutralized.

keep playing and practicing , looking good regards
Queeg

[Steve B]

As far as the GM goes , I think it is maybe 1900 elo and I have 2 so I will probably play Dr. and put a 24 MHZ crystal in it making it maybe 2000 elo. Which will send the townsfolk, pitchforks in hand,( and Steve B) to my house to burn it down.

Hard to burn down the house when i dont even know your name or address


your a tad high on the GM rating though
Selective Search has it at 1814 Elo which would make it about 1930 ish USCF

Regards From Talking Heads
Steve

[Queegmeister]

Yes perhaps lower than 1900 elo but from a playing perspective against human competition , I have seen these units pull off some moves I would expect from a stronger player. Especially on the defensive side. I have had to dig deep to come up with the crusher in winning positions on a few occasions. All my intuition was telling me - it's there but with the clock ticking away and one misstep - your attack vaporizes like a fart in the wind.

So I give it a bonus vs. humans. Simply because it will not blunder a piece like a class A player does sometimes. It does make some big mistakes on occasion but due to it's fairly consistent play and seeing the IGOR 24 MHZ on the test so far (not tallied in yet ) on games 2 and 3 - I suspect maybe a little underrated.

Wiki debating regards
Queeg

[Steve B]

Wiki debating regards
Queeg

lol
i think i am finished with the Wiki debates
everytime i gain a few yards they
Move The Goal Posts Regards
Steve

[Jon_Doh]

This is my favorite computer chess board. It's tournament size board and pieces are pleasurable to play.