| |
Main
Date: 15 Jan 2009 00:57:11
From: Sanny
Subject: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
Today GetClub was further improved. And now GetClub is one of the
Strongest program in the world. As it is playing as good as Rybka
Which is worlds fastest Chess Program.
Now GetClub is playing as good as Rybka program (Free Version)
Play Chess at: http://www.GetClub.com/Chess.html
So now the new ratings to GetClub levels are
Beginner: 2000+
Easy: 2200+
Normal: 2400+
Master: 2600+
Anyone having Fritz/ Chess Master and other strong Chess Programs Play
GetClub against them and let me know if they are able to beat GetClub
or not.
Give same time to both Players. Play against Easy Level: (30 sec /
move ).
Do not use dual core / quad core as GetClub utilize only 1 thread So
if you have a multicore then give GetClub 2-4 times extra time
depending on the number of multiprocessors you have
At the moment GetClub is as strong as the Free Version of Rybka.
Now I am wishing to see lots of Wins at GetClub. Lets see if Help Bot
can still beat the Beginner Level or not.
No need to play with Easy Level as Beginner will play a lot stronger
game. What do Taylor Kingston say about the new Strength. Now even he
will agree that GetClub is unbeatable.
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
|
|
| |
Date: 19 Jan 2009 05:24:22
From:
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 19, 5:51 am, Martin Brown <
|
| |
Date:
From: Martin Brown
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
|
| |
Date: 18 Jan 2009 04:59:22
From: help bot
Subject: Re: GetClub doesn't acknowledge draw
|
On Jan 18, 3:51=A0am, [email protected] wrote:
> > =A0 Are you aware of the fact that the process has
> > the undesireable side-effect of pushing down the
> > ratings (and rankings) of older computers over
> > time as newer, faster processors are invented?
>
> > =A0 What was once entered in tournaments to
> > determine a realistic USCF rating, may later be
> > said to be rated 1700 or 1500... it all depends.
> Your comment definitely made me stop and think. So I did a little
> poking around that marvelous thing called the Internet. The oldest
> SSDF list that I could find is exactly 13 years old -- from January
> 1996:
>
> http://www.geocities.com/CapeCanaveral/Launchpad/2640/ssdf/1996/ssdf9...
>
> It lists the Par Excellence at 1835, 100 points above where it is now.
>
> The Par Excellence was released in 1986, ten years before the above
> list. So I grant a small possibility that it might have been
> reasonably estimated 2000+ OTB by the USCF (although probably not by
> an organization that did more thorough testing).
There is no need to "grant" me anything on this;
the facts are *published*, in one of the big-name
magazines! Do yourself a favor and look this up.
Don't wait for me to do it, just so I can thumb my
nose at all those still living in denial.
In addition to Chess Life and god-knows what
other major magazines, one can find the old
numbers in "Computer Chess Reports" by ICD
(which incidentally, had its own ratings that
were claimed to be better than USCF's).
> I did find one post on a message board, written one year ago, where
> somebody said that the USCF rated the Par Excellence at 2100.
Bingo. Now that I think about it, it was the
Novag which was "2015" or thereabouts. The
Par Excellence was /claimed/ to be 2100 in
strength; that's why people bought it.
I remember it like it was only yesterday...
the tiny chessmen, something like a Drueke
set in its design. The reliability... not like a
hot-headed Novag Super-Constellation.
> The SSDF was created in 1984, so if somebody could find an earlier
> list, it would be interesting to see the first entry for the Par
> Excellence.
The trouble with SSDF ratings is that they are
calculated only based on results against other
computers. What we have been trying to do
with GetClub is put Sanny's ridiculous claims
into human perspective, along with the ludicrous
attacks of his multitudinous critics.
For this purpose, the perfect tool is USCF
ratings... of real people. The introduction of
other computers is merely to protect the fragile
egos of many of those same critics.
> But all of this is pretty much irrelevant; that was then and this is
> now.
> If Sanny is going to say that today his engine is 2000+, then we
> have to compare it to today's ratings -- and the Par Excellence is
> 1735.
Wrong. Sanny has never, ever refered to
SSDF or other computer-only ratings. All his
critics are talking about USCF (or FIDE) OTB
ratings. In fact, Sanny does not really know
what ratings are, or how they are calculated.
Where computer-ratings come into it is in
discussions of Rybka and other too-high-up
engines, which have no official USCF or FIDE
ratings. People do not know how the 3000+
Rybka numbers fit together with USCF's
ratings; some imagine that anything that big
must be "inflated".
But we don't need to tackle that issue at all,
for we now have the official USCF rating of one
of the players at GetClub-- the Par Excellence.
What's more, it cannot have "improved" since
it is a fixed, tabletop model. Problem solved.
Now all we need are a variety of games, with
Black and with White, on various levels and
in different openings.
It would also help if Sanny could get his
monstrosity to play legal chess and rate
draws as draws... .
-- help bot
|
| |
Date: 18 Jan 2009 02:30:55
From: zzz
Subject: Re: GetClub doesn't acknowledge draw
|
On 18 jan, 09:51, [email protected] wrote:
> On Jan 17, 10:20=A0pm, help bot <[email protected]> wrote:
>
>
>
> > On Jan 18, 12:15=A0am, [email protected] wrote:
>
> > > > =A0 If I recall correctly, this may be the same beast
> > > > that was "certified" by the sinister USCF as being
> > > > a true Expert in OTB tournament play against
> > > > humans. =A0Forgive me a cruel laugh-- bwahahaha.
> > > The SSDF tested the Par Excellence in over 1000 games and rated it at
> > > 1735. So, once again, GetClub fails to live up to Sanny's billing, no
> > > matter which setting "zzz" was using.
>
> > =A0 Are you aware of the fact that the process has
> > the undesireable side-effect of pushing down the
> > ratings (and rankings) of older computers over
> > time as newer, faster processors are invented?
>
> > =A0 What was once entered in tournaments to
> > determine a realistic USCF rating, may later be
> > said to be rated 1700 or 1500... it all depends.
>
> Your comment definitely made me stop and think. So I did a little
> poking around that marvelous thing called the Internet. The oldest
> SSDF list that I could find is exactly 13 years old -- from January
> 1996:
>
> http://www.geocities.com/CapeCanaveral/Launchpad/2640/ssdf/1996/ssdf9...
>
> It lists the Par Excellence at 1835, 100 points above where it is now.
>
> The Par Excellence was released in 1986, ten years before the above
> list. So I grant a small possibility that it might have been
> reasonably estimated 2000+ OTB by the USCF (although probably not by
> an organization that did more thorough testing).
>
> I did find one post on a message board, written one year ago, where
> somebody said that the USCF rated the Par Excellence at 2100.
>
> The SSDF was created in 1984, so if somebody could find an earlier
> list, it would be interesting to see the first entry for the Par
> Excellence.
>
> But all of this is pretty much irrelevant; that was then and this is
> now. If Sanny is going to say that today his engine is 2000+, then we
> have to compare it to today's ratings -- and the Par Excellence is
> 1735.
>
> And I'm still curious to know which setting in GetClub "zzz" was
> playing against.
>
> jm
GetClub was set to "Easy" level.
Par Excellence was set to "level 8", which plays 60 moves in 30
minutes. Any additional time registered is me being a slow operator.
If you want to compare the hardware:
Par Excellence runs on a 5MHz 65C02 CPU, with 32Kb ROM and 8Kb RAM.
The GetClub applet was running on a 64-bit Vista machine with a Q9550
CPU, each core running at 2.83 GHz, with 8Gb RAM installed.
|
| |
Date: 18 Jan 2009 01:33:36
From: Sanny
Subject: Re: GetClub doesn't acknowledge draw
|
> And I'm still curious to know which setting in GetClub "zzz" was
> playing against.
He played with Easy Level which thinks 30-40 sec / move. and the game
was a DRAW !!!
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
|
| |
Date: 18 Jan 2009 00:51:44
From:
Subject: Re: GetClub doesn't acknowledge draw
|
On Jan 17, 10:20=A0pm, help bot <[email protected] > wrote:
> On Jan 18, 12:15=A0am, [email protected] wrote:
>
> > > =A0 If I recall correctly, this may be the same beast
> > > that was "certified" by the sinister USCF as being
> > > a true Expert in OTB tournament play against
> > > humans. =A0Forgive me a cruel laugh-- bwahahaha.
> > The SSDF tested the Par Excellence in over 1000 games and rated it at
> > 1735. So, once again, GetClub fails to live up to Sanny's billing, no
> > matter which setting "zzz" was using.
>
> =A0 Are you aware of the fact that the process has
> the undesireable side-effect of pushing down the
> ratings (and rankings) of older computers over
> time as newer, faster processors are invented?
>
> =A0 What was once entered in tournaments to
> determine a realistic USCF rating, may later be
> said to be rated 1700 or 1500... it all depends.
Your comment definitely made me stop and think. So I did a little
poking around that marvelous thing called the Internet. The oldest
SSDF list that I could find is exactly 13 years old -- from January
1996:
http://www.geocities.com/CapeCanaveral/Launchpad/2640/ssdf/1996/ssdf9601.ht=
m
It lists the Par Excellence at 1835, 100 points above where it is now.
The Par Excellence was released in 1986, ten years before the above
list. So I grant a small possibility that it might have been
reasonably estimated 2000+ OTB by the USCF (although probably not by
an organization that did more thorough testing).
I did find one post on a message board, written one year ago, where
somebody said that the USCF rated the Par Excellence at 2100.
The SSDF was created in 1984, so if somebody could find an earlier
list, it would be interesting to see the first entry for the Par
Excellence.
But all of this is pretty much irrelevant; that was then and this is
now. If Sanny is going to say that today his engine is 2000+, then we
have to compare it to today's ratings -- and the Par Excellence is
1735.
And I'm still curious to know which setting in GetClub "zzz" was
playing against.
jm
|
| | |
Date: 18 Jan 2009 12:23:56
From: Andy Walker
Subject: Re: GetClub doesn't acknowledge draw
|
[email protected] wrote:
> [...] The oldest
> SSDF list that I could find is exactly 13 years old -- from January
> 1996:
> http://www.geocities.com/CapeCanaveral/Launchpad/2640/ssdf/1996/ssdf9601.htm
> It lists the Par Excellence at 1835, 100 points above where it is now.
[...]
> The SSDF was created in 1984, so if somebody could find an earlier
> list, it would be interesting to see the first entry for the Par
> Excellence.
My copies of ICCAJ are somewhat scattered; the earliest I could
find easily was vol 10 #3, September 1987, wherein the "Swedish Rating
List" puts "Par Excellence 5 MHz" 9th at 1894. Top was "Mephisto Dallas
68020 14 MHz" at 2102, and bottom the "Turbo 16K", 35th at 1513. The
5MHz machines were all between 1873 and 1914.
How any of these relate to human Elo ratings is a whole new can
of worms, of course. Likewise, reasons why computer ratings deflate.
--
Andy Walker
Nottingham
|
| |
Date: 18 Jan 2009 00:04:43
From: help bot
Subject: Re: GetClub doesn't acknowledge draw
|
On Jan 18, 2:34=A0am, Sanny <[email protected] > wrote:
> > =A0 If GetClub can manage to hold its own versus
> > the Par Excellence, then we will know that it
> > has finally reached the Expert level of OTB
> > chess in the USA, circa 1985. =A0 Granted, this
> > is mainly on account of raw processor speed,
> > not any clever eval function or pruning scheme.
> One more thing I would like to add. When people here say Old Program.
>
> They are running old program on a new Machine. Todays machines are
> 10-20 times faster than the machines in 1990. So That old program is
> also running on fast computers.
>
> The programming technique is the same today and then. If I say x=3D5+10;
> Then even if the old program says so or the new program. The computer
> will take same amount of time whether it is new or old.
>
> So if Par Excellence was 1700+ in 1990 Then on todays fast computer it
> will be playing 10-20 times faster So having a rating of +400.
Um, no. The Fidelity Par Excellence is a
tabletop machine, and it runs the same speed
now as it did back then.
Now, if people talk about "Chess Genius" or
"M-chess" or "Fritz 5.32" today, they are
very likely running an old chess software on
their current machine, with a big increase in
raw speed.
> So on modern computer The rating of par Excellence is 1700+400 =3D
> 2100+.
No, the USCF-rating of an old Par Excellence
tabletop is roughly the same now as it was when
the USCF tested it in official, rated tournaments.
It is possible that there has been a bit of ratings
inflation, and later, deflation in these ratings, but
they mostly cancel one another out.
By the way, anybody with old issues of Chess
Life magazine can easily look the exact number
up, as this machine was quite heavily advertised
at the time. Odds are that only a Mr. Kingston
type would have any difficulty whatever in
"finding" the precise number.
> Say I stop improving GetClub. After 50 years when computers will be 20
> times faster than today
More like twenty million times faster...
> my GetClub program will automatically play 20
> times faster. So the rating of GetClub Chess will keep increasing with
> the speed of computation increased.
Sanny, fifty years from now, your program
would not be *allowed* to contaminate the
cloud computing networks (which of course
will have almost "solved" chess, and moved
on to bigger and better things). Bugs may
be in fashion today, but things change.
-- help bot
|
| |
Date: 17 Jan 2009 23:38:26
From: Sanny
Subject: Re: GetClub doesn't acknowledge draw
|
On Jan 17, 9:47=A0pm, zzz <[email protected] > wrote:
> I'm currently operating a game between Par Excellence and GetClub.
> Par Excellence has correctly claimed a draw for the 3-fold repetition
> of a position, but GetClub doesn't acknowledge this. The game is
> completely stuck. Both sides are repeating a sequence of moves, which
> leads to the same positions over and over. Neither side is willing to
> deviate, and rightly so, because doing so would mean losing. The only
> way to finish this game would be to let them keep repeating moves
> until the 50-move rule kicks in (provided this has been implemented!).
> I don't have the time or the patience for this, so please Sanny,
> cancel this game (username: Par Excellence).
The game was automatically drawn. In such games you have to play till
100 moves then the "Resign Now" Button turns into "Draw Offer" Button
and then you can draw the game by pressing it.
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
|
| |
Date: 17 Jan 2009 23:34:14
From: Sanny
Subject: Re: GetClub doesn't acknowledge draw
|
> =A0 If GetClub can manage to hold its own versus
> the Par Excellence, then we will know that it
> has finally reached the Expert level of OTB
> chess in the USA, circa 1985. =A0 Granted, this
> is mainly on account of raw processor speed,
> not any clever eval function or pruning scheme.
One more thing I would like to add. When people here say Old Program.
They are running old program on a new Machine. Todays machines are
10-20 times faster than the machines in 1990. So That old program is
also running on fast computers.
The programming technique is the same today and then. If I say x=3D5+10;
Then even if the old program says so or the new program. The computer
will take same amount of time whether it is new or old.
So if Par Excellence was 1700+ in 1990 Then on todays fast computer it
will be playing 10-20 times faster So having a rating of +400.
So on modern computer The rating of par Excellence is 1700+400 =3D
2100+.
Say I stop improving GetClub. After 50 years when computers will be 20
times faster than today my GetClub program will automatically play 20
times faster. So the rating of GetClub Chess will keep increasing with
the speed of computation increased.
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
|
| |
Date: 17 Jan 2009 22:20:26
From: help bot
Subject: Re: GetClub doesn't acknowledge draw
|
On Jan 18, 12:15=A0am, [email protected] wrote:
> > =A0 If I recall correctly, this may be the same beast
> > that was "certified" by the sinister USCF as being
> > a true Expert in OTB tournament play against
> > humans. =A0Forgive me a cruel laugh-- bwahahaha.
> The SSDF tested the Par Excellence in over 1000 games and rated it at
> 1735. So, once again, GetClub fails to live up to Sanny's billing, no
> matter which setting "zzz" was using.
Are you aware of the fact that the process has
the undesireable side-effect of pushing down the
ratings (and rankings) of older computers over
time as newer, faster processors are invented?
What was once entered in tournaments to
determine a realistic USCF rating, may later be
said to be rated 1700 or 1500... it all depends.
But at the time it played rated chess, the
results were, such as I can recall, that Fidelity's
Par Excellence was rated pretty much the best
bang for your buck; the closest thing to a real
Expert, without paying for a fancy wooden
chess board (circa $1,000 and up).
Maybe the PE was "only" 2015 or so? I
don't recall the exact number, but it wasn't
17-something. I believe it was advertised in
Chess Life magazine as a USCF Expert class
machine, based on actual performance against
real, live humans. This was an official USCF
rating, not some mere illusion.
Many times in later years, I looked up the
top-rated programs (for there was a distinct
shift from chess-playing machines to software)
but then scanned the "complete list" only to
find that my outmoded machines were being
pushed ever lower, even crossing class bound-
ries in the process. I am not surprised that
someone now has this old machine rated two
classes lower, for it stands no chance against
Genius, M-chess, Fritz or even Chessmaster
these days.
But all this means is that the newer programs
are utilizing the faster and more-powerful hard-
ware we have today; it cannot change the fact
that PE competed at the Expert, or at the very
least, Class A level in USCF-rated competition,
against humans like you and me.
If GetClub can manage to hold its own versus
the Par Excellence, then we will know that it
has finally reached the Expert level of OTB
chess in the USA, circa 1985. Granted, this
is mainly on account of raw processor speed,
not any clever eval function or pruning scheme.
-- help bot
|
| |
Date: 17 Jan 2009 21:15:39
From:
Subject: Re: GetClub doesn't acknowledge draw
|
On Jan 17, 8:28=A0pm, help bot <[email protected] > wrote:
> On Jan 17, 11:47=A0am, zzz <[email protected]> wrote:
>
> > I'm currently operating a game between Par Excellence and GetClub.
> > Par Excellence has correctly claimed a draw
>
> =A0 Get that, you ninnies? =A0 Fidelity's Par Excellence
> was held to a draw by Sanny's godawful GetClub
> monstrosity!
>
> =A0 If I recall correctly, this may be the same beast
> that was "certified" by the sinister USCF as being
> a true Expert in OTB tournament play against
> humans. =A0Forgive me a cruel laugh-- bwahahaha.
>
> =A0 -- help bot
The SSDF tested the Par Excellence in over 1000 games and rated it at
1735. So, once again, GetClub fails to live up to Sanny's billing, no
matter which setting "zzz" was using.
jm
|
| |
Date: 17 Jan 2009 20:28:51
From: help bot
Subject: Re: GetClub doesn't acknowledge draw
|
On Jan 17, 11:47=A0am, zzz <[email protected] > wrote:
> I'm currently operating a game between Par Excellence and GetClub.
> Par Excellence has correctly claimed a draw
Get that, you ninnies? Fidelity's Par Excellence
was held to a draw by Sanny's godawful GetClub
monstrosity!
If I recall correctly, this may be the same beast
that was "certified" by the sinister USCF as being
a true Expert in OTB tournament play against
humans. Forgive me a cruel laugh-- bwahahaha.
-- help bot
|
| |
Date: 17 Jan 2009 08:47:20
From: zzz
Subject: GetClub doesn't acknowledge draw
|
I'm currently operating a game between Par Excellence and GetClub.
Par Excellence has correctly claimed a draw for the 3-fold repetition
of a position, but GetClub doesn't acknowledge this. The game is
completely stuck. Both sides are repeating a sequence of moves, which
leads to the same positions over and over. Neither side is willing to
deviate, and rightly so, because doing so would mean losing. The only
way to finish this game would be to let them keep repeating moves
until the 50-move rule kicks in (provided this has been implemented!).
I don't have the time or the patience for this, so please Sanny,
cancel this game (username: Par Excellence).
Sanny, please make your program play according to the rules and
implement the draw rules correctly.
For more information please check article 9 in the Laws of Chess
(http://www.fide.com/fide/handbook?id=124&view=article)
Thank you.
|
| |
Date: 16 Jan 2009 21:05:10
From: help bot
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 16, 11:12=A0am, Sanny <[email protected] > wrote:
> > =A0 Generally speaking, a crude chess pro-
> > gram ought to do well in simple endings,
> > via brute-force calculation. =A0 =A0But in the
> > case of GetClub, there seems to be a
> > definite speed-up near the end, along with
> > lousy play and some strange behaviors,
> > such as willingly trapping its own King on
> > the back rank (delusion of King "safety")
> > when it should be fighting madly, as in a
> > Mel Gibson movie.
> Should King be allowed to move on foraward ranks?
Um, no. If you enable the King to move around
freely, I may then have difficulty in thumping the
program with ease in the endgame because I will
no longer have a titanic advantage.
> There is huge
> penalty for King leaving the Back Rank in the middle game. Its done so
> that king remain safe. Should I change it and allow the King to move
> arround?
How do you define "middle game"? In many
of my "complaints", I have discussed end game
positions where only a few pieces remain on the
board, and I don't recall having often slammed
GC for keeping its King "safe" in a complex
Queen ending.
This is not a simple matter; there is a fine line
between "exposure to attack" and "piece activity"
when it comes to the Kings, and chess literature
is filled with examples of grandmasters getting
it wrong in serious play. In fact, Pal Benko rou-
tinely filled his CL column with such examples--
purportedly writing about the endgame but in
some cases, there are quite a few pieces left on
the board at the beginning.
> Since you heard Play as good as Rybka I suppose your game Style has
> also improved. Now you are playing better moves than earlier.
Thank you. Personally, I suspect that all
that's happened is that I've lucked into some
positions where the Queens and other pieces
can easily be exchanged, then used my
amazing skills to win the won-games that
result, but if you think I'm getting better, well,
it must be so.
> I never see you playing the Beginner & Easy Levels
I took you at your word! You said that GC
was now playing as well as Rybka, so it
seemed ludicrous for me to face any of the
higher levels, or to not take odds, just like
the numerous GMs and IMs have been
doing lately (with little success, I might add).
Rybka, as everyone should know by now,
offers GMs such odds as pawn-and-move, or
White in every game, and still prevails! She
also has given everything short of Knight
odds to IMs (and pretend-GMs), generally
coming out on top here as well. The only
question seems to be: WHEN, not if, Rybka
will start giving weaker GMs QN odds in a
serious match.
My feeling is that there has been something
seriously wrong with the way the openings
are handled by nit-witted humans, who have
been manipulating this part of Rybka's play.
I am also less than impressed by the way in
which "contempt-factor" has been managed
by Rybka's handlers. Perhaps, were these
idiots to just get out of the way and leave
Rybka to herself, we would already be at the
point of QN-odds parity against the average
GM; but this assumes a certain cleverness
in the way of strategy, for the position must
be imbalanced, and most piece trades
avioded. How best to implement such a
strategy? For humans, it is fairly easy, as
we adapt "on the fly".
-- help bot
|
| |
Date: 16 Jan 2009 11:32:13
From:
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 16, 12:22=A0pm, [email protected] wrote:
> On Jan 16, 2:12=A0am, help bot <[email protected]> wrote:
>
> > On Jan 16, 4:17=A0am, Martin Brown <
|
| |
Date: 16 Jan 2009 09:22:00
From:
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 16, 2:12=A0am, help bot <[email protected] > wrote:
> On Jan 16, 4:17=A0am, Martin Brown <
|
| |
Date: 16 Jan 2009 09:13:37
From:
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 16, 6:42=A0am, [email protected] wrote:
>
> You should be ashamed for not learning that nothing you say to
> him will make any difference. =A0He has adequately demonstrated his
> lack of consideration to the readers here AND to those that give
> him good, accurate advice.
>
> K
I have to admit that, to a large degree, you are right. Having worked
with several brilliant chess people (Grandmasters and engine
programmers) over the years on Chessmaster and Majestic Chess, Sanny's
claims about his engine just get my blood boiling because they are so
obviously false as to compel a heated response. (The Occam's Razor
part of my brain is still screaming at me that this has all been a
hoax from the beginning -- I hope that it's right because then "Sanny"
would be brilliant, instead of a complete moron).
Sanny says that Beginner is "2000+". And yet it drew a game against a
1300-rated Chessmaster personality (that plays instantly) because his
program does not know the rules of chess. This despite years of
programming and advice from many people who were initially trying to
help. He now only has help bot who tries to assist his improvements,
but I can even tell that help bot sometimes has a difficult time
hiding his amusement in how bad the GetClub engine is.
I guess I just need to ignore this board entirely, because the only
two people who post appear to be completely insane (Sanny and Sam
Sloan). Let's hope I don't miss an interesting Chessmaster post. :-)
jm
|
| |
Date: 16 Jan 2009 08:12:32
From: Sanny
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
> =A0 Generally speaking, a crude chess pro-
> gram ought to do well in simple endings,
> via brute-force calculation. =A0 =A0But in the
> case of GetClub, there seems to be a
> definite speed-up near the end, along with
> lousy play and some strange behaviors,
> such as willingly trapping its own King on
> the back rank (delusion of King "safety")
> when it should be fighting madly, as in a
> Mel Gibson movie.
Should King be allowed to move on foraward ranks? There is huge
penalty for King leaving the Back Rank in the middle game. Its done so
that king remain safe. Should I change it and allow the King to move
arround?
> =A0 Against these lower levels, I haven't
> noticed any particular improvement to
> match the hype. =A0 After being pronouced
> "as good as Rybka", I gave myself odds
> and crushed GetClub easily; this would
> never happen against Rybka... er, unless
> the odds were Q, R, R, B, B, N, N, and
> I somehow kept the intitiative. =A0 ;>D
Since you heard Play as good as Rybka I suppose your game Style has
also improved. Now you are playing better moves than earlier.
I never see you playing the Beginner & Easy Levels
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
|
| |
Date: 16 Jan 2009 06:57:52
From: Taylor Kingston
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 16, 2:16=A0am, billbrock <[email protected] > wrote:
> On Jan 15, 11:46=A0pm, Sanny <[email protected]> wrote:
>
>
>
>
>
> > On Jan 15, 11:22=A0pm, "Wlodzimierz Holsztynski (Wlod)"
>
> > <[email protected]> wrote:
> > > On Jan 15, 12:57=A0am, Sanny <[email protected]> wrote:
>
> > > > Today GetClub was further improved.
>
> > > The sign of times. We used to have
> > > Berlinger and his crowd on rgc--now
> > > we have Sanny.
>
> > Who was Berlinger?
>
> > Bye
> > Sanny
>
> > Play Chess at:http://www.GetClub.com/Chess.html
>
> He played a famous postal game in the Ulvestagd Variation against
> Estring.
No, wasn't that Gstring?
|
| |
Date: 16 Jan 2009 06:48:01
From:
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 16, 4:17 am, Martin Brown <
|
| |
Date: 16 Jan 2009 06:42:18
From:
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 15, 9:26 pm, [email protected] wrote:
> On Jan 15, 12:57 am, Sanny <[email protected]> wrote:
>
>
>
> > No need to play with Easy Level as Beginner will play a lot stronger
> > game. What do Taylor Kingston say about the new Strength. Now even he
> > will agree that GetClub is unbeatable.
>
> As always, Sanny, you are a big stupid liar, and you are promoting a
> big stupid program.
And you continue to fuel his visits here by feeding his posts.
Killfile
and ignore. Please. He might just stop.
> You should be incredibly ashamed of even calling your atrocity a
> "chess program", when it doesn't even know the rules of chess.
You should be ashamed for not learning that nothing you say to
him will make any difference. He has adequately demonstrated his
lack of consideration to the readers here AND to those that give
him good, accurate advice.
K
|
| |
Date: 16 Jan 2009 02:12:05
From: help bot
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 16, 4:17=A0am, Martin Brown <
|
| |
Date:
From: Martin Brown
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
|
| |
Date: 15 Jan 2009 23:36:52
From: Sanny
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 16, 12:16=A0pm, billbrock <[email protected] > wrote:
> On Jan 15, 11:46=A0pm, Sanny <[email protected]> wrote:
>
>
>
>
>
> > On Jan 15, 11:22=A0pm, "Wlodzimierz Holsztynski (Wlod)"
>
> > <[email protected]> wrote:
> > > On Jan 15, 12:57=A0am, Sanny <[email protected]> wrote:
>
> > > > Today GetClub was further improved.
>
> > > The sign of times. We used to have
> > > Berlinger and his crowd on rgc--now
> > > we have Sanny.
>
> > Who was Berlinger?
>
> > Bye
> > Sanny
>
> > Play Chess at:http://www.GetClub.com/Chess.html
>
> He played a famous postal game in the Ulvestagd Variation against
> Estring.- Hide quoted text -
>
Yes, I got more details about him at wikipedia.
http://en.wikipedia.org/wiki/Hans_Berliner
He developed a Chess Program "Hitech" His program beat defeated GM
Arnold Denker 3=BD-=BD in a match in 1988.
It also win North American Computer Chess Championship in 1989.
Does anyone has "Hitech" Chess Program play it against GetClub and
show us the game.
Bye
Sanny
Play Chess at:http://www.GetClub.com/Chess.html
|
| |
Date: 15 Jan 2009 23:16:32
From: billbrock
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 15, 11:46=A0pm, Sanny <[email protected] > wrote:
> On Jan 15, 11:22=A0pm, "Wlodzimierz Holsztynski (Wlod)"
>
> <[email protected]> wrote:
> > On Jan 15, 12:57=A0am, Sanny <[email protected]> wrote:
>
> > > Today GetClub was further improved.
>
> > The sign of times. We used to have
> > Berlinger and his crowd on rgc--now
> > we have Sanny.
>
> Who was Berlinger?
>
> Bye
> Sanny
>
> Play Chess at:http://www.GetClub.com/Chess.html
He played a famous postal game in the Ulvestagd Variation against
Estring.
|
| |
Date: 15 Jan 2009 21:46:43
From: Sanny
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 15, 11:22=A0pm, "Wlodzimierz Holsztynski (Wlod)"
<[email protected] > wrote:
> On Jan 15, 12:57=A0am, Sanny <[email protected]> wrote:
>
> > Today GetClub was further improved.
>
> The sign of times. We used to have
> Berlinger and his crowd on rgc--now
> we have Sanny.
Who was Berlinger?
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
|
| |
Date: 15 Jan 2009 21:44:44
From: Sanny
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 16, 1:33=A0am, Offramp <[email protected] > wrote:
> Does anyone remember this:
> 1. Nb1-c3 d7-d5 2. d2-d4 f7-f6 3. e2-e3 g7-g5 4. Bf1-d3 Nb8-d7 5.Qh5#
There was a bug that time. That was a game played 2 months back. Today
GetClub plays much stronger Games.
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
|
| |
Date: 15 Jan 2009 21:43:35
From: Sanny
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
> Once again, Beginner Level could not beat a 1300-rated personality in
> Chessmaster. This time, I will admit, Beginner had the game won
> against Lacey, being up a Knight+Pawn, and was actually performing as
> advertised in terms of average seconds per move.
Thats good that beginner was playing fast.
GetClub was a Knight+Pawn = 3+1 = 4 Points up
In such cases GetClub do not draw. When you play repeated moves it
changes its move. So that Three fold repetition do not take place.
> But then came the endgame, in which your BIG STUPID PROGRAM still
> doesn't know what a 3-fold repetition draw is.
> 54.Kf2 Kg5 55.Kg1 Kg4 56.Kh2 Ng3 57.Kg1
> Kf4 58.Kh2 Kg4 59.Kg1 Nf5 60.Kh2 Ng3 1/2-1/2
Here What White played. (WHITE)
55. Kg1
56 Kh2
57 Kg1
58. Kh2
49. Kg1
60. Kh2
So White is repeating same moves for last 5 moves. Now consider Black
Moves
and GetClub Played. (BLACK)
55 ..... Kg4
56 ..... Ng3
57 ..... Kf4
58 ..... Kg4
59 ..... Nf5
60..... Ng3
So you can see Black (GetClub) is not repeating the same moves. Three
FOLD REPETITION requires both sides to repeat their same moves. You
can see above that White was repeating its moves but Black was always
changing its move. GetClub was changing moves because it was Knight +
Pawn up. So it wanted some way to win the game instead of drawing it.
Incase you want to draw, snatch its extra Knight & Pawn then it will
allow Draw by three fold Repetition.
Or bring End game where only Knight is left then it will Draw the
game.
One more thisg can be done Play till 100 moves after that Computer
will allow you to draw the game by pressing "DRAW GAME BUTTON". For
that you have to continue the game till 100 moves. Then the "Resign
Now" Button converts into "Draw Now Button".
I am glad that GetClub Beginner Level good challenge to Lacey.
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
> You should be incredibly ashamed of even calling your atrocity a
> "chess program", when it doesn't even know the rules of chess. There
> are probably several hundred programs that are stronger than yours,
> and also actually know the rules!
>
> My God, you're an idiot.
>
> jm
|
| |
Date: 15 Jan 2009 19:26:17
From: help bot
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 15, 11:09=A0am, J=FCrgen R. <[email protected] > wrote:
> OK I'll play against your great program. Perhaps you know that
> the great Sam Sloan has accused me of being unable to play chess.
> If I could only get past your interface I would beat your 2600+
> master level and the Loon would have to eat its words.
>
> However, when I click on master level a page with some ads shows -
> but no chess board. And if I then click on 'play free' I'm back on
> the start page.
>
> So how are you supposed to play against this award-winning robot?- Hide q=
uoted text -
One of the many improvements recently
implemented is a feature which magically
detects the skill level of prospective
opponents. In your case, it is very likely
that you were "locked out" after the GC
program detected it was not even in your
class, chess-wise. This new feature is
but one of numerous advancements in the
field-- a field now led by Sanny and his
crack chess programming team.
-- help bot
|
| |
Date: 15 Jan 2009 18:26:07
From:
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 15, 12:57=A0am, Sanny <[email protected] > wrote:
>
> No need to play with Easy Level as Beginner will play a lot stronger
> game. What do Taylor Kingston say about the new Strength. Now even he
> will agree that GetClub is unbeatable.
As always, Sanny, you are a big stupid liar, and you are promoting a
big stupid program.
Once again, Beginner Level could not beat a 1300-rated personality in
Chessmaster. This time, I will admit, Beginner had the game won
against Lacey, being up a Knight+Pawn, and was actually performing as
advertised in terms of average seconds per move.
But then came the endgame, in which your BIG STUPID PROGRAM still
doesn't know what a 3-fold repetition draw is.
[Date "2009.1.15"]
[White "Lacey"]
[Black "Beginner"]
[Result "1/2-1/2"]
1.d4 d5 2.Nf3 Nf6 3.Bf4 e6 4.e3 Ne4 5.Nbd2 Bd6 6.Nxe4 dxe4 7.Bb5+ Bd7
8.Bxd6 Bxb5 9.a4 exf3 10.axb5 fxg2 11.Rg1 Qxd6 12.Rxg2 Qb4+ 13.c3
Qxb2
14.Rxg7 Qxc3+ 15.Kf1 Nd7 16.Kg1 e5 17.dxe5 Qxe5 18.Qd4 Qxd4 19.exd4
Nb6
20.Re1+ Kf8 21.Rg5 Rg8 22.f4 Rd8 23.Rxg8+ Kxg8 24.Ra1 Rxd4 25.Rxa7
Rxf4
26.Rxb7 Rb4 27.Rxc7 Rb1+ 28.Kf2 Rb2+ 29.Kf3 Rb3+ 30.Ke4 Rb2 31.Kf3
Rb3+
32.Ke4 Rxb5 33.Rb7 f5+ 34.Kf3 h5 35.h4 Rb4 36.Kf2 f4 37.Kf3 Rb3+
38.Ke2
f3+ 39.Kf2 Kh8 40.Rf7 Nc4 41.Rxf3 Rxf3+ 42.Kxf3 Kg7 43.Ke4 Kg6 44.Kd4
Nd6 45.Ke5 Nf5 46.Ke4 Kf6 47.Kd5 Nxh4 48.Ke4 Nf5 49.Kf4 Kg6 50.Ke5 h4
51.Ke4 Ng7 52.Ke3 h3 53.Kf3 Nf5 54.Kf2 Kg5 55.Kg1 Kg4 56.Kh2 Ng3
57.Kg1
Kf4 58.Kh2 Kg4 59.Kg1 Nf5 60.Kh2 Ng3 1/2-1/2
You should be incredibly ashamed of even calling your atrocity a
"chess program", when it doesn't even know the rules of chess. There
are probably several hundred programs that are stronger than yours,
and also actually know the rules!
My God, you're an idiot.
jm
|
| |
Date: 15 Jan 2009 12:33:28
From: Offramp
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
Does anyone remember this:
1. Nb1-c3 d7-d5 2. d2-d4 f7-f6 3. e2-e3 g7-g5 4. Bf1-d3 Nb8-d7 5.Qh5#
|
| |
Date: 15 Jan 2009 10:22:40
From: Wlodzimierz Holsztynski (Wlod)
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 15, 12:57=A0am, Sanny <[email protected] > wrote:
> Today GetClub was further improved.
The sign of times. We used to have
Berlinger and his crowd on rgc--now
we have Sanny.
Wlod
|
| |
Date: 15 Jan 2009 08:32:10
From: Sanny
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
> OK I'll play against your great program. Perhaps you know that
> the great Sam Sloan has accused me of being unable to play chess.
> If I could only get past your interface I would beat your 2600+
> master level and the Loon would have to eat its words.
Lets hope so.
>
> However, when I click on master level a page with some ads shows -
> but no chess board. And if I then click on 'play free' I'm back on
> the start page.
Do not click Just Scroll Down for playing with the Applet below the
Ads.
>
> So how are you supposed to play against this award-winning robot?- Hide quoted text -
I found you started a game with Easy Level.
But for some reasion you were unable to play that game.
Once the page is loaded with some Ads pages. The Chess applet is below
the Ads. It takes 30 sec to download the Java Applet. The size of
Applet is quite large.
You can see the applet below "Java is required to play the game."
Here are the steps.
http://www.GetClub.com/Chess.html
1. Login First.
2. Click on any buttons to start the Applet Page
3. Wait for Applet to download.
Once Applet is downloaded
1. Choose your Color
2. Choose your Level.
Initially you can play on with Beginner & Easy Level as your ratings
is low. Once you get high ratings You can play with higher Levels.
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
|
| |
Date: 15 Jan 2009 07:38:19
From: Sanny
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 15, 4:35=A0pm, [email protected] wrote:
> I really must ask ....
>
> Are you really just this deluded or are you
> simply trolling for ridicule? GitClub is not
> and has never been a strong program by
> any stretch of the imagination. To this day
What was the last time you played at GetClub? The game was much
improved this month. Whats your username at GetClub? Have you played a
game recently at GetClub?
Play Chess at: http://www.GetClub.com/Chess.html
> Somewhere
> in the depths of my attic I have an old
> 486 laptop with a venerable copy of
> Chess Genius on it (circa 1994 I believe).
> One of these days, when I am overcome
> with boredom I will drag it out and I am
> utterly confident it will beat down GitClub
> like the carrot that it is, not that I have
> any hope that such imperical evidence
> will remove the fog you seem to view the
> strength of your program through.
Old is Gold, So may be that sleeping beast beat the GetClub. Do not
give 486 to GetClub. GetClub needs atleast a pentium 500 to play
reasionably.
>
> As an exercise in the personal development
> of a browser accesable, java based chess
> program GitClub was of at least some interest
> to me early on. Even with it's severe limitations
> the concept was still laudable. Unfortunately
> your utterly insane proclamations regarding
> it's playing strength have rendered it a laughing
> stock.
Java is slower than downloaded app. So it is difficult to have
comparision of an Applet to downloaded program.
> If you truely do believe most of the things that
> you post here I would strongly recommend
> spending all of that energy seeking some
> professional help with recovering a grasp of
> reality.
Help bot plays daily so he knows what progress has been made recently.
Earlier he used to beat the Master & Advance Level. Now he finds it
difficult winning the Easy Level.
So help bot knows the progress GetClub has made.
Bye
Sanny
Play Chess at: http://www.GetClub.com/Chess.html
|
| | |
Date: 15 Jan 2009 17:09:41
From: =?iso-8859-1?Q?J=FCrgen_R.?=
Subject: AW: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
Sanny wrote:
> On Jan 15, 4:35 pm, [email protected] wrote:
>> I really must ask ....
>>
>> Are you really just this deluded or are you
>> simply trolling for ridicule? GitClub is not
>> and has never been a strong program by
>> any stretch of the imagination. To this day
>
> What was the last time you played at GetClub? The game was much
> improved this month. Whats your username at GetClub? Have you played a
> game recently at GetClub?
>
> Play Chess at: http://www.GetClub.com/Chess.html
>
>> Somewhere
>> in the depths of my attic I have an old
>> 486 laptop with a venerable copy of
>> Chess Genius on it (circa 1994 I believe).
>> One of these days, when I am overcome
>> with boredom I will drag it out and I am
>> utterly confident it will beat down GitClub
>> like the carrot that it is, not that I have
>> any hope that such imperical evidence
>> will remove the fog you seem to view the
>> strength of your program through.
>
> Old is Gold, So may be that sleeping beast beat the GetClub. Do not
> give 486 to GetClub. GetClub needs atleast a pentium 500 to play
> reasionably.
>
>>
>> As an exercise in the personal development
>> of a browser accesable, java based chess
>> program GitClub was of at least some interest
>> to me early on. Even with it's severe limitations
>> the concept was still laudable. Unfortunately
>> your utterly insane proclamations regarding
>> it's playing strength have rendered it a laughing
>> stock.
>
> Java is slower than downloaded app. So it is difficult to have
> comparision of an Applet to downloaded program.
>
>> If you truely do believe most of the things that
>> you post here I would strongly recommend
>> spending all of that energy seeking some
>> professional help with recovering a grasp of
>> reality.
>
> Help bot plays daily so he knows what progress has been made recently.
> Earlier he used to beat the Master & Advance Level. Now he finds it
> difficult winning the Easy Level.
>
> So help bot knows the progress GetClub has made.
>
> Bye
> Sanny
>
> Play Chess at: http://www.GetClub.com/Chess.html
OK I'll play against your great program. Perhaps you know that
the great Sam Sloan has accused me of being unable to play chess.
If I could only get past your interface I would beat your 2600+
master level and the Loon would have to eat its words.
However, when I click on master level a page with some ads shows -
but no chess board. And if I then click on 'play free' I'm back on
the start page.
So how are you supposed to play against this award-winning robot?
|
| |
Date: 15 Jan 2009 03:35:06
From:
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
I really must ask ....
Are you really just this deluded or are you
simply trolling for ridicule? GitClub is not
and has never been a strong program by
any stretch of the imagination. To this day
it remains unable to even comply with all
of the game's rules. Any significant real
chess program can beat the stuffing out
of GitClub nor matter how many processor
cores are thrown into the mix. Somewhere
in the depths of my attic I have an old
486 laptop with a venerable copy of
Chess Genius on it (circa 1994 I believe).
One of these days, when I am overcome
with boredom I will drag it out and I am
utterly confident it will beat down GitClub
like the carrot that it is, not that I have
any hope that such imperical evidence
will remove the fog you seem to view the
strength of your program through.
As an exercise in the personal development
of a browser accesable, java based chess
program GitClub was of at least some interest
to me early on. Even with it's severe limitations
the concept was still laudable. Unfortunately
your utterly insane proclamations regarding
it's playing strength have rendered it a laughing
stock.
If you truely do believe most of the things that
you post here I would strongly recommend
spending all of that energy seeking some
professional help with recovering a grasp of
reality.
|
| |
Date: 15 Jan 2009 01:17:36
From: help bot
Subject: Re: Worlds Strongest Chess Program - GetClub Chess is one of them.
|
On Jan 15, 3:57=A0am, Sanny <[email protected] > wrote:
> Today GetClub was further improved. And now GetClub is one of the
> Strongest program in the world. As it is playing as good as Rybka
> Which is worlds fastest Chess Program.
The folks at Rybka once claimed they were
about average in terms of raw speed, but had
a better evaluation function-- especially with
regard to unbalanced (or was it double-edged?)
positions; average among the top programs,
that is.
> Now GetClub is playing as good as Rybka program (Free Version)
Uh oh. I will have to accept odds from now on.
> Play Chess at:http://www.GetClub.com/Chess.html
>
> So now the new ratings to GetClub levels are
>
> Beginner: 2000+
> Easy: 2200+
> Normal: 2400+
> Master: 2600+
That's odd; these numbers don't match up with
Rybka's (except on a Radio Shack TSR-80,
perhaps).
> Anyone having Fritz/ Chess Master and other strong Chess Programs Play
> GetClub against them and let me know if they are able to beat GetClub
> or not.
Why bother? If what Sanny says is true,
throw all those inferior programs away and
just "go with the best, forget the rest"!
> Give same time to both Players. Play against Easy Level: (30 sec /
> move ).
>
> Do not use dual core / quad core as GetClub utilize only 1 thread
Same here! In fact, whenever I try to
analyze twelve games at once, as with a
Pillsbury or a Najdorf blindfold simul, I
lose all the threads, the spools, the
needles-- everything.
> So if you have a multicore then give GetClub 2-4 times extra time
> depending on the number of multiprocessors you have
>
> At the moment GetClub is as strong as the Free Version of Rybka.
Hmm... I can still play Baby level, accepting
Rook odds and taking White. The plan will
be to "do nothing, and do it well", while I wait
for more bugs to appear.
> No need to play with Easy Level as Beginner will play a lot stronger
> game. What do Taylor Kingston say about the new Strength. Now even he
> will agree that GetClub is unbeatable.
I predict that TK will either say nothing at
all, or will suddenly appear with another of
his computer-assisted wins, played under
an alternate moniker to protect his delicate
ego if he should somehow lose.
The real test will be if Zeb the computer
operator can still smash GetClub, as before.
As far as I could tell, he was probably using
some other, inferior program, so if GC is
now "as strong as Rybka", Zeb will be in
for quite a shock!
-- help bot
|
|
