Sam Sloan's Chess Lesson Video Posted on the Internet

Main

Date: 26 May 2005 01:27:32
From: Sam Sloan
Subject: Sam Sloan's Chess Lesson Video Posted on the Internet

I have posted as a streaming video my Chess Lesson #2 as broadcast on
Cable TV last week.

The address is
http://www.ishipress.com/samchess2.asf

Lesson 2 covers some basic opening traps such as the basic trap in the
Albin Counter Gambit and basic rook and pawn endgames such as the
Lucena Position. It also demonstrates the infamous Keres-Botvinnik
1948 World Championship Game where Keres dumped the game to insure
that Botvinnik and not Reshevsky would be World Chess Champion.

The title of this episode is "Chess: Basic Openings and Basc
Endgames".

I need to thank Gary Popkin for providing both the inspiration and the
technical expertise for this show. II also need to thank Leshaun
Fossett for capturing and recording the streaming video for me.

Anybody who views this video, please post feedback here and tell me
what you think. Among other things, I want to know whether you feel
that this show is ready for prime time.

Sam Sloan

Date: 30 May 2005 18:53:13
From: David Richerby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

[Ludicrous crosspost snipped.]

Sam Sloan <[email protected] > wrote:
> Lesson 2 covers some basic opening traps such as the basic trap in the
> Albin Counter Gambit

I'm confused. After 1.d4 d5 2.c4 e5 3.dxe5 d4 4.e3 Bb4+ 5.Bd2 dxe3 6.Bxb4
exf2+ 7.Ke2 fxg1=N+, you declare that ``There is no square for the king to
move to'', while pointing at e1. Ke1 looks like a perfectly reasonable
move to me -- te worst I can see for White is 8... Qh5+ 9.Kd2 and life,
while not pleasant, continues. Presumably Black can't extricate the Ng1
but will win the pawn on e5 so, materially, White will only be a pawn
down, albeit with a dangerously airy king. Certainly better than losing
the queen.

After 1.d4 f5 2.Bg5 h6 3.Bh4 g5 4.Bg3 f4?? {The Noah's Ark trap is so
called because it's about as old as Noah's Ark, to the best of my
knowledge, not because of any alleged resemblance of the shape of the
pieces to the Ark.} 5.e3 h5 6.Bd3 it would be a good idea to mention that
6... d6 completely busts White's checkmate leaving him with nothing better
than 7.exf4 gxf4 8.Bxf4, a pawn up and in a very nice position.

Your analysis is *very* hard to follow because you refer to the pieces by
the wrong name as often as the right one. The most spectacular example is
``White moves his pawn up to this rank. This is called building a
bridge,'' said as you move a black rook.

Dave.

--
David Richerby Flammable Indelible Projector (TM):
www.chiark.greenend.org.uk/~davidr/ it's like a 16mm film projector
but it can't be erased and it burns
really easily!

Date: 31 May 2005 08:48:50
From: Frisco Del Rosario
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

In article <PHm*[email protected] >, David Richerby
<[email protected] > wrote:

> After 1.d4 f5 2.Bg5 h6 3.Bh4 g5 4.Bg3 f4?? {The Noah's Ark trap is so
> called because it's about as old as Noah's Ark, to the best of my
> knowledge, not because of any alleged resemblance of the shape of the
> pieces to the Ark.} 5.e3 h5 6.Bd3 it would be a good idea to mention that
> 6... d6 completely busts White's checkmate leaving him with nothing better
> than 7.exf4 gxf4 8.Bxf4, a pawn up and in a very nice position.

Not 7...gf, but 7...h4.

--
Frisco Del Rosario
A First Book of Morphy -- http://www.amazon.com/exec/obidos/ASIN/1412039061

Date: 31 May 2005 13:48:06
From: =?ISO-8859-1?Q?Claus-J=FCrgen_Heigl?=
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Frisco Del Rosario wrote:
>>After 1.d4 f5 2.Bg5 h6 3.Bh4 g5 4.Bg3 f4?? {The Noah's Ark trap is so
>>called because it's about as old as Noah's Ark, to the best of my
>>knowledge, not because of any alleged resemblance of the shape of the
>>pieces to the Ark.} 5.e3 h5 6.Bd3 it would be a good idea to mention that
>>6... d6 completely busts White's checkmate leaving him with nothing better
>>than 7.exf4 gxf4 8.Bxf4, a pawn up and in a very nice position.
>
>
> Not 7...gf, but 7...h4.

Absolutely, this looks almost like a refutation. The black king finds
safety on c7.

Better than 6. Bd3 could be 6. Be2. 6...Nf6 looks forced (6...g4 7. Bxf4
and Black is down a pawn for less than nothing), and after 7. exf4 h4 8.
fxg5 hxg3 9. gxf6 White has his piece back with one or two extra pawns.

The original Noah's ark trap is in the Ruy Lopez.

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 d6 5. d4? b5 6. Bb3 Nxd4 7. Nxd4
exd4 8. Qxd4? (recommended by Alekhine who claimed this leads to a draw)
8...c5 9. Qd5 Be6 10. Qc6+ Bd7 11. Qd5 (expecting Be6 and draw by
repetition) 11...c4! Refutation courtesy of Capablanca, in his game
against Steiner.

Claus-Juergen

Date: 31 May 2005 11:03:41
From: David Richerby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Frisco Del Rosario <[email protected] > wrote:
> David Richerby <[email protected]> wrote:
>> After 1.d4 f5 2.Bg5 h6 3.Bh4 g5 4.Bg3 f4?? {The Noah's Ark trap is so
>> called because it's about as old as Noah's Ark, to the best of my
>> knowledge, not because of any alleged resemblance of the shape of the
>> pieces to the Ark.} 5.e3 h5 6.Bd3 it would be a good idea to mention
>> that 6... d6 completely busts White's checkmate leaving him with
>> nothing better than 7.exf4 gxf4 8.Bxf4, a pawn up and in a very nice
>> position.
>
> Not 7...gf, but 7...h4.

Good point. So White is completely busted. He should probably slot in
7.Bg6+ Kd7 to at least prevent Black from castling his piece-up king.

Dave.

--
David Richerby Indelible Portable Widget (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a thingy but you can take it
anywhere and it can't be erased!

Date: 31 May 2005 19:15:27
From: Frisco Del Rosario
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

In article <Xyr*[email protected] >, David Richerby
<[email protected] > wrote:

> >> After 1.d4 f5 2.Bg5 h6 3.Bh4 g5 4.Bg3 f4?? 5.e3 h5 6.Bd3 it would be
a good idea to mention
> >> that 6... d6 completely busts White's checkmate leaving him with
> >> nothing better than 7.exf4 gxf4 8.Bxf4, a pawn up and in a very nice
> >> position.
> >
> > Not 7...gf, but 7...h4.
>
> Good point. So White is completely busted. He should probably slot in
> 7.Bg6+ Kd7 to at least prevent Black from castling his piece-up king.

I'm not saying that Black's game is much improved by 7...h4, but 7...h4
seems to be reason Black played 5...h5: to deal with the mate threat while
putting ...h4 in store (a misguided idea in itself, since ...h4xg3
develops the white rook for free).

If Sloan used this in his TV show, I hope he showed the stem game
Teed-Del, New York 1910, which ended 6. Bd3 Rh6 7. Qh5 Resigns. How
Del, who won the NY state championship a couple of times, stumbled into
all this is a mystery.

--
Frisco Del Rosario
A First Book of Morphy -- http://www.amazon.com/exec/obidos/ASIN/1412039061

Date: 29 May 2005 06:22:30
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

You're *still* missing the point.

Follow the bouncing ball, and all will be revealed.

Date: 29 May 2005 05:42:23
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

You're missing the point, Angelo. Use your intelligence.

Date: 29 May 2005 03:07:20
From: Patrick Volk
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Thu, 26 May 2005 01:27:32 GMT, [email protected] (Sam Sloan)
wrote:

>I have posted as a streaming video my Chess Lesson #2 as broadcast on
>Cable TV last week.
>
>The address is
>http://www.ishipress.com/samchess2.asf
>
>Lesson 2 covers some basic opening traps such as the basic trap in the
>Albin Counter Gambit and basic rook and pawn endgames such as the
>Lucena Position. It also demonstrates the infamous Keres-Botvinnik
>1948 World Championship Game where Keres dumped the game to insure
>that Botvinnik and not Reshevsky would be World Chess Champion.
>
>The title of this episode is "Chess: Basic Openings and Basc
>Endgames".
>
>I need to thank Gary Popkin for providing both the inspiration and the
>technical expertise for this show. II also need to thank Leshaun
>Fossett for capturing and recording the streaming video for me.
>
>Anybody who views this video, please post feedback here and tell me
>what you think. Among other things, I want to know whether you feel
>that this show is ready for prime time.
>
>Sam Sloan

First, lay off the coffee before you go on. Your dialog isn't matching
what you're showing on the board, and your presentation is a little on
the hectic side. Maybe try vodka! Slow down a little bit.

Second, I would recommend you use something on the computer, loading
up the sequences you want to show, and you could put additional
graphics in there (say showing attacking lanes, or potential future
moves without mucking the board all up). I think that would help you
calm down on the speaking side of it.
Between sequences, you can cut back to yourself so you could set up
for the next position.

I could tell that wasn't the original king, it was off-white.

Finally, you don't need to script it word-for-word, but have some sort
of order in the presentation. You seemed to jump around a bit. Maybe
that was the coffee...

But what you described here, to be objective, you seemed to back into
in your show... like you happened upon it. The other thing I think I
picked up on was that you say it's basic, but mentally you seem to be
having a tough time keeping it at that level... you seem to not be
comfortable explaining stuff at that level.

If there's anything, the hectic nature of the presentation probably
spared the public any pretentiousness, or self-promotion ;)

Date: 27 May 2005 19:05:16
From: Thomas Ewald
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

This is a multi-part message in MIME format.

------=_NextPart_000_0133_01C562EF.046682E0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Is Lesson 1 available anywhere on the Net?

Thanks.

--=20
THOMAS EWALD
"Sam Sloan" <[email protected] > wrote in message =
news:[email protected]...
I have posted as a streaming video my Chess Lesson #2 as broadcast on
Cable TV last week.

The address is
http://www.ishipress.com/samchess2.asf

Lesson 2 covers some basic opening traps such as the basic trap in the
Albin Counter Gambit and basic rook and pawn endgames such as the
Lucena Position. It also demonstrates the infamous Keres-Botvinnik
1948 World Championship Game where Keres dumped the game to insure
that Botvinnik and not Reshevsky would be World Chess Champion.

The title of this episode is "Chess: Basic Openings and Basc
Endgames".

I need to thank Gary Popkin for providing both the inspiration and the
technical expertise for this show. II also need to thank Leshaun
Fossett for capturing and recording the streaming video for me.

Anybody who views this video, please post feedback here and tell me
what you think. Among other things, I want to know whether you feel
that this show is ready for prime time.

Sam Sloan
------=_NextPart_000_0133_01C562EF.046682E0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" >
<HTML ><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1" >
<META content=3D"MSHTML 6.00.2900.2627" name=3DGENERATOR >
<STYLE ></STYLE>
</HEAD >
<BODY bgColor=3D#ffffff >
<DIV ><FONT face=3DArial size=3D2>Is Lesson 1 available anywhere on the=20
Net?</FONT ></DIV>
<DIV ><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV ><FONT face=3DArial size=3D2>Thanks.</FONT></DIV>
<DIV ><BR>-- <BR>THOMAS EWALD</DIV>
<DIV >"Sam Sloan" <<A=20
href=3D"mailto:[email protected]" >[email protected]</A>> wrote in =
message=20
<A=20
href=3D"news:[email protected]" >news:429523c1.68105843@=
ca.news.verio.net</A >...</DIV>I=20
have posted as a streaming video my Chess Lesson #2 as broadcast =
on<BR >Cable TV=20
last week.<BR ><BR>The address is<BR><A=20
href=3D"http://www.ishipress.com/samchess2.asf" >http://www.ishipress.com/=
samchess2.asf</A ><BR><BR>Lesson=20
2 covers some basic opening traps such as the basic trap in the<BR >Albin =
Counter=20
Gambit and basic rook and pawn endgames such as the<BR >Lucena Position. =
It also=20
demonstrates the infamous Keres-Botvinnik<BR >1948 World Championship =
Game where=20
Keres dumped the game to insure<BR >that Botvinnik and not Reshevsky =
would be=20
World Chess Champion.<BR ><BR>The title of this episode is "Chess: Basic =
Openings=20
and Basc<BR >Endgames".<BR><BR>I need to thank Gary Popkin for providing =
both the=20
inspiration and the<BR >technical expertise for this show. II also need =
to thank=20
Leshaun<BR >Fossett for capturing and recording the streaming video for=20
me.<BR ><BR>Anybody who views this video, please post feedback here and =
tell=20
me<BR >what you think. Among other things, I want to know whether you=20
feel<BR >that this show is ready for prime time.<BR><BR>Sam =
Sloan</BODY ></HTML>

------=_NextPart_000_0133_01C562EF.046682E0--

Date: 27 May 2005 21:45:41
From: Sam Sloan
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

So far, my Chess Lesson Video has had 273 downloads since I posted it
two days ago.

On the other hand, my io video has had none.

Sam Sloan

Date: 28 May 2005 01:38:00
From: J�rgen R.
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Fri, 27 May 2005 21:45:41 GMT, [email protected] (Sam Sloan)
wrote:

>So far, my Chess Lesson Video has had 273 downloads since I posted it
>two days ago.

Great! Nobody who has seen you live would ever consider voting for you
for anything. So the USCF is safe from the Sloan attack for another
round.

>
>On the other hand, my io video has had none.
>
>Sam Sloan

Date: 27 May 2005 13:49:48
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

LOL - I'll give you that one! Nice post, Angelo!

You're still missing the point, I fear.

Think about it. How justified are my posts to you?

.......

Date: 27 May 2005 13:17:58
From: Liam Too
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

k Houlsby wrote:
> Wrong on all counts.
>
> First of all, it *need not* require tablebases to solve chess. Brute
> force can do it.

You must be kidding, name the computer that can just do brute force
and solve chess. If you can't name one then you're just writing BS
as always.

> Second of all, the universe may not be a universe at all, but a
> multiverse. If it *is* a universe, and it *is* a closed system, then it
> is, nevertheless, finite, at least in terms of the amount of matter it
> contains (for that is what is relevant, here).
>
> Keep banging those rocks together, jackass.

You cannot even discuss things without using profanities. You are worse
than Jason Repa!

Date: 27 May 2005 15:32:56
From: Ubiquitous
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

In article <[email protected] >, [email protected]
wrote:

>I have posted as a streaming video my Chess Lesson #2 as broadcast on
>Cable TV last week.

And this is related to this newsgroup because...?

--
It is simply breathtaking to watch the glee and abandon with which
the liberal media and the Angry Left have been attempting to turn
our military victory in Iraq into a second Vietnam quagmire. Too bad
for them, it's failing.

Date: 27 May 2005 10:45:18
From: Larry Tapper
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

>Am I the only one here who spent 15 minutes googlling "games, infinite
boars" because it sounded darn interesting?

- Geof

Cf A.E. Wardrop's classic work, Hypermodern Pigsticking.

Date: 27 May 2005 10:44:26
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Great minds think alike (and so do ours!)

- k

Date: 27 May 2005 10:27:20
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

>>Well, I'm a maths undergraduate, so I will do my best ;-)
>>It would certainly depend on what order of infinity you are talking

> about here.

> I think that beginning to discuss orders of infinity may be a red
> herring. Chess is what it is. It's not hypothetical.

>>Let us hypothetically, drop the 3 move repitition and 50-move rule, so the game can be infinite. Nevertheless, you can still

> calculate optimum moves and the number of *positions* are finite.

>> The number of positions is finite, sure, but if said positions may
>> repeat, ad infinitum, then, other than Kk, how does the game end?

>If there is no forced win, then the result must be a draw, yes?

Just to be clear... if we're discussing an infinite board with only one
corner at a1, if there's no 50-move rule and no threefold repetition,
then, providing a pawn or other piece of either colour remains on the
board, the game will *never* end... so...

no.

>>I believe it *may* occur in games with infinite boars as well, as long as there can be a clear endpoint to the game.

>Oops! That should have been "boards"

Oh...... so an infinite number of 8x8 boards? Well, let's discuss one
at a time, shall we?!

> Well, if you have an infinite board, no threefold repetition and no
> 50-move rule, other than Kk, how can the game end, since checkmate is
> clearly impossible?

>It is? You have not demonsrated that point to me.

Ok... to be clear... if we're discussing the *practice* of playing
chess, then the *solution* equals *perfect* play for *both sides*. Any
line which allows either king anywhere near either the a-file or the
first rank must be less than optimum, therefore. That was discussing an
infinite board... not an infinite *number* of 8x8 boards.

>>For instance let us

> consider an unbouded chess board on 2 sides, so that a1 is the only
> corner. Then place a white king on a1, a black king an a3, a black
> queen on b3, and a white bishop on a2; white to move.

You cannot say that for certain, besides, this is an extreme example.

>> To get into this fix, white would have to have played less than
> perfectly (to put it mildly).

>You cannot say that for certain, besides, this is an extreme example.

Certainly I can. Any move which allows checkmate is a decisive blunder,
and therefore doesn't represent best play.

>>White is lost as the (set of moves) - (the set of moves which prevent

> checkmate next move) is the empty set. As long as a forced win is
> possible, the game is solved.

> Nope. Not remotely.

>Let me remind you of the object of the game, the object of the game is
to win, by chechmating the enemy king before your opponent does. If
you
can see a forced mate, there is no point finding a better one.

Ah... but that's not true. The object of the game is not *to win*, the
object of the game is *not to lose*. Perfect play, by definition,
avoids losing, since, if it lost, it would not be perfect, would it?
Therefore, with perfect play, there is *no* forced mate for either
side.

>>>Chess is a closed system. Therefore, it's *finite*. It's very, very
>>>large, but it's finite. It can be solved. There even exists an
>>>algorithm to solve chess.

>>Yes, but we may never find it, there may exist more possibilities than

> atoms on the universe, and If I remember my science correctly it would
> therefore take more time than the universe has to calculate it.
> We *have* found the algorithm. We *may not* see the solution, if we
> become extinct first, but, other than that, we shall.

>No, the universe (unless our algorithim found a way to avoid calculating
every position) would end first, it is fundemental theory on physics.

You think that physics theories are perfect? That they describe the
universe (or possibly multiverse) perfectly? Why do you suppose they
are called "theories"?

Unless we become extinct, chess *will* be solved. Already 16 of white's
possible 20 first choices have been more-or-less discarded. BTW I do
think that we will become extinct first.

>>Well, the endgame tables, or tablebases, represent the extent to which
>>chess *has been solved*. If we had 32-man tablebases, we would have the
>>*practical* solution, since we would know precisely best play from both
>>sides. Equally, if one had the "brute force" algorithm playing against
>>itself on a superfast computer inconceivable by today's standards, then
>>it, too, would *generate* a solution.

>>See above

> Huh? Where?

>The bit with universe in.

Could you narrow that down? Post a quote, even?

>>>If this is to 'save-time' then a further cheat is revealed, see Tal

>>anecdote.

>>No, it's irrelevant. See above.

>>>>When chess is solved, it *could* be that the Sicilian Defence (1...c5)
>>>>refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
>>>>against the Sicilian with best play. Now, *if* the Sicilian refutes
>>>>1.e4, then *by definition* it *also* refutes *all* other white openings
>>>>beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
>>>>unplayable, then it's been refuted.

>>It would hardly be a forced loss though if 2.Nf3 led to a forced loss

> and f4 didn;t say, anyway, I agree with Taylor, it's a prevention.
> And you can only refute once IMO, you can't refute the open sicillian
> because it's just part of the same refutation, your move refutes the
> kings pawn opening.

> So, what part of everything which follows 1.e4 for white will not have
> been refuted, in that case? What exactly are you trying to argue?

>I admit I didn't phrase that very well (must be all the caffine). I
will admit that it does depend on what you mean by refutation in a way.

I will try to put it more clearly then, 1. d4 d5 2.Nf3 e5 is not an
Albin as far as I know, so 2. Nf3 cannot be a refutation to it. I take

refutation to be "a counter to an opening continuation has been found
that analysis has proven results in a better position for the oppenent"

Note the word *to* in my definition, not *for*

oh ok

>If you define refutation as "to reach equality or a slightly better
position" then I can understand where you might be coming from, even
then though, it it difficult to say really to any great degree whether
Nf3 is better than c4 or not.

Uh huh.

>>>>>A refutation would have to

>>>>occur after move 3 since you don't have an Albin until Black has played
>>>>...e5.

>>>>Nope. See above.

>>>I think Taylor made a good point between a refutation : making move

>>unplayable

>><snip>

>>I think it's irrelevant.

You are of course entitled to your own opinion ;-)

>>>>>Enough with insulting Kramnik, already!

>>>>But everyone knows *real* men play Kd2

>>>Real Vermont Men often skip this sissy 'flatlander' step and immediately

>>play 1. Kd2 taking their own pawn!

>>Uh huh!

>>>>LOL yeah!

>>>Mano-a-Mano chess, true Yankee style.

>>Ah, thats how I lost my right eye and both my livers! Still won the

> game though.

>>>Phil Innes

>>Bitter Lemonly
>>k

>>Hoping I've made sense,

> Chris Barnett.

> Certainly, most of the time, and thanks for writing.

> k Houlsby

I hope this makes more sense then
Chris Barnett

A little

k

Date: 27 May 2005 10:25:22
From:
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Chris Barnett wrote:

[snip]

>>I believe it *may* occur in games with infinite boars as well, as long as there can be a clear endpoint to the game.
>>Oops! That should have been "boards"

[snip]

Am I the only one here who spent 15 minutes googlling "games, infinite
boars" because it sounded darn interesting?

- Geof

Date: 27 May 2005 09:38:36
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

>k Houlsby wrote:
>>"k Houlsby" <[email protected]> wrote in message

>>news:[email protected]...

>>>>k Houlsby wrote:

>>>>>>That's not technically a refutation of the Albin.

>>>>>Yes it is, moron, it renders ...e5 unsound.

>>>>You can't refute an opening when it it hasn't been played 2...e5 does

>>not make the Albin 1. d5 d5 2. c4 e5 does

>>>Sure you can. Think about it this way: one day, chess will be solved.

>>Chess is an infinite game, and I should like to hear from a mathematician if

> infinite games are solvable.

>Well, I'm a maths undergraduate, so I will do my best ;-)
>It would certainly depend on what order of infinity you are talking
about here.

I think that beginning to discuss orders of infinity may be a red
herring. Chess is what it is. It's not hypothetical.

> Let us hypothetically, drop the 3 move repitition and 50-move rule, so the game can be infinite. Nevertheless, you can still
calculate optimum moves and the number of *positions* are finite.

The number of positions is finite, sure, but if said positions may
repeat, ad infinitum, then, other than Kk, how does the game end?

>I believe it *may* occur in games with infinite boars as well, as long as there can be a clear endpoint to the game.

Well, if you have an infinite board, no threefold repetition and no
50-move rule, other than Kk, how can the game end, since checkmate is
clearly impossible?

>For instance let us
consider an unbouded chess board on 2 sides, so that a1 is the only
corner. Then place a white king on a1, a black king an a3, a black
queen on b3, and a white bishop on a2; white to move.

To get into this fix, white would have to have played less than
perfectly (to put it mildly).

>White is lost as the (set of moves) - (the set of moves which prevent
checkmate next move) is the empty set. As long as a forced win is
possible, the game is solved.

Nope. Not remotely.

>> Chess is a closed system. Therefore, it's *finite*. It's very, very
>> large, but it's finite. It can be solved. There even exists an
>> algorithm to solve chess.

>Yes, but we may never find it, there may exist more possibilities than
atoms on the universe, and If I remember my science correctly it would
therefore take more time than the universe has to calculate it.

We *have* found the algorithm. We *may not* see the solution, if we
become extinct first, but, other than that, we shall.

>>Secondly, what does this term 'solve' mean? Does it indicate a forced win
> draw or loss for the player making the first move, provided that both
> sides
> make 'best moves' thereafter?

> Yes.

>>With human beings, the performance of chess is within a certain time frame -

> and as Tal pointed out, many of his more spectacular combinations are
> refuted the next day, month or year - but within the timeframe allowed,
> was
> it solvable? Chess is a performance activity not a theoretical one.

>> Agreed. It's *still* finite, however.

>>Computers are actually a very long way from solving chess - and to test this
>>hypothesis turn off the opening book and also the endgame tables [both
>>cheating!]. There are no higher level computer ratings which are legal,
>>since all GM matches against computers are played with book/tables=on, and
>>when the book is on then by definition the program is referring to notes,
>>and playing moves it couldn't calculate itself. If it could calculate the
>>moves then , ipso facto, why have opening books and endgame tables?

> Well, the endgame tables, or tablebases, represent the extent to which
> chess *has been solved*. If we had 32-man tablebases, we would have the
> *practical* solution, since we would know precisely best play from both
> sides. Equally, if one had the "brute force" algorithm playing against
> itself on a superfast computer inconceivable by today's standards, then
> it, too, would *generate* a solution.

>See above

Huh? Where?

>>If this is to 'save-time' then a further cheat is revealed, see Tal

> anecdote.

> No, it's irrelevant. See above.

>>>When chess is solved, it *could* be that the Sicilian Defence (1...c5)
>>>refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
>>>against the Sicilian with best play. Now, *if* the Sicilian refutes
>>>1.e4, then *by definition* it *also* refutes *all* other white openings
>>>beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
>>>unplayable, then it's been refuted.

>It would hardly be a forced loss though if 2.Nf3 led to a forced loss
and f4 didn;t say, anyway, I agree with Taylor, it's a prevention.
And you can only refute once IMO, you can't refute the open sicillian
because it's just part of the same refutation, your move refutes the
kings pawn opening.

So, what part of everything which follows 1.e4 for white will not have
been refuted, in that case? What exactly are you trying to argue?

>>>>A refutation would have to

>>>occur after move 3 since you don't have an Albin until Black has played
>>>...e5.

>>>Nope. See above.

>>I think Taylor made a good point between a refutation : making move

> unplayable

> <snip>

> I think it's irrelevant.

>>>>Enough with insulting Kramnik, already!

>>>But everyone knows *real* men play Kd2

>>Real Vermont Men often skip this sissy 'flatlander' step and immediately

> play 1. Kd2 taking their own pawn!

> Uh huh!

>>>LOL yeah!

>>Mano-a-Mano chess, true Yankee style.

>Ah, thats how I lost my right eye and both my livers! Still won the
game though.

>>Phil Innes

> Bitter Lemonly
> k

>Hoping I've made sense,
Chris Barnett.

Certainly, most of the time, and thanks for writing.

k Houlsby

Date: 27 May 2005 18:09:46
From: Chris Barnett
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

>>Well, I'm a maths undergraduate, so I will do my best ;-)
>>It would certainly depend on what order of infinity you are talking
>
> about here.
>
> I think that beginning to discuss orders of infinity may be a red
> herring. Chess is what it is. It's not hypothetical.
>
>
>>Let us hypothetically, drop the 3 move repitition and 50-move rule, so the game can be infinite. Nevertheless, you can still
>
> calculate optimum moves and the number of *positions* are finite.
>
> The number of positions is finite, sure, but if said positions may
> repeat, ad infinitum, then, other than Kk, how does the game end?

If there is no forced win, then the result must be a draw, yes?

>
>>I believe it *may* occur in games with infinite boars as well, as long as there can be a clear endpoint to the game.

Oops! That should have been "boards"

>
> Well, if you have an infinite board, no threefold repetition and no
> 50-move rule, other than Kk, how can the game end, since checkmate is
> clearly impossible?
>

It is? You have not demonsrated that point to me.

>
>>For instance let us
>
> consider an unbouded chess board on 2 sides, so that a1 is the only
> corner. Then place a white king on a1, a black king an a3, a black
> queen on b3, and a white bishop on a2; white to move.

You cannot say that for certain, besides, this is an extreme example.

> To get into this fix, white would have to have played less than
> perfectly (to put it mildly).

You cannot say that for certain, besides, this is an extreme example.

>>White is lost as the (set of moves) - (the set of moves which prevent
>
> checkmate next move) is the empty set. As long as a forced win is
> possible, the game is solved.
>
> Nope. Not remotely.
>

Let me remind you of the object of the game, the object of the game is
to win, by chechmating the enemy king before your opponent does. If you
can see a forced mate, there is no point finding a better one.

>
>
>>>Chess is a closed system. Therefore, it's *finite*. It's very, very
>>>large, but it's finite. It can be solved. There even exists an
>>>algorithm to solve chess.
>
>
>
>>Yes, but we may never find it, there may exist more possibilities than
>
> atoms on the universe, and If I remember my science correctly it would
> therefore take more time than the universe has to calculate it.

> We *have* found the algorithm. We *may not* see the solution, if we
> become extinct first, but, other than that, we shall.

No, the universe (unless our algorithim found a way to avoid calculating
every position) would end first, it is fundemental theory on physics.

>
>
>>Well, the endgame tables, or tablebases, represent the extent to which
>>chess *has been solved*. If we had 32-man tablebases, we would have the
>>*practical* solution, since we would know precisely best play from both
>>sides. Equally, if one had the "brute force" algorithm playing against
>>itself on a superfast computer inconceivable by today's standards, then
>>it, too, would *generate* a solution.
>
>
>
>
>>See above
>
>
> Huh? Where?
>

The bit with universe in.

>>>If this is to 'save-time' then a further cheat is revealed, see Tal
>
>
>>anecdote.
>
>
>
>>No, it's irrelevant. See above.
>
>
>
>>>>When chess is solved, it *could* be that the Sicilian Defence (1...c5)
>>>>refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
>>>>against the Sicilian with best play. Now, *if* the Sicilian refutes
>>>>1.e4, then *by definition* it *also* refutes *all* other white openings
>>>>beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
>>>>unplayable, then it's been refuted.
>
>
>
>
>>It would hardly be a forced loss though if 2.Nf3 led to a forced loss
>
> and f4 didn;t say, anyway, I agree with Taylor, it's a prevention.
> And you can only refute once IMO, you can't refute the open sicillian
> because it's just part of the same refutation, your move refutes the
> kings pawn opening.
>
> So, what part of everything which follows 1.e4 for white will not have
> been refuted, in that case? What exactly are you trying to argue?

I admit I didn't phrase that very well (must be all the caffine). I
will admit that it does depend on what you mean by refutation in a way.
I will try to put it more clearly then, 1. d4 d5 2.Nf3 e5 is not an
Albin as far as I know, so 2. Nf3 cannot be a refutation to it. I take
refutation to be "a counter to an opening continuation has been found
that analysis has proven results in a better position for the oppenent"
Note the word *to* in my definition, not *for*

If you define refutation as "to reach equality or a slightly better
position" then I can understand where you might be coming from, even
then though, it it difficult to say really to any great degree whether
Nf3 is better than c4 or not.

>
>>>>>A refutation would have to
>
>
>
>>>>occur after move 3 since you don't have an Albin until Black has played
>>>>...e5.
>
>
>
>>>>Nope. See above.
>
>
>
>>>I think Taylor made a good point between a refutation : making move
>
>
>
>>unplayable
>
>
>
>><snip>
>
>
>
>>I think it's irrelevant.
>

You are of course entitled to your own opinion ;-)

>
>>>>>Enough with insulting Kramnik, already!
>
>
>
>>>>But everyone knows *real* men play Kd2
>
>
>
>>>Real Vermont Men often skip this sissy 'flatlander' step and immediately
>
>
>
>>play 1. Kd2 taking their own pawn!
>
>
>
>>Uh huh!
>
>
>
>>>>LOL yeah!
>
>
>
>>>Mano-a-Mano chess, true Yankee style.
>
>
>
>
>>Ah, thats how I lost my right eye and both my livers! Still won the
>
> game though.
>
>
>
>>>Phil Innes
>
>
>
>>Bitter Lemonly
>>k
>
>
>
>
>>Hoping I've made sense,
>
> Chris Barnett.
>
> Certainly, most of the time, and thanks for writing.
>
> k Houlsby
>

I hope this makes more sense then
Chris Barnett

Date: 27 May 2005 09:14:06
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

>If the memory of a computer to store the EGTB, that is required to
solve chess is larger than the universe, then chess is deemed to be
infinite as to its solvability. The universe can be a closed system,
but it's infinite.

Wrong on all counts.

First of all, it *need not* require tablebases to solve chess. Brute
force can do it.

Second of all, the universe may not be a universe at all, but a
multiverse. If it *is* a universe, and it *is* a closed system, then it
is, nevertheless, finite, at least in terms of the amount of matter it
contains (for that is what is relevant, here).

Keep banging those rocks together, jackass.

Date: 27 May 2005 09:00:52
From: Liam Too
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

k Houlsby wrote:
> Chess is a closed system. Therefore, it's *finite*. It's very, very
> large, but it's finite. It can be solved. There even exists an
> algorithm to solve chess.

If the memory of a computer to store the EGTB, that is required to
solve chess is larger than the universe, then chess is deemed to be
infinite as to its solvability. The universe can be a closed system,
but it's infinite.

Date: 29 May 2005 18:41:39
From: Patrick Volk
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On 27 May 2005 09:00:52 -0700, "Liam Too" <[email protected] >
wrote:

>k Houlsby wrote:
>> Chess is a closed system. Therefore, it's *finite*. It's very, very
>> large, but it's finite. It can be solved. There even exists an
>> algorithm to solve chess.
>
>If the memory of a computer to store the EGTB, that is required to
>solve chess is larger than the universe, then chess is deemed to be
>infinite as to its solvability. The universe can be a closed system,
>but it's infinite.

If it is finite, it is considered solvable. Whether or not it is
feasible doesn't matter from a scientific standpoint.

The universe is not a closed system either.

Given that at any point in a chess match there are a finite number of
board positions, there are a finite number of solutions that is less
than the number of board positions. There likely are multiple
solutions, but not infinite.
There can conceivably be an infinite number of moves, but any move
maps to a given board position.

Date: 30 May 2005 15:36:54
From: David Richerby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Patrick Volk <[email protected] > wrote:
> If it is finite, it is considered solvable. Whether or not it is
> feasible doesn't matter from a scientific standpoint.

Of course it matters. There's a world of a difference between `soluble in
principle' and `practically soluble'. Any `scientific' theory that omits
this distinction is missing something important.

> There can conceivably be an infinite number of moves, but any move
> maps to a given board position.

There cannot be an infinite number of moves. There are a finite number of
board positions and, in each of those one of at most 32 pieces can move to
one of at most 63 squares. Hence, there are only finitely many moves.

Dave.

--
David Richerby Solar-Powered Chair (TM): it's like a
www.chiark.greenend.org.uk/~davidr/ chair but it doesn't work in the dark!

Date: 27 May 2005 08:56:06
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Taylor Kingston wrote:
> k, in all cordiality, I strongly urge that you reconsider.

Ok. In all cordiality, I strongly urge you to take note that you're
jumping into the middle of a flame war between Angelo and me. If I'm
using defamatory language, it's *always* been provoked. Do you see?

>In this case Mr. DePalma is correct.

No, he isn't. Get the *context* right, Taylor. In *context* he's
totally wrong.

>1.d4 d5 2.Nf3 is a *prevention* of
the Albin, but not a *refutation*.

Huh? So if 1...c5 refutes 1.e4, then it's not a refutation? Lines and
moves are either refuted, or playable.

>A refutation can only stem from the
position after 1.d4 d5 2.c4 e5. To call 1.Nf3 or 2.Nf3 a refutation of
the Albin is like saying 1.e4 a6 is a refutation of the Ruy L=F3pez
since it prevents 3.Bb5.

No. It doesn't *prevent* 3.Bb5 or even 2.Bb5, it just renders it rather
silly (unless it proves to be a deep piece sac).

> In some circumstances even 1.Nf3 may not be a successful anti-Albin
approach. I recall Reuben Fine talking about a game with Weaver Adams,
an Albin expert. To avoid Adams' pet line, Fine opened (IIRC) 1.Nf3,
and the game proceeded 1...Nc6 2.d4 d5 3.c4 and after 3...e5!? guess
what: it was an Albin Counter Gambit.

Yes, I know the game:

Fine,R (2600) - Adams,W [D08]
New York ch, 1939

1=2ENf3 Nc6 2.c4 e5 3.d4 d5 4.dxe5 d4 5.a3 a5 6.g3 Be6 7.Nbd2 Bc5 8.Bg2
Nge7 9.0-0 0-0 10.Qc2 Ng6 11.Nb3 Ba7 12.Bg5 Qd7 13.Rad1 Ngxe5 14.Nxe5
Nxe5 15.c5 d3 16.exd3 Qa4 17.Na1 Qxc2 18.Nxc2 Bb3 19.d4 Bxc2 20.Rd2 Bb3
21.dxe5 Bxc5 22.Rc1 Bb6 23.Bxb7 Rae8 24.Bf4 Rd8 25.Be4 Rxd2 26.Bxd2 Re8
27.Bc3 f6 28.Re1 fxe5 29.Bf3 Rf8 30.Kg2 Bxf2 31.Kxf2 Bd5 32.Rxe5 Bxf3
33.Bxa5 c6 34.Bb4 1-0

Once again, please try to avoid jumping into the middle of my
discussions.

Warmest regards,
k

Date: 27 May 2005 16:45:40
From: Angelo DePalma
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

I strongly urge everyone who posts here to ask k "k My Holy Cornhole"
Houlsby before posting. This is his newsgroup. These are his discussions. He
doesn't approve of dissent, logic, common sense. Please don't get him
excited or he will forget to take his medicine and thereby violate his
probation.

"k Houlsby" <[email protected] > wrote

>Once again, please try to avoid jumping into the middle of my
>discussions.

>Warmest regards,
> k

Date: 27 May 2005 08:48:03
From: Taylor Kingston
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Chess One wrote:
> I think Taylor made a good point between a refutation : making move
> unplayable

Thank you, Phil. BTW, I found where Fine describes his Albin Counter
Gambit game with Weaver Adams. It is in "Great Moments in Modern Chess"
(aka "The World's a Chessboard), Dover 1948, page 280, in his notes to
the game Zita-Bronstein, Prague-Moscow Match 1946:

"A player anxious to transpose into his pet opening can often do so
in devious ways. Once when I was scheduled to play White against Weaver
Adams, I wanted to avoid his Albin Counter Gambit. Accordingly I opened
with the non-committal 1.Nf3, to be countered with the mysterious
1...Nc6. I continued with 2.c4, and he answered with 2...e5. English, I
thought. Triumphantly I played 3.d4, reaching, I felt, one of my
favorite lines. Whereupon he replied 3...d5! and I discovered we were
back at the Albin!"

Date: 27 May 2005 08:42:21
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

>"k Houlsby" <[email protected]> wrote in message

>news:[email protected]...

>> >k Houlsby wrote:
>>>>>That's not technically a refutation of the Albin.

>>>> Yes it is, moron, it renders ...e5 unsound.

>>>You can't refute an opening when it it hasn't been played 2...e5 does
> not make the Albin 1. d5 d5 2. c4 e5 does

>> Sure you can. Think about it this way: one day, chess will be solved.

>Chess is an infinite game, and I should like to hear from a mathematician if
infinite games are solvable.

Chess is a closed system. Therefore, it's *finite*. It's very, very
large, but it's finite. It can be solved. There even exists an
algorithm to solve chess.

>Secondly, what does this term 'solve' mean? Does it indicate a forced win
draw or loss for the player making the first move, provided that both
sides
make 'best moves' thereafter?

Yes.

>With human beings, the performance of chess is within a certain time frame -
and as Tal pointed out, many of his more spectacular combinations are
refuted the next day, month or year - but within the timeframe allowed,
was
it solvable? Chess is a performance activity not a theoretical one.

Agreed. It's *still* finite, however.

>Computers are actually a very long way from solving chess - and to test this
>hypothesis turn off the opening book and also the endgame tables [both
>cheating!]. There are no higher level computer ratings which are legal,
>since all GM matches against computers are played with book/tables=on, and
>when the book is on then by definition the program is referring to notes,
>and playing moves it couldn't calculate itself. If it could calculate the
>moves then , ipso facto, why have opening books and endgame tables?

Well, the endgame tables, or tablebases, represent the extent to which
chess *has been solved*. If we had 32-man tablebases, we would have the
*practical* solution, since we would know precisely best play from both
sides. Equally, if one had the "brute force" algorithm playing against
itself on a superfast computer inconceivable by today's standards, then
it, too, would *generate* a solution.

>If this is to 'save-time' then a further cheat is revealed, see Tal
anecdote.

No, it's irrelevant. See above.

>> When chess is solved, it *could* be that the Sicilian Defence (1...c5)
>> refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
>> against the Sicilian with best play. Now, *if* the Sicilian refutes
>> 1.e4, then *by definition* it *also* refutes *all* other white openings
>> beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
>> unplayable, then it's been refuted.

>>>A refutation would have to
>> occur after move 3 since you don't have an Albin until Black has played
>> ...e5.

>> Nope. See above.

>I think Taylor made a good point between a refutation : making move
unplayable

<snip >

I think it's irrelevant.

>>> Enough with insulting Kramnik, already!

>>But everyone knows *real* men play Kd2

>Real Vermont Men often skip this sissy 'flatlander' step and immediately
play 1. Kd2 taking their own pawn!

Uh huh!

>> LOL yeah!

>Mano-a-Mano chess, true Yankee style.

>Phil Innes

Bitter Lemonly
k

Date: 27 May 2005 17:20:34
From: Chris Barnett
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

k Houlsby wrote:
>>"k Houlsby" <[email protected]> wrote in message
>
>
>
>>news:[email protected]...
>
>
>
>>>>k Houlsby wrote:
>>>>
>>>>>>That's not technically a refutation of the Albin.
>
>
>>>>>Yes it is, moron, it renders ...e5 unsound.
>
>
>
>>>>You can't refute an opening when it it hasn't been played 2...e5 does
>>
>>not make the Albin 1. d5 d5 2. c4 e5 does
>
>
>
>>>Sure you can. Think about it this way: one day, chess will be solved.
>
>
>>Chess is an infinite game, and I should like to hear from a mathematician if
>
> infinite games are solvable.

Well, I'm a maths undergraduate, so I will do my best ;-)
It would certainly depend on what order of infinity you are talking
about here. Let us hypothetically, drop the 3 move repitition and 50
move rule, so the game can be infinite. Nevertheless, you can still
calculate optimum moves and the number of *positions* are finite.

I believe it *may* occur in games with infinite boars as well, as long
as there can be a clear endpoint to the game. For instance let us
consider an unbouded chess board on 2 sides, so that a1 is the only
corner. Then place a white king on a1, a black king an a3, a black
queen on b3, and a white bishop on a2; white to move.

White is lost as the (set of moves) - (the set of moves which prevent
checkmate next move) is the empty set. As long as a forced win is
possible, the game is solved.

> Chess is a closed system. Therefore, it's *finite*. It's very, very
> large, but it's finite. It can be solved. There even exists an
> algorithm to solve chess.

Yes, but we may never find it, there may exist more possibilities than
atoms on the universe, and If I remember my science correctly it would
therefore take more time than the universe has to calculate it.

>
>>Secondly, what does this term 'solve' mean? Does it indicate a forced win
>
> draw or loss for the player making the first move, provided that both
> sides
> make 'best moves' thereafter?
>
> Yes.
>
>
>>With human beings, the performance of chess is within a certain time frame -
>
> and as Tal pointed out, many of his more spectacular combinations are
> refuted the next day, month or year - but within the timeframe allowed,
> was
> it solvable? Chess is a performance activity not a theoretical one.
>
> Agreed. It's *still* finite, however.
>
>
>>Computers are actually a very long way from solving chess - and to test this
>>hypothesis turn off the opening book and also the endgame tables [both
>>cheating!]. There are no higher level computer ratings which are legal,
>>since all GM matches against computers are played with book/tables=on, and
>>when the book is on then by definition the program is referring to notes,
>>and playing moves it couldn't calculate itself. If it could calculate the
>>moves then , ipso facto, why have opening books and endgame tables?
>
>
> Well, the endgame tables, or tablebases, represent the extent to which
> chess *has been solved*. If we had 32-man tablebases, we would have the
> *practical* solution, since we would know precisely best play from both
> sides. Equally, if one had the "brute force" algorithm playing against
> itself on a superfast computer inconceivable by today's standards, then
> it, too, would *generate* a solution.
>

See above

>>If this is to 'save-time' then a further cheat is revealed, see Tal
>
> anecdote.
>
> No, it's irrelevant. See above.
>
>
>>>When chess is solved, it *could* be that the Sicilian Defence (1...c5)
>>>refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
>>>against the Sicilian with best play. Now, *if* the Sicilian refutes
>>>1.e4, then *by definition* it *also* refutes *all* other white openings
>>>beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
>>>unplayable, then it's been refuted.

It would hardly be a forced loss though if 2.Nf3 led to a forced loss
and f4 didn;t say, anyway, I agree with Taylor, it's a prevention.
And you can only refute once IMO, you can't refute the open sicillian
because it's just part of the same refutation, your move refutes the
kings pawn opening.

>
>>>>A refutation would have to
>>>
>>>occur after move 3 since you don't have an Albin until Black has played
>>>...e5.
>
>
>
>>>Nope. See above.
>
>>I think Taylor made a good point between a refutation : making move
>
> unplayable
>
> <snip>
>
> I think it's irrelevant.
>
>
>>>>Enough with insulting Kramnik, already!
>
>
>>>But everyone knows *real* men play Kd2
>
>
>
>
>>Real Vermont Men often skip this sissy 'flatlander' step and immediately
>
> play 1. Kd2 taking their own pawn!
>
> Uh huh!
>
>
>>>LOL yeah!
>
>
>
>>Mano-a-Mano chess, true Yankee style.

Ah, thats how I lost my right eye and both my livers! Still won the
game though.

>
>>Phil Innes
>
>
> Bitter Lemonly
> k

Hoping I've made sense,
Chris Barnett.

Date: 30 May 2005 16:29:09
From: David Richerby
Subject: Mathematical game theory (was Re: Sam Sloan's Chess Lesson Video Posted on the Internet)

Chris Barnett <[email protected] > wrote:
> As long as a forced win is possible, the game is solved.

Within the context of game theory, there are several definitions of
`solved' and yours is the weakest. The key notion in discussing this is
the `winning strategy' -- a player is said to have a winning strategy if
they can force the win no matter what the opponent does.

In any finite two-player game of perfect information without draws, one
player has a winning strategy. Chess has draws but the situation here is
only slightly more complicated: we can introduce two new games called
whitechess and blackchess. These have exactly the same rules as ordinary
chess except that games that would be drawn in ordinary chess are declared
as a win for white in whitechess and a win for black in blackchess. If
white has a winning strategy for blackchess, ordinary chess is a win for
white with best play; if black has a winning strategy for whitechess,
ordinary chess is won for black; otherwise, ordinary chess is a draw with
best play.

So, within this framework, chess (like any finite game) is solved in the
sense you give above: either it is a forced win for white or a forced win
for black or a draw. Every finite game has this property but there are
some infinite games that do not. One example is (as far as I recall) the
following: player 1 writes `0.' on a piece of paper and the players then
take it in turns to write either a zero or a one on the end of the number
written so far. So, at any stage of the game, the `position' is a real
number between zero and one, written in binary. Play continues infinitely
(if it were finite, player 1 would win every time so would have a trivial
winning strategy) and player 1 wins if the infinite binary number is
rational (can be expressed as a/b for integers a and b) and player 2 wins
if it is not.

The next level of `solvedness' is games where we know which player has the
winning strategy but do not know what the strategy is. This class of
games is said to be `ultra-weakly solved'. An example of this is the game
of `chomp'. The board is a chocolate bar divided into squares in n rows
and m columns. The two players take it in turns to choose a square and
eat that and every square below and to the left of it. The top-right
square has been laced with deadly poison so whoever eats that loses. It
is known that the first player has a winning strategy but nobody currently
knows what it is. It's easy for nx1 chocolate bars -- the first player
eats everything but the last square. nxn bars aren't much harder -- the
first player eats everything but the top row and right column on his first
move, leaving an L-shape; whenever the second player eats k squares on one
leg, the first player eats k squares on the other leg and, again, the
second player is left with the poison.

In general, though, nobody knows what the winning strategy is but the
first player must win by the following argument. Suppose the second
player has a winning strategy. This means that, whatever player 1 does as
his first move, player 2 can put him in a position where he is forced to
lose. So, suppose player 1 eats just the bottom-left square: player 2
eats square (x,y) and wins. But then player 1 could have won the game by
starting with (x,y) instead of (1,1), contradicting the assumption that
player 2 has a winning strategy. Since the game is finite, one player or
the other must have a winning strategy; since player 2 does not, it must
be player 1.

The next level is knowing what the winning strategy is. This is, of
course, the most interesting case, but also the hardest to demonstrate.
Most of the games in this category are very simple. Awari is a good
example. Strictly, there are two levels here: a game is `weakly solved'
if there is a known algorithm for winning the game from the initial
position and `strongly solved' if there is an algorithm for producing
optimal play from any position (which may, of course, already be lost for
the player who has the winning strategy from the start position).

When people talk about `solving chess', it's unclear whether they mean
`weakly solving' or `strongly solving'.

Dave.

--
David Richerby Homicidal Carnivorous Vomit (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a pile of puke but it's full of
teeth and it wants to kill you!

Date: 27 May 2005 17:23:18
From: Chess One
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

"Chris Barnett" <[email protected] > wrote in message
news:[email protected]...
> k Houlsby wrote:
>>>>Sure you can. Think about it this way: one day, chess will be solved.
>>
>>
>>>Chess is an infinite game, and I should like to hear from a mathematician
>>>if
>>
>> infinite games are solvable.
>
> Well, I'm a maths undergraduate, so I will do my best ;-)
> It would certainly depend on what order of infinity you are talking about
> here. Let us hypothetically, drop the 3 move repitition and 50 move rule,
> so the game can be infinite. Nevertheless, you can still calculate
> optimum moves and the number of *positions* are finite.
>
> I believe it *may* occur in games with infinite boars as well

Ah - we have several of them here!

But seriously, there is an interesting book by James. P. Carse titled Finite
and Infinite Games which is actually a philosophical work, by a professor of
religion. In a brutal precis of his work, he states "The rules of a finite
game may not change; the rules of an infinite game must change." But as I
said, this is a philosophical work which looks at the energy of systems,
rather than the mathematics of counting within systems.

>, as long as there can be a clear endpoint to the game. For instance let
>us consider an unbouded chess board on 2 sides, so that a1 is the only
>corner. Then place a white king on a1, a black king an a3, a black queen
>on b3, and a white bishop on a2; white to move.
>
> White is lost as the (set of moves) - (the set of moves which prevent
> checkmate next move) is the empty set. As long as a forced win is
> possible, the game is solved.

In another piece of speculation - this one provable, some games are either
won or lost by having the first move. It is interesting in chess that this
is less clearly the case. Most people think that having the 'extra tempo' is
better. Yet in chess the threat is often greater than the execution, so
spending that tempo may be disadvantageous :)

Is there a proof that having the first move in chess even provides the first
player with an advantage? Therefore, after white's first move, can it be
"black to move and win?"

>> Chess is a closed system. Therefore, it's *finite*. It's very, very
>> large, but it's finite. It can be solved. There even exists an
>> algorithm to solve chess.
>
> Yes, but we may never find it, there may exist more possibilities than
> atoms on the universe, and If I remember my science correctly it would
> therefore take more time than the universe has to calculate it.

In which case we should not call it finite, nor infinite, but say that we do
not know if it is either, since there is no proof that a solution could be
found given, literally, all the time in the universe.

>>
>>>Secondly, what does this term 'solve' mean? Does it indicate a forced win
>>
>> draw or loss for the player making the first move, provided that both
>> sides
>> make 'best moves' thereafter?
>>
>> Yes.

I notice this is k's point. But do we know which result is a proof? How
do know it?

Let's say hypothetically that there comes into being a proof that chess is
drawn - by following this procedure you could draw against any level of
player, and you or anyone else could repeat the draw by following the
procedure. Does this proof obviate a subsequent proof that chess can be won?

>>>With human beings, the performance of chess is within a certain time
>>>frame -
>>
>> and as Tal pointed out, many of his more spectacular combinations are
>> refuted the next day, month or year - but within the timeframe allowed,
>> was
>> it solvable? Chess is a performance activity not a theoretical one.
>>
>> Agreed. It's *still* finite, however.

Yes - 2 hours on your clock is finite, if you can solve THIS position in the
2 hours without notes.

>>
>>>Computers are actually a very long way from solving chess - and to test
>>>this
>>>hypothesis turn off the opening book and also the endgame tables [both
>>>cheating!]. There are no higher level computer ratings which are legal,
>>>since all GM matches against computers are played with book/tables=on,
>>>and
>>>when the book is on then by definition the program is referring to notes,
>>>and playing moves it couldn't calculate itself. If it could calculate the
>>>moves then , ipso facto, why have opening books and endgame tables?
>>
>>
>> Well, the endgame tables, or tablebases, represent the extent to which
>> chess *has been solved*.

This is a subtle distinction - not solved. The program would not need the
tables if it had an algorithm that solved chess. Right? Previous games can
be known, and procedures copied, but these are just that, copying - they are
not solved in any sense of a rendered algorithm.

>> If we had 32-man tablebases, we would have the
>> *practical* solution, since we would know precisely best play from both
>> sides. Equally, if one had the "brute force" algorithm playing against
>> itself on a superfast computer inconceivable by today's standards, then
>> it, too, would *generate* a solution.

Unfortunately, if the rules of the game is changed it is no longer chess.
Your solution is correct if an algorithm can achieve a result, but otherwise
all that is happening is exactly the same process as any other Turing Engine
achieves.

> See above
>
>
>>>If this is to 'save-time' then a further cheat is revealed, see Tal
>>
>> anecdote.
>>
>> No, it's irrelevant. See above.

If speeding up the processor without improving the evaluation of the
resulting positions does not resolve in automatic mate - then you have more
positions, not a solved position. Looking at either opening books or
table-bases does not solve anything by definition, although it may optimise
play of the chess algorith, which could not evaluate the material looked up.

Phil Innes

Date: 27 May 2005 08:11:34
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

>k Houlsby wrote:
>>>That's not technically a refutation of the Albin.

>> Yes it is, moron, it renders ...e5 unsound.

>You can't refute an opening when it it hasn't been played 2...e5 does
not make the Albin 1. d5 d5 2. c4 e5 does

Sure you can. Think about it this way: one day, chess will be solved.
When chess is solved, it *could* be that the Sicilian Defence (1...c5)
refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
against the Sicilian with best play. Now, *if* the Sicilian refutes
1.e4, then *by definition* it *also* refutes *all* other white openings
beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
unplayable, then it's been refuted.

>>A refutation would have to
> occur after move 3 since you don't have an Albin until Black has played ...e5.

Nope. See above.

>>..and if you don't have an Albin, that demonstrates its being no good
> against this particular move. DUH!

>The question was a refutation to the Albin, not to 2...e5

Yes, it was. See above.

>>>1. Nf3 and 2. Nf3 are part of the girlie-man system of avoiding anything

> interesting after 1. d4.

>> Enough with insulting Kramnik, already!

>But everyone knows *real* men play Kd2

LOL yeah!

Date: 27 May 2005 16:40:48
From: Angelo DePalma
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Further evidence that His Trollness, k-My-Holy Hole, doesn't appreciate
even the most basic meanings of English words. You can only "refute" the
Albin, or any opening or move, after it has been played.

To normal players, 2. Nf3 *prevents* the Albin, it does not refute it. If
Black went ahead and played 2...e5 anyway, which is not the Albin at all (it
is known as the k ("k My Holy Hole") Houlsby Defense) then the
"refutation" is not 3. Nf3, but 3. Nxe5.

Similarly, to normal people 1. e4 e5 2. Nc3 Nf6 *prevents* 3. Qh5. Normal
people wouldn't play Qh5 in this position. However, if it Qh5 is played,
then Nf3 is not the "refutation," 3...Nxh5 is. By the way, this opening is
known as the k ("k my Holy Hole") Houlsby Attack.

It helps to know something about the language in which you're posting.

"k Houlsby" <[email protected] > wrote in message
news:[email protected]...
> >k Houlsby wrote:
>>>>That's not technically a refutation of the Albin.
>
>>> Yes it is, moron, it renders ...e5 unsound.
>
>>You can't refute an opening when it it hasn't been played 2...e5 does
> not make the Albin 1. d5 d5 2. c4 e5 does
>
> Sure you can. Think about it this way: one day, chess will be solved.
> When chess is solved, it *could* be that the Sicilian Defence (1...c5)
> refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
> against the Sicilian with best play. Now, *if* the Sicilian refutes
> 1.e4, then *by definition* it *also* refutes *all* other white openings
> beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
> unplayable, then it's been refuted.
>
>>>A refutation would have to
>> occur after move 3 since you don't have an Albin until Black has played
>> ...e5.
>
> Nope. See above.
>
>
>>>..and if you don't have an Albin, that demonstrates its being no good
>> against this particular move. DUH!
>
>
>
>>The question was a refutation to the Albin, not to 2...e5
>
> Yes, it was. See above.
>
>>>>1. Nf3 and 2. Nf3 are part of the girlie-man system of avoiding anything
>
>> interesting after 1. d4.
>
>
>>> Enough with insulting Kramnik, already!
>
>
>
>>But everyone knows *real* men play Kd2
>
> LOL yeah!
>

Date: 28 May 2005 14:23:28
From: StanB
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

"Angelo DePalma" <[email protected] > wrote in message
news:[email protected]...

> To normal players, 2. Nf3 *prevents* the Albin, it does not refute it.

Nonsense. I've never lost to the Albin after playing 2.Nf3.

Date: 29 May 2005 00:23:18
From: Angelo DePalma
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Are you thick, or what? You've never FACED the goddamned Albin after Nf3
because the opening is impossible after that move. Besides, the refutation,
if there is one, of ...e5 is either de5 or Ne5, not Nf3. Look up the word
"refute" in the dictionary.

"StanB" <[email protected] > wrote in message
news:[email protected]...
>
> "Angelo DePalma" <[email protected]> wrote in message
> news:[email protected]...
>
>> To normal players, 2. Nf3 *prevents* the Albin, it does not refute it.
>
> Nonsense. I've never lost to the Albin after playing 2.Nf3.
>
>

Date: 27 May 2005 15:30:00
From: Chess One
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

"k Houlsby" <[email protected] > wrote in message
news:[email protected]...
> >k Houlsby wrote:
>>>>That's not technically a refutation of the Albin.
>
>>> Yes it is, moron, it renders ...e5 unsound.
>
>>You can't refute an opening when it it hasn't been played 2...e5 does
> not make the Albin 1. d5 d5 2. c4 e5 does
>
> Sure you can. Think about it this way: one day, chess will be solved.

Chess is an infinite game, and I should like to hear from a mathematician if
infinite games are solvable.

Secondly, what does this term 'solve' mean? Does it indicate a forced win
draw or loss for the player making the first move, provided that both sides
make 'best moves' thereafter?

With human beings, the performance of chess is within a certain time frame -
and as Tal pointed out, many of his more spectacular combinations are
refuted the next day, month or year - but within the timeframe allowed, was
it solvable? Chess is a performance activity not a theoretical one.

Computers are actually a very long way from solving chess - and to test this
hypothesis turn off the opening book and also the endgame tables [both
cheating!]. There are no higher level computer ratings which are legal,
since all GM matches against computers are played with book/tables=on, and
when the book is on then by definition the program is referring to notes,
and playing moves it couldn't calculate itself. If it could calculate the
moves then , ipso facto, why have opening books and endgame tables?

If this is to 'save-time' then a further cheat is revealed, see Tal
anecdote.

> When chess is solved, it *could* be that the Sicilian Defence (1...c5)
> refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
> against the Sicilian with best play. Now, *if* the Sicilian refutes
> 1.e4, then *by definition* it *also* refutes *all* other white openings
> beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
> unplayable, then it's been refuted.
>
>>>A refutation would have to
>> occur after move 3 since you don't have an Albin until Black has played
>> ...e5.
>
> Nope. See above.

I think Taylor made a good point between a refutation : making move
unplayable

<snip >

>>> Enough with insulting Kramnik, already!
>
>
>
>>But everyone knows *real* men play Kd2

Real Vermont Men often skip this sissy 'flatlander' step and immediately
play 1. Kd2 taking their own pawn!

> LOL yeah!

Mano-a-Mano chess, true Yankee style.

Phil Innes

Date: 30 May 2005 17:40:21
From: David Richerby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Chess One <[email protected] > wrote:
> Computers are actually a very long way from solving chess - and to test
> this hypothesis turn off the opening book and also the endgame tables
> [both cheating!].

Um, it's also against the rules of chess to use a computer to choose your
moves in competition.

Dave.

--
David Richerby Old-Fashioned Sword (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a razor-sharp blade but it's perfect
for your grandparents!

Date: 30 May 2005 17:36:41
From: Chess One
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

"David Richerby" <[email protected] > wrote in message
news:rK*[email protected]...
> Chess One <[email protected]> wrote:
>> Computers are actually a very long way from solving chess - and to test
>> this hypothesis turn off the opening book and also the endgame tables
>> [both cheating!].
>
> Um, it's also against the rules of chess to use a computer to choose your
> moves in competition.

True, even if you play left handed. Where would we be without all this
mathematical advice?

>
> Dave.
>
> --
> David Richerby Old-Fashioned Sword (TM): it's
> like
> www.chiark.greenend.org.uk/~davidr/ a razor-sharp blade but it's
> perfect
> for your grandparents!

Date: 31 May 2005 20:27:32
From: michael adams
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Chess One wrote:
>
> "David Richerby" <[email protected]> wrote in message
> news:rK*[email protected]...
> > Chess One <[email protected]> wrote:
> >> Computers are actually a very long way from solving chess - and to test
> >> this hypothesis turn off the opening book and also the endgame tables
> >> [both cheating!].
> >
> > Um, it's also against the rules of chess to use a computer to choose your
> > moves in competition.
>
> True, even if you play left handed. Where would we be without all this
> mathematical advice?

A long way from Alice, I'd hazard - even Timbukto too, but y'know, all
this pales somewhat when you consider whether windmills rotate (<-

Date: 31 May 2005 11:10:56
From: Chess One
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

"michael adams" <[email protected] > wrote in message
news:[email protected]...
> Chess One wrote:
>>
>> "David Richerby" <[email protected]> wrote in message
>> news:rK*[email protected]...
>> > Chess One <[email protected]> wrote:
>> >> Computers are actually a very long way from solving chess - and to
>> >> test
>> >> this hypothesis turn off the opening book and also the endgame tables
>> >> [both cheating!].
>> >
>> > Um, it's also against the rules of chess to use a computer to choose
>> > your
>> > moves in competition.
>>
>> True, even if you play left handed. Where would we be without all this
>> mathematical advice?
>
> A long way from Alice, I'd hazard - even Timbukto too, but y'know, all
> this pales somewhat when you consider whether windmills rotate (<-

Date: 31 May 2005 23:35:30
From: michael adams
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Chess One wrote:
>
> "michael adams" <[email protected]> wrote in message
> news:[email protected]...
> > Chess One wrote:
> >>
> >> "David Richerby" <[email protected]> wrote in message
> >> news:rK*[email protected]...
> >> > Chess One <[email protected]> wrote:
> >> >> Computers are actually a very long way from solving chess - and to
> >> >> test
> >> >> this hypothesis turn off the opening book and also the endgame tables
> >> >> [both cheating!].
> >> >
> >> > Um, it's also against the rules of chess to use a computer to choose
> >> > your
> >> > moves in competition.
> >>
> >> True, even if you play left handed. Where would we be without all this
> >> mathematical advice?
> >
> > A long way from Alice, I'd hazard - even Timbukto too, but y'know, all
> > this pales somewhat when you consider whether windmills rotate (<-

Date: 31 May 2005 22:52:55
From: michael adams
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Date: 29 May 2005 18:31:57
From: Patrick Volk
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Fri, 27 May 2005 15:30:00 GMT, "Chess One" <[email protected] >
wrote:

>
>"k Houlsby" <[email protected]> wrote in message
>news:[email protected]...
>> >k Houlsby wrote:
>>>>>That's not technically a refutation of the Albin.
>>
>>>> Yes it is, moron, it renders ...e5 unsound.
>>
>>>You can't refute an opening when it it hasn't been played 2...e5 does
>> not make the Albin 1. d5 d5 2. c4 e5 does
>>
>> Sure you can. Think about it this way: one day, chess will be solved.
>
>Chess is an infinite game, and I should like to hear from a mathematician if
>infinite games are solvable.

Chess has a finite number of board positions. It's extremely large,
but it is finite. It's been a while since school, but this falls into
the NP-complete set of solutions (NP meaning non-polynomial). They are
considered solvable, but not feasible to solve.

There are a few other examples:

Best route through a directed graph (e.g. mapquest). Euler looked at
this about 200 years ago or so. It started to get feasible in the
mid-70's with small sets. Now, places offer it as a free service.

The Towers of Hanoi - The solution has been known for at least 50
years. However, the number of moves required is exponential (2^n moves
for n discs). The recursive solution is very elegant, maybe 15 lines
of code.

Packing problems - Surpisingly tough. You run a distillery, and your
product is in 750-pound casks. You have to store them for anywhere
from 7 to 20 years. Without wasting more than half of my warehouse
space for loading and unloading each individual cask, can I pack them,
and move them efficiently when I have to take them out of storage?
This one was solved probably in the last 7 years. It's also
NP-complete.

Decryption (to decrypt anything is infinite, but given parameters,
it's NP-complete)

An infinite game would indicate an infinite number of positions.

Tic-Tac-Toe (Draughts and Noughts) is a small example of a closed
game. It was solved before the days of computers.

The challenge of chess is that it is a closed game (board and pieces
are finite, finite number of positions), but it is extremely dense in
moves. The number of moves can possibly be infinite, but I would say
that even so, the number of positions is finite.

Brute force is possible, but it would require orders of magnitude
more computing storage and power to not be able to win only by the
computers' opponent dying of old age (and subsequent generations).

So, they employ shortcuts. Opening books. Endgame tablebases.
Alpha-beta pruning (scoring potential positions, and throwing away the
obviously bad ones, and processing more on the obviously better
ones... This is really where human players do damage to computer
opponents, because the computer will stop searching on a given
position, not seeing the danger).

>
>Secondly, what does this term 'solve' mean? Does it indicate a forced win
>draw or loss for the player making the first move, provided that both sides
>make 'best moves' thereafter?

The term 'solve' means the computer has the ability to evaluate all
positions. Reasonably this would mean the worst the computer would do
is draw against a perfect player. If chess was solved, and the
computer played itself, it would always draw (and probably always play
the same game... although if there are two equal lines, it might
change up).

Towers of Hanoi is 2^n. Chess is many orders of magnitude more
complex.

>
>With human beings, the performance of chess is within a certain time frame -
>and as Tal pointed out, many of his more spectacular combinations are
>refuted the next day, month or year - but within the timeframe allowed, was
>it solvable? Chess is a performance activity not a theoretical one.

Computing improves about an order of magnitude every 18 months.
Although this trend from a physics standpoint might not hold out,
other improvements are taking place in computing (parallel processing
for example)

>
>Computers are actually a very long way from solving chess - and to test this
>hypothesis turn off the opening book and also the endgame tables [both
>cheating!]. There are no higher level computer ratings which are legal,
>since all GM matches against computers are played with book/tables=on, and
>when the book is on then by definition the program is referring to notes,
>and playing moves it couldn't calculate itself. If it could calculate the
>moves then , ipso facto, why have opening books and endgame tables?

Opening books make the computer stronger. It doesn't have to
brute-force the calculations. Pretty much the same reason human
players use opening books.

Where the computer has shown its' usefulness to the gameof chess is
endgame studies. The board is more simplified, which allows it to
examing depth of moves faster.
I would venture to say that endgame tables merely give a value. I
doubt they help much on how to play the line. But what they do is
given a position, the computer can know whether to press for the win,
ask for a draw, or resign.

>
>If this is to 'save-time' then a further cheat is revealed, see Tal
>anecdote.

One of my first computer teachers gave me a great analogy. A computer
is like a dopey little brother who can only follow simple
instructions, and you have to write them down on index cards for him.

A computer does simple things fast. The 'simple' brute-force approach
you suggest isn't either feasible or complete at this point. Even if
you reduce the complexity by allowing for transpositions and things
like that, it's still terribly complex.

Cheating? Not really. By taking such shortcuts, the program follows
human play a little better. Opening books are used by human and
computer alike, memorized, to try and put the opponent at a
disadvantage. Closed game or open? Center or flanks?

I think that is a bad analogy, that opening books encoded are cheating
for a computer. It 'memorizes' better than a human, but both do
memorize them.
>
>> When chess is solved, it *could* be that the Sicilian Defence (1...c5)
>> refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
>> against the Sicilian with best play. Now, *if* the Sicilian refutes
>> 1.e4, then *by definition* it *also* refutes *all* other white openings
>> beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
>> unplayable, then it's been refuted.
>>
>>>>A refutation would have to
>>> occur after move 3 since you don't have an Albin until Black has played
>>> ...e5.
>>
>> Nope. See above.
>
>I think Taylor made a good point between a refutation : making move
>unplayable
>
><snip>
>
>>>> Enough with insulting Kramnik, already!
>>
>>
>>
>>>But everyone knows *real* men play Kd2
>
>Real Vermont Men often skip this sissy 'flatlander' step and immediately
>play 1. Kd2 taking their own pawn!
>
>> LOL yeah!
>
>Mano-a-Mano chess, true Yankee style.
>
>Phil Innes
>

Date: 30 May 2005 17:19:02
From: David Richerby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Patrick Volk <[email protected] > wrote:
> Chess has a finite number of board positions. It's extremely large,
> but it is finite. It's been a while since school, but this falls into
> the NP-complete set of solutions (NP meaning non-polynomial).

Chess is almost certainly not NP-complete and NP doesn't mean `non-poly-
nomial' but `nondeterministic polynomial time', i.e., things that could be
solved with a polynomial number of steps on a nondeterministic machine.
(One that has more than one possible next state from a given current
state, unlike real computers.)

> They are considered solvable, but not feasible to solve.

This is true. Well, not feasible to solve on large instances. For
example, the following problem of determining whether lessons at a school
can be timetabled is NP-complete but every school on the planet manages to
solve it without massive computer systems.

> There are a few other examples:
>
> Best route through a directed graph (e.g. mapquest). Euler looked at
> this about 200 years ago or so. It started to get feasible in the
> mid-70's with small sets. Now, places offer it as a free service.

Nope. Finding the best route from A to B in a weighted graph (e.g., a
road map) is polynomial-time solvable. The algorithm in fact computes the
shortest route between every possible pair of towns in n^3 steps, where n
is the number of towns. Look up `Floyd-Warshall algorithm' or `all-pairs
shortest path' in any good book on algorithms.

You're probably thinking of the travelling salesman problem, which is
NP-complete: given a set of towns and the distances between them, what is
the shortest path that visits each town exactly once? (Strictly, the
problem as I've phrased it, isn't NP-complete because NP is a class of
decision problems, i.e., problems with yes/no answers. The decision
version of this is `is there a path of length at most d' and that is
NP-complete.)

> The Towers of Hanoi - The solution has been known for at least 50
> years. However, the number of moves required is exponential (2^n moves
> for n discs). The recursive solution is very elegant, maybe 15 lines
> of code.

That's a completely different kettle of fish. The standard algorithm
gives a list of moves that will transfer the discs from one peg to another
in 2^n moves. What it won't do is compute the shortest possible sequence
of moves from any given position.

> Packing problems - Surpisingly tough. You run a distillery, and your
> product is in 750-pound casks. You have to store them for anywhere
> from 7 to 20 years. Without wasting more than half of my warehouse
> space for loading and unloading each individual cask, can I pack them,
> and move them efficiently when I have to take them out of storage?
> This one was solved probably in the last 7 years. It's also
> NP-complete.

That's not the usual packing problem. Indeed, it's so vaguely stated that
I can't work out what exactly the problem is. When stating these things,
it's a good idea to give an explicit list of the inputs to the problem and
its output. For example, generalized chess has as its inputs a number n
and a list of types and positions of chess pieces (i.e., a position on an
nxn board) and as its output is one of the words `white', `black' and
`draw', indicating who wins from the given position, assuming best play.

> Decryption (to decrypt anything is infinite, but given parameters,
> it's NP-complete)

Depends on the encryption algorithm. Without the keystream, it's
trivially impossible to decrypt something encrypted with a one-time pad.
With the keystream, it takes a number of steps linear in the length of the
message -- much easier than NP-complete.

> An infinite game would indicate an infinite number of positions.

Not necessarily. There are plenty of infinite games played over a finite
number of positions. Chess is, in effect, one of these, but the infinite
plays are defined to be draws and the 50-move and repetition rules
truncate the infinite plays to finite ones.

> The challenge of chess is that it is a closed game (board and pieces
> are finite, finite number of positions), but it is extremely dense in
> moves. The number of moves can possibly be infinite, but I would say
> that even so, the number of positions is finite.

As I've said before, the number of moves is finite.

> The term 'solve' means the computer has the ability to evaluate all
> positions. Reasonably this would mean the worst the computer would do
> is draw against a perfect player. If chess was solved, and the
> computer played itself, it would always draw (and probably always play
> the same game... although if there are two equal lines, it might
> change up).

No, no, no. If we have a computer that is correctly able to evaluate all
positions, it can play perfect chess by always chosing the move leading to
the position with the best evaluation. However, if chess is a forced win
for white, the computer would lose as black against a perfect player. In
those circumstances, the computer would beat itself.

> Towers of Hanoi is 2^n. Chess is many orders of magnitude more complex.

This statement doesn't make sense. In ToH, `n' refers to the number of
discs on the initial peg and to say that the time complexity is 2^n means
that it takes 2^n steps to move n pegs. There is no `n' in chess -- it's
a fixed game played on an 8x8 board with 32 pieces of specified type. So
the time complexity of chess is `c'. Because it's a problem of fixed
finite size, there is a constant c such that, in c computation steps, the
computer can work out what the best move is from any position. The
problem is that c is so large that we can't run the computation.

> Computing improves about an order of magnitude every 18 months.
> Although this trend from a physics standpoint might not hold out,
> other improvements are taking place in computing (parallel processing
> for example)

As an aside, chess doesn't seem to be very amenable to parallelization.
Even if it were, having n processors lets you go at most n times as fast
as one processor, which doesn't help much in this case.

Dave.

--
David Richerby Swiss Metal Widget (TM): it's like a
www.chiark.greenend.org.uk/~davidr/ thingy that's made of steel but it's
made in Switzerland!

Date: 30 May 2005 22:12:55
From: Patrick Volk
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On 30 May 2005 17:19:02 +0100 (BST), David Richerby
<[email protected] > wrote:

>Patrick Volk <[email protected]> wrote:
>> Chess has a finite number of board positions. It's extremely large,
>> but it is finite. It's been a while since school, but this falls into
>> the NP-complete set of solutions (NP meaning non-polynomial).
>
>Chess is almost certainly not NP-complete and NP doesn't mean `non-poly-
>nomial' but `nondeterministic polynomial time', i.e., things that could be
>solved with a polynomial number of steps on a nondeterministic machine.
>(One that has more than one possible next state from a given current
>state, unlike real computers.)
>
>> They are considered solvable, but not feasible to solve.
>
>This is true. Well, not feasible to solve on large instances. For
>example, the following problem of determining whether lessons at a school
>can be timetabled is NP-complete but every school on the planet manages to
>solve it without massive computer systems.
>
>
>> There are a few other examples:
>>
>> Best route through a directed graph (e.g. mapquest). Euler looked at
>> this about 200 years ago or so. It started to get feasible in the
>> mid-70's with small sets. Now, places offer it as a free service.
>
>Nope. Finding the best route from A to B in a weighted graph (e.g., a
>road map) is polynomial-time solvable. The algorithm in fact computes the
>shortest route between every possible pair of towns in n^3 steps, where n
>is the number of towns. Look up `Floyd-Warshall algorithm' or `all-pairs
>shortest path' in any good book on algorithms.
>
>You're probably thinking of the travelling salesman problem, which is
>NP-complete: given a set of towns and the distances between them, what is
>the shortest path that visits each town exactly once? (Strictly, the
>problem as I've phrased it, isn't NP-complete because NP is a class of
>decision problems, i.e., problems with yes/no answers. The decision
>version of this is `is there a path of length at most d' and that is
>NP-complete.)

You are right there.

>
>
>> The Towers of Hanoi - The solution has been known for at least 50
>> years. However, the number of moves required is exponential (2^n moves
>> for n discs). The recursive solution is very elegant, maybe 15 lines
>> of code.
>
>That's a completely different kettle of fish. The standard algorithm
>gives a list of moves that will transfer the discs from one peg to another
>in 2^n moves. What it won't do is compute the shortest possible sequence
>of moves from any given position.

Is it? There are only 3 possible moves for any disc.

>
>
>> Packing problems - Surpisingly tough. You run a distillery, and your
>> product is in 750-pound casks. You have to store them for anywhere
>> from 7 to 20 years. Without wasting more than half of my warehouse
>> space for loading and unloading each individual cask, can I pack them,
>> and move them efficiently when I have to take them out of storage?
>> This one was solved probably in the last 7 years. It's also
>> NP-complete.
>
>That's not the usual packing problem. Indeed, it's so vaguely stated that
>I can't work out what exactly the problem is. When stating these things,
>it's a good idea to give an explicit list of the inputs to the problem and
>its output. For example, generalized chess has as its inputs a number n
>and a list of types and positions of chess pieces (i.e., a position on an
>nxn board) and as its output is one of the words `white', `black' and
>`draw', indicating who wins from the given position, assuming best play.

It was in Discover probably in the last year. What I remember is it
was trying to calculate the fewest number of moves to get the casks
out. Each cask has a date, and can be moved a number of directions,
possibly involving moving other casks in the process. Those would be
the inputs.

>
>
>> Decryption (to decrypt anything is infinite, but given parameters,
>> it's NP-complete)
>
>Depends on the encryption algorithm. Without the keystream, it's
>trivially impossible to decrypt something encrypted with a one-time pad.
>With the keystream, it takes a number of steps linear in the length of the
>message -- much easier than NP-complete.

I meant without the keystream.

>
>
>
>> An infinite game would indicate an infinite number of positions.
>
>Not necessarily. There are plenty of infinite games played over a finite
>number of positions. Chess is, in effect, one of these, but the infinite
>plays are defined to be draws and the 50-move and repetition rules
>truncate the infinite plays to finite ones.
>
>
>> The challenge of chess is that it is a closed game (board and pieces
>> are finite, finite number of positions), but it is extremely dense in
>> moves. The number of moves can possibly be infinite, but I would say
>> that even so, the number of positions is finite.
>
>As I've said before, the number of moves is finite.
>
>
>> The term 'solve' means the computer has the ability to evaluate all
>> positions. Reasonably this would mean the worst the computer would do
>> is draw against a perfect player. If chess was solved, and the
>> computer played itself, it would always draw (and probably always play
>> the same game... although if there are two equal lines, it might
>> change up).
>
>No, no, no. If we have a computer that is correctly able to evaluate all
>positions, it can play perfect chess by always chosing the move leading to
>the position with the best evaluation. However, if chess is a forced win
>for white, the computer would lose as black against a perfect player. In
>those circumstances, the computer would beat itself.

I corrected myself on that.

>
>
>> Towers of Hanoi is 2^n. Chess is many orders of magnitude more complex.
>
>This statement doesn't make sense. In ToH, `n' refers to the number of
>discs on the initial peg and to say that the time complexity is 2^n means
>that it takes 2^n steps to move n pegs. There is no `n' in chess -- it's
>a fixed game played on an 8x8 board with 32 pieces of specified type. So
>the time complexity of chess is `c'. Because it's a problem of fixed
>finite size, there is a constant c such that, in c computation steps, the
>computer can work out what the best move is from any position. The
>problem is that c is so large that we can't run the computation.

Right. Every move leads to many other moves, so as you go on, you have
an inordinate number of moves. Time to evaluate a position is
constant, right? The number of positions is the problem.

>
>
>> Computing improves about an order of magnitude every 18 months.
>> Although this trend from a physics standpoint might not hold out,
>> other improvements are taking place in computing (parallel processing
>> for example)
>
>As an aside, chess doesn't seem to be very amenable to parallelization.
>Even if it were, having n processors lets you go at most n times as fast
>as one processor, which doesn't help much in this case.

Not necessarily - a single processor incurs penalties for
multitasking.

My expectation would be a massively parallel computer gets to chug
away at it. A good parallel operation would be to go through a
movebase, and look for optimizations (handling things like equivilant
positions on different parts of the board and stuff like that). With
sufficient data, I would expect some optimizations, and possibly
improved algorithms which would make the system more efficient.

More is better... just ask the NSA.

>
>
>Dave.

Date: 31 May 2005 10:56:31
From: David Richerby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Patrick Volk <[email protected] > wrote:
> David Richerby <[email protected]> wrote:
>> Patrick Volk <[email protected]> wrote:
>>> The Towers of Hanoi - The solution has been known for at least 50
>>> years. However, the number of moves required is exponential (2^n moves
>>> for n discs). The recursive solution is very elegant, maybe 15 lines
>>> of code.
>>
>> That's a completely different kettle of fish. The standard algorithm
>> gives a list of moves that will transfer the discs from one peg to
>> another in 2^n moves. What it won't do is compute the shortest
>> possible sequence of moves from any given position.
>
> Is it? There are only 3 possible moves for any disc.

There are only two possible moves for any disc so, yes, it is possible to
compute the best move from any position in time at most 2^n, now I think
about it. But you say that ``the number of moves required is
exponential'' and referred to an elegant recursive solution, from which I
inferred that you were talking about the number of steps required to move
the tower from peg A to peg B, not the complexity of working out the best
move from a given position, which would be done by ugly brute-force
search.

>>> Decryption (to decrypt anything is infinite, but given parameters,
>>> it's NP-complete)
>>
>> Depends on the encryption algorithm. Without the keystream, it's
>> trivially impossible to decrypt something encrypted with a one-time
>> pad. With the keystream, it takes a number of steps linear in the
>> length of the message -- much easier than NP-complete.
>
> I meant without the keystream.

OK. Well, without the keystream, one-time pads are impossible to decrypt,
not NP-complete. There may be some algorithms for which the process is
NP-complete.

>>> Towers of Hanoi is 2^n. Chess is many orders of magnitude more complex.
>>
>> This statement doesn't make sense. In ToH, `n' refers to the number of
>> discs on the initial peg and to say that the time complexity is 2^n
>> means that it takes 2^n steps to move n pegs. There is no `n' in chess
>> -- it's a fixed game played on an 8x8 board with 32 pieces of specified
>> type.
>
> Right. Every move leads to many other moves, so as you go on, you have
> an inordinate number of moves. Time to evaluate a position is
> constant, right? The number of positions is the problem.

OK, so your `n' in chess is the depth to which the move is evaluated? As
has been demonstrated in a thread over on rec.games.chess.computer, any
game of chess can be claimed drawn after 5500 moves or so, which means
that there's still a constant upper bound on the amount of work to be
done. However big that is, there's are infinitely Towers of Hanoi
instances that are harder to solve. In complexity-theoretic terms, chess
is trivial because it's a single instance of a finite problem. Practic-
ally speaking, chess is harder than TOH but, by introducing NP-complete-
ness, the discussion immediately became much more theoretical.

>>> Computing improves about an order of magnitude every 18 months.
>>> Although this trend from a physics standpoint might not hold out,
>>> other improvements are taking place in computing (parallel processing
>>> for example)
>>
>> As an aside, chess doesn't seem to be very amenable to parallelization.
>> Even if it were, having n processors lets you go at most n times as
>> fast as one processor, which doesn't help much in this case.
>
> Not necessarily - a single processor incurs penalties for
> multitasking.

Just the same as a pile of processors incur penalties because they still
have to talk to each other.

> My expectation would be a massively parallel computer gets to chug
> away at it. A good parallel operation would be to go through a
> movebase, and look for optimizations (handling things like equivilant
> positions on different parts of the board and stuff like that). With
> sufficient data, I would expect some optimizations, and possibly
> improved algorithms which would make the system more efficient.

The problem here is all the locking and communication overhead that would
be required to have some processors producing a game tree and other
processors changing that game tree as it's being searched, to say ``ignore
this bit -- it's been pruned / it's a transposition / whatever''. I'm not
saying it can't be made to work because there are a lot of very st
people out there who really want it to work but it's a *very* difficult
problem. Parallelization isn't a magic bullet -- it makes it *harder* to
write good algorithms.

> More is better... just ask the NSA.

More is better in the sense that a thousand processors, working on a very
parallelizable task will be a few hundred times faster than the best
single processor available. But the return curve is not usually linear.

Dave.

--
David Richerby Indelible Lead Wine (TM): it's like a
www.chiark.greenend.org.uk/~davidr/ vintage Beaujolais that weighs a ton
but it can't be erased!

Date: 01 Jun 2005 00:22:46
From: Patrick Volk
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On 31 May 2005 10:56:31 +0100 (BST), David Richerby
<[email protected] > wrote:

>Patrick Volk <[email protected]> wrote:
>> David Richerby <[email protected]> wrote:
>>> Patrick Volk <[email protected]> wrote:
>>>> The Towers of Hanoi - The solution has been known for at least 50
>>>> years. However, the number of moves required is exponential (2^n moves
>>>> for n discs). The recursive solution is very elegant, maybe 15 lines
>>>> of code.
>>>
>>> That's a completely different kettle of fish. The standard algorithm
>>> gives a list of moves that will transfer the discs from one peg to
>>> another in 2^n moves. What it won't do is compute the shortest
>>> possible sequence of moves from any given position.
>>
>> Is it? There are only 3 possible moves for any disc.
>
>There are only two possible moves for any disc so, yes, it is possible to
>compute the best move from any position in time at most 2^n, now I think
>about it. But you say that ``the number of moves required is
>exponential'' and referred to an elegant recursive solution, from which I
>inferred that you were talking about the number of steps required to move
>the tower from peg A to peg B, not the complexity of working out the best
>move from a given position, which would be done by ugly brute-force
>search.

The solution is not complex. Adding a disc to the tower doubles the
number of moves required to move the whole tower.

>
>
>>>> Decryption (to decrypt anything is infinite, but given parameters,
>>>> it's NP-complete)
>>>
>>> Depends on the encryption algorithm. Without the keystream, it's
>>> trivially impossible to decrypt something encrypted with a one-time
>>> pad. With the keystream, it takes a number of steps linear in the
>>> length of the message -- much easier than NP-complete.
>>
>> I meant without the keystream.
>
>OK. Well, without the keystream, one-time pads are impossible to decrypt,
>not NP-complete. There may be some algorithms for which the process is
>NP-complete.

>
>
>>>> Towers of Hanoi is 2^n. Chess is many orders of magnitude more complex.
>>>
>>> This statement doesn't make sense. In ToH, `n' refers to the number of
>>> discs on the initial peg and to say that the time complexity is 2^n
>>> means that it takes 2^n steps to move n pegs. There is no `n' in chess
>>> -- it's a fixed game played on an 8x8 board with 32 pieces of specified
>>> type.
>>
>> Right. Every move leads to many other moves, so as you go on, you have
>> an inordinate number of moves. Time to evaluate a position is
>> constant, right? The number of positions is the problem.
>
>OK, so your `n' in chess is the depth to which the move is evaluated? As
>has been demonstrated in a thread over on rec.games.chess.computer, any
>game of chess can be claimed drawn after 5500 moves or so, which means
>that there's still a constant upper bound on the amount of work to be
>done. However big that is, there's are infinitely Towers of Hanoi
>instances that are harder to solve. In complexity-theoretic terms, chess
>is trivial because it's a single instance of a finite problem. Practic-
>ally speaking, chess is harder than TOH but, by introducing NP-complete-
>ness, the discussion immediately became much more theoretical.
>
>
>>>> Computing improves about an order of magnitude every 18 months.
>>>> Although this trend from a physics standpoint might not hold out,
>>>> other improvements are taking place in computing (parallel processing
>>>> for example)
>>>
>>> As an aside, chess doesn't seem to be very amenable to parallelization.
>>> Even if it were, having n processors lets you go at most n times as
>>> fast as one processor, which doesn't help much in this case.
>>
>> Not necessarily - a single processor incurs penalties for
>> multitasking.
>
>Just the same as a pile of processors incur penalties because they still
>have to talk to each other.

Granted it's a variable, dependant on how much interaction. A good
parallel algorithm will do better than n * single processor speed.

>
>
>> My expectation would be a massively parallel computer gets to chug
>> away at it. A good parallel operation would be to go through a
>> movebase, and look for optimizations (handling things like equivilant
>> positions on different parts of the board and stuff like that). With
>> sufficient data, I would expect some optimizations, and possibly
>> improved algorithms which would make the system more efficient.
>
>The problem here is all the locking and communication overhead that would
>be required to have some processors producing a game tree and other
>processors changing that game tree as it's being searched, to say ``ignore
>this bit -- it's been pruned / it's a transposition / whatever''. I'm not
>saying it can't be made to work because there are a lot of very st
>people out there who really want it to work but it's a *very* difficult
>problem. Parallelization isn't a magic bullet -- it makes it *harder* to
>write good algorithms.

I don't deny it's difficult, and don't believe it will be solved in
my lifetime. Parallelization works best when there are concurrent
steps with heavy, independent calculation. It doesn't work best when
there is a single task to do.
My basis for the belief really is thinking how Enigma was cracked.
It started out with I think quadrillions of possibilities (26^3 * 120
for the wheels, and around 26^10 for the steckers). Analysis revealed
certain deficiencies (reflexivity of the alphabet, being able to
determine when a 'flop' took place, bombes looking for cribs and
stecker maps). Which got it from impossible to weeks, to finally
hours. Not only that, they managed to decipher the teletraffic.
I expect chess will be somewhat the same way. Heuristically, there
are some readily apparent ways to prune things. The larger the data
set, the more apparent.
Also, brute force is only necessary really once. If I do a deep
analysis, and I generate a map of positions. It no longer becomes a
computational problem, but a database one.
A petabyte (reasonable storage size) is 10^15, the number of valid
positions is 10^53 or so. If optimizations (not recording those
positions which lead to a losing position, or not recording moves
which would be illegal based on white's move, or based on same
situation on different parts of the board) can get that down to 10^30,
and databases scale up, it might be possible.
I realize I'm starting to sound like the underpants gnomes here
(step 1: collect underwear... step 2: ????... step 3: profit!), but I
know the bear is in the computation.
Probably the best way to go would be to start at both ends, until
they meet at common positions. When (if is more like it) all the
winning positons are run through, then start looking at the drawn
positions. I wonder how many different checkmate positions there are.
Then you'd probably have to do the same things for black.

>
>
>> More is better... just ask the NSA.
>
>More is better in the sense that a thousand processors, working on a very
>parallelizable task will be a few hundred times faster than the best
>single processor available. But the return curve is not usually linear.

Solving chess I imagine is a very parallelizable task. Would you
agree?

>
>
>Dave.

Pat

Date: 30 May 2005 01:14:44
From: Chess One
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

"Patrick Volk" <[email protected] > wrote in message
news:[email protected]...
> On Fri, 27 May 2005 15:30:00 GMT, "Chess One" <[email protected]>
> wrote:
>
>>
>>"k Houlsby" <[email protected]> wrote in message
>>news:[email protected]...
>>> >k Houlsby wrote:
>>>>>>That's not technically a refutation of the Albin.
>>>
>>>>> Yes it is, moron, it renders ...e5 unsound.
>>>
>>>>You can't refute an opening when it it hasn't been played 2...e5 does
>>> not make the Albin 1. d5 d5 2. c4 e5 does
>>>
>>> Sure you can. Think about it this way: one day, chess will be solved.
>>
>>Chess is an infinite game, and I should like to hear from a mathematician
>>if
>>infinite games are solvable.
>
> Chess has a finite number of board positions. It's extremely large,
> but it is finite. It's been a while since school, but this falls into
> the NP-complete set of solutions (NP meaning non-polynomial). They are
> considered solvable, but not feasible to solve.

I am glad that a PhD in mathematics is attending this thread - Andy Walker -
and to consider his messages. We seem to have an immediate clash here over
'positions' versus 'moves', and I only not that chess is played move by
move.

> There are a few other examples:
>
> Best route through a directed graph (e.g. mapquest). Euler looked at
> this about 200 years ago or so. It started to get feasible in the
> mid-70's with small sets. Now, places offer it as a free service.

I don't understand anything from that expression.

> The Towers of Hanoi - The solution has been known for at least 50
> years. However, the number of moves required is exponential (2^n moves
> for n discs). The recursive solution is very elegant, maybe 15 lines
> of code.

Mathematical abstraction as analogy?

> Packing problems - Surpisingly tough. You run a distillery, and your
> product is in 750-pound casks. You have to store them for anywhere
> from 7 to 20 years. Without wasting more than half of my warehouse
> space for loading and unloading each individual cask, can I pack them,
> and move them efficiently when I have to take them out of storage?
> This one was solved probably in the last 7 years. It's also
> NP-complete.

Metaphor.

> Decryption (to decrypt anything is infinite, but given parameters,
> it's NP-complete)
>
> An infinite game would indicate an infinite number of positions.

More than that. See original post. It would indicate a valency of rules.

To return to the subject, what seems presented here is a variety of
quantitative analysis, but what is not explained is any qualitative
improvement in that analysis - to wit; how does deeper search depth resolve
in better positional understanding?

> Tic-Tac-Toe (Draughts and Noughts) is a small example of a closed
> game. It was solved before the days of computers.
>
> The challenge of chess is that it is a closed game (board and pieces
> are finite, finite number of positions), but it is extremely dense in
> moves. The number of moves can possibly be infinite, but I would say
> that even so, the number of positions is finite.

And what significance do is there of poistions verus moves?

> Brute force is possible, but it would require orders of magnitude
> more computing storage and power to not be able to win only by the
> computers' opponent dying of old age (and subsequent generations).

A more-ism argument. But if the evaluation algorithm is stupid, then the
result is more stupidity, no?

> So, they employ shortcuts.

Your argument then is that computers cannot adequately solve chess at the
moment, so they use 'look-ups'. But this is against the rules of chess, and
is no prescription for a future chess engine. It is simply an expediency to
gain time, which is unfortunately illegal.

> Opening books. Endgame tablebases.
> Alpha-beta pruning (scoring potential positions, and throwing away the
> obviously bad ones, and processing more on the obviously better
> ones... This is really where human players do damage to computer
> opponents, because the computer will stop searching on a given
> position, not seeing the danger).
>
>
>>
>>Secondly, what does this term 'solve' mean? Does it indicate a forced win
>>draw or loss for the player making the first move, provided that both
>>sides
>>make 'best moves' thereafter?
>
> The term 'solve' means the computer has the ability to evaluate all
> positions. Reasonably this would mean the worst the computer would do
> is draw against a perfect player.

A nonsense logically. You assume the evaluation is okay? Is it?

>If chess was solved, and the
> computer played itself, it would always draw (and probably always play
> the same game... although if there are two equal lines, it might
> change up).

Another assumption without a predicate. Why is the result always a draw?
Should not White or Black win?

> Towers of Hanoi is 2^n. Chess is many orders of magnitude more
> complex.
>
>>
>>With human beings, the performance of chess is within a certain time
>>frame -
>>and as Tal pointed out, many of his more spectacular combinations are
>>refuted the next day, month or year - but within the timeframe allowed,
>>was
>>it solvable? Chess is a performance activity not a theoretical one.
>
> Computing improves about an order of magnitude every 18 months.
> Although this trend from a physics standpoint might not hold out,
> other improvements are taking place in computing (parallel processing
> for example)

The point addresses the time allowed, not theoretical space.

>>Computers are actually a very long way from solving chess - and to test
>>this
>>hypothesis turn off the opening book and also the endgame tables [both
>>cheating!]. There are no higher level computer ratings which are legal,
>>since all GM matches against computers are played with book/tables=on, and
>>when the book is on then by definition the program is referring to notes,
>>and playing moves it couldn't calculate itself. If it could calculate the
>>moves then , ipso facto, why have opening books and endgame tables?
>
> Opening books make the computer stronger.

No. they make the chess engine weaker. Opening books avoid the program
actually playing chess - it simply trots out moves by GMs, which, if it were
that good itself, it would play - so it wouldn't need opening books.

This is a logical truism. You are admitting that computers are weak in the
opening by your statement that they can't play them!

> It doesn't have to
> brute-force the calculations. Pretty much the same reason human
> players use opening books.
>
> Where the computer has shown its' usefulness to the gameof chess is
> endgame studies. The board is more simplified, which allows it to
> examing depth of moves faster.

Faster than... ?

So it takes advantage of the time it can calculate by look-ups, which is
denied the human player, and whcih is incidentally not chess - it is illegal
to look up notes. So the computer cheats with time.

> I would venture to say that endgame tables merely give a value. I
> doubt they help much on how to play the line. But what they do is
> given a position, the computer can know whether to press for the win,
> ask for a draw, or resign.
>
>>
>>If this is to 'save-time' then a further cheat is revealed, see Tal
>>anecdote.
>
> One of my first computer teachers gave me a great analogy. A computer
> is like a dopey little brother who can only follow simple
> instructions, and you have to write them down on index cards for him.
>
> A computer does simple things fast. The 'simple' brute-force approach
> you suggest isn't either feasible or complete at this point. Even if
> you reduce the complexity by allowing for transpositions and things
> like that, it's still terribly complex.

I did not advocate brute force. I did state the conditions of a Turing
Engine.

> Cheating? Not really. By taking such shortcuts, the program follows
> human play a little better. Opening books are used by human and
> computer alike, memorized, to try and put the opponent at a
> disadvantage. Closed game or open? Center or flanks?

Wrong! During the game it is illegal to look up opening books. Period.

> I think that is a bad analogy, that opening books encoded are cheating
> for a computer. It 'memorizes' better than a human, but both do
> memorize them.

If a computer can memorise better than a human it should not need to look up
stuff during the game. Ipso facto, this is not memorisation!

ROFL. Its cheating!

Phil Innes

>>> When chess is solved, it *could* be that the Sicilian Defence (1...c5)
>>> refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
>>> against the Sicilian with best play. Now, *if* the Sicilian refutes
>>> 1.e4, then *by definition* it *also* refutes *all* other white openings
>>> beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
>>> unplayable, then it's been refuted.
>>>
>>>>>A refutation would have to
>>>> occur after move 3 since you don't have an Albin until Black has played
>>>> ...e5.
>>>
>>> Nope. See above.
>>
>>I think Taylor made a good point between a refutation : making move
>>unplayable
>>
>><snip>
>>
>>>>> Enough with insulting Kramnik, already!
>>>
>>>
>>>
>>>>But everyone knows *real* men play Kd2
>>
>>Real Vermont Men often skip this sissy 'flatlander' step and immediately
>>play 1. Kd2 taking their own pawn!
>>
>>> LOL yeah!
>>
>>Mano-a-Mano chess, true Yankee style.
>>
>>Phil Innes
>>
>

Date: 30 May 2005 01:12:09
From: Patrick Volk
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Mon, 30 May 2005 01:14:44 GMT, "Chess One" <[email protected] >
wrote:

>
>"Patrick Volk" <[email protected]> wrote in message
>news:[email protected]...
>> On Fri, 27 May 2005 15:30:00 GMT, "Chess One" <[email protected]>
>> wrote:
>>
>>>
>>>"k Houlsby" <[email protected]> wrote in message
>>>news:[email protected]...
>>>> >k Houlsby wrote:
>>>>>>>That's not technically a refutation of the Albin.
>>>>
>>>>>> Yes it is, moron, it renders ...e5 unsound.
>>>>
>>>>>You can't refute an opening when it it hasn't been played 2...e5 does
>>>> not make the Albin 1. d5 d5 2. c4 e5 does
>>>>
>>>> Sure you can. Think about it this way: one day, chess will be solved.
>>>
>>>Chess is an infinite game, and I should like to hear from a mathematician
>>>if
>>>infinite games are solvable.
>>
>> Chess has a finite number of board positions. It's extremely large,
>> but it is finite. It's been a while since school, but this falls into
>> the NP-complete set of solutions (NP meaning non-polynomial). They are
>> considered solvable, but not feasible to solve.
>
>I am glad that a PhD in mathematics is attending this thread - Andy Walker -
>and to consider his messages. We seem to have an immediate clash here over
>'positions' versus 'moves', and I only not that chess is played move by
>move.

I just popped over and read his stuff. Interesting stuff.

>
>> There are a few other examples:
>>
>> Best route through a directed graph (e.g. mapquest). Euler looked at
>> this about 200 years ago or so. It started to get feasible in the
>> mid-70's with small sets. Now, places offer it as a free service.
>
>I don't understand anything from that expression.

I was trying to give an example where a solution was known, but it
wasn't feasible on computers until fairly recently.

>
>> The Towers of Hanoi - The solution has been known for at least 50
>> years. However, the number of moves required is exponential (2^n moves
>> for n discs). The recursive solution is very elegant, maybe 15 lines
>> of code.
>
>Mathematical abstraction as analogy?

A case of a problem with a very simple solution, but not feasible to
solve on computers because of the number of moves.

>
>> Packing problems - Surpisingly tough. You run a distillery, and your
>> product is in 750-pound casks. You have to store them for anywhere
>> from 7 to 20 years. Without wasting more than half of my warehouse
>> space for loading and unloading each individual cask, can I pack them,
>> and move them efficiently when I have to take them out of storage?
>> This one was solved probably in the last 7 years. It's also
>> NP-complete.
>
>Metaphor.

I don't know the algorithm for that, but I know it exists. Call it
what you will.

The point I was trying to make was that 2 of 3 of these are examples
that were too complex to solve in the near past. The Towers of Hanoi
is one where the answer is known, and is easy, but not feasible.

>
>> Decryption (to decrypt anything is infinite, but given parameters,
>> it's NP-complete)
>>
>> An infinite game would indicate an infinite number of positions.
>
>More than that. See original post. It would indicate a valency of rules.

Chess is not an infinite game. I chose to describe it by the number of
positions. Since the premise of your original post is brute-force, I
chose the brute-force method of evaluating the game. By the position.
Turn does not matter.
If you wish to proclaim time and move rules, fine, but irrelevant
when it comes to considering how to do a brute-force solve.

Chess is *not* an infinite game. It has one axis which is infinite,
but a finite alphabet.

Positions are nodes on a map. The moves are the steps you take along
that map. To wit: A winning move can be the same as a losing move, the
important thing is the position. I don't care whether you're a
computer or a human, you may read the moves, but what you really
evaluate is the position.

>
>To return to the subject, what seems presented here is a variety of
>quantitative analysis, but what is not explained is any qualitative
>improvement in that analysis - to wit; how does deeper search depth resolve
>in better positional understanding?

If you think for any given position, associate a score with it. This
currently is calculated based on the pieces, and expressed in pawns. A
deeper search allows the computer to find positions with higher
scores.

Brute force, there is no understanding. Only an eventuality.
Incomplete brute force, that eventuality can be incorrect.

>
>> Tic-Tac-Toe (Draughts and Noughts) is a small example of a closed
>> game. It was solved before the days of computers.
>>
>> The challenge of chess is that it is a closed game (board and pieces
>> are finite, finite number of positions), but it is extremely dense in
>> moves. The number of moves can possibly be infinite, but I would say
>> that even so, the number of positions is finite.
>
>And what significance do is there of poistions verus moves?

Please see the above.

>
>> Brute force is possible, but it would require orders of magnitude
>> more computing storage and power to not be able to win only by the
>> computers' opponent dying of old age (and subsequent generations).
>
>A more-ism argument. But if the evaluation algorithm is stupid, then the
>result is more stupidity, no?

What is currently used it heuristics. Ster than brute force,
because it tries to use human-type evaluations.

But brute force knows all positions, and you won't surprise it. At
that point, essentially it is looking at a database.

>
>> So, they employ shortcuts.
>
>Your argument then is that computers cannot adequately solve chess at the
>moment, so they use 'look-ups'. But this is against the rules of chess, and
>is no prescription for a future chess engine. It is simply an expediency to
>gain time, which is unfortunately illegal.

How do you figure? Humans have the benefit of study, and can put
things in there memory. What you're saying is that is a no-no because
the computer can out-book a person.

An opening book is an example of a heuristic. Instead of evaluating,
go to this table, and play these moves.

>
>> Opening books. Endgame tablebases.
>> Alpha-beta pruning (scoring potential positions, and throwing away the
>> obviously bad ones, and processing more on the obviously better
>> ones... This is really where human players do damage to computer
>> opponents, because the computer will stop searching on a given
>> position, not seeing the danger).
>>
>>
>>>
>>>Secondly, what does this term 'solve' mean? Does it indicate a forced win
>>>draw or loss for the player making the first move, provided that both
>>>sides
>>>make 'best moves' thereafter?
>>
>> The term 'solve' means the computer has the ability to evaluate all
>> positions. Reasonably this would mean the worst the computer would do
>> is draw against a perfect player.
>
>A nonsense logically. You assume the evaluation is okay? Is it?

Nonsense? No. Merely the end result. Although I was mistaken. If it is
found a solution which leads to a losing position, that is a
possibility.

Brute force, there is no evaluation. The program identifies where it
is, and already has a path to a solution.

>
>>If chess was solved, and the
>> computer played itself, it would always draw (and probably always play
>> the same game... although if there are two equal lines, it might
>> change up).
>
>Another assumption without a predicate. Why is the result always a draw?
>Should not White or Black win?

I stand corrected on that, indeed it is possible. However, the result
would be invariant. Most people believe finding the 'solution' to
chess will lead to a draw.

>
>> Towers of Hanoi is 2^n. Chess is many orders of magnitude more
>> complex.
>>
>>>
>>>With human beings, the performance of chess is within a certain time
>>>frame -
>>>and as Tal pointed out, many of his more spectacular combinations are
>>>refuted the next day, month or year - but within the timeframe allowed,
>>>was
>>>it solvable? Chess is a performance activity not a theoretical one.
>>
>> Computing improves about an order of magnitude every 18 months.
>> Although this trend from a physics standpoint might not hold out,
>> other improvements are taking place in computing (parallel processing
>> for example)
>
>The point addresses the time allowed, not theoretical space.

My point was that what is theoretical today, is often feasible
tomorrow.

>
>>>Computers are actually a very long way from solving chess - and to test
>>>this
>>>hypothesis turn off the opening book and also the endgame tables [both
>>>cheating!]. There are no higher level computer ratings which are legal,
>>>since all GM matches against computers are played with book/tables=on, and
>>>when the book is on then by definition the program is referring to notes,
>>>and playing moves it couldn't calculate itself. If it could calculate the
>>>moves then , ipso facto, why have opening books and endgame tables?
>>
>> Opening books make the computer stronger.
>
>No. they make the chess engine weaker. Opening books avoid the program
>actually playing chess - it simply trots out moves by GMs, which, if it were
>that good itself, it would play - so it wouldn't need opening books.
>
>This is a logical truism. You are admitting that computers are weak in the
>opening by your statement that they can't play them!

I never denied computers are weak in the opening. There are two ways
to look at the problem of computer chess. The first is the brute-force
approach, which is what I am referring to when I say chess is finite.
A brute-force solution either is running calculations until you find
the best result, or traversing a map until you do.

Where computers are now is heuristics. Using an opening book is a
heuristic. Deep Blue used 64 processors, each evaluating a different
aspect of the game. Using heuristics often does not guarantee complete
solvability. It's getting away from hard rules, to rules of thumb.

That aspect of heuristics is exactly why chess computers would be
lousy in opening games without books. A player studies an opening
book, and gets the first 5-12 moves. At that point in the game is
where there are the most possibilities.

I would point out that a brute-force solution effectively is a book,
end-to-end.

>
>> It doesn't have to
>> brute-force the calculations. Pretty much the same reason human
>> players use opening books.
>>
>> Where the computer has shown its' usefulness to the gameof chess is
>> endgame studies. The board is more simplified, which allows it to
>> examing depth of moves faster.
>
>Faster than... ?

It can in the middlegame, where more pieces = more moves to analyze.
Human GM have the ability to see 10-15 moves in advance. Even
supercomputers cannot see that far, especially when there are many
more moves to analyze.

That is another aspect of heuristics - how broad do you go versus how
many ply in moves you look ahead.

>
>So it takes advantage of the time it can calculate by look-ups, which is
>denied the human player, and whcih is incidentally not chess - it is illegal
>to look up notes. So the computer cheats with time.

I still disagree with that. Humans can study opening books and
databases and memorize them. They save time as well in doing so.

What you call notes to a computer is memory to a human player.

>
>> I would venture to say that endgame tables merely give a value. I
>> doubt they help much on how to play the line. But what they do is
>> given a position, the computer can know whether to press for the win,
>> ask for a draw, or resign.
>>
>>>
>>>If this is to 'save-time' then a further cheat is revealed, see Tal
>>>anecdote.
>>
>> One of my first computer teachers gave me a great analogy. A computer
>> is like a dopey little brother who can only follow simple
>> instructions, and you have to write them down on index cards for him.
>>
>> A computer does simple things fast. The 'simple' brute-force approach
>> you suggest isn't either feasible or complete at this point. Even if
>> you reduce the complexity by allowing for transpositions and things
>> like that, it's still terribly complex.
>
>I did not advocate brute force. I did state the conditions of a Turing
>Engine.

Brute force would be a deterministic Turing machine, while the
heuristic ones would be a non-deterministic.

>
>> Cheating? Not really. By taking such shortcuts, the program follows
>> human play a little better. Opening books are used by human and
>> computer alike, memorized, to try and put the opponent at a
>> disadvantage. Closed game or open? Center or flanks?
>
>Wrong! During the game it is illegal to look up opening books. Period.

As soon as you can find a copy of the opening book the computer takes
to the table, I'll believe you!

>
>> I think that is a bad analogy, that opening books encoded are cheating
>> for a computer. It 'memorizes' better than a human, but both do
>> memorize them.
>
>If a computer can memorise better than a human it should not need to look up
>stuff during the game. Ipso facto, this is not memorisation!

A computers' program is in memory. If it cannot consult it's memory,
it cannot move. Therefore it must consult it's memory. It cannot help
it. I'll see your ipso facto, and raise you a Q.E.D:

I believe it is illegal for a chess player to use any guide outside of
itself. That is what an opening book is, right?
So where does the computer keep this? Does it go outside of itself,
during game time, to recall the information?

>
>ROFL. Its cheating!

>
>Phil Innes
>
>
>>>> When chess is solved, it *could* be that the Sicilian Defence (1...c5)
>>>> refutes 1.e4: that is, anyone who plays 1.e4 will *inevitably* lose
>>>> against the Sicilian with best play. Now, *if* the Sicilian refutes
>>>> 1.e4, then *by definition* it *also* refutes *all* other white openings
>>>> beginning with 1.e4. So, Stan's point, and mine, is this: if ...e5 is
>>>> unplayable, then it's been refuted.
>>>>
>>>>>>A refutation would have to
>>>>> occur after move 3 since you don't have an Albin until Black has played
>>>>> ...e5.
>>>>
>>>> Nope. See above.
>>>
>>>I think Taylor made a good point between a refutation : making move
>>>unplayable
>>>
>>><snip>
>>>
>>>>>> Enough with insulting Kramnik, already!
>>>>
>>>>
>>>>
>>>>>But everyone knows *real* men play Kd2
>>>
>>>Real Vermont Men often skip this sissy 'flatlander' step and immediately
>>>play 1. Kd2 taking their own pawn!
>>>
>>>> LOL yeah!
>>>
>>>Mano-a-Mano chess, true Yankee style.
>>>
>>>Phil Innes
>>>
>>
>

Date: 30 May 2005 17:37:13
From: David Richerby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Patrick Volk <[email protected] > wrote:
> Brute force would be a deterministic Turing machine, while the
> heuristic ones would be a non-deterministic.

No, heuristics are perfectly deterministic. They give different results
to brute force because they are approximations. A nondeterministic
computer can make more than one choice simultaneously, leading to a
tree of parallelism of unbounded width. Then, there is some mechanism for
combining these parallel results into a single answer. Needless to say,
this isn't the sort of thing that can be done on a real computer without
simulating the parallel threads in series at huge cost in time.

Dave.

--
David Richerby Enormous Electronic Spoon (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a piece of cutlery but it uses
electricity and it's huge!

Date: 30 May 2005 11:56:20
From: Chess One
Subject: On Computers Solving Chess

"Patrick Volk" <[email protected] > wrote in message
news:[email protected]...

>>To return to the subject, what seems presented here is a variety of
>>quantitative analysis, but what is not explained is any qualitative
>>improvement in that analysis - to wit; how does deeper search depth
>>resolve
>>in better positional understanding?
>
> If you think for any given position, associate a score with it. This
> currently is calculated based on the pieces, and expressed in pawns. A
> deeper search allows the computer to find positions with higher
> scores.
>
> Brute force, there is no understanding. Only an eventuality.
> Incomplete brute force, that eventuality can be incorrect.

Dear Patrick,

This is the heart of the matter - in chess and in AI.

There is no increase in the evaluation of the positon which is obtained by
faster speed. In other words, there is no qualitative difference in the
result. There are only more results at deeper ply-depths.

The consequence of this directly challenges claims made for computer
solutions to chess. If the evaluation is based only on the numeric value of
pawns and pieces [with other numeric sophistications, examples are such as
an increased score for 2 bishops, or even for 2 bishops in an open position]

One would have to claim that this rule-of-thumb method of evaluating chess
is sufficient to solve chess. It is not evident that this is true.

-------

>>> Brute force is possible, but it would require orders of magnitude
>>> more computing storage and power to not be able to win only by the
>>> computers' opponent dying of old age (and subsequent generations).
>>
>>A more-ism argument. But if the evaluation algorithm is stupid, then the
>>result is more stupidity, no?
>
> What is currently used it heuristics. Ster than brute force,
> because it tries to use human-type evaluations.
>
> But brute force knows all positions, and you won't surprise it. At
> that point, essentially it is looking at a database.

It doesn't know anything. It evaluates the result of all possibilities, but
if the evaluation is not profound, it does not come to a profound
conclusion!

-----
>>Your argument then is that computers cannot adequately solve chess at the
>>moment, so they use 'look-ups'. But this is against the rules of chess,
>>and
>>is no prescription for a future chess engine. It is simply an expediency
>>to
>>gain time, which is unfortunately illegal.
>
> How do you figure? Humans have the benefit of study, and can put
> things in there memory. What you're saying is that is a no-no because
> the computer can out-book a person.

You are using an analogy and not addressing the issue of looking up material
from notes during the game. If the computer's algorithm could generate by
itself data stored in the computer, then it could refer to that sort of
'memory', but that is not what is happening - the computer consults GM level
play during the game, and this is in no sense 'memory', and it is not the
result of its own evaluation.

> An opening book is an example of a heuristic. Instead of evaluating,
> go to this table, and play these moves.

Yes, it is not evaluating. It is just assuming an advantage over the
opponent.

-----------

>>> The term 'solve' means the computer has the ability to evaluate all
>>> positions. Reasonably this would mean the worst the computer would do
>>> is draw against a perfect player.
>>
>>A nonsense logically. You assume the evaluation is okay? Is it?
>
> Nonsense? No. Merely the end result. Although I was mistaken. If it is
> found a solution which leads to a losing position, that is a
> possibility.
>
> Brute force, there is no evaluation. The program identifies where it
> is, and already has a path to a solution.

I don't think you are answering the question of the worth of the evaluation,
which is to provide a qualitative result as the measure for the search. In
the above segment you claim that the worst result would be to draw against a
perfect player - but its not clear that one statement supports the other.

>>
>>>If chess was solved, and the
>>> computer played itself, it would always draw (and probably always play
>>> the same game... although if there are two equal lines, it might
>>> change up).
>>
>>Another assumption without a predicate. Why is the result always a draw?
>>Should not White or Black win?
>
> I stand corrected on that, indeed it is possible. However, the result
> would be invariant. Most people believe finding the 'solution' to
> chess will lead to a draw.

Interesting! Perhaps Andy Walker could say if this 'belief' is based in
anything mathematical. I mean - why do we believe this?

------

>>The point addresses the time allowed, not theoretical space.
>
> My point was that what is theoretical today, is often feasible
> tomorrow.

But there is no evidence presented here that brute-force increases the
evaluation of the positon. Is brute-force solution a false hypothesis for
solving chess?

So - I reject both these ideas of brute-force solution and also analogies on
memory and look-ups. It seems clear that no Turing-Engine [which can be a
paper calculation!] yet contains an algorithm which can play chess
sufficiently to survive against master level play without using resources of
opening books and end-tables - both of which are illegal.

The strange factor of their incorporation into chess actually has the
partadoxial result of keeping a potential chess playing algorithm which
might solve chess from coming into existance.

I have also discussed these matter extensively with Bob Hyatt and with Chris
Whittington - and Bob argues his case much as you have Patrick, while Chris
agrees with my statement in the previous paragraph.

Cordially, Phil Innes

----------

>>> Opening books make the computer stronger.
>>
>>No. they make the chess engine weaker. Opening books avoid the program
>>actually playing chess - it simply trots out moves by GMs, which, if it
>>were
>>that good itself, it would play - so it wouldn't need opening books.
>>
>>This is a logical truism. You are admitting that computers are weak in the
>>opening by your statement that they can't play them!
>
> I never denied computers are weak in the opening. There are two ways
> to look at the problem of computer chess. The first is the brute-force
> approach, which is what I am referring to when I say chess is finite.
> A brute-force solution either is running calculations until you find
> the best result, or traversing a map until you do.

<balance snipped >

Date: 27 May 2005 17:16:47
From: Dr A. N. Walker
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

In article <Y1Hle.6522$Ib.2967@trndny03 >, Chess One <[email protected]> wrote:
>Chess is an infinite game,

Chess is; but only because there is no absolute requirement
that claimable draws [by repetition or by the 50-move rule] actually
*be* claimed. For the purposes of analysis, any position in which a
draw *can* be claimed by either side is drawn, which turns chess into
a finite game.

FWIW, there is no great difficulty in writing a program of a
few hundred lines in any standard computer language that would "solve"
chess using only a few thousand bytes of storage. Sadly, in the
current state of knowledge, you would be unwise to expect it to
give you "the answer" within the next few years ....

> and I should like to hear from a mathematician if
>infinite games are solvable.

*Some* infinite games are solvable. A trivial example is the
game "My dad has more than your dad", often used to resolve disputes
in the playground, and which is an easy win for the last player. There
are some much more interesting examples -- see "Winning Ways", by
Berlekamp, Conway and Guy. There are also different sorts of infinity
in this context, depending on whether the game is "wide" [infinite
choice of moves] or "long" [unbounded length] or both. Games such as
"Sylver Coinage" [which you can google for] are not only both
unboundedly wide and long, but also may remain so for unboundedly many
moves, yet must terminate.

However, many standard games have generalised forms that are
PSpace-complete [meaning, roughly, that any problem whatsoever that
can be solved using only a polynomially-bounded amount of space can
be represented as a chess/go/whatever problem but on a (very) large
board], which, in our present state of knowledge, means that it seems
very unlikely that they will have general solutions, only those that
can be obtained by brute force. [But this tells us nothing about
chess as played normally on an 8x8 board.]

>Secondly, what does this term 'solve' mean? Does it indicate a forced win
>draw or loss for the player making the first move,

Yes.

> provided that both sides
>make 'best moves' thereafter?

No. A side that is losing has only losing moves, and there is
no game-theoretic concept of "best move" in such a case [though, as a
matter of practical chess against less-than-perfect opposition, there
may well be]. A side that is drawing has at least one drawing move but
no winning move, and will draw or win as long as it plays one of the
drawing moves; it will win only if the opponent makes a mistake. Again,
there is no game-theoretic distinction between the drawing moves, but
there may well be as a matter of practical chess [on the "give enough
rope" principle, for example]. A side that is winning has at least one
winning move, and at least one of them will force a win, no matter what
the opponent plays. In some games, there is a potential trap here, in
that it is not sufficient to have a won position, you have actually to
win it; at least one of your moves will do this, but there may be
other moves which lead merely to other won positions and may leave you
"going round in circles". [You see this often when beginners play
endings such as KRvK -- in chess, the "going round" is eventually
prevented by the 50-move rule, but there are other games where this
does not happen.]

>With human beings, the performance of chess is within a certain time frame -
>and as Tal pointed out, many of his more spectacular combinations are
>refuted the next day, month or year - but within the timeframe allowed, was
>it solvable? Chess is a performance activity not a theoretical one.

Oh, indeed.

>Computers are actually a very long way from solving chess

Well, we know that.

> - and to test this
>hypothesis turn off the opening book and also the endgame tables [both
>cheating!].

But this is not a very good test. *Some* programs indeed
rely very directly on these features, but others do not. The effect
of turning off the opening book is usually, in such cases, that they
play rather routine, perhaps passive, opening moves, but not that
they play *badly*. If it were otherwise, then the "anti-computer"
ploy of playing 1 e3 [eg] would work much better than in fact it
does, as GK found out. Bronstein's idea of playing gambits, so
that the computer doesn't *understand* its book, seems better. As
for the endings, well 'tis true that computers can make a dreadful
pig's ear of some positions, but they can also play *amazingly*
well [without tablebases] in some others -- being able to see some
20 or 30 moves [40-60 ply] ahead in N&P endings compensates for an
awful lot of knowledge .... There are plenty of programs around
that can win KBNvK and KQvKR by brute force.

--
Andy Walker, School of MathSci., Univ. of Nott'm, UK.
[email protected]

Date: 27 May 2005 08:11:02
From: Taylor Kingston
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

k Houlsby wrote:
> >That's not technically a refutation of the Albin.
>
> Yes it is, moron, it renders ...e5 unsound.

k, in all cordiality, I strongly urge that you reconsider. In
this case Mr. DePalma is correct. 1.d4 d5 2.Nf3 is a *prevention* of
the Albin, but not a *refutation*. A refutation can only stem from the
position after 1.d4 d5 2.c4 e5. To call 1.Nf3 or 2.Nf3 a refutation of
the Albin is like saying 1.e4 a6 is a refutation of the Ruy L=F3pez
since it prevents 3.Bb5.
In some circumstances even 1.Nf3 may not be a successful anti-Albin
approach. I recall Reuben Fine talking about a game with Weaver Adams,
an Albin expert. To avoid Adams' pet line, Fine opened (IIRC) 1.Nf3,
and the game proceeded 1...Nc6 2.d4 d5 3.c4 and after 3...e5!? guess
what: it was an Albin Counter Gambit.

Date: 27 May 2005 07:48:55
From: Mark Houlsby
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

>That's not technically a refutation of the Albin.

Yes it is, moron, it renders ...e5 unsound.

>A refutation would have to
occur after move 3 since you don't have an Albin until Black has played

...e5.

..and if you don't have an Albin, that demonstrates its being no good
against this particular move. DUH!

>1. Nf3 and 2. Nf3 are part of the girlie-man system of avoiding anything
interesting after 1. d4.

Enough with insulting Kramnik, already!

Date: 27 May 2005 16:31:06
From: Angelo DePalma
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Ladies and Gentleman,

I present to you a troll. Someone with nothing more attractive in his
pathetic life than commenting on my posts.

Angelo DePalma

"k Houlsby" <[email protected] > wrote in message
news:[email protected]...
> >That's not technically a refutation of the Albin.
>
> Yes it is, moron, it renders ...e5 unsound.
>
>>A refutation would have to
> occur after move 3 since you don't have an Albin until Black has played
>
> ...e5.
>
> ..and if you don't have an Albin, that demonstrates its being no good
> against this particular move. DUH!
>
>>1. Nf3 and 2. Nf3 are part of the girlie-man system of avoiding anything
> interesting after 1. d4.
>
> Enough with insulting Kramnik, already!
>

Date: 27 May 2005 16:01:18
From: Chris Barnett
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

k Houlsby wrote:
>>That's not technically a refutation of the Albin.
>
>
> Yes it is, moron, it renders ...e5 unsound.

You can't refute an opening when it it hasn't been played 2...e5 does
not make the Albin 1. d5 d5 2. c4 e5 does

>
>>A refutation would have to
>
> occur after move 3 since you don't have an Albin until Black has played
>
> ...e5.
>
> ..and if you don't have an Albin, that demonstrates its being no good
> against this particular move. DUH!

The question was a refutation to the Albin, not to 2...e5

>>1. Nf3 and 2. Nf3 are part of the girlie-man system of avoiding anything
>
> interesting after 1. d4.
>
> Enough with insulting Kramnik, already!

But everyone knows *real* men play Kd2

Date: 27 May 2005 10:37:30
From: Jerzy
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Sam Sloan wrote:
> I have posted as a streaming video my Chess Lesson #2 as broadcast on
> Cable TV last week.
>
> The address is
> http://www.ishipress.com/samchess2.asf
>
> Lesson 2 covers some basic opening traps such as the basic trap in the
> Albin Counter Gambit and basic rook and pawn endgames such as the
> Lucena Position. It also demonstrates the infamous Keres-Botvinnik
> 1948 World Championship Game where Keres dumped the game to insure
> that Botvinnik and not Reshevsky would be World Chess Champion.

Finally I have watched all the movie and I must say :
Sam, the ending Keres- Botvinnik from 1948 WCC tournament was very well
covered by you !
Thank you !

I think it should be clear now to all disputants here that Keres was forced
to lose the game by Soviet apparatchiks.

Regards,

Jerzy

Date: 27 May 2005 12:20:56
From: J�rgen R.
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Fri, 27 May 2005 10:37:30 +0200, "Jerzy" <[email protected] > wrote:

>Sam Sloan wrote:
>> I have posted as a streaming video my Chess Lesson #2 as broadcast on
>> Cable TV last week.
>>
>> The address is
>> http://www.ishipress.com/samchess2.asf
>>
>> Lesson 2 covers some basic opening traps such as the basic trap in the
>> Albin Counter Gambit and basic rook and pawn endgames such as the
>> Lucena Position. It also demonstrates the infamous Keres-Botvinnik
>> 1948 World Championship Game where Keres dumped the game to insure
>> that Botvinnik and not Reshevsky would be World Chess Champion.
>
>Finally I have watched all the movie and I must say :
>Sam, the ending Keres- Botvinnik from 1948 WCC tournament was very well
>covered by you !

Except that the analysis is completely wrong. The a-pawns are by no
means irrelevant.

>Thank you !
>
>I think it should be clear now to all disputants here that Keres was forced
>to lose the game by Soviet apparatchiks.

Please note that the cold war has ended - remember, Ronnie Raygun won
it. It is no longer in fashion to make up nonsense about the evils of
Russian Communism. It's Russian Capitalism that's the problem now.

>
>Regards,
>
>Jerzy
>

Date: 27 May 2005 09:06:36
From: Jerzy
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Date: 26 May 2005 23:58:22
From: Sam Sloan
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

I just did a search of my database, and to my surprise I found the
actual game. On the TV Show, I was speaking entirely from memory of a
game that was played 46 years ago.

Here is the game. As you will see, I did not actually fall into the
trap. However, I did not see it coming either. My opponent showed me
the trap after the game was over and I have never forgotten it.

At the time of this game, I was 14 years old and was only a 1600 or
1700 player. Still, I held onto my position fairly well and may have
had the advantage at one point. My opponent later on became a famous
master.

Sam Sloan

[Event "National Capital Open"]
[Site "Washington DC"]
[Date "1959.??.??"]
[Round "?"]
[White "Sloan, Sam"]
[Black "Hill, Myron"]
[Result "0-1"]
[ECO "D08"]

1.d4 d5 2.c4 e5 3.dxe5 d4 4.Nf3 Nc6 5.e3 Bb4+ 6.Bd2 dxe3 7.fxe3
Bxd2+ 8.Nbxd2 f6 9.exf6 Nxf6 10.Be2 Bf5 11.O-O Ng4 12.Qb3 Qd7
13.Rfd1 O-O-O 14.Qc3 Qe7 15.Nf1 Rxd1 16.Rxd1 Re8 17.Bd3 Nxe3
18.Re1 Qc5 19.Bxf5+ Nxf5+ 20.Kh1 Rxe1 21.Nxe1 Ne5 22.Nd2 Qe3
23.Qxe3 Nxe3 24.b3 a5 25.Kg1 Nd1 26.Kf1 Kd7 27.a3 a4 28.bxa4 Ne3+
29.Ke2 N3xc4 30.Nb1 Ke6 31.Nf3 Nxf3 32.Kxf3 Kf5 33.h3 c5 34.g3
Ke5 35.g4 g5 36.Nc3 h6 37.Ne4 b6 38.Ng3 Nd6 39.Ne2 Kd5 40.Ng3 Kc4
41.Nf5 Nf7 42.Kg3 Kd3 43.h4 0-1

Date: 26 May 2005 21:40:56
From: David Johnston
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Thu, 26 May 2005 01:27:32 GMT, [email protected] (Sam Sloan)
wrote:

>I have posted as a streaming video my Chess Lesson #2 as broadcast on
>Cable TV last week.
>

I see no reason why people who aren't chess players would give a shit.

Date: 26 May 2005 13:24:14
From: Taylor Kingston
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Angelo DePalma wrote:
> It's generally a very entertaining, informative program.
> What it lacks is production, for example the board is difficult to see from
> up above.

True. I would suggest more rehearsal, and review of the rehearsal
takes, before making the final product. Lack of these made it rather
less than optimally informative. The moves of the Albin Counter Gambit
game Sloan presents are:

1. d4 d5
2. c4 e5
3. dxe5 d4
4. e3?? Bb4+
5. Bd2 dxe3
6. Bxb4 exf2+
7. Ke2 fxg1=N+
8. Rxg1 Bg4+ etc. and wins

Sloan ran through these moves twice. The first time, he said 5...dxe3
was check, which it is not. The second time through, in addition to the
white king being absent from the board, 4...Bb4+ was described as
"Queen up, check" and 8...Bg4+ as "King up, check" (or maybe it was
"king up" then "queen up"), when of course both checks were by bishops.
Combined with the absence of the king, the difficulty of telling one
piece from another, and Sloan's undue haste, I cannot imagine an
inexperienced player being anything but hopelessly confused by all
this, while to an experienced player the trap is old news.
I recall watching "Koltanowski on Chess" on public TV in the 1960s.
Kolty was always very careful and deliberate in showing moves. I
recommend his technique to Mr. Sloan.

Date: 26 May 2005 21:25:47
From: T Mark Hall
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

In message <[email protected] >, Mike Murray
<[email protected] > writes
>On Thu, 26 May 2005 15:23:02 -0400, "Angelo DePalma"
><[email protected]> wrote:
>
>>It's generally a very entertaining, informative program.
>
>>What it lacks is production, for example the board is difficult to see from
>>up above. And it looks like Sam did it all in one take, probably a result of
>>having a budget of $12 or less for production.
>
>He should have re-recorded the audio -- too many bloopers where he
>referred to Rooks as Pawns, etc. I'd guess this would confuse the
>target audience. Then, as you mentioned, he should have redone the
>video to shoot the board at about a 45 degree angle. Also, as was
>pointed out, he should have redone the part with the missing King.
>And it took a while to get the lighting adjusted.
>
>And there are some errors in Sam's analysis. In the Albin CG game,
>after 1 d4 d4, 2 c4 e5, 4 dxc5 d4, 4 e3 Bb4ch, 5 Bd2 dxe3, 6 fxe3 he
>states that White has a lost game. Now, White's earlier play is
>revealed as stupid and his position sucks, but it doesn't look lost to
>me. Continuing after 6 Bxb4? exf2ch 7 Ke2 fxg1Nch, he says the king
>has no place to go so White has to take the Knight. However, 8 Ke1 is
>actually the lesser of evils, although White's obviously lost either
>way.
>
>Other than that, it was perfect.
>
While incredibly interesting to Chess players and people fascinated by
Sam Sloan's personal life, can you please stop cross posting this crap
to rec.games.go (as I have done), since it is absolutely of no interest
to any player of the game of Go.

Best wishes.
--
T k Hall
www.gogod.demon.co.uk
www.gogod.demon.co.uk/NewInGo/NewInGo.htm

Date: 27 May 2005 00:22:23
From: Chris Barnett
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

T k Hall wrote:
> In message <[email protected]>, Mike Murray
> <[email protected]> writes
>
>> On Thu, 26 May 2005 15:23:02 -0400, "Angelo DePalma"
>> <[email protected]> wrote:
>>
>>> It's generally a very entertaining, informative program.
>>
>>
>>> What it lacks is production, for example the board is difficult to
>>> see from
>>> up above. And it looks like Sam did it all in one take, probably a
>>> result of
>>> having a budget of $12 or less for production.
>>
>>
>> He should have re-recorded the audio -- too many bloopers where he
>> referred to Rooks as Pawns, etc. I'd guess this would confuse the
>> target audience. Then, as you mentioned, he should have redone the
>> video to shoot the board at about a 45 degree angle. Also, as was
>> pointed out, he should have redone the part with the missing King.
>> And it took a while to get the lighting adjusted.
>>
>> And there are some errors in Sam's analysis. In the Albin CG game,
>> after 1 d4 d4, 2 c4 e5, 4 dxc5 d4, 4 e3 Bb4ch, 5 Bd2 dxe3, 6 fxe3 he
>> states that White has a lost game. Now, White's earlier play is
>> revealed as stupid and his position sucks, but it doesn't look lost to
>> me. Continuing after 6 Bxb4? exf2ch 7 Ke2 fxg1Nch, he says the king
>> has no place to go so White has to take the Knight. However, 8 Ke1 is
>> actually the lesser of evils, although White's obviously lost either
>> way.
>>
>> Other than that, it was perfect.
>>
> While incredibly interesting to Chess players and people fascinated by
> Sam Sloan's personal life, can you please stop cross posting this crap
> to rec.games.go (as I have done), since it is absolutely of no interest
> to any player of the game of Go.
>
> Best wishes.
Come to think of it, rec.games.chess.politics is a bit of a stretch as
well, especially as most people will be subscibed to all the
rec.games.chess newsgroups. However, people talking about it are
presumably what Sam wants, even if 80% of it is negative. :-)

Date: 26 May 2005 15:23:02
From: Angelo DePalma
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

It's generally a very entertaining, informative program.

What it lacks is production, for example the board is difficult to see from
up above. And it looks like Sam did it all in one take, probably a result of
having a budget of $12 or less for production.

I look forward to the next show.

"Sam Sloan" <[email protected] > wrote in message
news:[email protected]...
>I have posted as a streaming video my Chess Lesson #2 as broadcast on
> Cable TV last week.
>
> The address is
> http://www.ishipress.com/samchess2.asf
>
> Lesson 2 covers some basic opening traps such as the basic trap in the
> Albin Counter Gambit and basic rook and pawn endgames such as the
> Lucena Position. It also demonstrates the infamous Keres-Botvinnik
> 1948 World Championship Game where Keres dumped the game to insure
> that Botvinnik and not Reshevsky would be World Chess Champion.
>
> The title of this episode is "Chess: Basic Openings and Basc
> Endgames".
>
> I need to thank Gary Popkin for providing both the inspiration and the
> technical expertise for this show. II also need to thank Leshaun
> Fossett for capturing and recording the streaming video for me.
>
> Anybody who views this video, please post feedback here and tell me
> what you think. Among other things, I want to know whether you feel
> that this show is ready for prime time.
>
> Sam Sloan

Date: 26 May 2005 12:58:10
From: Mike Murray
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Thu, 26 May 2005 15:23:02 -0400, "Angelo DePalma"
<[email protected] > wrote:

>It's generally a very entertaining, informative program.

>What it lacks is production, for example the board is difficult to see from
>up above. And it looks like Sam did it all in one take, probably a result of
>having a budget of $12 or less for production.

He should have re-recorded the audio -- too many bloopers where he
referred to Rooks as Pawns, etc. I'd guess this would confuse the
target audience. Then, as you mentioned, he should have redone the
video to shoot the board at about a 45 degree angle. Also, as was
pointed out, he should have redone the part with the missing King.
And it took a while to get the lighting adjusted.

And there are some errors in Sam's analysis. In the Albin CG game,
after 1 d4 d4, 2 c4 e5, 4 dxc5 d4, 4 e3 Bb4ch, 5 Bd2 dxe3, 6 fxe3 he
states that White has a lost game. Now, White's earlier play is
revealed as stupid and his position sucks, but it doesn't look lost to
me. Continuing after 6 Bxb4? exf2ch 7 Ke2 fxg1Nch, he says the king
has no place to go so White has to take the Knight. However, 8 Ke1 is
actually the lesser of evils, although White's obviously lost either
way.

Other than that, it was perfect.

Date: 26 May 2005 19:47:51
From: Sam Sloan
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Thu, 26 May 2005 15:23:02 -0400, "Angelo DePalma"
<[email protected] > wrote:

>It's generally a very entertaining, informative program.
>
>What it lacks is production, for example the board is difficult to see from
>up above. And it looks like Sam did it all in one take, probably a result of
>having a budget of $12 or less for production.
>
>I look forward to the next show.

Thank you for the positive review, especially since my volunteer crew
seems to be getting discouraged by the negative sniping by the other
chess players.

However, you have greatly overstated the costs of the production.

Sam Sloan

Date: 26 May 2005 14:53:15
From: Bugsy
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Sam Sloan wrote:
> I have posted as a streaming video my Chess Lesson #2 as broadcast on
> Cable TV last week.
>
> The address is
> http://www.ishipress.com/samchess2.asf
>
> Lesson 2 covers some basic opening traps such as the basic trap in the
> Albin Counter Gambit and basic rook and pawn endgames such as the
> Lucena Position. It also demonstrates the infamous Keres-Botvinnik
> 1948 World Championship Game where Keres dumped the game to insure
> that Botvinnik and not Reshevsky would be World Chess Champion.
>
> The title of this episode is "Chess: Basic Openings and Basc
> Endgames".
>
> I need to thank Gary Popkin for providing both the inspiration and the
> technical expertise for this show. II also need to thank Leshaun
> Fossett for capturing and recording the streaming video for me.
>
> Anybody who views this video, please post feedback here and tell me
> what you think. Among other things, I want to know whether you feel
> that this show is ready for prime time.
>
> Sam Sloan

Sam you remind of Newman on Seinfeld -:)

Date: 26 May 2005 21:15:52
From: Taylor
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

---
http://tv.groups.yahoo.com/group/My_Name_Is_Earl/

[email protected]

Bugsy wrote:

> Sam Sloan wrote:
>
>> I have posted as a streaming video my Chess Lesson #2 as broadcast on
>> Cable TV last week.
>>
>> The address is
>> http://www.ishipress.com/samchess2.asf
>>
>> Lesson 2 covers some basic opening traps such as the basic trap in the
>> Albin Counter Gambit and basic rook and pawn endgames such as the
>> Lucena Position. It also demonstrates the infamous Keres-Botvinnik
>> 1948 World Championship Game where Keres dumped the game to insure
>> that Botvinnik and not Reshevsky would be World Chess Champion.
>>
>> The title of this episode is "Chess: Basic Openings and Basc
>> Endgames".
>>
>> I need to thank Gary Popkin for providing both the inspiration and the
>> technical expertise for this show. II also need to thank Leshaun
>> Fossett for capturing and recording the streaming video for me.
>>
>> Anybody who views this video, please post feedback here and tell me
>> what you think. Among other things, I want to know whether you feel
>> that this show is ready for prime time.
>>
>> Sam Sloan
>
>
> Sam you remind of Newman on Seinfeld -:)

Nuh-uh! He reminds me of that Asian Chinese food delivery guy who sued
Eliane-- kind of annoying and a non-entity. Just go away, Sam. I think
his name was Ping, right? "Ping", go away, Sam.

Date: 26 May 2005 16:56:07
From: Sam Sloan
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

At 10:33 AM 5/26/2005 EDT, [email protected] wrote:

>1. Is your show available on videocassette? My computer won't oblige me with
>streaming video, but the subject matter does sound interesting.
>
>2. Is there an outright refutation of the Albin? You'd think there would be;
>it looks so weak on its face.
>
>Joseph Dobrian
>

Dear Joseph,

I did not realize that you were a chess player.

I have the show on video cassette and I would be most happy to give
one to you, but I need to find a place that will copy it.

No. There is no refutation to the Albin Counter Gambit. To the
contrary, Grandmaster Morozevich of Russia, one of the top-10 players
in the world, has been playing it against the world's leading
grandmasters and winning with it. Go to http://www.chessbase.com and
you will find some recent grandmaster games with this revived opening.

Sam Sloan

Date: 26 May 2005 23:50:20
From: StanB
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

"Sam Sloan" <[email protected] > wrote in message
news:[email protected]...

> No. There is no refutation to the Albin Counter Gambit.

1. d4, Nf6 2. Nf3

Date: 27 May 2005 10:27:49
From: Angelo DePalma
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

That's not technically a refutation of the Albin. A refutation would have to
occur after move 3 since you don't have an Albin until Black has played
...e5.

1. Nf3 and 2. Nf3 are part of the girlie-man system of avoiding anything
interesting after 1. d4.

"StanB" <[email protected] > wrote in message
news:[email protected]...
>
> "Sam Sloan" <[email protected]> wrote in message
> news:[email protected]...
>
>> No. There is no refutation to the Albin Counter Gambit.
>
> 1. d4, Nf6 2. Nf3
>
>

Date: 26 May 2005 10:43:22
From: J�rgen R.
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Thu, 26 May 2005 01:27:32 GMT, [email protected] (Sam Sloan)
wrote:

>I have posted as a streaming video my Chess Lesson #2 as broadcast on
>Cable TV last week.
>
>The address is
>http://www.ishipress.com/samchess2.asf

Christ Almighty, what a stunningly bad program. I actually watched 8
minutes of it.

Sloan, who evidently has terminal syphilis, has ugly growths on his
face and his manipulatory skills are vanishing. His hair is uncombed
and unwashed, but he did borrow a shirt and a tie from somebody, God
knows why. He uses a cheap USCF chess set to demonstrate a cheap
opening trap. But his unweighted pieces keep tipping over. At one
point the King falls off the table, and even though Sloan crawls
around on the floor he can't find it - the syph has begun to affect
his eyes - but, no problem, he simply continues showing his stupid
trap without the white King - the white K is the one being attacked.
You think I'm making this up? Go see for yourself.

Sloan, you are an abject fool.

>
>Lesson 2 covers some basic opening traps such as the basic trap in the
>Albin Counter Gambit and basic rook and pawn endgames such as the
>Lucena Position. It also demonstrates the infamous Keres-Botvinnik
>1948 World Championship Game where Keres dumped the game to insure
>that Botvinnik and not Reshevsky would be World Chess Champion.
>
>The title of this episode is "Chess: Basic Openings and Basc
>Endgames".
>
>I need to thank Gary Popkin for providing both the inspiration and the
>technical expertise for this show. II also need to thank Leshaun
>Fossett for capturing and recording the streaming video for me.
>
>Anybody who views this video, please post feedback here and tell me
>what you think. Among other things, I want to know whether you feel
>that this show is ready for prime time.
>
>Sam Sloan

Date: 26 May 2005 22:20:48
From: Angelo DePalma
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

This is a stunningly funny review.

S'lud!

"J�rgen R." <[email protected] > wrote in message
news:[email protected]...
> On Thu, 26 May 2005 01:27:32 GMT, [email protected] (Sam Sloan)
> wrote:
>
>>I have posted as a streaming video my Chess Lesson #2 as broadcast on
>>Cable TV last week.
>>
>>The address is
>>http://www.ishipress.com/samchess2.asf
>
> Christ Almighty, what a stunningly bad program. I actually watched 8
> minutes of it.
>
> Sloan, who evidently has terminal syphilis, has ugly growths on his
> face and his manipulatory skills are vanishing. His hair is uncombed
> and unwashed, but he did borrow a shirt and a tie from somebody, God
> knows why. He uses a cheap USCF chess set to demonstrate a cheap
> opening trap. But his unweighted pieces keep tipping over. At one
> point the King falls off the table, and even though Sloan crawls
> around on the floor he can't find it - the syph has begun to affect
> his eyes - but, no problem, he simply continues showing his stupid
> trap without the white King - the white K is the one being attacked.
> You think I'm making this up? Go see for yourself.
>
> Sloan, you are an abject fool.
>
>>
>>Lesson 2 covers some basic opening traps such as the basic trap in the
>>Albin Counter Gambit and basic rook and pawn endgames such as the
>>Lucena Position. It also demonstrates the infamous Keres-Botvinnik
>>1948 World Championship Game where Keres dumped the game to insure
>>that Botvinnik and not Reshevsky would be World Chess Champion.
>>
>>The title of this episode is "Chess: Basic Openings and Basc
>>Endgames".
>>
>>I need to thank Gary Popkin for providing both the inspiration and the
>>technical expertise for this show. II also need to thank Leshaun
>>Fossett for capturing and recording the streaming video for me.
>>
>>Anybody who views this video, please post feedback here and tell me
>>what you think. Among other things, I want to know whether you feel
>>that this show is ready for prime time.
>>
>>Sam Sloan
>

Date: 26 May 2005 21:20:16
From: LSD
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

"J�rgen R." <[email protected] > wrote in message
news:[email protected]...
> On Thu, 26 May 2005 01:27:32 GMT, [email protected] (Sam Sloan)
> wrote:
>
>>I have posted as a streaming video my Chess Lesson #2 as broadcast on
>>Cable TV last week.
>>
>>The address is
>>http://www.ishipress.com/samchess2.asf
>
> Christ Almighty, what a stunningly bad program. I actually watched 8
> minutes of it.
>
> Sloan, who evidently has terminal syphilis, has ugly growths on his
> face and his manipulatory skills are vanishing. His hair is uncombed
> and unwashed, but he did borrow a shirt and a tie from somebody, God
> knows why. He uses a cheap USCF chess set to demonstrate a cheap
> opening trap. But his unweighted pieces keep tipping over. At one
> point the King falls off the table, and even though Sloan crawls
> around on the floor he can't find it - the syph has begun to affect
> his eyes - but, no problem, he simply continues showing his stupid
> trap without the white King - the white K is the one being attacked.
> You think I'm making this up? Go see for yourself.

Dude, that's merciless! I had to watch it myself. I am glad you are not my
critic.

Date: 26 May 2005 20:22:39
From: Warp
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

In rec.games.go "J�rgen R." <[email protected] > wrote:
> He uses a cheap USCF chess set to demonstrate a cheap
> opening trap.

The editing and lighting problems, bloopers and terminology (pawn,
rook...) and in general the lack of professionalism was very obvious,
of course, and as someone suggested it would be better if the video
was redone with more care. However, I found the lesson in question
rather interesting. It actually kept me interested to the end.
I played quite a lot of chess as a kid, but then left it. I can't
call myself a very experienced player anymore, but I do remember
something and I could perfectly follow what he was saying and
demonstrating on the board. I found the traps and anecdotes quite
interesting. If the video is redone with a bit more professionalism
it would be, in my opinion, a quite good one, specially for
beginners.

--
- Warp

Date: 26 May 2005 14:47:13
From: Bugsy
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

J�rgen R. wrote:
> On Thu, 26 May 2005 01:27:32 GMT, [email protected] (Sam Sloan)
> wrote:
>
>
>>I have posted as a streaming video my Chess Lesson #2 as broadcast on
>>Cable TV last week.
>>
>>The address is
>>http://www.ishipress.com/samchess2.asf
>
>
> Christ Almighty, what a stunningly bad program. I actually watched 8
> minutes of it.
>
> Sloan, who evidently has terminal syphilis, has ugly growths on his
> face and his manipulatory skills are vanishing. His hair is uncombed
> and unwashed, but he did borrow a shirt and a tie from somebody, God
> knows why. He uses a cheap USCF chess set to demonstrate a cheap
> opening trap. But his unweighted pieces keep tipping over. At one
> point the King falls off the table, and even though Sloan crawls
> around on the floor he can't find it - the syph has begun to affect
> his eyes - but, no problem, he simply continues showing his stupid
> trap without the white King - the white K is the one being attacked.
> You think I'm making this up? Go see for yourself.
>
> Sloan, you are an abject fool.

I wasn't going to watch it, but now I will -:)

Date: 26 May 2005 17:58:40
From: Sam Sloan
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

At 09:50 AM 5/26/2005 -0700, gary popkin wrote:
>
>On Thu, 26 May 2005 10:43:22 +0200, J�Egen R. wrote:
>>
>>Christ Almighty, what a stunningly bad program. I actually watched 8
>>minutes of it.
>
>
>Should I be insulted that there is no mention of the production values in the review, no comment on the lack of opening music, the lack of dissolves, the lighting of Sloan, the lighting of the chess pieces, the camera positions, the sudden blackout at the end with no credits or anything? We were using an extraordinarily ill-equipped studio, not the one I use for Hardfire, for it was the best studio that Sloan, the producer, cared to try to obtain..

Oh. Well. Nobody is perfect.

Do not be concerned about the negative reks of Jurgen R. His
negative review will get me a few hundred viewers I otherwise would
not have gotten. Here is one of his best known quotes:

"Sam Sloan is nearly always right; the trouble is that his insights
are so deep, and so far ahead of the time, that they appear incredible
to the educated layman. I myself have occasionally made the mistake of
thinking that Sam Sloan had made an error. Invariably, sometimes a
year or more later, I had to admit my mistake.

"Jurgen R."

Date: 26 May 2005 16:10:53
From: Sam Sloan
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

On Thu, 26 May 2005 10:43:22 +0200, J�rgen R. <[email protected] > wrote:

>On Thu, 26 May 2005 01:27:32 GMT, [email protected] (Sam Sloan)
>wrote:
>
>>I have posted as a streaming video my Chess Lesson #2 as broadcast on
>>Cable TV last week.
>>
>>The address is
>>http://www.ishipress.com/samchess2.asf
>
>Christ Almighty, what a stunningly bad program. I actually watched 8
>minutes of it.
>
>Sloan, who evidently has terminal syphilis, has ugly growths on his
>face and his manipulatory skills are vanishing. His hair is uncombed
>and unwashed, but he did borrow a shirt and a tie from somebody, God
>knows why. He uses a cheap USCF chess set to demonstrate a cheap
>opening trap. But his unweighted pieces keep tipping over. At one
>point the King falls off the table, and even though Sloan crawls
>around on the floor he can't find it - the syph has begun to affect
>his eyes - but, no problem, he simply continues showing his stupid
>trap without the white King - the white K is the one being attacked.
>You think I'm making this up? Go see for yourself.
>
>Sloan, you are an abject fool.

Thank you for your prompt posting and for your kind and complimentary
reks.

Just to let you know, when the white king fell on the floor, I did not
crawl around trying to pick it up. Instead, I opened another chess set
and got out a new king.

I do not have editing capability for these shows yet. I will be
working on that soon.

The mole on my lower left cheek is not the result of some loathsome
disease. I have had that mole all my life. Here is a picture of me
from my school yearbook when I was in the fifth grade. You will see
the mole was there, when I was ten years old.

http://www.samsloan.com/sammy.htm

For those of you who do not know J�rgen R., he always posts
complimentary things like this about me. J�rgen R. is one of my
biggest fans.

Sam Sloan

Date: 25 May 2005 21:11:08
From: Not from Chicago
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Regarding your Albin Counter Gambit example:

After 7... PxN(N), you said that the white King has nowhere to go, so the
checking black Knight must be captured.

But it appears to me that 8. Ke1 would be a legal move, and would save the
queen as well.

Am I missing something?

Date: 26 May 2005 13:32:54
From: Chris Barnett
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

Not from Chicago wrote:
> Regarding your Albin Counter Gambit example:
>
> After 7... PxN(N), you said that the white King has nowhere to go, so the
> checking black Knight must be captured.
>
> But it appears to me that 8. Ke1 would be a legal move, and would save the
> queen as well.
>
> Am I missing something?
>
>
Well, as far as I can see, after Ke1, White isn;t losing outright, but
the opening certainly has gone badly for him. He has an isolated pawn
he can't realistically maintain, Black can (and probably will swap
queen, and use the time to develop quickly, and White will at best be
fighting for a a draw.

The point of the trap (4. e3?) was not that it cost the queen, but that
white's winning chances has been blown from the water.

Date: 26 May 2005 07:02:20
From: Not from Chicago
Subject: Re: Sam Sloan's Chess Lesson Video Posted on the Internet

"Chris Barnett" wrote

> The point of the trap (4. e3?) was not
> that it cost the queen, but that white's
> winning chances has been blown from the water.

Good explanation. Makes sense. Thanks.