Main
Date: 15 Nov 2008 21:24:55
From: Sanny
Subject: Peter win Master Level at GetClub.
Peter played with Master Level using his computer program. Glaurung
2.1

Glaurung 2.1 was thinking for 2 sec / move
GetClub Master Level was thinking: 500 sec / move.

So GetClub was taking 250 times more time, Still GetClub lost the
game.

6. Nf3xd4{392} Ng4xf2{16}

Here on 6th move Glaurung gave sacrifice of its knight for a pawn.

After 40 ply it recover a extra Bishop. Since Master Level cannot see
40 ply deep it overlooked this move.

I am amazed to see how Glaurung find a sacrifice 40 ply deep.

Game Played between peter and master at GetClub.com

--------------------------------------------------------------------------------
peter: (Black)
master: (White)
Game Played at: http://www.GetClub.com/Chess.html
View Recorded Game: http://www.getclub.com/playgame.php?id=DM30550&game=Chess
--------------------------------------------------------------------------------

White -- Black
(master) -- (peter)

1. c2-c4{0} e7-e5{10}
2. Nb1-c3{0} Ng8-f6{20}
3. e2-e4{1096} Bf8-c5{210}
4. Ng1-f3{328} Nf6-g4{14}
5. d2-d4{692} e5xd4{12}
6. Nf3xd4{392} Ng4xf2{16}
7. Ke1-f2{346} Qd8-f6{72}
8. Kf2-e3{358} Nb8-c6{14}
9. Nc3-b5{536} Ke8-g8{14}
10. Bf1-e2{436} Rf8-e8{14}
11. a2-a3{336} Qf6-e5{14}
12. Be2-f3{490} a7-a5{28}
13. Ra1-a2{1440} f7-f5{6064}
14. g2-g3{0} f5-e4{18}
15. Bf3-g2{832} d7-d5{1012}
16. Rh1-f1{0} Nc6-d4{14}
17. Nb5-d4{456} Bc8-g4{62}
18. Qd1-d2{402} Bg4-f3{84}
19. b2-b3{412} d5-c4{16}
20. Ra2-c2{598} b7-b5{20}
21. b3-b4{492} Bc5-b6{78}
22. b4-a5{1010} Ra8-a5{20}
23. Qd2-c3{334} Re8-a8{18}
24. Rc2-a2{456} Bf3-g2{252}
25. Rf1-f4{598} g7-g5{16}
26. Rf4-f2{560} Bg2-h3{18}
27. Ra2-d2{662} Bh3-g4{18}
28. Rf2-f1{396} Ra5-a6{18}
29. Bc1-b2{406} Bb6-c5{18}
30. a3-a4{844} Ra6-a4{26}
31. Rf1-f2{930} Ra8-d8{20}
32. Rf2-f1{744} Ra4-b4{16}
33. Bb2-a1{608} Rb4-b3{14}
34. Qc3-b3{654} c4-b3{12}
35. Rf1-f2{1330} b3-b2{24}
36. Ba1-b2{580} Qe5-e6{8}
37. Rf2-f1{1346} Qe6-b3{90}
38. Ke3-f2{138} e4-e3{14}
39. Kf2-g2{380} Qb3-d5{14}
40. Kg2-g1{996}
--------------------------------------------------------------------------------
peter: (Black)
master: (White)
Game Played at: http://www.GetClub.com/Chess.html
View Recorded Game: http://www.getclub.com/playgame.php?id=DM30550&game=Chess

Can anyone spot any mistake in GetClub's game?

Bye
Sanny

Play Chess at: http://www.GetClub.com/Chess.html






 
Date: 16 Nov 2008 20:33:05
From: Sanny
Subject: Re: Peter win Master Level at GetClub.
On Nov 17, 9:17=A0am, help bot <[email protected] > wrote:
> On Nov 16, 12:24=A0am, Sanny <[email protected]> wrote:
>
>
>
>
>
> > Peter played with Master Level using his computer program. Glaurung
> > 2.1
>
> > Glaurung 2.1 was thinking for 2 sec / move
> > GetClub Master Level was thinking: 500 sec / move.
>
> > So GetClub was taking 250 times more time, Still GetClub lost the
> > game.
>
> > 6. Nf3xd4{392} Ng4xf2{16}
>
> > Here on 6th move Glaurung gave sacrifice of its knight for a pawn.
>
> > After 40 ply it recover a extra Bishop. Since Master Level cannot see
> > 40 ply deep it overlooked this move.
>
> > I am amazed to see how Glaurung find a sacrifice 40 ply deep.
>
> > Game Played between peter and master at GetClub.com
>
> > -----------------------------------------------------------------------=
----=AD=AD-----
> > peter: (Black)
> > master: (White)
> > Game Played at:http://www.GetClub.com/Chess.html
> > View Recorded Game:http://www.getclub.com/playgame.php?id=3DDM30550&gam=
e=3DChess
> > -----------------------------------------------------------------------=
----=AD=AD-----
>
> > White -- Black
> > (master) -- (peter)
>
> > 1. c2-c4{0} e7-e5{10}
> > 2. Nb1-c3{0} Ng8-f6{20}
> > 3. e2-e4{1096} Bf8-c5{210}
> > 4. Ng1-f3{328} Nf6-g4{14}
> > 5. d2-d4{692} e5xd4{12}
> > 6. Nf3xd4{392} Ng4xf2{16}
> > 7. Ke1-f2{346} Qd8-f6{72}
> > 8. Kf2-e3{358} Nb8-c6{14}
> > 9. Nc3-b5{536} Ke8-g8{14}
> > 10. Bf1-e2{436} Rf8-e8{14}
> > 11. a2-a3{336} Qf6-e5{14}
> > 12. Be2-f3{490} a7-a5{28}
> > 13. Ra1-a2{1440} f7-f5{6064}
> > 14. g2-g3{0} f5-e4{18}
> > 15. Bf3-g2{832} d7-d5{1012}
> > 16. Rh1-f1{0} Nc6-d4{14}
> > 17. Nb5-d4{456} Bc8-g4{62}
> > 18. Qd1-d2{402} Bg4-f3{84}
> > 19. b2-b3{412} d5-c4{16}
> > 20. Ra2-c2{598} b7-b5{20}
> > 21. b3-b4{492} Bc5-b6{78}
> > 22. b4-a5{1010} Ra8-a5{20}
> > 23. Qd2-c3{334} Re8-a8{18}
> > 24. Rc2-a2{456} Bf3-g2{252}
> > 25. Rf1-f4{598} g7-g5{16}
> > 26. Rf4-f2{560} Bg2-h3{18}
> > 27. Ra2-d2{662} Bh3-g4{18}
> > 28. Rf2-f1{396} Ra5-a6{18}
> > 29. Bc1-b2{406} Bb6-c5{18}
> > 30. a3-a4{844} Ra6-a4{26}
> > 31. Rf1-f2{930} Ra8-d8{20}
> > 32. Rf2-f1{744} Ra4-b4{16}
> > 33. Bb2-a1{608} Rb4-b3{14}
> > 34. Qc3-b3{654} c4-b3{12}
> > 35. Rf1-f2{1330} b3-b2{24}
> > 36. Ba1-b2{580} Qe5-e6{8}
> > 37. Rf2-f1{1346} Qe6-b3{90}
> > 38. Ke3-f2{138} e4-e3{14}
> > 39. Kf2-g2{380} Qb3-d5{14}
> > 40. Kg2-g1{996}
> > -----------------------------------------------------------------------=
----=AD=AD-----
> > =A0peter: (Black)
> > master: (White)
> > Game Played at:http://www.GetClub.com/Chess.html
> > View Recorded Game:http://www.getclub.com/playgame.php?id=3DDM30550&gam=
e=3DChess
>
> > Can anyone spot any mistake in GetClub's game?
>
> =A0 First of all, Glaurund did not see 40 plys ahead,
> (and even if it could, it would not be looking at the
> inferior line played by the GetClub program).
>
> =A0 This opening is very similar to something folks
> around here used to call the Fried Liver Attack,
> possibly also known as the Wilkes-Barre.
>
> =A0 White erred first of all by playing both e4 and
> c4, which weakens the dark squares; careful
> defense could alleviate such a problem, but
> instead GC tried to calculate its way through
> a maze of sharp variations, and lost its way.
>
> =A0 Tactically, the simple 4. Be2 would have
> circumvented the Knight sacrifice, but this is
> a poor arrangement for White. =A0I prefer the
> idea of a Kingside fianchetto, even if it may
> become necessary to play p-h3 to keep the
> Black steed at bay.
>
> =A0 Anyway, after the sacrifice on f2 White was
> under the gun the rest of the way, and by
> mid-game it was clear what the result would
> likely be. =A0Take a gander at GetClub's horrific
> 13. Ra2; this is the sort of thing which so
> often comes up in tactical skirmishes, and
> is the hallmark of a very confused mind-- er,
> chess engine.
>
> =A0 In attempting to defend such a precarious
> position, a /coherent development/ of the
> pieces must be figured out, but GC always
> jumbles its men about in a fog of confusion.
> This in turn creates numerous opportunities
> for the opponent to land a decisive tactical
> blow-- or perhaps two or three.
>
> =A0 My machine spotted a forced checkmate
> as early as the move 35. ... b2, vacating the
> b3 square for its Queen. =A0But even before
> that, it had Black a piece or so ahead for
> many moves. =A0 I did not realize that the
> attack was quite so powerful; in fact, I had
> the impression (though I never played
> this line from either side except in casual
> blitz games) that White was supposed to
> emerge on top-- if he defended perfectly.
> After analysing this game, I have changed
> my mind.
>
> =A0 So then, how does one properly play
> this sort of opening? =A0Well, it's hard to go
> wrong with the old adage "Knights before
> Bishops"; in other words, the early moves
> might go:
>
> =A01. c4 =A0e5
>
> =A02. Nc3 =A0Nf6
>
> =A03. Nf3
>
> =A0 I suspect that apart from fetching moves
> by rote from a database, this sort of thing
> can be accomplished by giving a small
> bonus for piece-development in the early
> opening. =A0Note how moves like 3. e4 do
> nothing to promote castling, whereas the
> move 3. Nf3 gets one obstruction out of
> the White King's way (if he castles King-
> side).
>
> =A0 One more thing; the move 18. Qd2 is
> yet another symptom of a disasterous
> approach to defense. =A0Chopping the
> Bishop on g4 would have led to a lost
> endgame, but guess what? =A0You said
> that Glaurung was thinking just two
> seconds per move; suppose it got into
> a position which it could not solve in just
> two seconds? =A0There is a possibility of
> a draw.
>
> =A0 Keeping all the pieces on just makes
> it that much easier for the attacker to
> stumble upon a forced mate-- even
> with just two seconds per move; and
> the defender? =A0The extra pieces get in
> one another's way.
>
> =A0 My suggestion is to simply remove
> GetClub's opponents' QN before
> starting play. =A0Then, instead of one
> side losing the whole way, we would
> see a true contest of sorts. =A0This
> would also effectively eliminate the
> whole issue of computers playing
> openings by rote; both sides would
> calculate and evaluate and only then
> decide upon their own moves.

Thanks for your Analysis.

In this game I find a few weaknesses and removed them So that GetClub
plays stronger games.

Bye
Sanny

Play Chess at: http://www.GetClub.com/Chess.html


 
Date: 16 Nov 2008 20:17:03
From: help bot
Subject: Re: Peter win Master Level at GetClub.
On Nov 16, 12:24=A0am, Sanny <[email protected] > wrote:

> Peter played with Master Level using his computer program. Glaurung
> 2.1
>
> Glaurung 2.1 was thinking for 2 sec / move
> GetClub Master Level was thinking: 500 sec / move.
>
> So GetClub was taking 250 times more time, Still GetClub lost the
> game.
>
> 6. Nf3xd4{392} Ng4xf2{16}
>
> Here on 6th move Glaurung gave sacrifice of its knight for a pawn.
>
> After 40 ply it recover a extra Bishop. Since Master Level cannot see
> 40 ply deep it overlooked this move.
>
> I am amazed to see how Glaurung find a sacrifice 40 ply deep.
>
> Game Played between peter and master at GetClub.com
>
> -------------------------------------------------------------------------=
--=AD-----
> peter: (Black)
> master: (White)
> Game Played at:http://www.GetClub.com/Chess.html
> View Recorded Game:http://www.getclub.com/playgame.php?id=3DDM30550&game=
=3DChess
> -------------------------------------------------------------------------=
--=AD-----
>
> White -- Black
> (master) -- (peter)
>
> 1. c2-c4{0} e7-e5{10}
> 2. Nb1-c3{0} Ng8-f6{20}
> 3. e2-e4{1096} Bf8-c5{210}
> 4. Ng1-f3{328} Nf6-g4{14}
> 5. d2-d4{692} e5xd4{12}
> 6. Nf3xd4{392} Ng4xf2{16}
> 7. Ke1-f2{346} Qd8-f6{72}
> 8. Kf2-e3{358} Nb8-c6{14}
> 9. Nc3-b5{536} Ke8-g8{14}
> 10. Bf1-e2{436} Rf8-e8{14}
> 11. a2-a3{336} Qf6-e5{14}
> 12. Be2-f3{490} a7-a5{28}
> 13. Ra1-a2{1440} f7-f5{6064}
> 14. g2-g3{0} f5-e4{18}
> 15. Bf3-g2{832} d7-d5{1012}
> 16. Rh1-f1{0} Nc6-d4{14}
> 17. Nb5-d4{456} Bc8-g4{62}
> 18. Qd1-d2{402} Bg4-f3{84}
> 19. b2-b3{412} d5-c4{16}
> 20. Ra2-c2{598} b7-b5{20}
> 21. b3-b4{492} Bc5-b6{78}
> 22. b4-a5{1010} Ra8-a5{20}
> 23. Qd2-c3{334} Re8-a8{18}
> 24. Rc2-a2{456} Bf3-g2{252}
> 25. Rf1-f4{598} g7-g5{16}
> 26. Rf4-f2{560} Bg2-h3{18}
> 27. Ra2-d2{662} Bh3-g4{18}
> 28. Rf2-f1{396} Ra5-a6{18}
> 29. Bc1-b2{406} Bb6-c5{18}
> 30. a3-a4{844} Ra6-a4{26}
> 31. Rf1-f2{930} Ra8-d8{20}
> 32. Rf2-f1{744} Ra4-b4{16}
> 33. Bb2-a1{608} Rb4-b3{14}
> 34. Qc3-b3{654} c4-b3{12}
> 35. Rf1-f2{1330} b3-b2{24}
> 36. Ba1-b2{580} Qe5-e6{8}
> 37. Rf2-f1{1346} Qe6-b3{90}
> 38. Ke3-f2{138} e4-e3{14}
> 39. Kf2-g2{380} Qb3-d5{14}
> 40. Kg2-g1{996}
> -------------------------------------------------------------------------=
--=AD-----
> =A0peter: (Black)
> master: (White)
> Game Played at:http://www.GetClub.com/Chess.html
> View Recorded Game:http://www.getclub.com/playgame.php?id=3DDM30550&game=
=3DChess
>
> Can anyone spot any mistake in GetClub's game?


First of all, Glaurund did not see 40 plys ahead,
(and even if it could, it would not be looking at the
inferior line played by the GetClub program).

This opening is very similar to something folks
around here used to call the Fried Liver Attack,
possibly also known as the Wilkes-Barre.

White erred first of all by playing both e4 and
c4, which weakens the dark squares; careful
defense could alleviate such a problem, but
instead GC tried to calculate its way through
a maze of sharp variations, and lost its way.

Tactically, the simple 4. Be2 would have
circumvented the Knight sacrifice, but this is
a poor arrangement for White. I prefer the
idea of a Kingside fianchetto, even if it may
become necessary to play p-h3 to keep the
Black steed at bay.

Anyway, after the sacrifice on f2 White was
under the gun the rest of the way, and by
mid-game it was clear what the result would
likely be. Take a gander at GetClub's horrific
13. Ra2; this is the sort of thing which so
often comes up in tactical skirmishes, and
is the hallmark of a very confused mind-- er,
chess engine.

In attempting to defend such a precarious
position, a /coherent development/ of the
pieces must be figured out, but GC always
jumbles its men about in a fog of confusion.
This in turn creates numerous opportunities
for the opponent to land a decisive tactical
blow-- or perhaps two or three.

My machine spotted a forced checkmate
as early as the move 35. ... b2, vacating the
b3 square for its Queen. But even before
that, it had Black a piece or so ahead for
many moves. I did not realize that the
attack was quite so powerful; in fact, I had
the impression (though I never played
this line from either side except in casual
blitz games) that White was supposed to
emerge on top-- if he defended perfectly.
After analysing this game, I have changed
my mind.

So then, how does one properly play
this sort of opening? Well, it's hard to go
wrong with the old adage "Knights before
Bishops"; in other words, the early moves
might go:

1. c4 e5

2. Nc3 Nf6

3. Nf3

I suspect that apart from fetching moves
by rote from a database, this sort of thing
can be accomplished by giving a small
bonus for piece-development in the early
opening. Note how moves like 3. e4 do
nothing to promote castling, whereas the
move 3. Nf3 gets one obstruction out of
the White King's way (if he castles King-
side).

One more thing; the move 18. Qd2 is
yet another symptom of a disasterous
approach to defense. Chopping the
Bishop on g4 would have led to a lost
endgame, but guess what? You said
that Glaurung was thinking just two
seconds per move; suppose it got into
a position which it could not solve in just
two seconds? There is a possibility of
a draw.

Keeping all the pieces on just makes
it that much easier for the attacker to
stumble upon a forced mate-- even
with just two seconds per move; and
the defender? The extra pieces get in
one another's way.

My suggestion is to simply remove
GetClub's opponents' QN before
starting play. Then, instead of one
side losing the whole way, we would
see a true contest of sorts. This
would also effectively eliminate the
whole issue of computers playing
openings by rote; both sides would
calculate and evaluate and only then
decide upon their own moves.


-- help bot