Whitehat a strong player was beaten by easy level. However nomorechess
was able to win with easy level.

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

That game had bugs when nomorechess won. Now the bugs have been
removed. I feel It would be a lot difficult to win Easy & Normal
Levels.

Now the GetClub Chess plays in time.

Beginner - 10-30 sec
Easy - 30-90 sec
Normal 2-4 min

So have fun.

Bye
Sanny

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

On Sep 20, 11:02 am, Sanny wrote:
Whitehat a strong player was beaten by easy level. However nomorechess
was able to win with easy level.

Here is the recorded game where Easy Level beat Whitehat

http://www.getclub.com/playgame.php?...724&game=Chess

Easy Level won the match after a lot of struggle.

Can someone analyze and tell how GetClub Chess be further improved?

Bye
Sanny

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

Now the GetClub Chess plays in time.

Beginner - 10-30 sec
Easy - 30-90 sec
Normal 2-4 min

So have fun.

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

Today Zebediah played a very nice game and defeated the Master Level
in just 28 moves. Master had no choice but to give away its bishop for
one pawn.

Recorded game: http://www.getclub.com/playgame.php?...742&game=Chess

Bye
Sanny

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

On Sep 20, 8:09 am, Sanny wrote:

Whitehat a strong player was beaten by easy level.

Looking at the game, Whitehat is obviously not a strong player but
indeed a rather weak player (at least presented himself as a weak
one) . He made several severe mistakes no strong player would make.

Ralf

PS: somebody asked you already not to pollute the groups: your
postings belong to rec.games.chess.computer, and not to four more.

[F'up to rgcc]

On Sep 20, 6:40 pm, Ralf Callenberg wrote:

Looking at the game, Whitehat is obviously not a strong player but
indeed a rather weak player

.... The way he played somehow looked a bit like a weak computer
program. Maybe somebody let your program play against itself!?

Ralf

Sanny wrote:
Now the GetClub Chess plays in time.

Beginner - 10-30 sec
Easy - 30-90 sec
Normal 2-4 min

So have fun.

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

Today Zebediah played a very nice game and defeated the Master Level
in just 28 moves. Master had no choice but to give away its bishop for
one pawn.

Recorded game: http://www.getclub.com/playgame.php?...742&game=Chess


"Master"????

Hint: it's impossible to play a "very nice game" when your opponent
rolls over and plays dead.

--
Kenneth Sloan
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://www.cis.uab.edu/sloan/

On Sep 20, 1:02 am, Sanny wrote:
Whitehat a strong player was beaten by easy level. However nomorechess
was able to win with easy level.

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

That game had bugs when nomorechess won. Now the bugs have been
removed. I feel It would be a lot difficult to win Easy & Normal
Levels.

Yes, I found that the only way to beat GetClub is by
triple-clicking on the square where its Queen rests;
this changes a White Queen to a Black one, or vice
versa, whereupon I am just barely able to get a win,
after a long, hard struggle.

-------------

Sanny, your program is playing fairly well now, but
there are a few things which even the best commercial
programs get wrong; one of these is misplaying the
opening once out of book; somehow or other, even
top level programs are programmed wrong and will
commit beginner-like positional errors here, if they
are not allowed to retrieve a by-rote correct move in
the opening.

In the middle game, things are different; here the
brute power of calculation reigns supreme, as shown
in the recent odds match where Rybka bested GM
Benjamin repeatedly from very bad positions.

But once the endgame is reached, again the power
of calculation can give way to a much more subtle
understanding based on ideas, not calculations; one
example was the fairly simple Rook ending I won at
GetClub recently, which ought to have been drawn
and if anything, the program had better chances on
account of its one-pawn advantage in conjunction
with my desperate attempts to avoid the
embarrassment of not winning. At one point, I moved
my Rook between two of my own pawns, to avoid any
further material losses; surely, even a cow could draw
this position, but no.


-- help bot

On Sep 20, 1:02 am, Sanny wrote:

Whitehat a strong player was beaten by easy level.

Okay, I replayed that game and it looked like GetClub
was floundering a bit; Whitehat made several errors
which no "strong" chess player would ever make, but
even when the attack was overwhelming GetClub
somehow managed to nearly get back rank mated,
luckily losing just its Queen instead.

The ending, cut off at about move 68, had GetClub
doing some spite checks, apparently having no idea
how to win with K & Q vs. K; luckily, we are spared
the spectacle, as I suppose Whitehat got tired of
this and just resigned; in any case, sooner or later
the program could have "accidentally" promoted a
pawn and then lucked into a checkmate having two
Queens to play with.


-- help bot

On Sep 20, 6:11 am, Sanny wrote:

Today Zebediah played a very nice game and defeated the Master Level
in just 28 moves. Master had no choice but to give away its bishop for
one pawn.

Once again, the GetClub program voluntarily sacrificed
a piece for fewer than three pawns.

Never mind the Knights dancing about the board, or any
strategical considerations; when is this "bug"/feature
going to get fixed? If the program justs gives material
away like this, even patzers like IM Innes and Gnarly
Fognewt will beat it. Tell the programmers to make it
hang on to every last piece until it is pried from the
machine's cold, dead hands. Then we may see a battle
in which humans attempt to "outplay" the program
strategically, but where it has winning chances if they
make even a single serious mistake.

Right now, I am still seeing numerous cases in which
the program allows elementary pawn forks, or sac's a
Knight for a pawn and a couple of spite checks which
are of no real value. Selective search does not work as
a *replacement* for brute force calculations; it may be
useful in conjunction with it, but tactics are 90% of the
game.


-- help bot

In rec.games.chess.computer help bot wrote:
On Sep 20, 6:11 am, Sanny wrote:

Today Zebediah played a very nice game and defeated the Master Level
in just 28 moves. Master had no choice but to give away its bishop for
one pawn.

Once again, the GetClub program voluntarily sacrificed
a piece for fewer than three pawns.

Never mind the Knights dancing about the board, or any
strategical considerations; when is this "bug"/feature
going to get fixed? If the program justs gives material
away like this, even patzers like IM Innes and Gnarly
Fognewt will beat it. Tell the programmers to make it
hang on to every last piece until it is pried from the
machine's cold, dead hands. Then we may see a battle
in which humans attempt to "outplay" the program
strategically, but where it has winning chances if they
make even a single serious mistake.


Patzers like "IM Innes"?

Come on? Give Sanny a little credit if that is your only counter at this
point.

Right now, I am still seeing numerous cases in which
the program allows elementary pawn forks, or sac's a
Knight for a pawn and a couple of spite checks which
are of no real value. Selective search does not work as
a *replacement* for brute force calculations; it may be
useful in conjunction with it, but tactics are 90% of the
game.


Constructive criticism is good, as is the preceding paragraph.

--
Thomas T. Veldhouse

You will be advanced socially, without any special effort on your part.